Next Article in Journal
The Leakage Current Characteristics of High-Gradient MOA Plate and Its Heating Analysis with Coatings under High-Frequency Overvoltage
Previous Article in Journal
The Development of Cement-Based, Intumescent and Geopolymer Fire-Retardation Coatings for Metal Structures: A Review
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Coatings
Volume 13
Issue 3
10.3390/coatings13030496
coatings-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Review

An Overview of Technological Parameter Optimization in the Case of Laser Cladding

by
Kaiming Wang
1,2,3,4,
Wei Liu
1,
Yuxiang Hong
2,
H. M. Shakhawat Sohan
1,
Yonggang Tong
1,*,
Yongle Hu
1,
Mingjun Zhang
1,
Jian Zhang
1,
Dingding Xiang
3,4,
Hanguang Fu
5 and
Jiang Ju
6,*
1
College of Automotive and Mechanical Engineering, Changsha University of Science and Technology, Changsha 410114, China
2
Key Laboratory of Intelligent Manufacturing Quality Big Data Tracing and Analysis of Zhejiang Province, China Jiliang University, Hangzhou 310018, China
3
State Key Laboratory of Tribology in Advanced Equipment, Department of Mechanical Engineering, Tsinghua University, Beijing 100084, China
4
Key Laboratory of Vibration and Control of Aero-Propulsion Systems, Ministry of Education, Northeastern University, Shenyang 110819, China
5
Key Laboratory of Advanced Functional Materials, Ministry of Education, Department of Materials Science and Engineering, Beijing University of Technology, Beijing 100124, China
6
Department of Materials Science and Engineering, City University of Hong Kong, Hong Kong 999077, China
*
Authors to whom correspondence should be addressed.
Coatings 2023, 13(3), 496; https://doi.org/10.3390/coatings13030496
Submission received: 1 February 2023 / Revised: 16 February 2023 / Accepted: 22 February 2023 / Published: 23 February 2023
(This article belongs to the Section Laser Coatings)
Browse Figures
Versions Notes

Abstract

:
This review examines the methods used to optimize the process parameters of laser cladding, including traditional optimization algorithms such as single-factor, regression analysis, response surface, and Taguchi, as well as intelligent system optimization algorithms such as neural network models, genetic algorithms, support vector machines, the new non-dominance ranking genetic algorithm II, and particle swarm algorithms. The advantages and disadvantages of various laser cladding process optimization methods are analyzed and summarized. Finally, the development trend of optimization methods in the field of laser cladding is summarized and predicted. It is believed that the result would serve as a foundation for future studies on the preparation of high-quality laser cladding coatings.

1. Introduction

Laser cladding is a technique for the deposition of a protective layer on a substrate [1,2]. It is the process of depositing a material called a “cladding material” onto a substrate with the help of the thermal energy provided by a laser beam [3,4,5]. The cladding material could be applied to the substrate by wire feeding, powder injection, or a preset powder [6,7,8]. Among them, the wire feeding system is very suitable for processing with high deposition rates [9]. However, most cladding materials still use powder because the powder-feeding method is more flexible than wire feeding [10,11,12]. Laser cladding is an advanced material processing technology in many parts of the business world [13,14,15]—for example, the metallurgical industry, mining machinery industry, marine industry, aerospace industry, automotive industry, and biomedical industry [16,17,18,19,20,21]. The laser cladding aims to increase the substrate’s hardness, wear resistance, corrosion resistance, and oxidation resistance [22,23,24,25,26]. It offers significant advantages over conventional surface modification techniques, such as smaller heat-affected zones, lower dilution, good metallurgical bonding with the substrate, and a finer grain size [27,28,29,30,31]. However, the process parameters in the cladding process have an important influence on the quality of the laser cladding coating. If the process parameters are not appropriate, the coating will have defects such as holes and cracks. Therefore, the process parameters must be effectively controlled to obtain the desired performance [32,33,34].
The laser cladding process is affected by many factors [35,36,37]. Among them, the process parameter is one of the critical factors affecting the formation and quality of the cladding layer [38,39,40]. The cladding process is complex and non-linear due to laser cladding involving more than 19 process parameters [41,42]. Laser power, scanning speed, and powder feeding rate have the most significant influence on the quality of the cladding layer [43]. These three parameters offer the broadest potential window in terms of changing and improving the quality of the cladding layer [44]. Experimental and theoretical studies have shown that the proper selection of input process parameters can significantly improve the properties of the cladding layer [45,46,47]. To obtain high-quality coatings, many scholars have studied and improved the optimization methods of laser cladding process parameters [48,49]. One approach is to investigate the effect of process parameters on the properties of the cladding layer using empirical–statistical and analytical models, such as the response surface method, linear regression method, and Taguchi method [50], to find the optimal process parameters. Another method is to use intelligent algorithms such as machine learning and metaheuristics to establish a predictive model of the mapping relationship between process parameters and the cladding layer, to achieve the optimization of the process parameters [51,52].
This paper reviews the current research status on the optimization of process parameters in laser cladding. It summarizes the optimization methods of various process parameters from traditional and intelligence optimization methods. Then, the advantages and disadvantages of various optimization methods are contrasted. Finally, we provide an outlook on the development trend of optimization methods for the laser cladding process. We hope to provide a reference for future research on the manufacturing of high-quality laser cladding coatings.

2. Traditional Optimization Methods

2.1. Single-Factor Experiment

The traditional single-factor experiment is the most common way to characterize the influence of each parameter on the coating properties [53]. Controlling a single parameter to obtain the desired property is more accessible than controlling multiple parameters [54].
In the study of laser cladding process parameters, many scholars have adopted single-factor experiments to study the process parameters in which they are interested. Among them, the more studied parameters are the laser power, powder feeding rate, and scanning speed [55,56,57,58]. Pornsak et al. [59] observed the cladding process in real time by using an infrared camera with image analysis software. The results showed that the laser spot, melt pool area, cladding height, and width all increased with increasing laser power. Zhan et al. [60] found that as the laser power increased, the clad height, width, and grain size also increased. Li [61] and Jiao et al. [62] found that the microhardness of the coating increased as the scanning speed increased. However, the wear properties of the coating will deteriorate sharply when the scanning speed is too large. Bartkowski et al. [63] prepared Stellite-6/WC metal-based composite coatings using different laser powers and powder feeding rates. Studies have found that an increase in laser power and powder feeding rate increases the coating thickness. The effects of the laser power, powder feeding rate, and scanning speed on cladding results are shown in Figure 1.
Some scholars have studied parameters such as the gas supply speed, laser beam diameter, and pre-placed powder layer thickness. Qi et al. [64] found that the clad width, depth, and height decreased with the increase in the laser beam diameter. Chryssolouris et al. [65] found that the clad depth decreased when increasing the gas supply. Qu et al. [66] analyzed the effects of different pre-coating thicknesses on the cladding layer’s microstructure evolution and mechanical properties. They found that the dilution rate of the coating decreased with the increase in the pre-placed powder layer thickness, and the average values of microhardness and fracture toughness increased with the thickness of the pre-coating. The influence of the gas supply speed, laser beam diameter, and pre-coating thickness on the cladding layer is shown in Figure 2.
Above all, a single-factor experiment is a trial-and-error approach. This method often requires extensive experimentation, which is costly and time-consuming. Moreover, since this method studies the effects of only one factor at a time, interactions between input parameters may be overlooked, so the results may not be optimal.

2.2. Regression Analysis

Regression analysis is mainly used to reveal the causal relationship between variables [67] and has been widely used in various types of statistical analysis [68]. It is primarily used as an analysis method in laser cladding to determine the relationship between the process parameters and coating properties. It is widely used in the empirical–statistical model [69,70]. Ansari [71], Shayanfar [72], Erfanmanesh [73], and Nabhani et al. [74] used regression analysis to correlate the main processing parameters (laser power (P), scanning rate (V), powder feeding rate (F)) with the coating’s geometrical characteristics (clad height, clad width, penetration depth, wetting angle, and dilution). Afterwards, the relationship between them was investigated using a combined parameter (PαVβFγ). Through this method, they obtained a laser cladding process drawing, from which the best process parameters could be obtained. Through experimental analysis, they discovered that the laser power influences the clad height and the clad width by the combined parameters of the laser power and scanning speed, and the clad depth through the combined parameters of the laser power, scanning speed, and powder feeding rate. However, they have different views on the dilution ratio. Since the laser power factor is present in the numerator and denominator of the dilution rate equation, Ansari [71] and Nabhani et al. [74] argue that its effect can be eliminated by simplifying the fraction, while Shayanfar [72] and Erfanmanesh et al. [73] do not adopt this approach. Ansari [71] and Nabhani et al. [74] believe that the dilution ratio is related to the scanning speed and powder feeding rate. Shayanfar [72] and Erfanmanesh et al. [73] believe that the dilution ratio is related to the combination of the scanning speed, powder feeding rate, and laser power. The combination of process parameters that affect the cladding result is shown in Table 1. The processing diagram of laser cladding is demonstrated in Figure 3.
Regression analysis can accurately measure the degree of correlation between each factor and the degree of the regression fit to improve the effectiveness of the prediction equation. However, the calculation is more complex. Moreover, for non-linear data or data with complex polynomials, there are difficulties in modeling using the regression analysis method.

2.3. Response Surface Methodology

The response surface methodology is a statistical method that integrates experimental design and mathematical modeling to solve multivariate problems [75,76]. It can be used to improve the secondary effects and interaction effects in different variables [77]. It is an essential branch of experimental design methods that can be used to develop, improve, and optimize processes [78].
In laser cladding, the response surface methodology is a powerful tool in analyzing the relationship between process parameters and cladding results, especially when the response target is affected by multiple process parameters [79,80]. Among them, the three process parameters of laser power, scanning speed, and powder feeding rate are more studied [81]. Cui et al. [82] used the response surface methodology to optimize a cobalt-based alloy coating’s process parameters. Their work laid the foundation for the setup of the laser cladding processing parameters for multi-track coatings made from cobalt. The response surface methodology was used by Li et al. [83] to produce a Ni60PTA coating with good properties. The coating did not have any cracks, dents, or holes. Wu et al. [84] studied the relationship between pore formation and multiple process parameters in Ni60A alloy coatings. They found that the powder feeding rate is the main factor affecting the porosity of the coating. The results showed the good surface forming quality of the specimens processed with the optimized process parameters.
Some researchers used the response surface methodology to optimize the laser frequency, pulse width, gas flow, overlap rate, and spot diameter. Khorram et al. [85] found that the laser frequency and pulse width positively affect the clad width, angle, and dilution ratio but negatively affect the clad height and hardness. Lian et al. [86] found that the clad width was inversely proportional to the overlap rate; the dilution ratio was inversely proportional to the gas flow. Within a specific range, the flatness ratio decreases with the increase in gas flow and overlap rate and then increases. Wu et al. [87] prepared a Ni60A-25% laser cladding layer with high microhardness on 42CrMo alloy structural steel. The experimental results show that the effect of the spot diameter on the dilution ratio and effective unit area is minimal. The optimization variables and response indexes used in the response surface methodology are shown in Table 2.
As a traditional optimization method, the response surface methodology has been successfully applied to various system optimization problems, especially to multivariate response problems. However, the premise of an experimental design using this method is that the designed experimental sites should contain optimal conditions. Otherwise, the optimization results will be biased. Therefore, reasonable experimental factors and levels must be determined when optimizing the response surface methodology.

2.4. Taguchi Method

The Taguchi method is a low-cost, high-efficiency experimental design method used in many engineering problems [88]. In the Taguchi method, orthogonal arrays can effectively determine the influence of variables and levels to achieve robust designs, which can significantly reduce the experimental time and cost [89].
Due to the simplicity and clarity of the experimental design method of the Taguchi method, in recent years, the Taguchi method has become a powerful tool to improve productivity in the R&D stage [90]. Xu et al. [91] investigated the effect of the process parameters on bond shear strength and microhardness using the Taguchi method. They found that the powder type had the most significant effect on the bond strength. Quazi et al. [92] improved the tribomechanical properties of a Ni-WC composite coating on the surface of AA5083 aluminum alloy by using the Taguchi method to optimize the laser cladding process parameters. The results show that the laser power, defocus amount, and scanning speed significantly affect the surface hardness and wear resistance.
Some scholars have improved the Taguchi method and proposed the mixed Taguchi method. Chen [93] and Shi et al. [94] combined the Taguchi method with the empirical statistical model and TOPSIS methods. The results show that both methods can significantly improve the microhardness of coatings. Zhang [95], Paul [96], Deng [97], and Yu et al. [98] combined grey correlation analysis and the Taguchi method to optimize the process parameters. This method can simplify complex multi-parameter optimization to a single-objective optimization problem, which significantly reduces the difficulty of optimization. The optimized cladding layer obtained by this method is significantly better than other cladding layers in terms of morphology and microstructure, which also verifies the feasibility of Taguchi’s grey correlation method. A grey relational graph of the process parameters; the influence of the process parameters on the cladding’s height, width, and dilution SNR; and the cross-sectional metallographic structure comparison of the specimen before and after optimization are shown in Figure 4.
Although the Taguchi method is a cost-effective experimental design, it must indicate the test. The best combination can only be a combination of a specific test level. The optimal solution can only be within the range of the chosen level. When an interaction between controllable factors is apparent, it can lead to inaccurate results from ANOVA.

2.5. Other Traditional Optimization Methods

In addition to optimization methods such as the response surface methodology, regression analysis, single-factor experiments, and Taguchi method, traditional optimization methods such as orthogonal experiments, principal component analysis, TOPSIS, grey relational analysis, the improved analytic hierarchy process approach, theoretical–empirical models, the desirability function approach, and Design Expert statistical software are also used to optimize the laser cladding process parameters.
Peng et al. [99] used the coating’s average hardness value and dilution ratio as evaluation indexes and obtained the optimal process parameters through orthogonal experiments. Marzban [100] and Wang et al. [101] combined principal component analysis (PCA) with the TOPSIS method and grey correlation relationship analysis, respectively. They achieved the multi-response optimization of the laser cladding process. Liang et al. [102] proposed a laser cladding coating quality evaluation method based on fuzzy comprehensive evaluation (PCE) and an improved analytic hierarchy process (IAHP) approach. The results show that the method can help to optimize the process parameters. Reddy et al. [103] studied the influence of the process parameters on the powder deposition efficiency, dilution, and porosity by establishing a theoretical–empirical model. The experimental results show that the laser power and powder feeding rate greatly influence the deposition efficiency and dilution, and the porosity is mainly related to the raw material. Meng [104], Menghani [105], Mohammed [106], and Liu et al. [107] used the desirability function approach to identify the optimum magnitude of the process parameters in the laser cladding process to obtain better cladding quality and geometry. Rasheedat [108] and Moradi et al. [109] designed a complete factorial experiment, ran many experiments with different process parameters, and obtained the best process parameters through the Design Expert statistical software. Table 3 lists the optimization methods and evaluation indexes. The microstructure at the optimum process parameters and a comparison with the substrate microstructure are shown in Figure 5. It can be seen from the figure that the coating and substrate bond well at the optimum process parameters. In addition, many equiaxed dendrites were formed at the top of the coating, the cladding layer was uniform and compact, the overall quality was good, and the wear resistance was improved. Compared to the substrate, the wear mechanism of the coated specimen changed to abrasive wear.
Traditional optimization methods are often specific to a given process, without considering physical changes such as thermal strain and the cooling rate during optimization. Optimization methods still need to be redesigned when new materials or processes are used, and extensive experiments are performed to optimize the process parameters. In addition, the use of traditional optimization methods often requires a specific mathematical foundation, and some mathematical operations are necessary for the optimization process and the mathematical analysis of the experimental results.

3. Intelligent Optimization Methods

3.1. Artificial Neural Network Model

The artificial neural network is a robust empirical modeling tool with significant advantages, such as high accuracy, low costs, and short times, and it is one of the most applicable models in non-linear analysis [110]. It has a high degree of non-linear fitting capability, which is impossible with traditional methods [111]. Among them, the BP neural network is well suited for physical applications, and it is currently one of the most widely used artificial neural networks [112,113].
In recent years, the artificial neural network has made remarkable achievements in materials science. It has begun to be applied to laser cladding to predict the cladding layer morphology and optimize the process parameters [114,115,116]. Song et al. [117] found, through an experimental comparison, that the artificial neural network model is more suitable for describing the non-linear relationship between process parameters and cladding geometry than the response surface model. Li et al. [118] used the BP neural network to establish a laser cladding AlCoCrFeNi coating dilution rate prediction model; they found that under the optimum process parameters, the coating microstructure was composed of a simple BCC solid solution phase, the grains were equiaxed, and no serious segregation of internal elements occurred. Thus, the coating crack could be effectively controlled. Caiazzo et al. [119] used an artificial neural network to study the relationship between the laser cladding process parameters and cladding layer size. The results show that this method can obtain the required process parameters for specific cladding layer sizes. Guo et al. [120] used the BP neural network model to study the influence of the process parameters on the coating quality of a high-power semiconductor laser cladding cobalt-based alloy. The results showed that the width, height, and depth of the cladding layer were related to the laser scanning speed, and the hardness of the cladding layer was related to the powder feeding rate. The neural network prediction errors are shown in Table 4. The BP neural network training process, a comparison between the BP neural network model and the RSM model prediction results, and the microstructure of AlCoCrFeNi HEA coatings under the optimal process parameters are shown in Figure 6.
A neural network is suitable for handling large data samples, and when the number of samples is too small, the results may need to be more accurate. In addition, neural network models tend to fall into local minima and take longer to learn and train. Therefore, there is a need to apply more optimization algorithms to the neural network.

3.2. Genetic Algorithm Optimizes BP Neural Network (GABP)

The genetic algorithm is a non-traditional optimization tool based on random search techniques [121]. The genetic algorithm has a good global optimization ability, which is suitable for dealing with multi-objective problems [122,123]. The artificial neural network has a long learning time and slow convergence speed, and quickly falls into local minima [124]. Its combination with genetic algorithms can effectively optimize the threshold value of the neural network and achieve a fast global optimization search [125].
Liu et al. [126] found that GABP neural networks have higher prediction accuracy than BP neural networks. At the same time, the combination of a double hidden layer and single-output neural network led to higher prediction accuracy. Pang et al. [127] found that the GABP neural network had better optimization ability than the response surface method. The process parameters obtained using the GABP neural network could improve the deposition rate for laser cladding. Huang [128] and Yang et al. [129] combined genetic algorithms with a neural network model to study the relationship between the laser cladding process parameters and cladding layer quality. The results show that the optimization method can be applied to the crack control of laser cladding molding and can provide some references for the optimization of the laser cladding process parameters. The GABP predictions and their comparison with the BPNN predictions are shown in Figure 7.
The programming implementation of genetic algorithms is complex. The search speed of the algorithm could be faster and requires a long training time. There are better solutions than a single genetic algorithm for large-scale computational problems, and they can easily fall into premature convergence.

3.3. Support Vector Machines (SVM)

Support vector machines (SVM) are a statistical learning method based on structural risk minimization. Support vector regression (SVR) is a vital application branch of support vector machines (SVM). Compared with artificial neural networks, SVM still has a strong generalization ability and global optimization ability when the training data are insufficient. In recent years, it has become a hotspot in various research and industrial process applications and has been successfully applied to modeling and regression analysis problems [130,131,132,133].
Chen et al. [134] used the support vector machine (SVM) model to study the influence of the process parameters on the coating quality characteristics. The results show that the preset powder thickness, laser spot diameter, and power are the most critical process parameters. Chen et al. [135] established a multi-output support vector regression (M-SVR) model. Through experiments, they found that its accuracy was higher than that of the single-output support vector regression (S-SVR) model and the BP neural network model. Yao et al. [136] established a support vector regression (SVR) model based on the Gaussian radial (RBF) kernel function. This model is more accurate than the BP neural network model and can help to select the process parameters. Zhang et al. [137] optimized the process parameters by combining the multi-objective slime mold algorithm (MOSMA) and support vector regression (SVR). The results show that compared with other methods, such as the response surface methodology, the best process parameters obtained by this method can lead to coatings with better performance. A comparison of the prediction results of support vector regression and the BP neural network is shown in Table 5. The prediction model framework for support vector machines, a comparison of the predicted and experimental values for SVR, and a comparison of the predicted values for SVR and BPNN are shown in Figure 8.
Support vector machines (SVM) can effectively solve regression problems and perform well when dealing with small sample data. However, they are only suitable when processing a small amount of data, and, when there are too many data sets, they require a long time to solve the problem and show poor results.

3.4. Novel Non-Dominated Sorting Genetic Algorithm II (NSGA-II)

The novel non-dominated sorting genetic algorithm II (NSGAII) is a well-known metaheuristic that can effectively solve non-linear multi-objective optimization problems [138]. It has the advantages of fast operation and good solution convergence.
Peng et al. [139] used a hybrid TS-GEP algorithm and NSGA-II to optimize the laser cladding process parameters. By using this method, they improved the energy and material utilization rate in the laser cladding process. Jiang [140] and Lin et al. [141] optimized the laser cladding process parameters using the NSGA-II algorithm. The results show that this method can effectively reduce the energy loss in the laser cladding process and improve the macro quality and microhardness of the cladding layer. Zhao et al. [142] used the NSGA-II algorithm to optimize the process parameters for coaxial powder-feeding laser cladding. The results showed that the use of the optimized parameters could reduce the depth of the heat-affected zone of the substrate and improve the cladding efficiency. The results of the response values before and after optimization using the NSGA-II algorithm are shown in Table 6.
The NSGA-II algorithm can achieve good results in low-dimensional multi-objective optimization problems. However, for high-dimensional multi-objective optimization problems, when there are too many optimization goals, the selection pressure will be too small, and the computational complexity will increase significantly. Therefore, the algorithm needs to continue to be improved.

3.5. Particle Swarm Optimization Algorithm (PSO)

Particle swarm optimization (PSO) is a class of metaheuristic search algorithms without derivatives [143]. The mode of operation of particle swarm algorithms is similar to that of genetic algorithms. However, particle swarm algorithms consider individuals in the form of populations, rather than focusing only on a single individual, as genetic algorithms do [144]. They are widely used in multi-objective optimization problems because of their fast convergence, simple operation, and few parameters [145].
Pant et al. [146] developed a prediction model considering the process parameters and powder capture efficiency, clad height, and width using the particle swarm algorithm to optimize the artificial neural network. The results showed that the PSO-ANN model’s prediction results were more accurate than those of the ANN model. Deng et al. [147] developed a BPNN-QPSO neural network prediction model for the laser cladding of Ti (C, N) ceramic coatings and compared the method with the Taguchi method, BPNN, and BPNN-PSO. The results show that the probability of the BPNN-QPSO model falling into the locally optimal solution is low, and the convergence speed is fast. Ma et al. [148] used a multi-objective quantum particle swarm optimization algorithm to find the minimum dilution rate and residual stress. They found that the defocus amount had the most excellent effect on the dilution rate, and the scanning speed had the most significant impact on the residual stress. Chukwubuikem et al. [149] used particle swarm optimization algorithms to optimize individual objective functions, obtained the optimal process parameters, and developed a user interface through which the parameters could be optimized directly using particle swarm optimization, without redesigning the optimization methods. Figure 9 shows a comparison of the performance of the neural network and the particle swarm optimization neural network models, and an iterative comparison of the coatings’ microhardness fitness values under different optimization methods.
The standard PSO algorithm may sometimes fail to converge to the optimal global solution. The multi-objective QPSO algorithm is superior in terms of convergence speed and accuracy compared to the standard PSO algorithm. However, the multi-objective QPSO algorithm that converges too quickly may lead to premature convergence, reducing the convergence accuracy.

3.6. Other Intelligent Optimization Methods

The most commonly used intelligent optimization algorithms in laser cladding are the neural network, GABP, support vector machines (SVM), and particle swarm algorithms (PSO). However, in addition to these algorithms, algorithms such as the adaptive neuro-fuzzy inference system (ANFIS), random forest (RF), grey wolf optimization (GWO), and the Bonobo optimization algorithm have also been applied to laser cladding.
Singh et al. [150] used four different methods (sine cosine algorithm, coyote optimization algorithm, Jaya algorithm, and Bonobo optimization algorithm) to optimize the process parameters in order to obtain the minimum dilution. They found that the Bonobo optimization algorithm converged the fastest when searching for the global optimum solution. At the optimum parameters, the microhardness of the clad layer was significantly increased due to the diffusion of WC particles during the cladding process. Zhou et al. [151] determined the optimal process parameters by combining the grey wolf optimization algorithm and the BP neural network. They found that the GWO-BPNN model was more accurate than the predictions of the BPNN model. Under the optimum process parameters, the wear resistance and corrosion resistance of the cladding layer were significantly improved, with wear in the form of adhesion and fatigue wear. Sohrabpoor et al. [152] optimized the process parameters by correlating the ANFIS response model with the Imperialist Competition Algorithm (ICA). The results show that the method effectively improves the powder collection efficiency and obtains the desired clad height and clad width. Liang et al. [153] used a random forest algorithm to construct a regression model of laser cladding process parameters (laser power, scanning speed, powder feeding rate) and a single-pass cladding layer size. The results show that this method can accurately estimate the laser cladding process parameters required to process the cross-sectional geometry of a specific single-pass cladding layer. The prediction results of ANFIS, RF optimization methods, and GWO optimization of BPNN before and after their application are shown in Table 7. The results of the optimized microstructure are shown in Figure 10.
At present, intelligent optimization algorithms have begun to be applied to laser cladding, but there are still some limitations. Optimizing the process parameters using intelligent optimization methods often requires extensive experimentation to obtain valid training data, which can also increase the costs. Moreover, a single intelligent algorithm will likely produce local convergence, failing to obtain the optimal process parameters. Therefore, in the future, hybrid intelligence algorithms should be used more often to optimize the process parameters.

4. Summary and Outlook

It can be seen from the above that the optimization of the process parameters of laser cladding is a complex multi-parameter optimization problem. Through the optimization of the laser cladding process, the parameters can effectively improve the quality of the coating. With the deepening of research, many optimization methods have been developed. In the early years, more traditional optimization methods were used. However, with the development of computer technology, intelligent algorithms began to be applied to optimization problems, but these optimization methods still have some defects. The authors believe that the future development trend of process parameter optimization methods is as follows.

4.1. A Deeper Look at Intelligent Algorithms

Intelligent algorithms have recently been widely used in various optimization fields. However, there are still relatively few applications for evolutionary algorithms in laser cladding, such as simulated annealing algorithms, ant colony algorithms, prohibited search algorithms, and artificial bee colony algorithms. Convolutional neural networks and radial basis neural networks are rarely reported. There are many other areas where improvements to the non-dominated sorting second-generation algorithm have been reported to significantly improve the algorithm’s performance. However, they have yet to be widely used in laser cladding. In addition, since hybrid intelligence algorithms can overcome the shortcomings of single algorithms, more research into hybrid algorithms should follow as well.

4.2. Optimization of Process Parameters under External Auxiliary Conditions

There are many reports that external conditions such as ultrasound and electromagnetic fields can effectively improve the quality of coatings. However, only a few scholars have investigated their effect in optimizing the process parameters. The optimization of the process parameters under external auxiliary conditions can be analyzed to optimize the process parameters more successfully and prepare better coatings with superior properties.

4.3. Optimization Studies Carried out on More Parameters, and More Evaluation Indicators Introduced

There are many process parameters in laser cladding. However, many scholars only optimize and study process parameters such as the laser power, scanning speed, and powder feeding rate. More research needs to be carried out to optimize other process parameters. The laser cladding results are generally evaluated by performance indicators such as the cladding width, height, depth, dilution rate, wear resistance, microhardness, etc. In contrast, the mechanical properties of the coating, such as bending, tensile, fatigue, compression, and shear properties, are rarely studied. It is essential to strengthen the research on the optimization of other process parameters and introduce more evaluation indicators to study the laser cladding process parameters.

4.4. Development of Software That Allows the Optimization of Laser Cladding Process Parameters

Both traditional and intelligent optimization methods require many experiments, which can waste time and increase costs. Developing software that can intelligently select the best process parameters is essential.

Author Contributions

Conceptualization, formal analysis and writing—original draft, K.W., W.L., J.J. and Y.T.; writing—original draft preparation, K.W. and W.L., Methodology, data curation and funding acquisition, K.W., M.Z. and H.F.; Project administration and supervision, K.W. and Y.T.; Date analysis, writing—review and editing, H.M.S.S., Y.H. (Yuxiang Hong), Y.H.(Yongle Hu), J.Z. and D.X. All authors have read and agreed to the published version of the manuscript.

Funding

The authors would like to acknowledge the financial support for this work from the National Natural Science Foundation of China (52205334), the Natural Science Foundation of Hunan Province (2022JJ40495), the Changsha Key Research and Development Project (kh2201275), the Changsha Municipal Natural Science Foundation (kq2202196), the R&D Program of Beijing Municipal Education Commission (KZ202210005004), the Open Fund of Key Laboratory of Intelligent Manufacturing Quality Big Data Tracing and Analysis of Zhejiang Province (ZNZZSZ-CJLU2022-02), the Tribology Science Fund of State Key Laboratory of Tribology in Advanced Equipment (SKLTKF21B08), and the Open Fund of Key Laboratory of Vibration and Control of Aeropropulsion System (VCAME202213).

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Wang, K.; Du, D.; Liu, G.; Chang, B.; Hong, Y. Microstructure and properties of WC reinforced Ni-based composite coatings with Y2O3 addition on titanium alloy by laser cladding. Sci. Technol. Weld. Join. 2019, 24, 517–524. [Google Scholar] [CrossRef]
  2. Ke, J.; Liu, X.-B.; Liang, J.; Liang, L.; Luo, Y.-S. Microstructure and fretting wear of laser cladding self-lubricating anti-wear composite coatings on TA2 alloy after aging treatment. Opt. Laser Technol. 2019, 119, 105599. [Google Scholar] [CrossRef]
  3. Chakraborty, S.S.; Dutta, S. Estimation of dilution in laser cladding based on energy balance approach using regression analysis. Sādhanā 2019, 44, 150. [Google Scholar] [CrossRef] [Green Version]
  4. Nenadl, O.; Ocelík, V.; Palavra, A.; De Hosson, J.T. The Prediction of Coating Geometry from Main Processing Parameters in Laser Cladding. Phys. Procedia 2014, 56, 220–227. [Google Scholar] [CrossRef] [Green Version]
  5. Wang, K.; Du, D.; Liu, G.; Chang, B.; Ju, J.; Sun, S.; Fu, H. Microstructure and property of laser clad Fe-based composite layer containing Nb and B4C powders. J. Alloy. Compd. 2019, 802, 373–384. [Google Scholar] [CrossRef]
  6. Li, F.; Gao, Z.; Li, L.; Chen, Y. Microstructural study of MMC layers produced by combining wire and coaxial WC powder feeding in laser direct metal deposition. Opt. Laser Technol. 2016, 77, 134–143. [Google Scholar] [CrossRef]
  7. Wang, K.; Du, D.; Liu, G.; Pu, Z.; Chang, B.; Ju, J. A study on the additive manufacturing of a high chromium Nickel-based superalloy by extreme high-speed laser metal deposition. Opt. Laser Technol. 2020, 133, 106504. [Google Scholar] [CrossRef]
  8. Bhatnagar, S.; Mullick, S.; Gopinath, M. A lumped parametric analytical model for predicting molten pool temperature and clad geometry in pre-placed powder laser cladding. Optik 2021, 247, 168015. [Google Scholar] [CrossRef]
  9. Frazier, W.E. Metal Additive Manufacturing: A Review. J. Mater. Eng. Perform. 2014, 23, 1917–1928. [Google Scholar] [CrossRef]
  10. Syed, W.U.H.; Pinkerton, A.J.; Li, L. A comparative study of wire feeding and powder feeding in direct diode laser deposition for rapid prototyping. Appl. Surf. Sci. 2005, 247, 268–276. [Google Scholar] [CrossRef]
  11. Dass, A.; Moridi, A. State of the Art in Directed Energy Deposition: From Additive Manufacturing to Materials Design. Coatings 2019, 9, 418. [Google Scholar] [CrossRef] [Green Version]
  12. de Oliveira, U.; Ocelík, V.; De Hosson, J. Analysis of coaxial laser cladding processing conditions. Surf. Coat. Technol. 2005, 197, 127–136. [Google Scholar] [CrossRef]
  13. Hao, M.; Sun, Y. A FEM model for simulating temperature field in coaxial laser cladding of TI6AL4V alloy using an inverse modeling approach. Int. J. Heat Mass Transf. 2013, 64, 352–360. [Google Scholar] [CrossRef]
  14. OuYang, C.-S.; Liu, X.-B.; Luo, Y.-S.; Liang, J.; Wang, M.; Chen, D.-Q. Preparation and high temperature tribological properties of laser in-situ synthesized self-lubricating composite coating on 304 stainless steel. J. Mater. Res. Technol. 2020, 9, 7034–7046. [Google Scholar] [CrossRef]
  15. Xixi, L.; Kaikai, Z.; Jing, C.; Guanghui, M.; Fangli, Y.; Yingying, Z.; Hongwei, Z.; Peiqing, L.; Hui, X. Effect of line energy density of the laser beam on the microstructure and wear resistance properties of the obtained Fe3Al laser cladding coatings. Optik 2022, 261, 169256. [Google Scholar]
  16. Ibrahim, M.Z.; Sarhan, A.A.; Yusuf, F.; Hamdi, M. Biomedical materials and techniques to improve the tribological, mechanical and biomedical properties of orthopedic implants—A review article. J. Alloy. Compd. 2017, 714, 636–667. [Google Scholar] [CrossRef]
  17. Weng, F.; Chen, C.; Yu, H. Research status of laser cladding on titanium and its alloys: A review. Mater. Des. 2014, 58, 412–425. [Google Scholar] [CrossRef]
  18. Chen, L.; Yu, T.; Xu, P.; Zhang, B. In-situ NbC reinforced Fe-based coating by laser cladding: Simulation and experiment. Surf. Coat. Technol. 2021, 412, 127027. [Google Scholar] [CrossRef]
  19. Wang, W.; Wang, M.; Jie, Z.; Sun, F.; Huang, D. Research on the microstructure and wear resistance of titanium alloy structural members repaired by laser cladding. Opt. Lasers Eng. 2008, 46, 810–816. [Google Scholar] [CrossRef]
  20. Sexton, L.; Lavin, S.; Byrne, G.; Kennedy, A. Laser cladding of aerospace materials. J. Mater. Process. Technol. 2002, 122, 63–68. [Google Scholar] [CrossRef]
  21. Zeng, C.; Tian, W.; Liao, W.H.; Hua, L. Microstructure and porosity evaluation in laser-cladding deposited Ni-based coatings. Surf. Coat. Technol. 2016, 294, 122–130. [Google Scholar] [CrossRef]
  22. Jin, G.; Cai, Z.; Guan, Y.; Cui, X.; Liu, Z.; Li, Y.; Dong, M.; Zhang, D. High temperature wear performance of laser-cladded FeNiCoAlCu high-entropy alloy coating. Appl. Surf. Sci. 2018, 445, 113–122. [Google Scholar] [CrossRef]
  23. Liu, X.-B.; Meng, X.-J.; Liu, H.-Q.; Shi, G.-L.; Wu, S.-H.; Sun, C.-F.; Wang, M.-D.; Qi, L.-H. Development and characterization of laser clad high temperature self-lubricating wear resistant composite coatings on Ti–6Al–4V alloy. Mater. Des. 2014, 55, 404–409. [Google Scholar] [CrossRef]
  24. Huang, C.; Zhang, Y.; Vilar, R.; Shen, J. Dry sliding wear behavior of laser clad TiVCrAlSi high entropy alloy coatings on Ti–6Al–4V substrate. Mater. Des. 2012, 41, 338–343. [Google Scholar] [CrossRef]
  25. Abioye, T.; McCartney, D.; Clare, A. Laser cladding of Inconel 625 wire for corrosion protection. J. Mater. Process. Technol. 2015, 217, 232–240. [Google Scholar] [CrossRef] [Green Version]
  26. Wang, G.; Liu, X.-B.; Zhu, G.-X.; Zhu, Y.; Liu, Y.-F.; Zhang, L.; Wang, J.-L. Tribological study of Ti3SiC2/Cu5Si/TiC reinforced Co-based coatings on SUS304 steel by laser cladding. Surf. Coat. Technol. 2022, 432, 128064. [Google Scholar] [CrossRef]
  27. El Cheikh, H.; Courant, B.; Branchu, S.; Hascoët, J.-Y.; Guillén, R. Analysis and prediction of single laser tracks geometrical characteristics in coaxial laser cladding process. Opt. Lasers Eng. 2012, 50, 413–422. [Google Scholar] [CrossRef] [Green Version]
  28. Hamedi, M.; Torkamany, M.; Sabbaghzadeh, J. Effect of pulsed laser parameters on in-situ TiC synthesis in laser surface treatment. Opt. Lasers Eng. 2010, 49, 557–563. [Google Scholar] [CrossRef]
  29. Ferreira, A.; Darabi, R.; Sousa, J.; Cruz, J.; Reis, A.; Vieira, M. Optimization of Direct Laser Deposition of a Martensitic Steel Powder (Metco 42C) on 42CrMo4 Steel. Metals 2021, 11, 672. [Google Scholar] [CrossRef]
  30. Shan, B.; Chen, J.; Chen, S.; Ma, M.; Ni, L.; Shang, F.; Zhou, L. Laser cladding of Fe-based corrosion and wear-resistant alloy: Genetic design, microstructure, and properties. Surf. Coat. Technol. 2022, 433, 128117. [Google Scholar] [CrossRef]
  31. Gan, Z.; Yu, G.; He, X.; Li, S. Numerical simulation of thermal behavior and multicomponent mass transfer in direct laser deposition of Co-base alloy on steel. Int. J. Heat Mass Transf. 2017, 104, 28–38. [Google Scholar] [CrossRef] [Green Version]
  32. Janicki, D. Laser cladding of Inconel 625-based composite coatings reinforced by porous chromium carbide particles. Opt. Laser Technol. 2017, 94, 6–14. [Google Scholar] [CrossRef]
  33. Fu, F.; Zhang, Y.; Chang, G.; Dai, J. Analysis on the physical mechanism of laser cladding crack and its influence factors. Optik 2016, 127, 200–202. [Google Scholar] [CrossRef]
  34. Balu, P.; Leggett, P.; Hamid, S.; Kovacevic, R. Multi-Response Optimization of Laser-based Powder Deposition of Multi-track Single Layer Hastelloy C-276. Mater. Manuf. Process. 2013, 28, 173–182. [Google Scholar] [CrossRef]
  35. Pant, P.; Chatterjee, D.; Nandi, T.; Samanta, S.K.; Lohar, A.K.; Changdar, A. Statistical modelling and optimization of clad characteristics in laser metal deposition of austenitic stainless steel. J. Braz. Soc. Mech. Sci. Eng. 2019, 41, 283. [Google Scholar] [CrossRef]
  36. Devojno, O.G.; Feldshtein, E.; Kardapolava, M.A.; Lutsko, N.I. On the formation features, structure, microhardness and tribological behavior of single tracks and coating layers formed by laser cladding of Al-Fe powder bronze. Surf. Coat. Technol. 2019, 358, 195–206. [Google Scholar] [CrossRef]
  37. Moeinfar, K.; Khodabakhshi, F.; Kashani-Bozorg, S.; Mohammadi, M.; Gerlich, A. A review on metallurgical aspects of laser additive manufacturing (LAM): Stainless steels, nickel superalloys, and titanium alloys. J. Mater. Res. Technol. 2022, 16, 1029–1068. [Google Scholar] [CrossRef]
  38. Zhu, L.; Xue, P.; Lan, Q.; Meng, G.; Ren, Y.; Yang, Z.; Xu, P.; Liu, Z. Recent research and development status of laser cladding: A review. Opt. Laser Technol. 2021, 138, 106915. [Google Scholar] [CrossRef]
  39. Zheng, Y.; Liu, F.; Gao, J.; Liu, F.; Huang, C.; Zheng, H.; Wang, P.; Qiu, H. Effect of different heat input on the microstructure and mechanical properties of laser cladding repaired 300M steel. J. Mater. Res. Technol. 2023, 22, 556–568. [Google Scholar] [CrossRef]
  40. Ebrahimnia, M.; Xie, Y.; Chi, C. Effect of cladding parameters on microstructure and defects in direct laser metal deposition of 24CrNiMo steel. J. Laser Appl. 2021, 33, 012007. [Google Scholar] [CrossRef]
  41. Muvvala, G.; Karmakar, D.P.; Nath, A.K. Online monitoring of thermo-cycles and its correlation with microstructure in laser cladding of nickel based super alloy. Opt. Lasers Eng. 2017, 88, 139–152. [Google Scholar] [CrossRef]
  42. Urbanic, R.J.; Saqib, S.M.; Aggarwal, K. Using Predictive Modeling and Classification Methods for Single and Overlapping Bead Laser Cladding to Understand Bead Geometry to Process Parameter Relationships. J. Manuf. Sci. Eng. 2016, 138, 051012. [Google Scholar] [CrossRef]
  43. Goodarzi, D.M.; Pekkarinen, J.; Salminen, A. Analysis of laser cladding process parameter influence on the clad bead geometry. Weld. World 2017, 61, 883–891. [Google Scholar] [CrossRef]
  44. Bax, B.; Rajput, R.; Kellet, R.; Reisacher, M. Systematic evaluation of process parameter maps for laser cladding and directed energy deposition. Addit. Manuf. 2018, 21, 487–494. [Google Scholar] [CrossRef]
  45. Siddiqui, A.A.; Dubey, A.K. Recent trends in laser cladding and surface alloying. Opt. Laser Technol. 2020, 134, 106619. [Google Scholar] [CrossRef]
  46. Cui, C.; Wu, M.; Miao, X.; Gong, Y.; Zhao, Z. The effect of laser energy density on the geometric characteristics, microstructure and corrosion resistance of Co-based coatings by laser cladding. J. Mater. Res. Technol. 2021, 15, 2405–2418. [Google Scholar] [CrossRef]
  47. Yao, K.; Zongde, L.; Quanbing, L. Effect of Laser Power on Microstructure and Properties of Ni-based Alloy Coatings on 30CrMnSiA Steel. J. Therm. Spray Technol. 2022, 31, 2136–2146. [Google Scholar] [CrossRef]
  48. Shamsaei, N.; Yadollahi, A.; Bian, L.; Thompson, S.M. An overview of Direct Laser Deposition for additive manufacturing; Part II: Mechanical behavior, process parameter optimization and control. Addit. Manuf. 2015, 8, 12–35. [Google Scholar] [CrossRef]
  49. Zhao, Y.; Sun, J.; Guo, K.; Li, J. Investigation on the effect of laser remelting for laser cladding nickel based alloy. J. Laser Appl. 2019, 31, 022512. [Google Scholar] [CrossRef]
  50. Pinkerton, A.J. Advances in the modeling of laser direct metal deposition. J. Laser Appl. 2015, 27, S15001. [Google Scholar] [CrossRef] [Green Version]
  51. Bakhtiyari, A.N.; Wang, Z.; Wang, L.; Zheng, H. A review on applications of artificial intelligence in modeling and optimization of laser beam machining. Opt. Laser Technol. 2020, 135, 106721. [Google Scholar] [CrossRef]
  52. Gao, J.; Wang, C.; Hao, Y.; Liang, X.; Zhao, K. Prediction of TC11 single-track geometry in laser metal deposition based on back propagation neural network and random forest. J. Mech. Sci. Technol. 2022, 36, 1417–1425. [Google Scholar] [CrossRef]
  53. Yu, D.; Wang, C.; Cheng, X.; Zhang, F. Optimization of hybrid PVD process of TiAlN coatings by Taguchi method. Appl. Surf. Sci. 2008, 255, 1865–1869. [Google Scholar] [CrossRef]
  54. Mahamood, R.; Akinlabi, E.; Shukla, M.; Pityana, S. Scanning velocity influence on microstructure, microhardness and wear resistance performance of laser deposited Ti6Al4V/TiC composite. Mater. Des. 2013, 50, 656–666. [Google Scholar] [CrossRef]
  55. Liu, J.; Li, J.; Cheng, X.; Wang, H. Effect of dilution and macrosegregation on corrosion resistance of laser clad AerMet100 steel coating on 300M steel substrate. Surf. Coat. Technol. 2017, 325, 352–359. [Google Scholar] [CrossRef]
  56. Weng, F.; Yu, H.; Chen, C.; Liu, J.; Zhao, L.; Dai, J.; Zhao, Z. Effect of process parameters on the microstructure evolution and wear property of the laser cladding coatings on Ti-6Al-4V alloy. J. Alloy. Compd. 2016, 692, 989–996. [Google Scholar] [CrossRef]
  57. Yi, P.; Zhan, X.; He, Q.; Liu, Y.; Xu, P.; Xiao, P.; Jia, D. Influence of laser parameters on graphite morphology in the bonding zone and process optimization in gray cast iron laser cladding. Opt. Laser Technol. 2018, 109, 480–487. [Google Scholar] [CrossRef]
  58. Gu, J.; Ju, J.; Wang, R.; Li, J.; Yu, H.; Wang, K. Effects of Laser Scanning Rate and Ti Content on Wear of Novel Fe-Cr-B-Al-Ti Coating Prepared via Laser Cladding. J. Therm. Spray Technol. 2022, 31, 2609–2620. [Google Scholar] [CrossRef]
  59. Srisungsitthisunti, P.; Kaewprachum, B.; Yang, Z.; Gao, G. Real-Time Quality Monitoring of Laser Cladding Process on Rail Steel by an Infrared Camera. Metals 2022, 12, 825. [Google Scholar] [CrossRef]
  60. Zhang, H.; Pan, Y.; Zhang, Y.; Lian, G.; Cao, Q.; Yang, J. Influence of laser power on the microstructure and properties of in-situ NbC/WCoB–TiC coating by laser cladding. Mater. Chem. Phys. 2022, 290, 126636. [Google Scholar] [CrossRef]
  61. Li, M.; Huang, J.; Zhu, Y.Y.; Li, Z.G.; Wu, Y.X. Effect of laser scanning speed on TiN/TiB–Ti based composite. Surf. Eng. 2013, 29, 346–350. [Google Scholar] [CrossRef]
  62. Jiao, X.; Wang, J.; Wang, C.; Gong, Z.; Pang, X.; Xiong, S. Effect of laser scanning speed on microstructure and wear properties of T15M cladding coating fabricated by laser cladding technology. Opt. Lasers Eng. 2018, 110, 163–171. [Google Scholar] [CrossRef]
  63. Bartkowski, D.; Kinal, G. Microstructure and wear resistance of Stellite-6/WC MMC coatings produced by laser cladding using Yb:YAG disk laser. Int. J. Refract. Met. Hard Mater. 2016, 58, 157–164. [Google Scholar] [CrossRef]
  64. Chao, Q.; Guo, T.; Jarvis, T.; Wu, X.; Hodgson, P.; Fabijanic, D. Direct laser deposition cladding of Al CoCrFeNi high entropy alloys on a high-temperature stainless steel. Surf. Coat. Technol. 2017, 332, 440–451. [Google Scholar] [CrossRef]
  65. Chryssolouris, G.; Zannis, S.; Tsirbas, K.; Lalas, C. An Experimental Investigation of Laser Cladding. CIRP Ann. 2002, 51, 145–148. [Google Scholar] [CrossRef]
  66. Qu, C.; Li, J.; Bai, L.; Shao, J.; Song, R.; Chen, J. Effects of the thickness of the pre-placed layer on microstructural evolution and mechanical properties of the laser-clad coatings. J. Alloy. Compd. 2015, 644, 450–463. [Google Scholar] [CrossRef]
  67. Fu, Y.; Downey, A.R.; Yuan, L.; Zhang, T.; Pratt, A.; Balogun, Y. Machine learning algorithms for defect detection in metal laser-based additive manufacturing: A review. J. Manuf. Process. 2022, 75, 693–710. [Google Scholar] [CrossRef]
  68. Davim, J.P.; Oliveira, C.; Cardoso, A. Predicting the geometric form of clad in laser cladding by powder using multiple regression analysis (MRA). Mater. Des. 2008, 29, 554–557. [Google Scholar] [CrossRef]
  69. Huang, Y.; Yuan, S. Modeling the geometric formation and powder deposition mass in laser induction hybrid cladding. J. Mech. Sci. Technol. 2012, 26, 2347–2351. [Google Scholar] [CrossRef]
  70. Shi, J.; Zhu, P.; Fu, G.; Shi, S. Geometry characteristics modeling and process optimization in coaxial laser inside wire cladding. Opt. Laser Technol. 2018, 101, 341–348. [Google Scholar] [CrossRef]
  71. Ansari, M.; Razavi, R.S.; Barekat, M. An empirical-statistical model for coaxial laser cladding of NiCrAlY powder on Inconel 738 superalloy. Opt. Laser Technol. 2016, 86, 136–144. [Google Scholar] [CrossRef]
  72. Shayanfar, P.; Daneshmanesh, H.; Janghorban, K. Parameters Optimization for Laser Cladding of Inconel 625 on ASTM A592 Steel. J. Mater. Res. Technol. 2020, 9, 8258–8265. [Google Scholar] [CrossRef]
  73. Erfanmanesh, M.; Abdollah-Pour, H.; Mohammadian-Semnani, H.; Shoja-Razavi, R. An empirical-statistical model for laser cladding of WC-12Co powder on AISI 321 stainless steel. Opt. Laser Technol. 2017, 97, 180–186. [Google Scholar] [CrossRef]
  74. Nabhani, M.; Razavi, R.S.; Barekat, M. An empirical-statistical model for laser cladding of Ti-6Al-4V powder on Ti-6Al-4V substrate. Opt. Laser Technol. 2018, 100, 265–271. [Google Scholar] [CrossRef]
  75. Javid, Y. Multi-response optimization in laser cladding process of WC powder on Inconel 718. CIRP J. Manuf. Sci. Technol. 2020, 31, 406–417. [Google Scholar] [CrossRef]
  76. Huang, Y.; Hu, Y.; Zhang, M.; Mao, C.; Wang, K.; Tong, Y.; Zhang, J.; Li, K. Multi-Objective Optimization of Process Parameters in Laser Cladding CoCrCuFeNi High-Entropy Alloy Coating. J. Therm. Spray Technol. 2022, 31, 1985–2000. [Google Scholar] [CrossRef]
  77. Bose, S.; Nandi, T. Microstructural characterization and measurement of laser responses of lens developed novel titanium matrix composite. Eur. Phys. J. Plus 2021, 136, 978. [Google Scholar] [CrossRef]
  78. Badkar, D.S.; Pandey, K.S.; Buvanashekaran, G. Application of the central composite design in optimization of laser transformation hardening parameters of commercially pure titanium using Nd:YAG laser. Int. J. Adv. Manuf. Technol. 2011, 59, 169–192. [Google Scholar] [CrossRef]
  79. Sun, Y.; Hao, M. Statistical analysis and optimization of process parameters in Ti6Al4V laser cladding using Nd:YAG laser. Opt. Lasers Eng. 2012, 50, 985–995. [Google Scholar] [CrossRef]
  80. Dada, M.; Popoola, P.; Mathe, N.; Pityana, S.; Adeosun, S. Parametric optimization of laser deposited high entropy alloys using response surface methodology (RSM). Int. J. Adv. Manuf. Technol. 2020, 109, 2719–2732. [Google Scholar] [CrossRef]
  81. Guo, C.; He, S.; Yue, H.; Li, Q.; Hao, G. Prediction modelling and process optimization for forming multi-layer cladding structures with laser directed energy deposition. Opt. Laser Technol. 2020, 134, 106607. [Google Scholar] [CrossRef]
  82. Cui, L.-J.; Zhang, M.; Guo, S.-R.; Cao, Y.-L.; Zeng, W.-H.; Li, X.-L.; Zheng, B. Multi-objective numerical simulation of geometrical characteristics of laser cladding of cobalt-based alloy based on response surface methodology. Meas. Control. 2020, 54, 1125–1135. [Google Scholar] [CrossRef]
  83. Li, T.; Long, H.; Qiu, C.; Wang, M.; Li, D.; Dong, Z.; Gui, Y. Multi-Objective Optimization of Process Parameters of 45 Steel Laser Cladding Ni60PTA Alloy Powder. Coatings 2022, 12, 939. [Google Scholar] [CrossRef]
  84. Wu, Z.; Li, T.; Li, Q.; Shi, B.; Li, X.; Wang, X.; Lu, H.; Zhang, H.-C. Process optimization of laser cladding Ni60A alloy coating in remanufacturing. Opt. Laser Technol. 2019, 120, 105718. [Google Scholar] [CrossRef]
  85. Khorram, A.; Jamaloei, A.D.; Paidar, M.; Cao, X. Laser cladding of Inconel 718 with 75Cr3C2 + 25(80Ni20Cr) powder: Statistical modeling and optimization. Surf. Coat. Technol. 2019, 378, 124933. [Google Scholar] [CrossRef]
  86. Lian, G.; Yao, M.; Zhang, Y.; Chen, C. Analysis and prediction on geometric characteristics of multi-track overlapping laser cladding. Int. J. Adv. Manuf. Technol. 2018, 97, 2397–2407. [Google Scholar] [CrossRef]
  87. Wu, S.; Liu, Z.; Huang, X.; Wu, Y.; Gong, Y. Process parameter optimization and EBSD analysis of Ni60A-25% WC laser cladding. Int. J. Refract. Met. Hard Mater. 2021, 101, 105675. [Google Scholar] [CrossRef]
  88. Ben-Arfa, B.A.; Salvado, I.M.M.; Frade, J.R.; Pullar, R.C. Fast route for synthesis of stoichiometric hydroxyapatite by employing the Taguchi method. Mater. Des. 2016, 109, 547–555. [Google Scholar] [CrossRef]
  89. Liu, Y.; Liu, C.; Liu, W.; Ma, Y.; Tang, S.; Liang, C.; Cai, Q.; Zhang, C. Optimization of parameters in laser powder deposition AlSi10Mg alloy using Taguchi method. Opt. Laser Technol. 2018, 111, 470–480. [Google Scholar] [CrossRef]
  90. Lin, C.-M. Parameter optimization of laser cladding process and resulting microstructure for the repair of tenon on steam turbine blade. Vacuum 2015, 115, 117–123. [Google Scholar] [CrossRef]
  91. Xu, M.; Li, J.; Jiang, J.; Li, B. Influence of Powders and Process Parameters on Bonding Shear Strength and Micro Hardness in Laser Cladding Remanufacturing. Procedia CIRP 2015, 29, 804–809. [Google Scholar] [CrossRef]
  92. Quazi, M.M.; Fazal, M.A.; Haseeb, A.S.M.A.; Yusof, F.; Masjuki, H.H.; Arslan, A. Laser Composite Surfacing of Ni-WC Coating on AA5083 for Enhancing Tribomechanical Properties. Tribol. Trans. 2016, 60, 249–259. [Google Scholar] [CrossRef]
  93. Chen, L.; Yu, T.; Chen, X.; Zhao, Y.; Guan, C. Process optimization, microstructure and microhardness of coaxial laser cladding TiC reinforced Ni-based composite coatings. Opt. Laser Technol. 2022, 152, 108129. [Google Scholar] [CrossRef]
  94. Shi, Y.; Li, Y.; Liu, J.; Yuan, Z. Investigation on the parameter optimization and performance of laser cladding a gradient composite coating by a mixed powder of Co50 and Ni/WC on 20CrMnTi low carbon alloy steel. Opt. Laser Technol. 2018, 99, 256–270. [Google Scholar] [CrossRef]
  95. Zhang, Z.; Kovacevic, R. Multiresponse Optimization of Laser Cladding Steel + VC Using Grey Relational Analysis in the Taguchi Method. JOM 2016, 68, 1762–1773. [Google Scholar] [CrossRef]
  96. Mondal, S.; Paul, C.P.; Kukreja, L.M.; Bandyopadhyay, A.; Pal, P.K. Application of Taguchi-based gray relational analysis for evaluating the optimal laser cladding parameters for AISI1040 steel plane surface. Int. J. Adv. Manuf. Technol. 2012, 66, 91–96. [Google Scholar] [CrossRef]
  97. Deng, D.; Li, T.; Huang, Z.; Jiang, H.; Yang, S.; Zhang, Y. Multi-response optimization of laser cladding for TiC particle reinforced Fe matrix composite based on Taguchi method and grey relational analysis. Opt. Laser Technol. 2022, 153, 108259. [Google Scholar] [CrossRef]
  98. Yu, T.; Yang, L.; Zhao, Y.; Sun, J.; Li, B. Experimental research and multi-response multi-parameter optimization of laser cladding Fe313. Opt. Laser Technol. 2018, 108, 321–332. [Google Scholar] [CrossRef]
  99. Fan, P.; Zhang, G. Study on process optimization of WC-Co50 cermet composite coating by laser cladding. Int. J. Refract. Met. Hard Mater. 2019, 87, 105133. [Google Scholar] [CrossRef]
  100. Marzbanrad, J.; Ghaseminejad, P.; Ahmadzadeh, M.H.; Teimouri, R. Experimental investigation and statistical optimization of laser surface cladding parameters. Int. J. Adv. Manuf. Technol. 2014, 76, 1163–1172. [Google Scholar] [CrossRef]
  101. Wang, Q.; Zeng, X.; Chen, C.; Lian, G.; Huang, X. An Integrated Method for Multi-Objective Optimization of Multi-Pass Fe50/TiC Laser Cladding on AISI 1045 Steel based on Grey Relational Analysis and Principal Component Analysis. Coatings 2020, 10, 151. [Google Scholar] [CrossRef] [Green Version]
  102. Liang, W.; Yang, Y.; Qi, K.; Jin, K.; Xiong, L. Quality evaluation of multi-path laser cladding coatings based on integrated fuzzy comprehensive evaluation and improved analytical hierarchy process method. Surf. Coat. Technol. 2021, 427, 127816. [Google Scholar] [CrossRef]
  103. Reddy, L.; Preston, S.; Shipway, P.; Davis, C.; Hussain, T. Process parameter optimisation of laser clad iron based alloy: Predictive models of deposition efficiency, porosity and dilution. Surf. Coat. Technol. 2018, 349, 198–207. [Google Scholar] [CrossRef]
  104. Meng, G.; Zhu, L.; Zhang, J.; Yang, Z.; Xue, P. Statistical analysis and multi-objective process optimization of laser cladding TiC-Inconel718 composite coating. Optik 2021, 240, 166828. [Google Scholar] [CrossRef]
  105. Menghani, J.; Vyas, A.; More, S.; Paul, C.; Patnaik, A. Parametric investigation and optimization for CO2 laser cladding of AlFeCoCrNiCu powder on AISI 316. High Temp. Mater. Process. 2021, 40, 265–280. [Google Scholar] [CrossRef]
  106. Mohammed, S.; Zhang, Z.; Kovacevic, R. Optimization of processing parameters in fiber laser cladding. Int. J. Adv. Manuf. Technol. 2020, 111, 2553–2568. [Google Scholar] [CrossRef]
  107. Liu, S.; Kovacevic, R. Statistical analysis and optimization of processing parameters in high-power direct diode laser cladding. Int. J. Adv. Manuf. Technol. 2014, 74, 867–878. [Google Scholar] [CrossRef]
  108. Mahamood, R.M.; Akinlabi, E.T. Processing Parameters Optimization for Material Deposition Efficiency in Laser Metal Deposited Titanium Alloy. Lasers Manuf. Mater. Process. 2015, 3, 9–21. [Google Scholar] [CrossRef]
  109. Moradi, M.; Hasani, A.; Pourmand, Z.; Lawrence, J. Direct laser metal deposition additive manufacturing of Inconel 718 superalloy: Statistical modelling and optimization by design of experiments. Opt. Laser Technol. 2021, 144, 107380. [Google Scholar] [CrossRef]
  110. Akbari, M.; Saedodin, S.; Panjehpour, A.; Hassani, M.; Afrand, M.; Torkamany, M.J. Numerical simulation and designing artificial neural network for estimating melt pool geometry and temperature distribution in laser welding of Ti6Al4V alloy. Optik 2016, 127, 11161–11172. [Google Scholar] [CrossRef]
  111. Shakeri, S.; Ghassemi, A.; Hassani, M.; Hajian, A. Investigation of material removal rate and surface roughness in wire electrical discharge machining process for cementation alloy steel using artificial neural network. Int. J. Adv. Manuf. Technol. 2015, 82, 549–557. [Google Scholar] [CrossRef]
  112. Yin, F.; Mao, H.; Hua, L. A hybrid of back propagation neural network and genetic algorithm for optimization of injection molding process parameters. Mater. Des. 2011, 32, 3457–3464. [Google Scholar] [CrossRef]
  113. Mohajernia, B.; Mirazimzadeh, S.E.; Pasha, A.; Urbanic, R.J. Machine learning approaches for predicting geometric and mechanical characteristics for single P420 laser beads clad onto an AISI 1018 substrate. Int. J. Adv. Manuf. Technol. 2021, 118, 3691–3710. [Google Scholar] [CrossRef]
  114. Mondal, S.; Bandyopadhyay, A.; Pal, P.K. Application of artificial neural network for the prediction of laser cladding process characteristics at Taguchi-based optimized condition. Int. J. Adv. Manuf. Technol. 2013, 70, 2151–2158. [Google Scholar] [CrossRef]
  115. Aggarwal, K.; Urbanic, R.J.; Saqib, S.M. Development of predictive models for effective process parameter selection for single and overlapping laser clad bead geometry. Rapid Prototyp. J. 2018, 24, 214–228. [Google Scholar] [CrossRef]
  116. Hongyu, L.; Hui, C.; Ying, W.; Yong, C.; Wei, Y. Prediction of two-dimensional topography of laser cladding based on neural network. Int. J. Mod. Phys. B 2019, 33, 1940034. [Google Scholar] [CrossRef]
  117. Song, C.; Liu, L.; Yang, Y.; Weng, C. Prediction on Geometrical Characteristics of Laser Energy Deposition Based on Regression Equation and Neural Network. IFAC-PapersOnLine 2020, 53, 89–96. [Google Scholar] [CrossRef]
  118. Li, Y.; Wang, K.; Fu, H.; Zhi, X.; Guo, X.; Lin, J. Prediction for Dilution Rate of AlCoCrFeNi Coatings by Laser Cladding Based on a BP Neural Network. Coatings 2021, 11, 1402. [Google Scholar] [CrossRef]
  119. Caiazzo, F.; Caggiano, A. Laser Direct Metal Deposition of 2024 Al Alloy: Trace Geometry Prediction via Machine Learning. Materials 2018, 11, 444. [Google Scholar] [CrossRef] [Green Version]
  120. Guo, S.; Chen, Z.; Cai, D.; Zhang, Q.; Kovalenko, V.; Yao, J. Prediction of Simulating and Experiments for Co-based Alloy Laser Cladding by HPDL. Phys. Procedia 2013, 50, 375–382. [Google Scholar] [CrossRef] [Green Version]
  121. Kumar, A.; Maji, K. Selection of Process Parameters for Near-Net Shape Deposition in Wire Arc Additive Manufacturing by Genetic Algorithm. J. Mater. Eng. Perform. 2020, 29, 3334–3352. [Google Scholar] [CrossRef]
  122. Yang, S.; Bai, H.; Li, C.; Shu, L.; Zhang, X.; Jia, Z. Numerical Simulation and Multi-Objective Parameter Optimization of Inconel718 Coating Laser Cladding. Coatings 2022, 12, 708. [Google Scholar] [CrossRef]
  123. Ahsan, N.; Khoda, B. AM optimization framework for part and process attributes through geometric analysis. Addit. Manuf. 2016, 11, 85–96. [Google Scholar] [CrossRef]
  124. Liu, J.C.; Ni, L.B. Prediction of laser clad parameters based on neural network. Mater. Technol. 2012, 27, 11–14. [Google Scholar] [CrossRef]
  125. Xu, Z.M.; Wu, H.B.; Hong, Z.H. Quality Prediction of Laser Cladding Based on Evolutionary Neural Network. Appl. Mech. Mater. 2010, 44–47, 1012–1017. [Google Scholar] [CrossRef]
  126. Liu, H.; Qin, X.; Huang, S.; Jin, L.; Wang, Y.; Lei, K. Geometry Characteristics Prediction of Single Track Cladding Deposited by High Power Diode Laser Based on Genetic Algorithm and Neural Network. Int. J. Precis. Eng. Manuf. 2018, 19, 1061–1070. [Google Scholar] [CrossRef]
  127. Pang, Y.; Fu, G.; Wang, M.; Gong, Y.; Yu, S.; Xu, J.; Liu, F. Parameter Optimization of High Deposition Rate Laser Cladding Based on the Response Surface Method and Genetic Neural Network Model. Chin. J. Lasers 2021, 48, 0602112. [Google Scholar] [CrossRef]
  128. Yu, J.; Sun, W.; Huang, H.; Wang, W.; Hu, Y. Crack Sensitivity Control of Nickel-Based Laser Coating Based on Genetic Algorithm and Neural Network. Coatings 2019, 9, 728. [Google Scholar] [CrossRef] [Green Version]
  129. Youwen, Y.; Zongjun, T.; Hu, P.; Dongsheng, W.; Lida, S. Geometry quality prediction of NI-based superalloy coating by laser cladding based on neural network and genetic algorithm. Trans. China Weld. Inst. 2013, 34, 78–82. [Google Scholar]
  130. Xiao, T.T.; Cai, C.Z.; Tang, J.L.; Huang, S.J. Modeling of transition temperature for pulsed laser deposition NdBa2Cu3O7-δ thin films via support vector regression. Int. J. Mod. Phys. B 2013, 27, 1362040. [Google Scholar] [CrossRef]
  131. Pan, X.; Pang, X.; Wang, H.; Xu, Y. A safe screening based framework for support vector regression. Neurocomputing 2018, 287, 163–172. [Google Scholar] [CrossRef]
  132. Wang, T.; Qi, J.; Xu, H.; Wang, Y.; Liu, L.; Gao, D. Fault diagnosis method based on FFT-RPCA-SVM for Cascaded-Multilevel Inverter. ISA Trans. 2016, 60, 156–163. [Google Scholar] [CrossRef] [PubMed]
  133. He, Z.; Zeng, Y.; Shao, H.; Hu, H.; Xu, X. Novel motor fault detection scheme based on one-class tensor hyperdisk. Knowl. Based Syst. 2023, 262, 110259. [Google Scholar] [CrossRef]
  134. Chen, T.; Wu, W.; Li, W.; Liu, D. Laser cladding of nanoparticle TiC ceramic powder: Effects of process parameters on the quality characteristics of the coatings and its prediction model. Opt. Laser Technol. 2019, 116, 345–355. [Google Scholar] [CrossRef]
  135. Chen, X.; Xiao, M.; Kang, D.; Sang, Y.; Zhang, Z.; Jin, X. Prediction of Geometric Characteristics of Melt Track Based on Direct Laser Deposition Using M-SVR Algorithm. Materials 2021, 14, 7221. [Google Scholar] [CrossRef] [PubMed]
  136. Yao, W.; Huang, Y.; Yang, Y. Size Prediction of Directed Energy Deposited Cladding Tracks Based on Support Vector Regression. Chin. J. Lasers 2020, 47, 0802007. [Google Scholar] [CrossRef]
  137. Zhang, Y.; Gong, B.; Tang, Z.; Cao, W. Application of a Bio-Inspired Algorithm in the Process Parameter Optimization of Laser Cladding. Machines 2022, 10, 263. [Google Scholar] [CrossRef]
  138. Jaber, A.; Lafon, P.; Younes, R. A branch-and-bound algorithm based on NSGAII for multi-objective mixed integer nonlinear optimization problems. Eng. Optim. 2021, 54, 1004–1022. [Google Scholar] [CrossRef]
  139. Peng, S.; Li, T.; Zhao, J.; Lv, S.; Tan, G.Z.; Dong, M.; Zhang, H. Towards energy and material efficient laser cladding process: Modeling and optimization using a hybrid TS-GEP algorithm and the NSGA-II. J. Clean. Prod. 2019, 227, 58–69. [Google Scholar] [CrossRef]
  140. Jiang, X.; Tian, Z.; Liu, W.; Tian, G.; Gao, Y.; Xing, F.; Suo, Y.; Song, B. An energy-efficient method of laser remanufacturing process. Sustain. Energy Technol. Assess. 2022, 52, 102201. [Google Scholar] [CrossRef]
  141. Shu, L.; Li, J.; Wu, H.; Heng, Z. Optimization of Multi-Track Laser-Cladding Process of Titanium Alloy Based on RSM and NSGA-II Algorithm. Coatings 2022, 12, 1301. [Google Scholar] [CrossRef]
  142. Zhao, K.; Liang, X.; Wang, W.; Yang, P.; Hao, Y.; Zhu, Z. Multi-Objective Optimization of Coaxial Powder Feeding Laser Cladding Based on NSGA-II. Chin. J. Lasers 2020, 47, 0102004. [Google Scholar] [CrossRef]
  143. Vahabli, E.; Rahmati, S. Improvement of FDM parts’ surface quality using optimized neural networks—medical case studies. Rapid Prototyp. J. 2017, 23, 825–842. [Google Scholar] [CrossRef]
  144. Adedeji, P.A.; Akinlabi, S.A.; Madushele, N.; Olatunji, O.O. Hybrid neurofuzzy wind power forecast and wind turbine location for embedded generation. Int. J. Energy Res. 2020, 45, 413–428. [Google Scholar] [CrossRef]
  145. Xu, S.-H.; Mu, X.-D.; Chai, D.; Zhao, P. Multi-objective quantum-behaved particle swarm optimization algorithm with double-potential well and share-learning. Optik 2016, 127, 4921–4927. [Google Scholar] [CrossRef]
  146. Pant, P.; Chatterjee, D. Prediction of clad characteristics using ANN and combined PSO-ANN algorithms in laser metal deposition process. Surf. Interfaces 2020, 21, 100699. [Google Scholar] [CrossRef]
  147. Deng, Z.; Chen, T.; Wang, H.; Li, S.; Liu, D. Process Parameter Optimization When Preparing Ti(C, N) Ceramic Coatings Using Laser Cladding Based on a Neural Network and Quantum-Behaved Particle Swarm Optimization Algorithm. Appl. Sci. 2020, 10, 6331. [Google Scholar] [CrossRef]
  148. Ma, M.; Xiong, W.; Lian, Y.; Han, D.; Zhao, C.; Zhang, J. Modeling and optimization for laser cladding via multi-objective quantum-behaved particle swarm optimization algorithm. Surf. Coat. Technol. 2019, 381, 125129. [Google Scholar] [CrossRef]
  149. Ngwoke, C.C.; Mahamood, R.M.; Aigbodion, V.S.; Jen, T.-C.; Adedeji, P.A.; Akinlabi, E.T. Soft computing-based process optimization in laser metal deposition of Ti-6Al-4 V. Int. J. Adv. Manuf. Technol. 2022, 120, 1079–1093. [Google Scholar] [CrossRef]
  150. Singh, A.K.; Bal, K.S.; Dey, D.; Das, A.K.; Pal, A.R.; Pratihar, D.K.; Choudhury, A.R. Experimental investigation and parametric optimization for minimization of dilution during direct laser metal deposition of tungsten carbide and cobalt powder mixture on SS304 substrate. Powder Technol. 2021, 390, 339–353. [Google Scholar] [CrossRef]
  151. Zhou, Z.; Du, Y.; He, G.; Xu, L.; Shu, L. Optimization and Characterization of Laser Cladding of 15-5PH Coating on 20Cr13 Stainless Steel. J. Mater. Eng. Perform. 2022, 32, 962–977. [Google Scholar] [CrossRef]
  152. Sohrabpoor, H. Analysis of laser powder deposition parameters: ANFIS modeling and ICA optimization. Optik 2016, 127, 4031–4038. [Google Scholar] [CrossRef]
  153. Liang, X.; Wang, W.; Zhao, K.; Hao, Y.; Yang, P.; Zhu, Z. Application of random forest re-gression analysis in trace geometry prediction of laser cladding. Chin. J. Nonferrous Met. 2020, 30, 1644–1652. [Google Scholar] [CrossRef]
Coatings 13 00496 g001 550
Figure 1. The influence of different laser power, scanning speed, and powder feeding rate on cladding results. (a) Effect of different laser powers on the melt pool. Reprinted from [59]. (b) Fracture toughness and fracture energy of coatings at different laser powers. Reprinted with permission from Ref. [60]. 2022, Elsevier. (c) Wear height losses of substrate and clad layers with different scanning speeds. Reprinted with permission from Ref. [61]. 2013, Taylor and Francis. (d) Effect of different powder feeding rates on the wear resistance of coatings. Reprinted with permission from Ref. [63]. 2016, Elsevier.
Figure 1. The influence of different laser power, scanning speed, and powder feeding rate on cladding results. (a) Effect of different laser powers on the melt pool. Reprinted from [59]. (b) Fracture toughness and fracture energy of coatings at different laser powers. Reprinted with permission from Ref. [60]. 2022, Elsevier. (c) Wear height losses of substrate and clad layers with different scanning speeds. Reprinted with permission from Ref. [61]. 2013, Taylor and Francis. (d) Effect of different powder feeding rates on the wear resistance of coatings. Reprinted with permission from Ref. [63]. 2016, Elsevier.
Coatings 13 00496 g001
Coatings 13 00496 g002 550
Figure 2. Influence of different laser beam diameters, gas supply speeds, and pre-coating thicknesses on cladding results. (a) Effect of different laser beam diameters on cladding results. Reprinted with permission from Ref. [64]. 2017, Elsevier. (b) The influence of different gas supply speeds on cladding results. Reprinted with permission from Ref. [65]. 2002, Elsevier. (c) Microhardness distribution along the depth direction of the coating under different pre-placed powder layer thicknesses. Reprinted with permission from Ref. [66]. 2015, Elsevier. (d) Variation in coating friction coefficient with sliding time for different pre-placed powder layer thicknesses. Reprinted with permission from Ref. [66]. 2015, Elsevier.
Figure 2. Influence of different laser beam diameters, gas supply speeds, and pre-coating thicknesses on cladding results. (a) Effect of different laser beam diameters on cladding results. Reprinted with permission from Ref. [64]. 2017, Elsevier. (b) The influence of different gas supply speeds on cladding results. Reprinted with permission from Ref. [65]. 2002, Elsevier. (c) Microhardness distribution along the depth direction of the coating under different pre-placed powder layer thicknesses. Reprinted with permission from Ref. [66]. 2015, Elsevier. (d) Variation in coating friction coefficient with sliding time for different pre-placed powder layer thicknesses. Reprinted with permission from Ref. [66]. 2015, Elsevier.
Coatings 13 00496 g002
Coatings 13 00496 g003 550
Figure 3. Processing map of laser cladding. (a) The processing map for coaxial laser cladding of NiCrAlY powder on Inconel 738 nickel-based alloys. Reprinted with permission from Ref. [71]. 2016, Elsevier. (b) The processing map for laser cladding of Inconel 625 powder on ASTM A592 steel. Reprinted with permission from Ref. [72]. 2020, Elsevier. (c) The processing map for laser cladding of WC-12Co powder on AISI 321 steel. Reprinted with permission from Ref. [73]. 2017, Elsevier. (d) The processing map for laser cladding of Ti-6Al-4V alloy on Ti-6Al-4V. Reprinted with permission from Ref. [74]. 2018, Elsevier.
Figure 3. Processing map of laser cladding. (a) The processing map for coaxial laser cladding of NiCrAlY powder on Inconel 738 nickel-based alloys. Reprinted with permission from Ref. [71]. 2016, Elsevier. (b) The processing map for laser cladding of Inconel 625 powder on ASTM A592 steel. Reprinted with permission from Ref. [72]. 2020, Elsevier. (c) The processing map for laser cladding of WC-12Co powder on AISI 321 steel. Reprinted with permission from Ref. [73]. 2017, Elsevier. (d) The processing map for laser cladding of Ti-6Al-4V alloy on Ti-6Al-4V. Reprinted with permission from Ref. [74]. 2018, Elsevier.
Coatings 13 00496 g003
Coatings 13 00496 g004 550
Figure 4. A grey relational graph of the process parameters, the influence of processing parameters on the signal-to-noise ratio of cladding, and the metallographic microstructure comparison of specimen cross-sections before and after optimization. (a) Effects of processing parameters on the S/N ratio of clad height. Reprinted with permission from Ref. [95]. 2016, Springer Nature. (b) Graph of grey relational grade. Reprinted with permission from Ref. [96]. 2013, Springer Nature. (c) Metallographic microstructure of the specimen cross-section under initial parameters. Reprinted with permission from Ref. [97]. 2022, Elsevier. (d) Metallographic microstructure of specimen cross-section under optimal parameters. Reprinted with permission from Ref. [97]. 2022, Elsevier. (e) Main effects plot for SNR of the dilution rate. Reprinted with permission from Ref. [98]. 2018, Elsevier. (f) Main effects plot for SNR of the clad width. Reprinted with permission from Ref. [98]. 2018, Elsevier.
Figure 4. A grey relational graph of the process parameters, the influence of processing parameters on the signal-to-noise ratio of cladding, and the metallographic microstructure comparison of specimen cross-sections before and after optimization. (a) Effects of processing parameters on the S/N ratio of clad height. Reprinted with permission from Ref. [95]. 2016, Springer Nature. (b) Graph of grey relational grade. Reprinted with permission from Ref. [96]. 2013, Springer Nature. (c) Metallographic microstructure of the specimen cross-section under initial parameters. Reprinted with permission from Ref. [97]. 2022, Elsevier. (d) Metallographic microstructure of specimen cross-section under optimal parameters. Reprinted with permission from Ref. [97]. 2022, Elsevier. (e) Main effects plot for SNR of the dilution rate. Reprinted with permission from Ref. [98]. 2018, Elsevier. (f) Main effects plot for SNR of the clad width. Reprinted with permission from Ref. [98]. 2018, Elsevier.
Coatings 13 00496 g004
Coatings 13 00496 g005 550
Figure 5. The microstructure at optimum process parameters and a comparison with the substrate microstructure. (a) Microstructure of the lower part of the cladding layer. Reprinted with permission from Ref. [99]. 2020, Elsevier. (b) Microstructure of the top of the cladding layer. Reprinted with permission from Ref. [99]. 2020, Elsevier. (c) 3D worn morphologies of substrate. Reprinted from [101]. (d) 3D worn morphologies of optimized clad layer [101]. (e) SEM morphologies of worn surfaces for HEA cladded SS-316 at optimum process parameters. Reprinted from [105]. (f) SEM morphologies of worn surfaces for untreated SS-316 [105].
Figure 5. The microstructure at optimum process parameters and a comparison with the substrate microstructure. (a) Microstructure of the lower part of the cladding layer. Reprinted with permission from Ref. [99]. 2020, Elsevier. (b) Microstructure of the top of the cladding layer. Reprinted with permission from Ref. [99]. 2020, Elsevier. (c) 3D worn morphologies of substrate. Reprinted from [101]. (d) 3D worn morphologies of optimized clad layer [101]. (e) SEM morphologies of worn surfaces for HEA cladded SS-316 at optimum process parameters. Reprinted from [105]. (f) SEM morphologies of worn surfaces for untreated SS-316 [105].
Coatings 13 00496 g005
Coatings 13 00496 g006 550
Figure 6. The BP neural network training process, the comparison between the BP neural network model and the RSM model prediction results, and the microstructure of AlCoCrFeNi HEA coatings under optimal process parameters. (a) The BP neural network training process. Reprinted from [118]. (b) Comparison of the prediction results of neural networks and RSM for clad height. Reprinted with permission from Ref. [117]. 2020, Elsevier. (c) Back-scattered SEM images of the AlCoCrFeNi coating microstructure under low magnification [118]. (d) Back-scattered SEM images of the AlCoCrFeNi coating microstructure under high magnification. At the region marked B in (d), the Fe element content was the highest, which was 30.40 at.%; and the Al element content was the lowest, which was 14.98 at.%. The black granular microstructure (marked A) had the highest content of Al and N elements, which were 38.41 at.% and 25.23 at.%, which indicates that the particles were the AlN phase. [118]. (e) EDS maps of AlCoCrFeNi coating [118].
Figure 6. The BP neural network training process, the comparison between the BP neural network model and the RSM model prediction results, and the microstructure of AlCoCrFeNi HEA coatings under optimal process parameters. (a) The BP neural network training process. Reprinted from [118]. (b) Comparison of the prediction results of neural networks and RSM for clad height. Reprinted with permission from Ref. [117]. 2020, Elsevier. (c) Back-scattered SEM images of the AlCoCrFeNi coating microstructure under low magnification [118]. (d) Back-scattered SEM images of the AlCoCrFeNi coating microstructure under high magnification. At the region marked B in (d), the Fe element content was the highest, which was 30.40 at.%; and the Al element content was the lowest, which was 14.98 at.%. The black granular microstructure (marked A) had the highest content of Al and N elements, which were 38.41 at.% and 25.23 at.%, which indicates that the particles were the AlN phase. [118]. (e) EDS maps of AlCoCrFeNi coating [118].
Coatings 13 00496 g006
Coatings 13 00496 g007 550
Figure 7. The GABP predictions and their comparison with the BPNN predictions. (a) Comparison of the prediction error of the triple-output-layer GABPNN with the single-output-layer GABPNN and BPNN for the clad width. Reprinted with permission from Ref. [126]. 2018, Springer Nature. (b) Comparison of the prediction error of the triple-output-layer GABPNN with the single-output-layer GABPNN and BPNN for the clad height. Reprinted with permission from Ref. [126]. 2018, Springer Nature. (c) Prediction results of BPNN [128]. (d) Prediction results of GABP [128].
Figure 7. The GABP predictions and their comparison with the BPNN predictions. (a) Comparison of the prediction error of the triple-output-layer GABPNN with the single-output-layer GABPNN and BPNN for the clad width. Reprinted with permission from Ref. [126]. 2018, Springer Nature. (b) Comparison of the prediction error of the triple-output-layer GABPNN with the single-output-layer GABPNN and BPNN for the clad height. Reprinted with permission from Ref. [126]. 2018, Springer Nature. (c) Prediction results of BPNN [128]. (d) Prediction results of GABP [128].
Coatings 13 00496 g007
Coatings 13 00496 g008 550
Figure 8. A prediction model framework for support vector machine, comparison of predicted and experimental values of SVR, and comparison of the predicted values of SVR and BPNN. (a) The prediction model framework of the support vector machine. Reprinted with permission from Ref. [134]. 2019, Elsevier. (b) Predicted results for microhardness and experimental values. Reprinted with permission from Ref. [134]. 2019, Elsevier. (c) Comparison of the predicted values of the clad width of SVR and BPNN [135]. (d) Comparison of the predicted values of the clad height of SVR and BPNN [135].
Figure 8. A prediction model framework for support vector machine, comparison of predicted and experimental values of SVR, and comparison of the predicted values of SVR and BPNN. (a) The prediction model framework of the support vector machine. Reprinted with permission from Ref. [134]. 2019, Elsevier. (b) Predicted results for microhardness and experimental values. Reprinted with permission from Ref. [134]. 2019, Elsevier. (c) Comparison of the predicted values of the clad width of SVR and BPNN [135]. (d) Comparison of the predicted values of the clad height of SVR and BPNN [135].
Coatings 13 00496 g008
Coatings 13 00496 g009 550
Figure 9. Performance comparison of a neural network model optimized by the particle swarm algorithm with a neural network model. (a) Comparison of experimental and ANN and PSO-ANN models’ predicted results for clad height. Reprinted with permission from Ref. [146]. 2020, Elsevier. (b) Relative error of ANN and PSO-ANN models for clad width. Reprinted with permission from Ref. [146]. 2020, Elsevier. (c) Performance comparison between the neural network and the PSOANN model. Reprinted with permission from Ref. [146]. 2020, Elsevier. (d) The iterative comparison of the coating microhardness fitness values under different optimization methods [147].
Figure 9. Performance comparison of a neural network model optimized by the particle swarm algorithm with a neural network model. (a) Comparison of experimental and ANN and PSO-ANN models’ predicted results for clad height. Reprinted with permission from Ref. [146]. 2020, Elsevier. (b) Relative error of ANN and PSO-ANN models for clad width. Reprinted with permission from Ref. [146]. 2020, Elsevier. (c) Performance comparison between the neural network and the PSOANN model. Reprinted with permission from Ref. [146]. 2020, Elsevier. (d) The iterative comparison of the coating microhardness fitness values under different optimization methods [147].
Coatings 13 00496 g009
Coatings 13 00496 g010 550
Figure 10. Microstructure of the cladding layer under optimal parameters, frictional wear properties, and corrosion resistance. (a) BSD image of the clad cross-section with optimal parameters. Reprinted with permission from Ref. [150]. 2021, Elsevier. (b) Microstructure of the cladding section with optimal parameters. 1 and 2 are showing spots for further EDS analysis. Reprinted with permission from Ref. [150]. 2021, Elsevier. (c) SEM image of the worn surface of a 15-5PH cladding layer under high magnification [151]. (d) SEM image of the worn surface of a 15-5PH cladding layer under low magnification [151]. (e) SEM image of 15-5PH cladding layer after electrochemical corrosion experiments [151]. (f) SEM image of 20Cr13 substrate after electrochemical corrosion experiments [151].
Figure 10. Microstructure of the cladding layer under optimal parameters, frictional wear properties, and corrosion resistance. (a) BSD image of the clad cross-section with optimal parameters. Reprinted with permission from Ref. [150]. 2021, Elsevier. (b) Microstructure of the cladding section with optimal parameters. 1 and 2 are showing spots for further EDS analysis. Reprinted with permission from Ref. [150]. 2021, Elsevier. (c) SEM image of the worn surface of a 15-5PH cladding layer under high magnification [151]. (d) SEM image of the worn surface of a 15-5PH cladding layer under low magnification [151]. (e) SEM image of 15-5PH cladding layer after electrochemical corrosion experiments [151]. (f) SEM image of 20Cr13 substrate after electrochemical corrosion experiments [151].
Coatings 13 00496 g010
Table
Table 1. A combination of process parameters that have an impact on cladding results.
Table 1. A combination of process parameters that have an impact on cladding results.
Authors
[Year/Reference]		SubstrateCladding MaterialClad HeightClad WidthClad DepthDilution Ratio
Ansari 2016 [71]Inconel 738 superalloyNiCrAlY powderP2V−3/2FP3/2V−1/3PV2/3F−2/3VF−1
Shayanfar 2020 [72]ASTM A592 steel625 powderP1/2V−2F2P1/2V−1/5P3V−1/2 F3/4P1/2V2F−3/4
Erfanmanesh 2017 [73]AISI 321 steelWC-12Co powderP2V−2F1/4PV−1/2P2V1/4F−1/4P1/2V2F−1
Nabhani 2017 [74]Ti-6Al-4VTi-6Al-4V powderPV−1F1/4PV−1/3PVF−1/8VF−1/2
P is laser power, V is scanning speed, and F is powder feeding rate.
Table
Table 2. Optimization variables and response indexes used in the response surface methodology.
Table 2. Optimization variables and response indexes used in the response surface methodology.
Authors
[Year/Reference]		SubstrateCladding MaterialResponse IndexesOptimization VariablesOptimal Process Parameters
Lujun Cui
2021 [82]		ZG310-570 (ZG45)Co-Cr-W alloy powerAspect ratio, dilution rate, clad width, clad height, clad depthLaser power 1400 W~1700 W
Powder feeding rate15 g/min~20 g/min
Scanning speed5 mm/s~6 mm/s
Tiankai Li
2022 [83]		45 steelNi60PTA alloy powderDilution rate, ratio of layer width to height,
contact angle		Laser power1477 W
Powder feeding rate17.5 mg/s
Scanning speed5 mm/s
Zupeng Wu
2019 [84]		45 steelNi60A alloy powerPorosity areaLaser power1524.8 W
Powder feeding rate5.20 g/min
Scanning speed6.72 mm/s
Ali Khorram
2019 [85]		Inconel 718 superalloy75Cr3C2 + 25(80Ni20Cr) powderClad width, clad height, clad angleLaser frequency20 Hz
Pulse width12.9 ms
Scanning speed5.43 mm/s
Guofu Lian
2018 [86]		AISI/SAE 1045 steelW6Mo5Cr4V2 powderMulti-track clad width, flatness ratio, dilution rateLaser power 1.5 kW
Scanning speed6 mm/s
Gas flow1018.81 L/h
Overlap rate23.47%
Sha Wu
2021 [87]		42CrMo alloyNi60A-25% WC powderDilution rate, unit effective areaLaser power2799.93 W
Scanning speed236.84 mm/min
Powder feeding rate5 g/min
Spot diameter3 mm
Table
Table 3. The optimization methods and evaluation indexes.
Table 3. The optimization methods and evaluation indexes.
Authors
[Year/Reference]		Optimization MethodSubstrateCladding MaterialEvaluation IndexesProcess ParametersOptimal Process Parameters
Pengfei Fan
2020 [99]		Orthogonal experiments15 M nNi4M o steelCo50 powder and WC powderClad depth, clad width, clad height, dilution rate, hardnessLaser power2.4 kW
Powder feeding rate0.5 g/s
Scanning speed7 mm/s
Javad Marzban
2015 [100]		PCA with TOPSISAISI 1040Ni-Cr-Mo powdersClad height, clad width, clad depthLaser power1 kw
Powder feeding rate8 mg/min
Scanning speed0.5 m/min
Qianting Wang
2020 [101]		PCA with GRAAISI 1045Fe50 powder and TiC powderClad width, flatness, non-fusion areaLaser power1.77 KW
Power ratio35.28%
Overlapping ratio24.06%
Defocus amount−0.44 mm
Wanxu Liang
2021 [102]		FCE with IAHP45 steel316 L stainless steel powderCoating profile, microstructure, mechanical propertiesLaser power≤1200 W
Scanning speed5~7 mm/s
Overlap rate30~40%
L. Reddy
2018 [103]		Theoretical–empirical model15Mo3SHS 7170 powderPowder deposition efficiency, dilution, porosityLaser power1000 W
Powder feeding rate4 g/min
Scanning speed300 mm/min
Jyoti Menghani
2021 [105]		Desirability function approachAISI 316AlFeCuCrCoNi high-entropy powderClad height, clad depth, clad width, percentage dilutionLaser power1.1 kW
Powder feeding rate4 g/min
Scanning speed500 mm/min
Mahmoud Moradi
2021 [109]		Design Expert statistical software4130
alloy steel		Inconel 718 powderClad height, clad width, standard deviation of microhardness, the stability of additively manufactured wallsScanning speed2.5 mm/s
Powder feeding rate28.52 g/min
Scanning strategies (unidirectional, bidirectional)Unidirectional
Table
Table 4. Neural network prediction objects and errors.
Table 4. Neural network prediction objects and errors.
Authors [Year/Reference]SubstrateCladding MaterialOptimization MethodPrediction Error
Changhui Song
2020 [117]		316 L stainless steel316 L stainless steel powderBPNNClad wightClad height
2.79%3.09%
Yutao Li
2021 [118]		40CrNiMo alloy steelAlCoCrFeNi high-entropy alloy powderBPNNDilution rate
5.89%
Fabrizia Caiazzo
2018 [119]		2024 aluminum alloy2024 aluminum alloy powderANNLaser powerScanning speedPowder feeding rate
2.0%5.8%5.5%
Table
Table 5. Comparison of support vector regression and BP neural network prediction results.
Table 5. Comparison of support vector regression and BP neural network prediction results.
Authors [Year/Reference]SubstrateCladding MaterialOptimization MethodPrediction Error
Xiyi Chen
2021 [135]		316 L stainless steel316 L stainless steel powderBPNNClad widthClad height
15%6%
M-SVRClad widthClad height
5%5%
Yao Wang
2020 [136]		316 L stainless steelFe powderBPNNClad widthClad height
6.72%7.96%
RBF-SVRClad widthClad height
4.58%5.33%
Table
Table 6. Comparison of response values before and after NSGA-II optimization.
Table 6. Comparison of response values before and after NSGA-II optimization.
Authors [Year/Reference]SubstrateCladding MaterialResponse Values before and after Optimization
Xingyu Jiang
2022 [140]		45 steel plate304 L powderResponseEnergy consumption (J)Powder utilizationMicrohardness (HV)Aspect ratio
Before2,972,340.40543%2213.6
After1,798,861.4346%2354.5
Linsen Shu
2022 [141]		TC4 plateTC4 alloy powderResponseWear depth (µm)Wear width (µm)Microhardness (HV)Average mean friction coefficient
Before74.542459.64429.50.374
After56.641884.79473.30.293
Zhao Kai
2020 [142]		20 steelInconel 625 powderResponseEfficiency (mm2·s−1)Heat-affected zone depthMicrohardness (HV)Dilution
Before15.240.855186.4330.518
After16.170.736218.3370.32
Table
Table 7. The prediction results of ANFIS, RF optimization methods, and GWO optimization of BPNN.
Table 7. The prediction results of ANFIS, RF optimization methods, and GWO optimization of BPNN.
Authors [Year/Reference]SubstrateCladding MaterialOptimization MethodPredicted Error
Zhijie Zhou
2022 [151]		20Cr13 stainless steel15-5PH powderBPNNClad height
MSE (10–3)		Clad weight MSE (10–3)Dilution
MSE (10–3)
1.0530.6424.969
GWO-BPNNClad height
MSE (10–3)		Clad weight MSE (10–3)Dilution
MSE (10–3)
0.1610.7150.267
Hamed Sohrabpoor
2016 [152]		A36 mild steelFe-based alloy powderANFISPowder catchment efficiencyClad heightClad width
7.62%8.36%−3.83%
Liang Xudong
2020 [153]		Stainless steelInconel 625 powderRFLaser powerScanning speedPowder feeding rate
1.17%3.43%3.51%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, K.; Liu, W.; Hong, Y.; Sohan, H.M.S.; Tong, Y.; Hu, Y.; Zhang, M.; Zhang, J.; Xiang, D.; Fu, H.; et al. An Overview of Technological Parameter Optimization in the Case of Laser Cladding. Coatings 2023, 13, 496. https://doi.org/10.3390/coatings13030496

AMA Style

Wang K, Liu W, Hong Y, Sohan HMS, Tong Y, Hu Y, Zhang M, Zhang J, Xiang D, Fu H, et al. An Overview of Technological Parameter Optimization in the Case of Laser Cladding. Coatings. 2023; 13(3):496. https://doi.org/10.3390/coatings13030496

Chicago/Turabian Style

Wang, Kaiming, Wei Liu, Yuxiang Hong, H. M. Shakhawat Sohan, Yonggang Tong, Yongle Hu, Mingjun Zhang, Jian Zhang, Dingding Xiang, Hanguang Fu, and et al. 2023. "An Overview of Technological Parameter Optimization in the Case of Laser Cladding" Coatings 13, no. 3: 496. https://doi.org/10.3390/coatings13030496

APA Style

Wang, K., Liu, W., Hong, Y., Sohan, H. M. S., Tong, Y., Hu, Y., Zhang, M., Zhang, J., Xiang, D., Fu, H., & Ju, J. (2023). An Overview of Technological Parameter Optimization in the Case of Laser Cladding. Coatings, 13(3), 496. https://doi.org/10.3390/coatings13030496

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop