1. Introduction
With the accelerated process of modernization and construction, there is an increasing demand for various large-scale engineering construction tools [
1]. In the construction process, which is affected by the complexity and variability of working conditions, frequent load changes, and other factors, construction machinery is often in a high-load operation state [
2,
3]. Therefore, it is necessary to ensure that the construction machinery is well lubricated and is reliable to ensure its regular operation [
4]. In recent years, with the development of artificial intelligence, the internet, and other technologies, the manufacturing industry has been gradually changing in the direction of intelligence [
5,
6]. Construction machinery is essential to modern manufacturing, and its intelligent development is also attracting attention [
7]. In the use of construction machinery, the lubrication system is one of its core components. Before the advent of intelligent lubrication, the lubrication field used three traditional lubricants, namely, single-wire progressive lubrication, double-wire and multi-wire lubrication, and double-wire and multi-wire hybrid lubrication. The traditional lubrication systems for construction machinery, with their increasingly obvious shortcomings, have not been able to meet the lubrication needs of modern construction machinery [
8,
9]. Therefore, intelligent lubrication systems came into being, which combine hydraulic technology, sensor technology, numerical control technology, other technologies, a high degree of intelligence in the control system, monitoring, and detection, and strong human–computer interaction, and they can realize multi-point simultaneous lubrication and can control the amount of oil supplied [
10,
11,
12,
13].
Intelligent lubrication systems are further developments in lubrication system intelligence and information technology, and the features of smart lubrication system mainly include the following: all lubrication points are maintained in a completely isolated state without any interference from external factors, and even if one lubrication point is blocked, the other lubrication points are not affected. Each lubrication point independently detects the grease displacement and can immediately reflect the lubrication point failure to the control cabinet. The independent detection of the grease displacement at each lubrication point can relay the lubrication point failure immediately to the control cabinet and directly to the central controller. The real-time monitoring of the lubricant supply, better control of the lubricant spraying volume, and the prevention of lubricant waste can be achieved. With the visualization of the monitoring interface, the user can adjust the cycle time and lubricant supply volume at each lubrication point according to the demand, and the same lubrication system can be adjusted on a large scale. However, in the use of construction machinery, friction-versus-material characteristics, environmental temperature changes, load conditions, and other factors will have an impact on the performance of the intelligent lubrication system [
14,
15], so it is crucial to evaluate the performance of the intelligent lubrication system, not only to repair the system in advance of potential failures but also to greatly improve the lubrication effect.
Both domestic and foreign research have been carried out on intelligent lubrication systems. For example, as early as 1997, Griffith et al. developed an automatic lubrication system for mining machinery that could detect the system function signal and automatically perform the lubrication system function, improving the efficiency of the mechanical equipment [
16]. The centralized lubrication system developed by Germany’s Fukuo Bird Vogel Company significantly improves upon the inadequacy of traditional manual lubrication, saving on lubricant by up to 80% and providing a more convenient service for users [
17]. In 2014, Wang [
18], of the Nanjing University of Aeronautics and Astronautics, researched and designed a multi-point quantitative lubrication pump structure, which increased the output performance of the lubrication systems for distributed construction machinery. In 2018, Wu [
19], of the Overseas Chinese University, proposed the method of working attitude detection for the automatic lubrication systems of excavators, which constitutes a new idea for construction machinery. The working lubrication strategy provides a unique idea. In 2019, Zhou [
20] proposed an intelligent lubrication optimization model to solve the problems encountered when there are several port equipment lubrication points that are not easy to maintain and repair. Based on this, he designed an intelligent lubrication equipment health management system, which realizes the on-demand and precise lubrication of port equipment. In 2021, Liu [
21], from the Central Plains Institute of Technology (CPIT), by studying the structure of lubrication systems, lubrication control systems, and end detection for research, designed a set of intelligent lubrication systems to realize the lubrication of construction machinery, which provides a new direction for the design of intelligent lubrication systems for construction machinery. In the same year, Lu et al. [
22] designed the software part of an intelligent track lubrication system using fuzzy group analysis based on the results of a curve radius and train speed simulation. In 2022, You Yintao [
23] and others, in order to solve the problems of roller press lubrication systems using traditional lubrication control methods, proposed an intelligent roller press bearing lubrication system based on PLC and remote control. The system can provide real-time remote monitoring and remote control, and it can detect the lubrication amount and operation status of on-line equipment, which solves the problems of traditional lubrication systems. In 2023, Chai Xiaohui [
24] and others proposed a PLC and remote control-based intelligent system for roller press bearing lubrication to solve the problem of using traditional lubrication control methods for roller press lubrication systems. The system is capable of real-time remote monitoring and remote control. It can detect the amount of lubrication and the operational status of the device online, solving the problems of conventional lubrication systems. In the same year, Zheng et al. [
25] reviewed the current research status of intelligent technology in construction machinery. They analyzed the current research status of Prognostic Health Management (PHM), environment sensing, and automation control in construction machinery, which indicates that construction machinery has been increasingly converging to intelligence. In the same year, Shi et al. [
26] developed a THED lubrication model for coupled bearings in diesel engines based on thermal pressure coupling effects. They experimentally verified that the lubrication characteristic parameter and the dynamical characteristic parameter of the bearing associated with the journal thrust increase significantly when thermal pressure coupling effects are taken into account.
In terms of evaluation methods, traditional evaluation methods mostly use the single-assignment method. For example, in 2020, Luo et al. [
27] constructed a model based on fuzzy synthesis evaluation to assess the construction risk of soft soil modified shield machines. It weights risk factors by triangular fuzzy numbers, which helps avoid subjectivity and better reflects the fuzzy nature of risk factors without the need for consistency tests as in hierarchical analysis methods. In 2021, C.X. et al. [
28] used the hierarchical analysis method to design the pumping station from the technical and economic aspects, but the hierarchical analysis method is more subjective and is better suited to evaluate subjective indicators and lacks objectivity. However, as evaluation methods become more mature and individual evaluation methods become susceptible to competency or objective influences, comprehensive evaluation methods are born in response. In 2019, Zhao et al. [
29] performed a comprehensive evaluation of AC/DC hybrid microgrid planning based on hierarchical analysis and the entropy weighting method. The effectiveness and rationality of the evaluation methods are also demonstrated through examples. The combination of hierarchical analysis method and entropy weight method for evaluation not only improved the objectivity of index weights, but also reduced the errors caused by insufficiency or omission so that the evaluation results are scientific and realistic. In 2022, Xie et al. [
30] developed an evaluation model based on fuzzy hierarchical analysis and hierarchical analysis to evaluate the adaptability of shield machines to address the difficulty of efficiently and quantitatively assessing the effectiveness of shield machines during construction. In 2023, Xue et al. [
31] constructed an energy efficiency assessment model for substation buildings using AHP and fuzzy synthesis theory. They chose a substation in Shandong, China as a case study to validate the effectiveness and feasibility of the model. However, the model is only applicable to substation buildings and cannot be generalized to other buildings.
Intelligent lubrication systems can monitor lubrication status in real-time, detect lubrication problems in a timely manner, and respond to them, thereby improving lubrication efficiency, prolonging device life, and reducing maintenance costs. [
32,
33]. While domestic and foreign research on the design of intelligent lubrication systems and their application to various mechanical devices is becoming more mature, research on the performance evaluation techniques of intelligent lubrication systems for construction machinery is also crucial. Until now, the study of evaluation techniques has been relatively decentralized, with research areas limited to one aspect of the lubrication system and few comprehensive evaluations of multiple aspects of the entire lubrication system. In this paper, we propose a multi-objective-based performance evaluation technique for intelligent lubrication systems in construction machinery. Its specific research content is as follows: First, we analyze the factors that affect the regular operation of intelligent lubrication systems of construction machinery, select suitable evaluation metrics, and develop a system of performance evaluation metrics for intelligent lubrication systems of construction machinery. We then develop a multi-objective performance evaluation model for intelligent lubrication systems of construction machinery based on the AHP–entropy weight method and a topological object model. Finally, the established evaluation model is used to evaluate intelligent lubrication systems with examples, and the evaluation results are analyzed to suggest improvements and measures.
3. Performance Evaluation Model of Intelligent Lubrication System for Construction Machinery
The performance evaluation of intelligent lubrication system of construction equipment can help enterprises discover the lubrication failure of equipment in time. In addition, corresponding maintenance measures can be taken to improve equipment reliability and stability. The selection of indicators should be reasonable and objective and in line with reality, so as to be conducive to obtaining accurate evaluation results. The weight allocation of different indicators will have an impact on the design, optimization, monitoring, fault diagnosis, degree of intelligence, and ease of use of the intelligent lubrication system of construction machinery, which will affect the performance evaluation results of the system. For the construction machinery intelligent lubrication system, performance weighting needs to be in line with the actual construction machinery, but also objective and fair, in line with the intelligent lubrication system in previous years, as well as the operation and maintenance of the characteristics.
This paper aims to provide a practical and feasible performance evaluation method for assessing the operation of intelligent lubrication systems for construction machinery. The construction machinery intelligent lubrication system needs to work properly under the conditions of harsh working environment, high load working condition, a long time operation, variable working conditions, and remote monitoring and control. The system involves multiple factors, and the factors are highly interrelated. The hierarchical analysis method can positively reflect the relationship between the factors within the Construction machinery intelligent lubrication system, but also serves to simplify the evaluation problem. In addition, most of the indicators are qualitative, making it difficult to measure them using quantitative indicators [
41]. The entropy weighting method relies on raw data to calculate weights, which are highly objective and accurate [
42]. The construction machinery intelligent lubrication system has objective operational data from previous years. For the multi-indicator factors, non-linear problems, the evaluation needs to be from multiple perspectives, multi-faceted, and combined with multiple factors to objectively give a realistic conclusion. Therefore, the comprehensive evaluation method combining AHP and the entropy weight method is chosen to determine the evaluation index weights, and the specific steps are as follows.
3.1. Determination of Indicator Weights
3.1.1. AHP Determines Subjective Weights
Hierarchical analysis was first proposed by Saaty, an American operations researcher in the 1970s, and it is a combination of qualitative and quantitative analysis methods [
43]. Its basic thinking is: firstly, based on the evaluation objectives, decompose the objectives into various constituent elements, then according to the degree of correlation and affiliation between elements, construct an evaluation hierarchy from high to low, then compare the factors of each layer, so as to construct a judgment matrix, and then, through the calculation of the hierarchical single ranking and consistency test, calculate the weights of the elements of this layer to the relative importance of the previous layer, and finally carry out a hierarchical total ranking and consistency test. Finally, calculate weights of factors in this layer in terms of their relative importance to the previous layer, and finally carry out total hierarchical ordering and the consistency test [
44]. The basic steps of AHP are shown in
Figure 1 below.
3.1.2. Entropy Weighting Method to Determine Objective Weights
The entropy weighting method is used to determine objective weights. Entropy is a measure of the degree of disorder in a system [
45]. The entropy weight method is an objective assignment method used to determine the degree of dispersion of a certain indicator through the entropy value. The smaller the value of the entropy information, the greater the degree of dispersion of the indicator, and the greater the corresponding weight. This method can effectively reflect the distribution of data and accurately assess the weight of indicators [
46]. The entropy weighting method processes the raw data collected and constructs them into an evaluation matrix, which can fully utilize the information contained in the indicators. This can then be used to analyze the variability between indicators by finding entropy values and coefficients of variance. The basic steps of the entropy weighting method are shown in
Figure 2 below.
3.1.3. AHP–Entropy Weighting Method to Determine the Combined Weights
The AHP–entropy weighting method to determine the combined weights; AHP relies on the experience of experts and the results obtained are limited by personal preferences, ignoring objective factors, and are prone to incorrectly exaggerate or minimize the impact of certain factors [
47]. The entropy weight method is supported by data and has strong objectivity. However, it is easy to ignore the correlation between indicators and focus too much on theoretical derivation to obtain results according to local conditions [
48], considering the diversity and complexity of the indicators of the construction machinery intelligent lubrication system and the advantages and disadvantages of the AHP–entropy weight method. Choose to combine the subjective weights obtained by AHP and the objective weights obtained by entropy weight method. The weights of the indicators of the construction machinery intelligent lubrication system are calculated using the comprehensive assignment method. The general formula for the AHP–entropy weighting method to calculate composite weights is:
Combined with the actual situation, the formula is improved by reviewing relevant information, and the formula for the integrated weight after improvement is:
3.2. Construction of Extension Matter–Element Model
Topology is a field of mathematics and statistics that describes and analyzes multidimensional data using extension models [
49], which can help us understand the characteristics and patterns of objective things more comprehensively. Conversely, the matter–element model is an evaluation model that combines qualitative and quantitative elements [
50]. Combining topology and the matter–element model to establish the matter–element extension model for evaluation of the construction machinery intelligent lubrication system. The specific computational procedure for the matter–element extension model is as follows.
3.2.1. Establishment of Object Element Model
An ordered triad
R = (
N,
C,
U) is introduced as the basic unit of the descriptor, and the
R object element possesses n-dimensional features.
3.2.2. Determine the Classical and Sectional Domains
The classical domain is a matter–element matrix consisting of each evaluation level, evaluation indicator and indicator value interval together, denoted as
R0.
Nok represents the kth rank, C1, … Cn represent the 1st to nth indicators, respectively, and Uoi = (aoi, boi) (i = 1, 2, … n) represents the range of quantitative values taken by each rank Nok to which each indicator Ci belongs.
The nodal domain is an interval of values about
C,
, Denoted as
Rp.
N denotes all evaluation levels.
3.2.3. Determination of the Matrix of Elements to Be Evaluated
According to the identified evaluation indicators, the quantitative values are processed and analyzed, and the results obtained are used to form the element-to-be-evaluated matrix, denoted as
Rq.
Uqi (
i = 1, 2, …
n) denotes the data obtained from the thing to be tested.
3.2.4. Determine the Correlation Function and the Degree of Correlation
3.2.5. Calculate the Overall Relevance and Evaluation Rating
The correlation of each level of the integrated evaluation is the weighted summary of the correlation function of each indicator. The expression is as follows:
wi is the indicator weight of Xi, indicating that the evaluation unit Ro belongs to evaluation level k, the level with the highest degree of relevance.
In summary, the comprehensive evaluation method combining the AHP–entropy weight method and topologizable object element model takes multiple factors into account, avoids the limitations of a single evaluation method, and improves the comprehensiveness of analysis and evaluation. The flow of the performance evaluation of the intelligent lubrication system for construction machinery is shown in
Figure 3 below.
5. Summary and Outlook
Construction machinery often operates under high loads and harsh environments, which place high demands on lubrication systems. Construction machinery requires regular maintenance and service, and lubrication is an important part of this; however, frequent maintenance and service work can increase operating costs and time costs. Moreover, the construction industry is increasingly demanding improved energy efficiency and environmental protection, and lubrication systems need to be continuously optimized in order to reduce energy consumption and environmental impact. Intelligent lubrication systems can accurately apply lubricants according to the working load and working conditions of the device, avoiding excessive or insufficient lubrication and achieving energy-efficient optimization. This can also enable remote monitoring and management of the lubrication system of construction machinery through Internet technology to improve maintenance efficiency and operation management. In addition, it can collect, store, and analyze data on lubrication work, provide predictive maintenance based on data analysis, and reduce unplanned downtime and maintenance costs. By evaluating the performance of intelligent lubrication systems, it is possible to continuously improve and optimize the lubrication system to increase the performance and competitiveness of the device. The following conclusions can be obtained by combining the findings of the whole paper:
By analyzing intelligent lubrication systems of construction machinery and screening the indicators, we have constructed a performance evaluation system applicable to smart lubrication systems of construction machinery. The system comprehensively describes the characteristics of an intelligent lubrication system for construction machinery while reducing the interference of irrelevant indicators, which can be applied effectively in practice.
Performance evaluation of intelligent lubrication systems for construction machinery using the AHP–entropy weight method and topological matter–element model. We have developed a model suitable for the performance evaluation of intelligent lubrication systems for construction machinery. This model combines subjective and objective aspects and can obtain more accurate evaluation results, which is promising for applications.
Based on the results obtained from the evaluation model, we analyze and conclude that the operating environment temperature, lubricant viscosity and failure rate have a large impact on the performance of the intelligent lubrication system for off-highway mining dump trucks, and suggest improvements. The model was reasonable, scientific, and adequate through example verification.
There are some differences in the working environment, structure, size, and operation of different construction machinery. In this paper, we have developed a performance evaluation system for intelligent lubrication systems for construction machinery, but how to select suitable evaluation metrics or modify the evaluation metrics for different construction machinery is a problem that needs to be considered. In addition, the established evaluation model is only applicable to construction machinery and not to other types of machinery, which brings certain limitations. The working environment of construction machinery is generally harsh, and the data required for the evaluation using the model are susceptible to objective conditions; thus, how to ensure the correctness of the data is also a consideration.