The Key Role of Laser Ultrasonics in the Context of Sustainable Production in an I 4.0 Value Chain

Abstract

The advancement of laser ultrasonics has increased rapidly in recent years, providing applications for materials characterization as well as for industrial utilization, as a quality control device. The wide-ranging capabilities for high-temperature in-situ analysis of a variety of microstructural characteristics offers a multitude of possibilities for usage in R&D. To date, this is the only known method that has been successfully deployed for in-situ materials characterization, as well as in the harsh environment of the metalworking industry. Combined with the enablers, introduced by the fourth industrial revolution, and the conjunction of a laser ultrasonic system with a Smart Production Lab, it has great potential to contribute to lower rejection rates, better recyclability, and consequently to a more sustainable production. In this review, the potential for systemic sustainability is explained throughout a part of the value chain, in the context of Industry 4.0. In addition, the integration of the methodology into a miniaturized Smart Production Lab is demonstrated, with the intention of incorporating it as a substantial part of the creation of a digital twin. Such a lab is designed to serve as an interface between laboratory and industry, in order to reveal the possibilities of digital transformation, Industry 4.0, and the application of highly flexible systems such as the laser-ultrasonic system for companies.

1. Introduction

The onset of the fourth industrial revolution has led to substantive changes in many manufacturing sectors. For metal processing facilities in particular, the enablers, occurring in conjunction with the digital transformation, provide enormous opportunities to enhance the quality of products, while simultaneously improving the sustainability of production processes [1,2,3,4]. The increasing digitization and automation of manufacturing processes empowers a large amount of process data to be generated, while these can be increasingly processed by Artificial Intelligence (AI) and Machine Learning (ML) supported algorithms, and thus contribute to an extensive competitive advantage [5,6]. One aspect that offers great potential in this context is Quality Control (QC), which is one of the most important stages within a manufacturing route [7,8,9,10,11,12]. Especially for suppliers of the aerospace industry, which often require 100% component inspection, is a laborious procedure that is difficult to maintain. In the majority of cases, component inspection is solely performed at the end of the manufacturing process, before delivery. In the event of a quality defect in the product, it has to pass through the entire manufacturing process before being sorted out, which leads to high expenses, especially for cost-intensive materials and complex manufacturing routes, that are frequently used in this industry branch. Therefore, especially in these sectors, there are efforts to adapt the possibilities of the new era, and thereby optimize economical production.

For instance, process-integrated real-time monitoring, which automatically screens the process data for discrepancies, would indicate production divergences at an early stage, and forward them to a superordinate system. Additional in-line Quality Inspection Systems, investigating the component for deviations (e.g., geometry, cracks, voids, etc.) between or during the processing steps would allow the defects to be corrected in further process steps. A significant innovation therefore, would be the implementation of a Digital Model (DM), a Digital Shadow (DS), or even a Digital Twin (DT) [13,14]. The latter one is able to detect deviations in real time, on the basis of simultaneously performed calculations, usually with the usage of finite element analysis (FEA), proposing appropriate measures for subsequent process routes, to eliminate the error or to remove the defective component [15].

In the aerospace industry, in addition to non-destructive testing of any safety-relevant component, a simulation of the manufacturing process with an integrated microstructure model is often required in order to predict the resulting microstructure prevailing in a component in advance, in order to be able to estimate mechanical properties [16,17,18,19]. For the purpose of creating a valid microstructure model of a material, an extensive amount of material parameters is required to be able to describe the essential microstructural changes during thermal and thermomechanical treatments [20,21,22]. Traditionally, experiments are carried out in a thermomechanical treatment simulator (TMTS) to determine the material parameters for static, dynamic, and metadynamic recrystallization (RX) and recovery phenomena, as well as for primary and secondary grain growth at different temperatures, strains and strain rates for different time steps. The analysis is subsequently carried out by an ex-situ analysis of the microstructural evolution i.e., by optical evaluation methods [23,24,25]. Accordingly, the generation of a set of material parameters for a valid microstructure model takes an average of three years.

As this is the most common development scenario, it is abundantly evident that it is time-, material-, and resource-intensive. Obviously, this is contrary to the actual development plan, both governmental and self-imposed by the industry, which is to emphasize sustainability and responsible CO₂-saving production [26,27,28]. Therefore, alternative methods to detect microstructural changes in real-time, in-situ, and also at high temperatures, under the respective manufacturing conditions, are intensively investigated. Some of the methods developed so far, coupled with the enablers of I 4.0, have the potential to become very powerful, reliable QC systems that point the way to the future. If these techniques can be integrated into the production process while preserving human resources, these methods represent a path to a sustainable future and a revolution in the metalworking industry.

For both of the above-mentioned scenarios, either in-line QC as well as for R&D purposes, in-situ material testing methods can provide an essential facilitation, cost reduction, and time saving. The focus in this context lies on the laser ultrasonic (LUS) testing method, which can be implemented in-line in the manufacturing process, as well as for the determination of material parameters, when mounted on a TMTS. The main advantage of this method is the contactless introduction of ultrasonic waves, providing information about material properties or product characteristics, depending on attenuation and velocity of sound. In addition, this system can be used at high temperatures and is virtually insensitive to a production facility’s harsh environment. Due to the flexible range of applications of the system, it has the potential to provide enormous improvements along the value chain. When integrated into a Smart Production Lab (SPL), such a system has the potential to exploit numerous possibilities for quality optimization and control. On the one hand, a multitude of microstructure data can be acquired rapidly when it is coupled to a TMTS. This material data in an automated workflow is capable of generating the required material parameters and integrating them into the microstructure model, which is used for the simulation of the manufacturing process. On this basis, a DM, DS, or even a DT can be created and incorporated into the value chain. It then manages the task of real-time monitoring of the processes of component manufacturing, on the basis of real-time process parameters and a FEA evaluation, operating in the background.

In the following sections, the most important enablers of Industry 4.0 (I 4.0) are outlined, in addition to their potential integration in a value chain, to ensure quality control. Furthermore, other in-situ quality control systems will be explained, both inline and for R&D purposes. The emphasis here lies on the LUS system, describing its setup and operating principle, and the possibilities for in-line as well as for DT integration in a SPL.

2. Enablers of the Fourth Industrial Revolution

The digital transformation induced by the advancement of I 4.0 has led to the establishment of new enabler technologies in the manufacturing industry [29,30]: (I) Industrial Internet of Things (IIoT); (II) Cyber Physical Production System (CPPS); (III) Cloud Computing (CC); (IV) DM, DS, and DT; (V) AI and ML; and (VI) Big Data. Due to the interconnected nature of I 4.0, the fundamental requirement for the successful and sustainable implementation of I 4.0 technologies requires an appropriate Information and Communication Technology (ICT) infrastructure [29]. The IIoT (I) interconnects physical entities through the internet, enabling real-time data exchange [30]. Generally, a CPPS (II) refers to a system composed of interconnected physical and virtual components. A more detailed definition of CPPSs, according to [31,32,33], concludes five characteristics, defining them as (i) superordinate systems within systems; (ii) consisting of cooperative components, capable of adjusting data transfer between multiple production environment layers; (iii) acting situationally appropriate and supporting decision-making processes; (iv) providing Human Machine Interfaces (HMIs); and (v) showcasing resilient design. CC (III) provides on-demand computing resources separated from the production site by the CC providers [34]. Hereby, different CC models provide different services, from data storage and processing, up to AI and ML [29]. The DM, DS, and DT (IV) describe digital depictions of physical entities, however differ depending on the automation of data transfer between digital and physical entities [35,36]. Here, the DM has no automated data transfer between both entities, whereas the DS features unilateral and the DT features bilateral data transfer [36]. AI (V) attempts to replicate human intelligence with suitable algorithms, enabling autonomous decision-making, and thus situationally appropriate acting. As a sub-category of AI, ML (V) aims to develop learning data-driven computational algorithms with the goal of linking data, recognizing correlations, drawing conclusions, and optimization [29,37,38,39]. Big Data (VI) refers to large volumes of heterogeneously structured data being gathered, analyzed, and disturbed at high speeds in order to extract valuable and trustworthy information [40,41]. These characteristics are summarized in the 5Vs of Big Data: Volume, Velocity, Variety, Veracity, and Value [42].

In order to fully profit from the benefits of I 4.0, the interconnection of value chain participants is imperative, enabling the optimization of products and processes along the entire product lifecycle, thus reducing waste along the value chain and improving the sustainability of production [29,43]. Per definition, a value chain includes all value creating steps in order to create a final product [44]. As for I 4.0 enabler technologies, suitable ICT infrastructure serves as a fundamental requirement for the networking of enterprises. Henceforth, transparency concerning data governance has to be emphasized, allowing transparent data sharing between enterprises in a value chain [45]. As a result of the transparent data sharing, the analysis of product and process data can be used to find correlations and draw conclusions, leading to the optimization of both [43]. Furthermore, the I 4.0 enabled omnipresent real-time monitoring leads to a drastic change of traditional QC methods [43,46]. Whereby traditional QC uses statistical methods to project the quality of a product sample, onto the entire production population, the new omnipresent data-driven QC allows the evaluation of each individual product, also incorporating advanced technologies, such as AI and ML [43,46,47,48]. Therefore, the implementation of I 4.0 based QC methods in a state-of-the-art DT value chain using technologies, such as Big Data, AI and ML, can result in a significant increase of flexibility, productivity, and sustainability of products and processes along the entire value chain and lifecycle of the product [46,49,50,51].

3. Quality Control Using In-Situ Methods

For in-line inspection, which in this context is regarded as QC in a manufacturing process, during or between the processing stages, there is a limited number of usable methods even today. While there are some established measuring systems for the quality inspection at the end of the component manufacturing route, which are also partially automated and connected in series, such as conventional ultrasonic technology, eddy current testing, etc., it is essential to find systems that include other restrictions.

The overriding goal of such systems is to observe the quality of a feedstock over the entire value chain until the finished component, in order to be able to react immediately in the event of any deviations. For this purpose, the targeted system requires complete networking and background calculations that can respond in real time, based on real process data. Computations of this kind can then decide, depending on the severity of the deviation, whether the existent deficiencies can be rectified by adjusting the production route, or whether it is rejected. Successful implementation of such systems can save a company an enormous amount of money, if a defective component is sorted out at an early stage. This does not only contribute to cost efficiency, but also to energy and resource efficiency, and thus leads to an enormous contribution to the sustainability of a company.

The criteria that a system has to fulfill in order to be a fundamental contributor are:

• Possibility of integration
• High measurement frequency
• Near real time data processing
• Robustness
• Reliability
• Innoxiousness
• Affordability
• Long service life

According to these criteria, most of the methodologies used for the final inspection are not suitable for an in-situ implementation, mainly due to the possibility of implementation, in combination with the low data acquisition frequency.

It is quite common for sensors to be installed in production facilities to monitor and control various process parameters in-situ, in a variety of ways. A decisive parameter in this case is the usually unavoidable temperature measurement, whether in the furnace, during the forming process, at the run-out table, or during galvanizing, to mention some examples. Pyrometers are mainly used for this purpose; thermal cameras are applied in rarer cases. Similarly developed are geometry measurements, which can also usually be checked in several manners, in a cost-effective manner, on the basis of a wide variety of measurement principles.

However, the situation is different when it comes to analyzing the microstructure prevailing in the component. For the quality of a component, there are several parameters that directly affect the material, which decide on good parts and rejects. On the one hand, for multiphase alloys or for composites, the phase composition or layer thickness inside would be an essential aspect. For correlating mechanical properties, grain size, aspect ratio, or orientation, is also essential. For thermomechanical treated steels, for example, it has been demonstrated that the prior austenite grain size and texture have an enormous influence on the morphology of the martensite produced in the cooling zone. This is ultimately reflected in the anisotropy of the mechanical properties (in rolling, normal, and transverse directions) [52]. In order to be able to determine the austenite grain size, which will not remain stable at lower temperatures, in the laboratory, elaborative stabilization procedures of the microstructure and complex microscopic analyses are often necessary. Consequently, the research for possibilities to investigate the microstructure of materials at high temperatures and in-situ, e.g., during corresponding deformations, is intensified. This in turn is achieved by means of different physical or chemical properties of, for example, phases, lattice structures, grain sizes, or orientations, which can be measurably separated from one another. Methods that make use of such properties are still predominantly found in the laboratory for off-line testing. In most cases, these are used to investigate how materials behave under certain conditions, in order to obtain the best possible process parameters. Particular focus is often placed on finding material parameters that can be used for microstructure models. Most of the microstructure models used in this aspect are based on mathematical formulations of real-physics based microstructure changes (e.g., RX, recovery, grain growth, and phase transformation) with material specific parameters, which are determined by the use of such methods [21,23].

3.1. In-Line Quality Control

Due to the increasing possibilities of integrating algorithms into data processing systems, there are now a large number of different approaches for the QC of products during manufacturing. This is specifically being enforced by the increasing accessibility to the usage of ML, AI, and Big Data. For example, neural networks are in focus for the integration in different systems. Especially for visual inspection, these methods are progressively used in the metalworking industry [53,54,55].

Surface inspection systems are particularly important in this respect. These are designed to detect and classify defects on the surface of the material, on the basis of images. This is usually done by the usage of training data sets, which show various defects on the surface, and which class they belong to. For this purpose there are countless approaches, for example [56] where ML algorithms like the support vector machine are applied to classify the defects. Here it was shown that the classification speed is sufficient to achieve acceptable defect detection. Another possibility to implement a statistically based approach into a neural network is the Principal Components Analysis (PCA), which is not only able to detect a defect using a large variation of defect images as input data, but also to classify the defect in order to understand its history. Other approaches can eventually be found in [57,58,59,60,61,62].

However, this is a very complex endeavor, since the variety of defects can be very high for different steel products alone. The defects occurring in hot-rolled steel products, for instance, are divided into nine main and 29 secondary classes [63]. An exact classification based on images is therefore an enormous challenge. Furthermore, it is a major hurdle to use image processing systems, due to the high temperatures and the exit speed of the strip of up to 100 km/h [56]. Even with increasing image acquisition frequency and improved computer performance, this type of inspection can be suppressed by oxidation and the harsh environment [64].

While the surface inspection systems are designed only for the detection of defects on the surface, there is a method that can also determine the microstructure of the product in-line. The 3MA (Multiparameter Micromagnetic Microstructure and stress Analyzer) technique is based on four micromagnetic principles. The detection of Barkhausen noise, the incremental permeability, the eddy current, and the tangential magnetic field method, which together provide many parameters which are used to gain information about the prevalent microstructure. For example, by measuring the coercive field, remanence, and maximum permeability, the change in the hysteresis can be related to dislocation density, and thus to the prevailing strain. The implementation in a rolling mill in three directions relative to the rolling direction (0°, 45° and 90°) allows anisotropy parameters to be determined, in addition to the common mechanical parameters [65,66]. Although this method has already been implemented in some steel rolling mills and is a promising method for in-line QC, there are still some drawbacks. The operating temperature is currently still limited to 300 °C and can only be realized by intensive cooling of the probe head. In a rolling mill, the distance between the probe and the cold strip is decisive for measuring accuracy. Therefore, the strip must be straightened by additional rollers to ensure an approximately constant lift-off. In addition, this method can only be used with ferromagnetic materials and requires complex calibration for each grade and strip thickness [65].

Eddy current inspection is a well-established method for detecting (near-) surface defects. In industry, these are also used for inline quality control, for example in wire rolling mills. Combined with high-speed cameras, they are able to detect manufacturing defects at temperatures of up to 1200 °C and wire speeds of up to 150 m/s. Here, a change in the signal recorded by the eddy current sensor triggers the camera, and a defect can be identified on the basis of the images. This system is likewise trained, as already discussed, with stored defect images and ML algorithms to detect defects autonomously [67].

The use of infrared (IR) cameras has increased in recent years due to their cheaper design. The advantage over visible-light imaging is that IR cameras are insensitive to smoke and can be used in dark environments. In addition, every object emits infrared radiation, and the intensity increases, the warmer the object is. In industry, these cameras are preferably used for temperature control as well as for defect detection, since a crack becomes detectable through a change in temperature. Especially for quality control in the field of welding and additive manufacturing, IR cameras are increasingly encountered. Another advantage is the image analysis, which can be adopted from the visible-light imaging [68,69].

In practice, some inspection methods based on X-rays are also employed. In particular, radioscopy and computed tomography (CT) have become widely established. The main difference lies in the two-dimensional image produced by radioscopy, while CT is used for more precise 3D analyses. The advantage of radioscopy is that image reconstruction is faster than CT, although it is not possible to determine the exact location or size of the defect, which is the case with CT. However, there are efforts to improve the use of CT in inline quality control by means of high speed area array detectors and a shorter exposure time [70,71].

3.2. In-Situ Methods for R&D Purposes

Approaches for the in-situ investigation of materials on a laboratory scale are widely existent. In order to characterize microstructural changes, the limiting factor is primarily the solution of applying measurement methodologies at elevated temperatures. One of the most prominent instruments in this case is dilatometry, which reproduces transformation kinetics via the abrupt change in thermal expansion. Phase fractions can also be determined by this technique. More comprehensive data can be obtained by structural analysis or high-resolution imaging instruments. Although the classical ultrasonic method is often used for microstructure characterization, the necessity of contacting the material restricts its use to low temperatures. The situation is similar for methodologies based on the eddy current principle. There are some approaches that equip established analytical methods like these with a heating chamber to apply the methodologies in-situ during heating, isothermal holding, or cooling. These are considered promising, as many are capable of simultaneously detecting and analyzing different changes in microstructure, but have certain limitations. Many of these methods can only be performed under vacuum or an inert gas atmosphere, are limited to surface examination, or can only be operated at a very low measurement frequency. Other methods, for example, based on synchrotron radiation or neutron scattering, have the disadvantage of limited accessibility. In the following, some promising methodologies for in-situ microstructure analysis are presented, which show great potential to analyze time- and temperature-resolved microstructural changes at high temperatures.

3.2.1. High Temperature X-ray Diffraction (HT-XRD)

This method is frequently used to perform in-situ qualitative or quantitative phase analysis, precipitation formation, as well as phase dissolution. The underlying principle is based on the classical XRD method, except that the sample can be heated in a heating chamber and XRD scans are continuously generated during the heating or holding time. Using the time-resolved generated diffractograms, the characteristic peak intensities and the peak shifts for individual phases can be evaluated, providing information about phase composition, transformation, phase precipitation, or dissolution kinetics. The peak intensities can also be used to obtain information about the quantitative fractions. On the other hand, the crystallite size, as well as the dislocation density, can also be analyzed via selective peak profile analysis. These analyses are mostly performed using either the modified Williamson-Hall or the modified Warren-Averbach procedure [72,73,74,75]. Difficulties may arise in the analysis of the diffractograms due to the difference in peak positions between room temperature and higher temperatures, as a consequence of thermal expansion and therefore a peak profile shift. In addition, with the latter analysis methods, there can be an overlap of the residual stresses and the crystallite size, where both are analyzed based on the peak profile, while residual stresses dissipate at elevated temperatures, as the crystallite size increases. Therefore, the analysis requires enormous expertise for the interpretation of the data obtained [76]. This method is used in the laboratory scale for characterizing the above-mentioned effects of a material. Various verified reaction kinetics are used to create thermokinetic models for simulation software, such as MatCalc.

A major disadvantage of this method is that characterization can only be performed in the near surface region. Consequently, this can lead to difficulties, if oxidation reactions occur, or if there are local differences in the chemical composition. Also, a grain size below 10µm and a statistical distribution is required for a reliable analysis. Furthermore, the sample preparation is a decisive criterion to ensure valid investigations [77].

3.2.2. High Temperature Scanning Electron Microscopy and Electron Backscatter Diffraction (HT-SEM/EBSD)

This method is also based on the classical SEM or EBSD method with an integrated heating chamber, where images are acquired at a certain frequency. In contrast to the HT-XRD method, actual images can be generated here, which are based on the interactions of the sample surface with the electron beam. This enables high-resolution imaging and tracking of changes in the microstructure. For example, the displacements of individual grain boundaries during grain growth or recrystallization phenomena or even twinning effects can be displayed [78]. By using the EBSD arrangement, the time- and temperature-resolved changes in grain size and grain orientation can be detected. In this context, the application of test facilities for the evaluation of mechanical properties and for the application at high temperatures is increasingly envisaged. To date, there are already a variety of mechanical testing equipment, such as tensile test equipment, which can be implemented in the chamber of an SEM. If such a setup is installed in a HT-SEM/EBSD, this opens up a wide range of new methods to apply mechanical properties in-situ, at high resolution [79,80,81,82]. Within the scope of deformation possibilities, such as the in-situ tensile test in the SEM, digital image correlation methods are also increasingly being used. By integrating camera systems, displacements of the previously applied patterns can be recorded with high precision, which allows conclusions to be drawn about strains in one or more spatial directions. Depending on the magnification adjusted in the SEM, strain localizations in the subgrain range can be determined. The combination of EBSD, for example, can also be used to identify the influence of the grain structure on the forming behavior [83,84].

The difficulty in this case is to obtain good image quality even at high temperatures. Similarly, as mentioned before, the sample preparation requires a lot of effort. Furthermore, the acquisition costs of such a system are enormously high.

3.2.3. In-Situ Methods Based on Synchrotron Radiation

The possibility to perform in-situ material characterization with High Energy X-rays opens up a great potential to investigate near-surface microstructural changes and to be able to investigate rapidly occurring processes in real time, due to high resolution capabilities at high measurement frequencies. One possibility involves implementing a TMTS, for example, in the form of a dilatometer that has windows mounted to transmit the High-Energy XRD (HEXRD) beam through the mounted specimen. The diffraction effects of the material can then be detected in complete Debye-Scherrer rings on a flat panel detector. Using such an arrangement, a variety of changes in microstructure (similar to HT-XRD) can be elicited simultaneously at high temperatures, and by using a deformation dilatometer, conclusions can also be made about the behavior during simultaneous deformation [85,86,87]. The main advantage of this method is that the high-energy radiation does not limit the investigation to the surface, but larger bulk materials can be investigated.

Another investigation method worth mentioning, which is often carried out with synchrotron radiation is Small-Angle X-ray Scattering. Investigations based on this principle provide information on the size distribution and precipitation or dissolution kinetics of precipitates. These analyses can be performed in-situ rapidly and quantitatively, and with time and spatial resolution [88].

3.2.4. High Temperature Laser Scanning Confocal Microscopy (HT-LSCM)

This method combines the classical LSCM with an infrared heated high temperature chamber. A focused laser beam scans the surface of the sample selectively. The intrinsic radiation is blocked out by a He-Ne laser, allowing the investigation to be carried out up to the fluid state. The thermal etching of the laser enables the observation of grain growth during a defined time-temperature profile. Thus, the effects of precipitates on grain growth behavior can be investigated as well. The main disadvantage of this method remains the restriction of the analysis to surfaces only. In addition, the oxygen content in the high-temperature chamber must be kept meticulously low, since the slightest oxidation leads to reduced observability [89].

3.2.5. Bainite Sensor

Another field of application for eddy current based testing is the so-called bainite sensor. This consists of an excitation coil, which generates an electric field in the component to be tested. In addition to the eddy current, this field also generates a magnetic field, which in turn generates a field in the reverse order to the first field. Thus, magnetic as well as electrical signals can be obtained by a coil system in the sensor. The signals obtained are analyzed according to their harmonic spectrum. In-situ testing of material properties is carried out by implementation in a TMTS, like the Gleeble. This allows phase fractions as well as transformation kinetics to be recorded during heat treatment or deformation. This is based on the ferromagnetic to paramagnetic change of steel during the transformation from low-temperature phases to austenite in the steel. So far, this method has been used primarily to study the bainite transformation in steel, which gives rise to the designation. This represents a simple technique to obtain in-situ information on transformation behavior in materials with allotropic phase transformation. However, this method is limited to this kind of investigation [90,91].

4. The Key Role of LUS in a Smart Manufacturing Production Site

Despite the multitude of established methods for in-line in-situ quality inspection and off-line in-situ material characterization methods mentioned, there are hardly any inspection methods that provide a reasonable application for both fields.

One possibility to investigate in a contactless manner, in-situ, and at high temperatures is the LUS method. This measurement technique has been developed over the last decades for many fields of application and can be used in various different configurations. This method can be applied to a variety of analyses known from the conventional ultrasonic technique. In particular, defect detection is one of the most common applications, especially the detection of defects in terms of pores, voids, or adhesion defects, especially in castings or welds [92]. However, such defect detections are also of great advantage in additive manufacturing. In this context, this method can also be applied in-situ, during the build-up process [93,94,95]. Furthermore, the ultrasonic method can also be used for the detection of corrosion phenomena and the associated fatigue analysis [77,96,97]. Damping analyses can be utilized to determine hardening depths in surface-hardened components. Most of the analysis options mentioned are based on approaches that can already be investigated with conventional ultrasonic measurement and are applied to static components not exposed to high temperatures.

Nevertheless, the important aspect here is the possibility for advanced in-situ microstructural characterizations at high temperatures or velocities, which is the core competence of LUS. Due to a high measurement frequency, this method can record time-resolved changes in the microstructure, even at temperatures above 1200 °C, thus providing a great contribution to an efficient characterization of metals, when implemented in-line for the inspection of e.g., dimensions, phase constitution, or grain size [98,99]. Unlike other systems, this method operates reliably in harsh environments as it is relatively insensitive to dirt and dust. The continuous development and improvement of the in-line system looks promising to extract progressively more microstructural information from, for example, hot-rolled steel strip and to contribute to a significant advance in QC. Currently tens of LUS systems are integrated in a production plant worldwide, the majority of them being oriented for wall thickness measurement in tube production plants, as described for in [100]. Recent LUS measurement devices are encountered in hot strip rolling mills. These are specialized for grain size measurement of austenite before the cooling section. The in-situ measurement of austenite grain size is an important factor in predicting the material properties of the steel strip. In [101], an example of a LUS system integrated in a hot strip mill is outlined. The excitation laser is a pulsed Q-switched laser operating at a frequency of 20 Hz, up to 200 mJ, and a pulse length of 6 ns, at a wavelength of 532 nm. The detection laser operates at a wavelength of 1064 nm and a pulse duration of 100 s, at a power of 600 W. A GaAs two-wave mixing interferometer was used as the interferometer. The hot strip mill rolls the pre-rolled strips in six slabs to a thickness of 2–15 mm. The LUS system is mounted behind the slabs and in front of the runout table on a linear displacement and can move between two strip guide rollers, at a distance of about half a meter from the strip. The detection laser is located at a distance of a few meters. This measurement method is very promising and will certainly be found in several rolling mills in the future. Further literature about LUS Systems implemented in industry can be found in [102,103,104,105,106].

Furthermore, such a LUS system can be coupled to a TMTS to study in-situ microstructural changes of samples during thermal and thermomechanical treatments. In the context discussed here, the usage of the method in the relevant areas of quality testing and the determination of material parameters will be specifically addressed. This section is intended to outline the operating principle and the analyzing methods that allow conclusions to be drawn about the prevailing microstructure.

4.1. Operating Principle

The laser ultrasonic method is based on the contactless introduction and detection of ultrasonic signals in the material. The core of the system is an excitation laser and a detection laser—most commonly Nd:YAG lasers are used for both—and integrated interferometry, such as a Fabry-Perot or a two-wave mixing interferometer. The pulsed excitation laser transmits up to several mJ of energy to the material surface, where either thermoelastic excitation or ablative excitation occurs. Both variants generate a more or less intense stress field, resulting in a broadband ultrasonic signal (usually in the range of 500–50 MHz). In the case of thermoelastic excitation, mainly surface and plane waves are generated, while ablative excitation tends to produce more pronounced longitudinal waves. These waves pass through the material and are reflected at the back side, creating an echo. The detection laser can be mounted either in transmission or reflection mode, where the smallest deflections of the surface, due to reflection lead to a frequency shift, which is processed by an interferometer into signals that can be evaluated [107]. Figure 1 shows a schematic visualization of a LUS system.

Apart from other types of waves generated by the excitation laser, such as surface waves or Lamb waves—an interaction of bulk waves occurring to (a-) symmetric vibrations in thin samples -, the essential ones here are the bulk waves. The bulk waves in a solid body can be subdivided depending on the mode of propagation. If the displacement of the involved particles is in the direction of propagation and results in compression, these waves are called longitudinal or compression waves; if the atomic motion is perpendicular to the direction of propagation, these waves are called transverse or shear waves. Bulk waves have the same frequency propagation, which means that they travel at the same speed, or in other words are non-dispersive. Since the compression waves propagate about twice as fast as the shear waves, these two types of waves can be clearly distinguished [107]. By analyzing the different types of waves in terms of propagation mode, direction and velocity, frequency and/or temperature dependence, or penetration depth into the material, different properties of the material can be determined.

The most important analysis parameters of the detected signals are the Time of Flight (ToF), i.e., the propagation velocity, and the amplitude or its decrease after passing through the material, i.e., the attenuation. These two parameters can be used to generate a wide range of information about the microstructural or geometric configuration of the component, by conducting suitable evaluation operations. The evaluation of the velocity of sound provides specific information about

• Geometry
• Phase transformation
• Phase composition
• Recrystallization
• Texture

while the evaluation of the attenuation covers the aspects of

• Grain size and grain growth
• Phase constitution
• Dislocations

In the following subsections the underlying principles of gaining insight into microstructural changes are briefly described.

4.1.1. Ultrasonic Velocity

The propagation speed of ultrasonic waves (especially longitudinal c_L and shear waves c_T) depends on the density and the stiffness tensor [108]. Assuming a polycrystalline, isotropic material, with a small grain size in relation to the sample size, the following relationships between the speed of sound, density and elastic constants (K, S, ν) for longitudinal and transverse waves can be assumed:

$$c_L=\sqrt{\frac{K}{\rho }·\frac{1-\nu }{(1+\nu )(1-2\nu )}}$$

(1)

for the sound velocity of longitudinal waves and

$$c_T=\sqrt{\frac{S}{\rho }}$$

(2)

for that of transverse waves. K is referred to as the Compression Module, S is the Shear Module, and ν is the Poisson ratio. In the simplest case, the propagation speed can be determined by ToF measurements. For a sample of thickness h, the propagation velocity can be derived from the time required for the wave to be reflected at the back of the sample and to be returned:

$$v=\frac{2h}{∆t}$$

(3)

If the density ρ of the inspected material is known, the Young’s modulus (via known relationships with the compressive and shear moduli) or Poisson’s ratio ν can be determined very accurately over temperature, by measuring the speed of sound.

Since polycrystalline metals generally exhibit preferential grain orientations as a result of the manufacturing process (solidification and preferential dendrite growth or forming processes), it is of particular importance to be able to determine the predominant orientation.

Due to the high sensitivity of laser ultrasonic measurements with respect to the velocity of sound, individual entries of the elastic stiffness tensor can further be determined. Since the propagation velocity differs depending on the direction of propagation, the preferred direction can be determined using reference crystals. Conversely, orientation distribution coefficients (ODC) can also be calculated from ultrasonic signals.

Recrystallization phenomena, respectively the recrystallized fraction, can similarly be measured by measuring the changes in texture, and in turn the change in ultrasonic velocity. Since recrystallization is preceded by a certain strain and thermal activation, the changes in the measured sound velocity can be assigned to the recrystallized fraction via the Johnson-Mehl-Avrami (JMAK) function.

Especially for phase transformation or allotropic transformations of metals, a very precise prediction can be obtained with this measuring system. Since the acoustic velocity differs strongly depending on the crystal structure, the phase composition rule (lever rule) can provide an exact determination of the predominant fractions [109]. The speed of sound is also changed considerably during a transition from ferromagnetic to paramagnetic material (Curie temperature).

4.1.2. Attenuation

The second important aspect, which can be analyzed from the ultrasonic signals and related to microstructural properties of the material, is attenuation. This manifests itself in the reduction of the amplitudes of subsequent echoes. The total attenuation can be attributed to three significant phenomena and added as:

$$\alpha \left(f,T\right)={\alpha }_{sc}\left(f,T\right)+{\alpha }_{IF}\left(f,T\right)+{\alpha }_D\left(f\right)$$

(4)

The three contributions are made by diffraction α_D, internal friction α_IF, and the contribution of grain scattering α_sc. The contribution of diffraction can be estimated qualitatively by the Fresnel parameter, while the contribution of internal friction (caused for example by magneto-mechanical damping, interstitial atoms, or dislocation motion) can be minimized by suitable algorithms, since this contribution is frequency independent to a large extent [110]. In turn, conclusions about recovery processes can be derived from the dislocation motion [111]. The contribution to be extracted is that of the grain boundary scattering, which provides information about the predominant grain size in the material. The contribution of grain scattering is primarily noticeable in metals with high elastic anisotropy and is the dominant contributor to the total damping, especially in metals of this type, such as steel, nickel, or cobalt. The grain size D is a temperature-dependent factor and damping increases as grain size increases. The contribution of the grain size to the total damping can be calculated via the power-law function

$${\alpha }_{Sc}\left(f,T\right)=C\left(T\right)D^{n-1}{f}^n$$

(5)

where C corresponds to a temperature dependent material constant. The value of the exponent n depends on the ratio of the wavelength to the grain size. Below are the following three regimes:

• Rayleigh regime $(\lambda \gg D):{\alpha }_{Sc}=C_r{D}^3{f}^4$
• Stochastic regime $\left(\lambda \approx D\right):{\alpha }_{Sc}=C_{S}D{f}^2$
• Diffusion regime $\left(\lambda \ll D\right):{\alpha }_{Sc}=C_D{D}^{-1}$

It is usually assumed that this ratio lies between the Rayleigh and the Stochasitic regime, which is why the value of 3 is usually chosen for n. However, this value can also be fitted specifically for a material, as described in [112].

To estimate the grain size, the power-law Formula (6) can be fitted into the measured attenuation curves:

$$a\left(f,T\right)=a+b{f}^n$$

(6)

Here, a represents a frequency-independent contribution that includes, for example, internal friction or external factors such as variations in laser intensity. The expression b represents the frequency dependent grain size contribution and can be assigned to the actual grain size via

$$b=\Gamma \left(T\right)D^{n-1}$$

(7)

with Γ(T) = $\sqrt{\frac{1}{C\left(T\right)}}$, which contains the material and temperature dependent Information.

Appropriate model calibration with ex-situ tests can be used to draw in-situ conclusions about grain size evolution in thermal and thermomechanical tests, based on the reference echo model or the single echo technique.

The correlations of damping as well as velocity with microstructure have been tested and published based on numerous investigations on a variety of materials. For example, grain size or grain growth investigations are discussed in [113,114,115,116,117,118,119,120]. Recrystallization effects can be found in [121,122,123,124,125,126,127,128,129], whereas texture measurements where conducted for instance in [130,131,132]. Further information of phase transformation and composition can be found in [133,134,135,136,137].

4.2. LUS in a Smart Production Lab

As already indicated, the LUS method is implemented to support the holistic construction of a SPL with regard to the creation of a DM/DS/DT, and to support it along the depicted value chain as a holistic auxiliary and quality inspection tool.

The MUL 4.0 project progressing at the Montanuniversität Leoben, in the form of a comprehensive SPL, encompasses parts of the holistic value chain. This value chain covers multiple technically different and geographically separated production sites, which are connected via a production network in order to transparently gather, store, and share data for analysis purposes. The starting point of the value chain, as depicted in Figure 2, is the Chair of Nonferrous Metallury (NFM), which specifically deals with aluminum alloys and has a miniature continuous casting facility. It enables the casting of a wide variety of common and newly developed alloys, as well as recycled material. This provides the feedstock for the Chair of Metal Forming (MF), where it is delivered. From here it can be further processed in a variety of methods both cold, or annealed to a specific temperature in a digitized industrial furnace. One possibility would be rolling in a miniature rolling mill that has been transformed into a CPPS. A black box machine learning approach allows rolling from a certain thickness to the desired final thickness in multiple passes, specifying the optimal pass schedule [32]. Another option would be forging in a hydraulic press capable of forces up to 1 MN. This press is being transformed into a fully integrated CPPS and is capable of recording data such as forces and temperature during upsetting, and automatically transferring them to a Supervisory Control and Data Acquisition (SCADA) system. The data is pre-processed and transmitted to a higher-level tracking system. Based on the raw material, this system can assign the process data to the respective product and provide information about the respective location. This data is additionally used throughout the value chain for input parameters for the FEA, in order to virtually represent the effects of the process conditions on the final product, and thus serve as a QC tool. Finally, the generated product can be heat treated to adjust the optimal material properties and then be machined to the final product. The value chain introduced here is illustrated in Figure 2.

The application of the LUS in the SPL is intended to provide a fundamental contribution to the creation of a DT. This consists of a FEA for the respectively considered heating or forming process of the specimen with microstructural information prevailing in the material. As already mentioned, the description of the microstructural behavior requires a large number of material parameters in order to be able to calculate various material responses as a function of strain, strain rate, and temperature, by means of a microstructure model. Therefore, the coupling of a LUS system with a TMTS, which records these characteristic values of the deformation as well as the corresponding time, is inevitable. This coupling enables the acquisition of data of the forming process via the TMTS Gleeble 3800, as well as in-situ data of the microstructural processes by a trigger signal from the same time and its correlation. Although the LUS system only provides information on the velocity and attenuation of the ultrasonic waves, these can be automatically transferred to various microstructural changes using programmed evaluation routines, as exemplified in Section 4. This makes it possible to obtain information about grain size, recrystallization behavior, or phase transformations during the experiment. Based on the corresponding recording of the flow curve and the flow behavior, a microstructural change from the LUS data can be assigned to the corresponding flow curve parameters. For data acquisition from the LUS, a high-frequency Data Aqusition System (DAQ) will be implemented and integrated into the IIoT network, which can record the analog signals from the LUS, at a sufficiently high frequency to ensure high-resolution signal mapping. For this purpose, the fiber optic based ibaPADU-4-AI-I with ibaFOB-Dexp, in combination with the ibaRackline SAS was chosen, which has measurement frequencies of up to 10 kHz with a resolution of 16bit, and thus fulfills the minimum sampling rate of 400 Hz set by the LUS. Since this DAQ has only four analog inputs, which are already occupied by Gleeble sensors, an identical ibaPADU-4-AI-I is implemented in a parallel synchronous operation, which is combined on the iba processing unit ibaRackline SAS and the software iba PDA & iba Analyzer. For improved traceability of the data recording, the Hierarchical Data Format (HDF) HDF5 is implemented here to record the metadata of each data set, and thus to get a more holistic insight into each measurement. These datasets are automatically synchronized by an open-source approach using Python, on a file server shared with the production network at the MF, in its Data processing layer and consequently stored in the production network’s relational open-source MySQL MariaDB database. This approach is visualized in Figure 3.

Consequently, the data stored in the database is used for further post processing. The quintessence of the LUS data is the derivation of the material parameters essential for semi-empirical microstructure modeling. In addition to material-specific activation energies, etc., the associated forming conditions are necessary for the derivation of these. A typical set of formulas for the description of the material behavior in terms of static, dynamic, and metadynamic recrystallization, as well as grain growth and phase transformation, as a function of the temperature, as well as the corresponding flow curve, contains about 40 material constants [17,23] for the complete description of the microstructural processes. These are derived in a deposited evaluation script and are stored in a database. The subroutine required for FEA of the forming process, with corresponding microstructure simulation, can then extract these determined assigned material parameters from this database, and thus perform an accurate simulation of the process for the specific material under investigation. Since the computation time of a forming process with integrated microstructure simulation is too high for simultaneous computation, this results in a DM that is not applicable for the in-situ implementation. Nevertheless, to create a DS or DT that can intervene in-situ in the process, several simulations with varying process parameters can be calculated, and the respective results in turn stored in a database for a number of process variations. These simulations can be performed locally or on a processing unit in the production network, in the form of CC, depending on the availability of computing resources required, and the results are always ultimately included in the superordinate database. For the implementation of a DS or DT, which is to intervene in-situ in a process, it is possible to fall back iteratively on the simulation results with the best-fitting process parameters, and to implement only individual data in the DS/DT. This saves an enormous amount of computational effort and still ensures an intervention in the process due to deviating microstructure conditions.

Furthermore, the LUS can be applied to characterize the raw material provided by NFM and the consequent material changes of the respective forming processes. Therefore, it can be used to rapidly identify significant deviations of a recycled material to a referenced primary material. In this way, it can be assured that the requirements for the use of a secondary metal meet specified tolerances, thereby contributing considerably to the avoidance of material waste. Thus, the LUS acts as an independent CPPS, capable of either using CC or local computing resources. As a result of the holistic data gathering, root causes of deviations in quality can be researched and evaluated, and thus be linked to the causing processes and parameters. In addition, ML can be applied in order to detect possible product quality deviations, due to unsuitable process parameters. Consequently, situationally appropriate decision-making is enhanced, and active measures can be taken, either adjusting the respective process or removing the affected specimen or product.

5. Conclusions and Outlook

The LUS method offers an immense variety of applications, whether in production, R&D, the laboratory or an interface, a so-called Smart Manufacturing Lab or SPL. In each of these areas, the method benefits from being applicable in-situ, contactless, at high temperatures, in harsh environments. By analyzing the resulting signals, a variety of different material or product characteristics can be derived in a rapid manner. Due to the constant further development, either of the internal systematics, for example, to increase the measuring frequency, or the more precise development of the models for the evaluation of any materials and alloys, the LUS will be indispensable for almost any scenario in the future. The possibility to reduce the development time of a microstructure model, including the whole required parameter set by about half, provides an enormous advantage. By implementing this methodology, it renders such microstructure models applicable, even in sectors where integration was not feasible for financial reasons. This development can provide a springboard for more economical and efficient production, since the effects of process conditions on the microstructure can be analyzed in advance.

Of course, the advantages on the production line are also of enormous importance. The possibility to inspect not only geometric parameters for tolerances in-line, but also to monitor the prevailing microstructure significantly, contributes to increasing efficiency and reducing scrap.

Especially the application as a CPPS in a digitalized production environment contributes to a considerable benefit. When combining the lab-scale utilization for a production facility, unimagined possibilities can still be considered with the assistance of the enablers of I 4.0. The interaction of DM, DS, and DTs and real-time process data integration with ML along the value chain can be a substantial contribution to the sustainable production of the future.

The main shortcomings in the implementation of the LUS in a SPL are that the evaluation of the obtained data is not yet mature for a variety of different materials. Even though a large number of investigations have already been carried out for microstructural changes of some various materials, it is necessary to investigate the behavior using the LUS for additional materials and alloys in a large-scale, creating a sub-ordinate database including microstructural data.