Process Monitoring System Research Articles

Photodiode-based process monitoring for the ultrashort-pulsed laser structuring of the diffusion media for fuel cells

In recent years, the incorporation of process monitoring systems has become an integral part across several laser material processing applications. This is due to the feasibility of inline evaluations, which enable the elimination of downstream quality checks that are typically time-consuming and costly. Simultaneous to recent sensor developments, ultrashort-pulsed (USP) laser material processing has become a promising technology for creating complex microstructures in a wide variety of materials due to its high precision and negligible thermal input. So far, process monitoring has yet to be established in USP applications as it is unknown which sensors are suitable, which parameters can be observed, and which process-relevant information can be detected within the data. Within this work, a state-of-the-art photodiode-based process monitoring system was used to collect spectral data from the process zone and extract relevant information from the process. The laser structuring of the two-layered diffusion media used in polymer electrolyte membrane fuel cells was selected as the use case. This structuring process is challenging due to production-related fluctuations in the material thickness and the surface quality. Within this study, information was acquired from a sensor system employing three photodiodes. Through the application of explorative data analysis techniques and machine learning, AI models were created, focusing on determining the focus position of the laser beam and identifying the transition between the two layers. The developed models were capable of determining the initial focus position with varying accuracies. The best performance was achieved by a convolutional neural network using spectrograms as input data, while the worst accuracy was achieved by a ridge regression model using manually selected features. Furthermore, a clustering model was used, demonstrating the capability to detect the layer transitions within a specified tolerance.

Acoustic process monitoring during the laser beam welding of stainless-steel foils using an adjustable ring mode laser beam source

Electrification of the mobility sector is vital to meet the targets for reducing greenhouse gas emissions. Besides battery-based mobility solutions, polymer electrolyte membrane fuel cells (PEMFCs) are a promising technology for electrifying drive trains, especially in heavy-duty applications, such as maritime or logistics. Bipolar plates, a key component of PEMFCs, can consist of two stainless-steel foils that must be welded to be gas-tight. In order to join the two metal foils, laser beam welding is the state-of-the-art technology. Current challenges include process instabilities at higher welding speeds, such as the humping effect, which can cause weld seam imperfections. Therefore, applying sensors for laser beam welding is a promising approach to monitor the welding process. AISI 316L foils were welded within the scope of this work with various process parameters using an adjustable ring mode laser beam source. Additionally, an optical microphone was used as a process monitoring system. By applying different parameter settings and due to the introduction of artificial faults, weld seam defects, such as a burn-through or a gap, were induced. After utilizing a noise reduction algorithm for the acoustic signals, numerous features in the time and frequency domains were extracted, with which multiple machine learning algorithms were trained and compared concerning their performance. A light gradient boosting machine was identified as a suitable machine learning model for weld seam classification. Finally, hyperparameter tuning was conducted, which resulted in a cross-validation accuracy of 94.78%, depending on the quality categories considered.

Completely noninvasive multi-analyte monitoring system for cell culture processes

Although online monitoring of dissolved O2, pH, and dissolved CO2 is critical in bioprocesses, nearly all existing technologies require some level of direct contact with the cell culture environment, posing risks of contamination. This study addresses the need for an accurate, and completely noninvasive technique for simultaneous measurement of these analytes. A “non-contact” technique for simultaneous monitoring of dissolved O2, pH, and dissolved CO2 was developed. Instead of direct contact with the culture media, the measurements were made through permeable membranes via either a sampling port in the culture vessel wall or a flow cell. The efficacy of the “non-contact” technique was validated in Escherichia coli (E.coli), Chinese hamster ovary (CHO) culture processes, and dynamic environments created by sparging gases in cell culture medium. The measurements obtained through the developed techniques were comparable to those obtained through control methods. The noninvasive monitoring system can offer accurate, and contamination-minimized monitoring of critical process parameters including dissolved O2, pH, and dissolved CO2. These advancements will enhance the control and optimization of cell culture processes, promising improved cell culture performance.

Detection Strategies for COM, WMI, and ALPC-Based Multi-Process Malware.

Behavioral malware detection is based on attributing malicious actions to processes. Malicious processes may try to hide by changing the behavior of other benign processes to achieve their goals. We showcase how Component Object Model (COM) and Windows Management Instrumentation (WMI) can be used to create such spoofing attacks. We discuss the internals of COM and WMI and Asynchronous Local Procedure Call (ALPC). We present multiple functional monitoring techniques to identify the spoofing and discuss the strong and weak points of each technique. We create a robust process monitoring system that can correctly identify the source of malicious actions spoofed via COM, WMI and ALPC with a low performance impact. Finally, we discuss how malicious actors use COM, WMI and ALPC by examining real-world malware detected by our monitoring system.

Intelligent process monitoring of smart polymer composites using large area graphene coated fabric sensor.

Herein, we report the development of an online process monitoring system for vacuum-assisted resin transfer molding (VARTM) process using large area graphene coated in-situ fabric sensor. Besides imparting excellent mechanical properties to the final composites, these sensors provide critical information during the composite processing including detecting defects and evaluating processing parameters. The obtained information can be used to create a digital passport of the manufacturing phase to develop a cost-effective production technique and fabricate high-quality composites. The fabric sensor was produced using a scalable dip-coating process by coating 1-, 3- or 5-layers of thermally reduced graphene oxide (rGO) onto glass fabric surface according to the number of dips of the fabrics into GO solution. The electrical resistances from all electrode pairs were simultaneously and continuously recorded during distinct stages of the VARTM process to determine the relative conductance. During the vacuum cycle, the range of relative conductance increased with the number of coated rGO layers, with the 5-layer rGO-coated sensor showing the highest conductance range of 16.9 %. Additionally, it was observed that the 5-layer coated sensor showed a consistent decrease in conductance during the infusion phase due to the fluid flow pressure dominating the resin electrical conductivity.

Efficient and lightweight layer-wise in-situ defect detection in laser powder bed fusion via knowledge distillation and structural re-parameterization

Powder bed quality is critical in Laser Powder Bed Fusion (LPBF) as defects in the powder bed will affect the quality of the forming parts. Recently, numerous powder bed defect detection methods based on semantic segmentation algorithms have been developed. However, due to the computational resource constraints and real-time requirements in industry scenarios, these heavy models face challenges in deploying into the LPBF process monitoring system. Meanwhile, existing lightweight models struggle to achieve sufficient performance on industrial data due to their limited feature extraction capabilities. To address the challenges in industrial deployment, this paper proposes a lightweight and efficient LPBF layer-wise defect detection paradigm to achieve a balance between performance and efficiency. Aiming at improving the performance of the lightweight model, this paper proposes a structured Knowledge Distillation framework, in which a pre-trained high-performance segmentation model (UNet++) is used to guide the training process of a lightweight model. To address the impact of model discrepancies on the effectiveness of Knowledge Distillation and further improve the inference speed, this paper constructs a specific lightweight model and applies Structural Re-parameterization to deeply compress the model. Moreover, we created a dataset comprising 406 images of powder-bed defects, with each image annotated at the pixel level. Massive experiments based on the dataset demonstrate the effectiveness of the proposed paradigm, with an excellent trade-off in performance and efficiency. The speed of the proposed model is approximately 5 times faster than that of high-performance models while maintaining comparable performance.

A Pragmatic Approach for Rapid, Non-Destructive Assessment of Defect Types in Laser Powder Bed Fusion Based on Melt Pool Monitoring Data.

Process monitoring systems, e.g., systems based on photodiodes, could be used in laser-based powder bed fusion (PBF-LB/M) to measure various process parameters and process signatures to eventually allow for a local, detailed analysis of the produced parts. Here, simple statements only concerning the occurrence of defects in parts are sufficient in many cases, especially with respect to industrial application. Therefore, a pragmatic approach to rapidly infer the occurrence of defects and their types based on in situ data obtained by commercially available process monitoring systems is introduced. In this approach, a color distribution in form of a histogram is determined for each produced part using layer-wise screenshots of the visualized data provided by the monitoring software. Assessment of the histograms of AlSi10Mg samples, which were processed with different parameter combinations, revealed characteristics depending on the prevailing defect types. These characteristics enable the prediction of the occurring defect types without the necessity to apply conventional downstream testing methods, and thus, a straightforward separation of parts with good quality from defective components. Since the approach presented uses the data visualization of the monitoring software, it can be used even when direct access to the raw data is not provided by the machine manufacturer.

Implementation of an intelligent process monitoring system for screw presses using the CRISP-DM standard

AbstractIncreasing the service life and process reliability of systems plays an important role in terms of sustainable and economical production. Especially in the field of energy-intensive bulk forming, low scrap rates and long tool lifetimes are business critical. This article describes a modular method for AI-supported process monitoring during hot forming within a screw press. With this method, the following deviations can be detected in an integrated process: the height of the semi-finished product, the positions of the die and the position of the semi-finished product. The method was developed using the CRISP-DM standard. A modular sensor concept was developed that can be used for different screw presses and dies. Subsequently a hot forming-optimized test plan was developed to examine individual and overlapping process deviations. By applying various methods of artificial intelligence, a method for process-integrated detection of process deviations was developed. The results of the investigation show the potential of the developed method and offer starting points for the investigation of further process parameters.

Prediction of Errors in the Field of Additive Manufacturing Technology

Additive manufacturing, also known as 3D printing, allows the formation of complex geometric structures layer by layer. Predicting errors in this process may help identify potential problems in a timely manner and minimise waste. A human may detect an additive manufacturing error, but cannot provide continuous monitoring or real-time correction. The article is focused on the design of a camera system design for online monitoring of the 3D printing process with the task of detecting process errors arising during 3D printing of objects. The article describes the methodology for tracking the occurrence of process errors in 3D printing, which are identified in the OctoPrint Nexus AI plug-in environment for the subsequent application of a suitable solution to minimize the occurrence of defects. The application of a real-time process monitoring system including the ability to correctly predict anomalous behaviour in the context of artificial intelligence has proven to be an appropriate solution to that particular problem.

Developing an IoT and Machine Learning-Based Monitoring System for Discrete Production Processes

Developing an IoT and Machine Learning-Based Monitoring System for Discrete Production Processes

Multi‐source data‐driven approach for prediction of melt density during polymer compounding

AbstractMelt density is a crucial quality indicator for polymer composites, yet real‐time measurement remains challenging due to processing complexities. While existing machine learning methods offer solutions, they often fall short in complex compounding scenarios. This study presents a novel multi‐source data‐driven approach for measuring melt density in polycarbonate/acrylonitrile butadiene styrene blends. By incorporating ultrasonic, near‐infrared, and Raman spectra data acquired during melt processing, a deep separable convolutional neural network model is developed to predict melt density accurately. The model effectively fuses multi‐source data to establish the mapping relationship between input data and melt density output. Results demonstrate the model's ability to monitor melt density in real‐time, achieving a prediction accuracy with RMSE and R2 indexes of 0.005 g/cm3 and 0.9841, respectively. The proposed approach outperforms existing methods, showcasing its effectiveness and superiority in melt density prediction for polymer compounding processes.Highlights Establishment of the real‐time monitoring system for polymer extrusion processes. Conversion of multi‐sensor signals into time‐frequency images using wavelet decomposition. Fusion of sensor data into a three‐channel tensor‐image. Development of a data‐driven DSCNN model for predicting melt density. Implementation for online monitoring and prediction in PC/ABS compounding system.

An interactive data-driven multiple-attribute decision-making technique via interval-valued intuitionistic fuzzy sets for teaching quality evaluation in higher education

Improving the quality of higher education teaching is a systematic project. The improvement and formulation of relevant laws, regulations, and measures at the macro level are the minimum and specific requirements for the overall private universities, and are the basic guarantee for controlling the healthy and orderly development of universities. At the micro level, school management needs to focus on two aspects: leadership level construction and teacher level construction. Only by scientifically controlling the above issues and comprehensively considering them can the persistent problem of low teaching quality be fundamentally and gradually solved. In short, the construction of the quality assurance and evaluation system for higher education teaching in China is still in the initial stage of development. Therefore, universities should start from the guarantee and evaluation system to promote the construction of teaching process monitoring and evaluation systems, and improve the level of education and teaching quality on the basis of highlighting higher education teaching reform and research. The teaching quality evaluation of higher education in the era of artificial intelligence is a MADM. In this study, in light with interval-valued intuitionistic fuzzy Hamacher interactive hybrid weighted geometric (IVIFHIHWG) technique and induced OWG (I-OWG) technique, the induced IVIFHIHWG (I-IVIFHIHWG) technique is administrated. Then, the I-IVIFHIHWG technique is exploited to manage the MADM under IVIFSs. Finally, the numerical example for teaching quality evaluation of higher education in the era of artificial intelligence is exploited to verify the I-IVIFHIHWG technique. Thus, the main research contributions are administrated: (1) the I-IVIFHIHWG technique is administrated in line with the IVIFHIHWG and I-OWG technique; (2) the I-IVIFHIHWG technique is exploited to manage the MADM under IVIFSs; (3) the numerical example for teaching quality evaluation of higher education in the era of artificial intelligence and some comparative studies were exploited to verify the I-IVIFHIHWG technique.

Pathways toward the Use of Non-Destructive Micromagnetic Analysis for Porosity Assessment and Process Parameter Optimization in Additive Manufacturing of 42CrMo4 (AISI 4140).

Laser-based powder bed fusion of metals (PBF-LB/M) is a widely applied additive manufacturing technique. Thus, PBF-LB/M represents a potential candidate for the processing of quenched and tempered (Q&T) steels such as 42CrMo4 (AISI 4140), as these steels are often considered as the material of choice for complex components, e.g., in the toolmaking industry. However, due to the presence of process-induced defects, achieving a high quality of the resulting parts remains challenging in PBF-LB/M. Therefore, an extensive quality inspection, e.g., using process monitoring systems or downstream by destructive or non-destructive testing (NDT) methods, is essential. Since conventionally used downstream methods, e.g., X-ray computed tomography, are time-consuming and cost-intensive, micromagnetic NDT measurements represent an alternative for ferromagnetic materials such as 42CrMo4. In this context, 42CrMo4 samples were manufactured by PBF-LB/M with different process parameters and analyzed using a widely established micromagnetic measurement system in order to investigate potential relations between micromagnetic properties and porosity. Using multiple regression modeling, relations between the PBF-LB/M process parameters and six selected micromagnetic variables and relations between the process parameters and the porosity were assessed. The results presented reveal first insights into the use of micromagnetic NDT measurements for porosity assessment and process parameter optimization in PBF-LB/M-processed components.

Enhancing Feature Extraction in Sensor Fault Detection

The large amount of data provided by sensors and sensor networks deployed in industrial plants presents an opportunity for advancing process monitoring and informed asset maintenance management. Data-driven anomaly detection systems can be developed for both process anomalies and monitoring systems but the approach must be tailored to effectively capture and categorize relevant features, establishing a robust approach. The use of multi-source signals facilitates the retention of maximal information without reducing the number of available features, as sensor signals may show high degrees of correlation even if the sensors measure different variables across different equipment. The scope of the paper is to execute a preliminary analysis of industrial data, obtaining a logical clustering from a physical one using Canonical Correlation Analysis. This data mining methodology is employed to gather highly correlated data for training an Artificial Neural Network aimed at identifying sensor anomalies. A benchmark is provided by comparing the performance of the ANN trained using logical clusters rather than physical clusters. This comparison serves to evaluate the effectiveness and advantages of utilizing logical clusters for anomaly detection and classification.

Reluctance Based Actuator for Discrete Machine Hammer Peening

In machine hammer peening, an actuator system moves a ram in oscillating motion, which is oriented in angular position, preferably perpendicular, to a workpiece surface. On the tip of the ram, a hammer head with application specific shapes can be mounted. The hammer head is moved across the targeted workpiece surface area by means of a machine tool or an industrial robot. The impacts in defined patterns enable modifications of the surface integrity like smoothing the surface roughness or increasing fatigue strength by induced compressive residual stresses. This paper presents a novel DC-driven actuator system using a highly dynamic short-stroke linear reluctance motor to provide a time or trigger controlled hammer head movement including a system for process monitoring. In contrast to conventional systems, Deterministic Impact Peening (DIP) can be realized by position or speed related triggering of the impacts. The reluctance-based actuator with its geometrically optimized mechatronic system design provides performance characteristics favorable to known machine hammer peening systems. The position-controlled triggering of the hammer head is able to achieve uniform or defined local elastoplastic deformations on the workpiece surface, thus resulting in deterministic residual stress distributions.

Using photodiodes and statistical evaluation methods for process monitoring tasks during the laser beam welding of stainless-steel foils

Besides battery-based mobility solutions, polymer electrolyte membrane fuel cells (PEMFCs) can be utilized to electrify powertrains. Metallic bipolar plates, usually consisting of two thin micro-formed metal foils, are a crucial component of PEMFCs. For the joining of the foils, laser beam welding is the state-of-the-art manufacturing process. Within the scope of this work, a stable parameter window for the welding of AISI 316L was identified. Therefore, an adjustable ring-mode laser beam source in combination with a photodiode-based sensor system for process monitoring was applied. During the experiments, artificial process faults (e.g., contamination of the samples) were introduced to understand their impact on the welding process and the photodiode signals. The welded samples were evaluated using metallographic cross-sections and surface topography scans. Finally, a statistical signal analysis was employed to interpret the data obtained from the photodiodes to distinguish between fault-free and defective weld seams based on specific signal features.

A Framework for Tool Condition Monitoring of Abrasive Waterjet Systems

The study presents an innovative method for developing a waterjet mixing tube process monitoring system, leveraging aluminium oxide abrasives for accelerated data collection. A novel method of recording data during the jet dwell cycle is proposed, which eliminates abrasive from the data collection step and allows for the mixing tube wear state to be predicted before machining takes place. Airflow sensors were shown to be capable of detecting changes in mixing tube exit diameter. The performance of machine learning models, utilizing airflow data, was compared against simpler models using wear time as a sole parameter. Both approaches were tested in their ability to predict the exit diameter and to classify the state of the tool. Although wear time-based models outperformed machine learning models in this study, their ability to adapt to process changes may be limited. Machine learning models, given a larger dataset, may be required for accurate wear detection. The research lays a promising foundation for developing a robust mixing tube monitoring system. Future work should focus on collecting more data, investigating the effect of mixing chamber and orifice wear on the airflow signal, and evaluating model performance trained with accelerated wear data, on tubes worn in a regular wear trial.

Processing and structural health monitoring of a composite overwrapped pressure vessel for hydrogen storage

A process and Structural Health Monitoring system was implemented on a Composite Overwrapped Pressure Vessel (COPV) for hydrogen storage at 350 bar to be used in a fuel-cell system of an Unmanned Aerial Vehicle. This work reports the embedment strategy of optical fibre Bragg grating (FBG) sensors to monitor the full life cycle of the vessel, consisting of an aluminium liner and a wound carbon fibre reinforced polymer composite overwrap. A FBG sensing array, bonded on the aluminium liner circumferential section, was covered with a localised unidirectional prepreg composite tape, enabling composite winding and curing monitoring. The sensing array strategy allowed to detect and locate Barely Visible Impact Damage resulting from drop-weight impact tests, based on the ratio of the residual strain amplitude between FBG sensor pairs. Errors as small as 17 mm and up to 56 mm were determined between the predicted and ‘real’ impact locations. To simulate the real-life operational pressure charging and discharging cycles, the COPV was subjected to cycling testing at different pressure ranges. The FBG sensors were able to monitor a total of 20 980 pressure cycles, revealing a linear response to the applied pressure, and remained operational after COPV failure. Furthermore, the FBG sensing array was able to detect the residual plastic strain caused in the aluminium liner by the autofrettage process that the COPV was subjected to prior to pressure cycling, at 600 bar for 2 min, to improve its fatigue performance. This manuscript also reports the COPV structural design by Finite Element Modelling (FEM), its manufacturing process and burst pressure testing for the FEM analysis validation. A small difference of 0.7% was found between the simulated and experimental determined burst pressure of 1061 ± 26 bar.

Design of intelligent manufacturing IoT sensing system for polymer process monitoring

Design of intelligent manufacturing IoT sensing system for polymer process monitoring

New approach to multivariate statistical process control and its application to wastewater treatment process monitoring

This paper presents a new process monitoring and fault diagnosis approach based on a modified Multivariate Statistical Process Control (MSPC) and evaluates its applicability to municipal wastewater treatment process monitoring. A conventional MSPC based on Principal Component Analysis (PCA) is firstly adjusted to have an easy-to-understand user interface and then a new yet simplified decomposable diagnostic model is introduced. The developed user interface is designed to seamlessly connect MSPC to existing process monitoring system adopting the so-called trend graphs. The proposed diagnostic model is derived in a constructive way by aggregating small size models with one input or two inputs to improve tractability of the diagnostic model. The effectiveness of the modified MSPC is illustrated through some offline and on-line experiments by using a set of real multivariate process data at a municipal wastewater treatment plant.