On-line Quality Monitoring Research Articles

Effect of laser parameters and shielding gas flow on co-axial photodiode-based melt pool monitoring signals in laser powder bed fusion

Melt pool monitoring using co-axial photodiodes is increasingly used for online quality monitoring in PBF-LB AM. However, the fundamental correlations between different processing conditions and the co-axial photodiode signals are not well understood. In this study the impact of laser parameters and the shielding gas flow speed on co-axial photodiode-based melt pool monitoring signals is established. It is shown that laser power has positive linear correlation with the photodiode signal, while the correlation with scanning speed is non-linear in the form of y=ax–b. The focal point position, scanning orientation and shielding gas flow speed have highly non-linear response on the photodiode signal, depending on the combinatorial effects of the parameters. These underlying physical correlations should be carefully assessed and taken into consideration when trying to establish correlations between photodiode-based melt pool monitoring signal and defect formation in the PBF-LB process.

Vision-Based Online Molten Pool Image Acquisition and Assessment for Quality Monitoring in Gas–Metal Arc Welding

In gas–metal arc welding (GMAW), the welding quality is influenced by various factors, including the welding conditions and external environment. Achieving a high welding quality requires careful observation and control. Welding monitoring is crucial for validating a planned welding process and assessing its quality. This study introduces a vision system for the online acquisition of clear molten pool images. It uses a small camera equipped with a complementary metal-oxide-semiconductor sensor during GMAW, as well as a method for constructing this system. The proposed vision system does not require triggering using devices such as field-programmable gate arrays, facilitating lower costs and online quality monitoring under various welding conditions during mild-steel GMAW. Accurate measurement of radiation from the weld is essential for clear observation of the molten pool during welding, emphasizing the importance of selecting an optical system with appropriate lenses and filters. In addition, appropriate digital camera parameters must be set for monitoring. High-quality images were successfully captured using the proposed method, and the image quality was evaluated using the blind/referenceless image spatial quality evaluator algorithm.

Autoencoder-Based Anomaly Detection System for Online Data Quality Monitoring of the CMS Electromagnetic Calorimeter

The CMS detector is a general-purpose apparatus that detects high-energy collisions produced at the LHC. Online data quality monitoring of the CMS electromagnetic calorimeter is a vital operational tool that allows detector experts to quickly identify, localize, and diagnose a broad range of detector issues that could affect the quality of physics data. A real-time autoencoder-based anomaly detection system using semi-supervised machine learning is presented enabling the detection of anomalies in the CMS electromagnetic calorimeter data. A novel method is introduced which maximizes the anomaly detection performance by exploiting the time-dependent evolution of anomalies as well as spatial variations in the detector response. The autoencoder-based system is able to efficiently detect anomalies, while maintaining a very low false discovery rate. The performance of the system is validated with anomalies found in 2018 and 2022 LHC collision data. In addition, the first results from deploying the autoencoder-based system in the CMS online data quality monitoring workflow during the beginning of Run 3 of the LHC are presented, showing its ability to detect issues missed by the existing system.

Near-Infrared Spectroscopy and Chemometrics for Effective Online Quality Monitoring and Process Control during Pelagic Fishmeal and Oil Processing.

Near-infrared spectroscopy has become a common quality assessment tool for fishmeal products during the last two decades. However, to date it has not been used for active online quality monitoring during fishmeal processing. Our aim was to investigate whether NIR spectroscopy, in combination with multivariate chemometrics, could actively predict the changes in the main chemical quality parameters of pelagic fishmeal and oil during processing, with an emphasis on lipid quality changes. Results indicated that partial least square regression (PLSR) models from the NIR data effectively predicted proximate composition changes during processing (with coefficients of determination of an independent test set at RCV2 = 0.9938, RMSECV = 2.41 for water; RCV2 = 0.9773, RMSECV = 3.94 for lipids; and RCV2 = 0.9356, RMSECV = 5.58 for FFDM) and were successful in distinguishing between fatty acids according to their level of saturation (SFA (RCV2=0.9928, RMSECV=0.24), MUFA (RCV2=0.8291, RMSECV=1.49), PUFA (RCV2=0.8588, RMSECV=2.11)). This technique also allowed the prediction of phospholipids (PL RCV2=0.8617, RMSECV=0.11, and DHA (RCV2=0.8785, RMSECV=0.89) and EPA content RCV2=0.8689, RMSECV=0.62) throughout processing. NIR spectroscopy in combination with chemometrics is, thus, a powerful quality assessment tool that can be applied for active online quality monitoring and processing control during fishmeal and oil processing.

AF-FTTSnet: An end-to-end two-stream convolutional neural network for online quality monitoring of robotic welding

Online welding quality monitoring (WQM) is crucial for intelligent welding, and deep learning approaches considering spatiotemporal features for WQM tasks show great potential. However, one of the important challenges for existing approaches is to balance the spatiotemporal representation learning capability and computational efficiency, which makes it challenging to adapt welding processes with complex and drastic molten pool dynamic behavior. This paper proposes a novel approach for WQM using molten pool visual sensing and deep learning considering spatiotemporal features, the proposed deep learning network called attention fusion based frame-temporality two-stream network (AF-FTTSnet). Firstly, a passive vision sensor is used to acquire continuous dynamic molten pool images. Meanwhile, temporal difference images are computed to provide novel features and temporal representations. Then, a two-stream feature extraction module is designed to concurrently extract rich spatiotemporal features from molten pool images and temporal difference images. Finally, an attention fusion module with the ability to automatically identify and weight the most relevant features is designed to achieve optimal fusion of the two-stream features. The shop welding experimental results indicate that the proposed AF-FTTSnet model can effectively and robustly recognize five typical welding states during helium arc welding, with an accuracy of 99.26%. This model has been demonstrated to exhibit significant performance improvements compared to mainstream temporal sequence models. Available: https://github.com/Just199806/TSCNN/tree/master.

Online monitoring of air quality using PCA-based sequential learning

Online monitoring of air quality using PCA-based sequential learning

Development of descriptive analysis module within a cloud-based quality analyser: Case study in a guitar industry

Along with the revolution of the manufacturing world, companies face a new challenge to continuously improving their production processes to meet the escalating demand for high-quality goods in a fiercely competitive market. The previous concept of Cloud-Based Quality Analyzer (CQA) can be used as a real-time monitoring software that provides information of the on-going process. This concept can be implemented as an effort to achieve high quality manufacturing. This study aimed to develop the descriptive analysis module to realise the concept of CQA and evaluate its ability in realising online quality monitoring process. By employing the waterfall methodology, this study developed and implemented the descriptive analysis module in the CQA environment. The module was implemented in a case study in guitar manufacturing. The result showed that this module worked perfectly in displaying the multivariate process control chart and successfully gave the warning when some processes were in out-ofcontrol state. Furthermore, the user acceptance test also showed a positive response from the users. The descriptive analysis module was believed able to enhance the quality of manufacturing process while reducing the dependencies to the human quality engineers. However, more case studies in various industries were required to evaluate the benefit of implementing this module.

Digital twin based online material defect detection for CNC-milled workpieces

Reliable lot size one capable online quality monitoring for CNC machined parts remains elusive. To address this challenge, the proposed approach aims to bridge the current gap in research by developing a cost-effective and reference-independent monitoring concept for material defect detection in CNC-machined parts. This paper presents a novel digital twin-based method, utilising machining vibrations and a g-code-based encoding of the cutting process. The objective is to detect material defects, such as blowholes, without the need for individual workpiece references. The proposed method aims to reduce barriers to entry, minimise waste, and enhance machine productivity by enabling automated early online quality control. To develop and validate the model, a dataset combining machining vibration with technological context data such as chip-shape is generated. The feasibility and potential of the approach is demonstrated in a job shop setting on a 3-axis CNC mill.

Analysis of Water Pollution in The Musi River Due to Community and Industrial Activities Using Onlimo

The Musi River is a natural resource which is one of the main trade routes and the largest air supplier for the people of South Sumatra. However, the condition of the Mus River has undergone changes due to dense settlements and industry. Water quality is an important factor in the sustainability of an ecosystem. Ecosystem damage can start from declining air quality which results in industrial waste that pollutes the Musi River. Water quality measurement parameters include measurements of oxygen, TSS, COD, BOD. Observation station sampling was carried out using Onlimo telemetry technology. Onlimo is a remote, online and real-time air quality monitoring technology developed by BPPT. Sampling was carried out during the transition season on the Musi River which was divided into station 1 located in the Musi River (PDAM Tirta Randik area, Balui Air Unit, Sanga Desa District). On May 10, 2023 there was a pollution index of 7.32 with moderate contamination with the critical parameters Chemical Oxygen Demand (COD) and Biological Oxygen Demand (BOD). There was an increase in the Pollution index of 7.51 due to an increase in Nitrate from 16 May 2023 to 17 May 2023 due to input from anthropogenic waste caused by differences in sampling time, besides that season and rainfall are reported to play an important role in dissolving a number of components in the waters. For this reason, to prevent and improve the function of the Musi River, namely the government must act decisively on violations committed by factories or companies or the public who dispose of waste carelessly

Online cloud management and quality monitoring of the whole process of undergraduate thesis

Abstract The undergraduate thesis is an important practical teaching part of the undergraduate teaching program, and it is important to manage its quality. In this paper, an online cloud management system for the thesis is designed according to the requirement analysis, and the ARIMA model in time series is used to predict the number of thesis supervisors by the supervisor and give a reminder for the supervisor’s guidance decision. For the risk management of dissertation progress, a Bayesian network-based assessment method is proposed. From the perspective of dissertation similarity, a process of dissertation quality analysis based on the SimHash algorithm is constructed. The average error rate of ARIMA is only 0.190, which is optimized to a certain extent compared with other algorithms. 20 samples of the inference operation results have no sample with the probability of the dissertation progress delay level at High below 40%.

Detection of apple moldy cores using transmittance spectroscopy combined with supervised classifier comparison and parameter optimization

It is difficult to determine which apples have moldy cores just by looking at the outside of the apple. In the present study, we investigated identifying moldy cores using near-infrared transmittance spectra. First, input spectral features selected by noise adjusted principal component analysis (NAPCA) for back propagation artificial neural network (BP ANN) was used to reduce the dimensions of the original data. Then, four factors and five levels uniform design of the input nodes, training functions, transfer layer functions and output layer functions for NAPCA-BP ANN optimization is proposed. And the original data were input into NAPCA-BP ANN to obtain the recognition accuracy and NAPCA-support vector machine (SVM) was as a comparative recognition model. The results showed that through the uniform design-based NAPCA-BP ANN optimization, the NAPCA method had higher identification accuracy, precision, recall and F1 score, than either full spectrum or principal component analysis. Being assessed by different ratio of model test, functions in the hidden layer and output layer of NAPCA-BP ANN, the proposed method achieved the best accuracy to 98.03%. The accuracy, precision, recall and F1 score based on NAPCA-BP ANN were 3.92%, 2.86%, 2.78% and 2.82% higher than those based on NAPCA-SVM, respectively. This method provides a theoretical basis for the development of on-line monitoring of the internal quality of apples.

Quantitative Analysis of Steel Alloy Elements Based on LIBS and Deep Learning of Multi-Perspective Features

The Si and Mn contents in steel alloys are important characteristic indexes that influence the plasticity and welding properties of these alloys. In this work, the quantitative analysis methods for trace elements under complex alloy matrices by laser-induced breakdown spectroscopy (LIBS) are studied, which provide a foundation for utilizing LIBS technology in the rapid online detection of steel alloy properties. To improve the quantitative analysis accuracy of LIBS, deep learning algorithm methods are introduced. Given the characteristics of LIBS spectra, we explore multi-perspective feature extraction and backward differential methods to extract the spatio-temporal characteristics of LIBS spectra. The Text Convolutional Neural Network (TextCNN) model, combined with multi-perspective feature extraction, displays good stability and lower average relative errors (6.988% for Si, 6.280% for Mn) in the test set compared to the traditional quantitative analysis method and deep neural network (DNN) model. Finally, the backward differential method is employed to optimize the two-dimensional LIBS spectral input matrix, and the results indicate that the average relative errors of Si and Mn elements in the test set decrease to 5.139% and 3.939%, respectively. The method proposed in this work establishes a theoretical basis and technical support for precise prediction and online quality monitoring.

In-situ crack and keyhole pore detection in laser directed energy deposition through acoustic signal and deep learning

Cracks and keyhole pores are detrimental defects in alloys produced by laser directed energy deposition (LDED). Laser-material interaction sound may hold information about underlying complex physical events such as crack propagation and pores formation. However, due to the noisy environment and intricate signal content, acoustic-based monitoring in LDED has received little attention. This paper proposes a novel acoustic-based in-situ defect detection strategy in LDED. The key contribution of this study is to develop an in-situ acoustic signal denoising, feature extraction, and sound classification pipeline that incorporates convolutional neural networks (CNN) for online defect prediction. Microscope images are used to identify locations of the cracks and keyhole pores within a part. The defect locations are spatiotemporally registered with acoustic signal. Various acoustic features corresponding to defect-free regions, cracks, and keyhole pores are extracted and analysed in time-domain, frequency-domain, and time-frequency representations. The CNN model is trained to predict defect occurrences using the Mel-Frequency Cepstral Coefficients (MFCCs) of the lasermaterial interaction sound. The CNN model is compared to various classic machine learning models trained on the denoised acoustic dataset and raw acoustic dataset. The validation results shows that the CNN model trained on the denoised dataset outperforms others with the highest overall accuracy (89%), keyhole pore prediction accuracy (93%), and AUC-ROC score (98%). Furthermore, the trained CNN model can be deployed into an in-house developed software platform for online quality monitoring. The proposed strategy is the first study to use acoustic signals with deep learning for insitu defect detection in LDED process.

Real-Time Quality Monitoring of Ultrathin Sheets Edge Welding Based on Microvision Sensing and SOCIFS-SVM

Ultrathin sheets edge welding is a critical technology for aircraft, aerospace, or much high-end equipment manufacturing industrial applications. However, its online quality monitoring is still challenging due to the complicated and multicoupled transport phenomena in the mesoscale molten pool. This article presents an intelligent methodology for real-time monitoring of edge welds forming quality using microvision sensing and support vector machine based on swarm optimized and computationally inexpensive floating selection (SOCIFS-SVM). The clear images of the mesoscale molten pool are continuously acquired by using the passive microvision sensor. Based on the analysis for the dynamic behavior of the mesoscale molten pool and the morphology of the formed edge welds, a robust image processing procedure is developed to extract the single-frame spatial features and interframe correlation features from the microvision image sequence of the molten pool. A Kalman filter is applied to suppress the noise caused by the pulse current, etc. Then, the SOCIFS-SVM-based approach is established to predict three typical types of welding states. Experimental results with multiple groups of different welding parameters show that the proposed method can effectively and robustly identify the lack of fusion, humping, and sound weld with the highest accuracy of 96.09%. In addition, the established model can still achieve mean accuracy of 94.98% for additional untrained data, indicating that the model has strong generalization ability. The proposed real-time quality monitoring method makes online defect diagnosis of ultrathin sheets edge welding possible and provides a basis for online quality closed-loop control.

Online visual monitoring method for liquid rocket engine nozzle welding based on a multi-task deep learning model

Online quality monitoring is important in liquid rocket engine nozzle (LREN) welding procedures. Convolutional neural networks (CNNs) are widely applied in welding quality monitoring. However, most CNNs perform single visual tasks and provide limited information. They are not competent in LREN welding quality monitoring. This paper presents an online visual sensing method for LREN welding quality monitoring based on a multi-task deep learning model. Weld-detection-segmentation network (WDS-net), a multi-task CNN for simultaneous detection and segmentation is proposed for weld defect detection and weld width measurement. WDS-net obtains more information from welding images and depicts the welding process more adequately compared with single-task models. A machine vision system is constructed to monitor the LREN welding process. Then, experiments are conducted to obtain LREN welding images and construct a dataset. WDS-net is trained and tested on the dataset and obtains a 95.85% mean average precision (mAP) for weld defect detection and a 92.63% intersection over union (IoU) for weld zone segmentation. Finally, WDS-net is tested on LREN welding image sequences, and the results are compared with the offline measurement results of corresponding welds. WDS-net detects, locates the defects without omission, and measures weld width with a mean error within 0.1 mm compared with the actual weld. WDS-net’s monitoring performance and its average inference rate of 147 frames per second meet the requirements of online monitoring of LREN welding.

Design of On-line Monitoring System for Photovoltaic Power Generation Power Quality Based on Stm32

Monitoring the quality of photovoltaic power generation in remote mountain areas is difficult, so this paper proposes a real-time online monitoring system to solve the problem by using AD7606 high-precision ADC chip and STM32 single-chip microcomputer. AD7606 realizes real-time acquisition and conversion of three-phase power data. STM32 performs wavelet transformation on the collected data to obtain power quality parameters. It stores the power quality parameters in the SD card and then uploads the power quality parameters to the host computer through Lora to display each power quality parameter in real time. Experiments show that the system can reliably realize on-line real-time monitoring of the power quality of photovoltaic power generation.

Monitoring sidewall tilting of pixelated nanogratings in 3D display.

Sidewall tilting is an important parameter to describe the grating morphology and would affect the diffraction efficiency of three-dimensional (3D) display devices based on pixelated nanogratings. However, there is currently a lack of a non-destructive measurement method that can accurately measure the sidewall tilting of the pixelated nanogratings. This is mainly because the kind of nanograting is manufactured in a micron-scale pixel region and the grating lines generally have various directions to ensure that the display device can display images smoothly. In this work, we propose to use a home-made imaging Mueller matrix ellipsometer (IMME) to monitor sidewall tilting of pixelated nanogratings. Simulation and experiments were carried out to characterize the sidewall tilting angle. Through the combination of Mueller matrix elements, we can quickly and qualitatively identify the tilting angle for the purpose of on-line quality monitoring of the device. Through the inverse calculation of the Mueller matrix, we can accurately and quantitatively obtain the value of the tilting, so as to meet the demands of the device design. It is expected the proposed method can provide guidance for the identification and detection of tilting in 3D display elements based on pixelated gratings.

A review of tensile, hardness, wear, and microstructural characteristics of friction stir welded/processed composites

In this review paper friction stir welding (FSW) of various aluminum matrix composites (AMCs) is highlighted. The use of traditional fusion welding processes such as the TIG/MIG process for joining of AMCs results in the formation of various defects such as porosity, solidification cracks, etc. such defects is formed mainly due to the melting of parent metals due to high heat supply during fusion welding. Such challenges could be overcome by using a solid-state welding process such as FSW, a novel joining process capable to obtain good-quality joints. In this paper, various AMCs joined by FSW are highlighted, especially emphasis given on the tensile, hardness, wear, and microstructural characteristics of AMCs joined by FSW. It is noted that tool material, tool pin profile, type of reinforcement, and selection of process parameters play a significant role in obtaining good-quality joints. FSW microstructure exhibits a homogeneous distribution of reinforcement as a result better joint efficiency is reported by different researchers. However, the integration of FSW with advanced technology tools such as artificial intelligence (AI) and machine learning (ML) is necessary for online quality inspection and monitoring. Such tools will help for obtaining better-quality joints by avoiding the formation of defects in FSW.

Predictive ratio CUSUM (PRC): A Bayesian approach in online change point detection of short runs

The online quality monitoring of a process with low volume data is a very challenging task and the attention is most often placed in detecting when some of the underline (unknown) process parameter(s) experience a persistent shift. Self-starting methods, both in the frequentist and the Bayesian domain aim to offer a solution. Adopting the latter perspective, we propose a general closed-form Bayesian scheme, where the testing procedure is built on a memory-based control chart that relies on the cumulative ratios of sequentially updated predictive distributions. The theoretic framework can accommodate any likelihood from the regular exponential family and the use of conjugate analysis allows closed form modeling. Power priors will offer the axiomatic framework to incorporate into the model different sources of information, when available. A simulation study evaluates the performance against competitors and examines aspects of prior sensitivity. Technical details and algorithms are provided as supplementary material.

Failure quantitative assessment approach to MOSFET power device by detecting parasitic parameters

With the emerging wide bandgap (WBG) semiconductor development, the increasing power density and efficiency of power electronic converters may cause more switching oscillation, electromagnetic interference noise, and additional power loss, further increasing the probability of device failure. Therefore, determining and quantifying the failure of a metal-oxide-semiconductor-field-effect transistor (MOSFET), which assembled using WBG semiconductor in some applications, is crucial to improving the reliability of a power converter. This study proposes a novel failure quantitative assessment approach based on MOSFET parasitic parameters. According to the two-port network theory, MOSFET is equivalent to some second-order RLC circuits composed of independent inductances, capacitances, and resistances in series. Then, the frequency-domain impedance associated with the physical failure of MOSFET is identified through frequency domain reflectometry. Accelerated aging and bond wires cut-off experiments are employed to obtain various quality states of the MOSFET device. Result shows that the MOSFET quality level and its number of bond wire lift-offs can be quantified effectively. Drain-to-source on-resistance (RDS(on)) that normally represents the MOSFET quality shows a positive linear function relationship on drain-to-source parasitic resistance (RD + RS) during the quality degradation proceeding. This finding matches with the correlation established between RDS (on) and RD + RS in theory. Meanwhile, source parasitic inductance (LS) increases with the severity of bond wires faults, and even the slight fault shows a high sensitivity. The proposed approach would be an effective quality screening technology for power semiconductor devices without power on treatment, which can effectively avoid the impact of junction temperature and test conditions (current and voltage) on test results, and does not need to design additional test circuits. The test frequency range we used in this approach was 10–300 MHz, which to some extent is suitable for providing an on-line quality monitoring technology for high-frequency WBG power devices manufacturing.