Control Chart Pattern Recognition Research Articles

Control chart pattern recognition for small samples based on Siamese Neural Network

Quality control for small samples is growing in popularity due to its broad application prospects and high research value, and one of the commonly used methods for quality control nowadays is control chart pattern recognition (CCPR). Deep learning is an effective way for CCPR, but its recognition rate of control chart patterns for small samples is low and the quality control capability is limited. Therefore, to solve these problems, this paper first proposes a Perceptron-Convolutional Siamese Neural Network (PCSNN) model to improve the pattern recognition rate of small sample control chart. And then, three pattern recognition models (Convolutional Neural Network, Perceptron-Convolutional Neural Network, and PCSNN) are compared to verify the validity of the proposed model. After that, simulation experiments are conducted, and results show that the proposed PCSNN model has a high recognition rate of control chart patterns for small samples and performs well when the sample quality parameters change within a certain range. Finally, the proposed model is applied to the manufacturing process of automatic boring of enterprise gearbox shell, and its validity is verified.

An imbalance-aware BiLSTM for control chart patterns early detection

Digital twins-based predictive models find their roots in smart manufacturing. However, their potential applications to control chart pattern recognition (CCPR) algorithms, which lie at the heart of advanced fault detection systems, remain underexplored. A key challenge in CCPR models arises from the intrinsic imbalance between classes, which can compromise the model’s performance if left untreated. Further, existing CCPR models are often trained over simulated control chart data in which abnormal patterns are generated separately from abnormal signals; the resulting classifiers, however, perform poorly in the early detection of abnormalities in real-time production environments. To address these challenges, we develop a cost-sensitive, bi-directional long short-term memory neural network for data sequences with mixed normal and abnormal signals. We further introduce a novel adaptive weighting strategy for data generation by enforcing the rates of abnormal signals within mini-batch distributions. Our model benefits from a new bi-objective early stopping technique, which optimally balances loss minimization and G-mean maximization for model training. Finally, we introduce a novel rolling window-based metric for evaluating CCPR classifier stability. We conduct a comprehensive experimental study of our model using both simulated data and two real-world datasets collected from biomanufacturing and wafer industries. The results of our study consistently demonstrate the superiority of our proposed stopping technique over traditional methods. Our experiments further show the effectiveness of our proposed model in maintaining the classifier stability and specifying optimal process monitoring window length within the datasets.

Research on the whole process monitoring and management of coal digital production based on IOT technology

Abstract This paper firstly explores the method and monitoring process of coal SPC production quality monitoring based on IOT technology and simulates different control chart patterns using the Monte Carlo method. Then a control chart pattern recognition model based on PCA-PSO-SVM is established to downscale the collected data using the PCA method and quickly discriminate whether the control chart is abnormal or not based on the OC-SVM algorithm. By identifying abnormal control chart patterns, real-time monitoring of the coal production process is realized. Finally, comparative analysis and fault source analysis are conducted to verify the sensitivity and effectiveness of the monitoring method in this paper. The difference in methane concentration between the experimental group and the comparison group is less than 10ppm, CO concentration is less than 5ppm, and dust concentration is less than 0.05mg/m3. It shows that the method of this paper is very effective for coal production monitoring.

A Novel Scheme of Control Chart Patterns Recognition in Autocorrelated Processes

Control chart pattern recognition (CCPR) can quickly recognize anomalies in charts, making it an important tool for narrowing the search scope of abnormal causes. Most studies assume that the observations are normal, independent and identically distributed (NIID), while the assumption of independence cannot always be satisfied under continuous manufacturing processes. Recent research has considered the existence of autocorrelation, but the recognition rate is overestimated. In this paper, a novel scheme is proposed to recognize control chart patterns (CCPs) in which the inherent noise is autocorrelated. By assuming that the inherent noise follows a first-order autoregressive (AR (1)) process, the one-dimensional convolutional neural network (1DCNN) is applied for extracting features in the proposed scheme, while the grey-wolf-optimizer-based support vector machine (GWOSVM) is employed as a classifier. The simulation results reveal that the proposed scheme can effectively identify seven types of CCPs. The overall accuracy is 89.02% for all the autoregressive coefficients, and the highest accuracy is 99.43% when the autoregressive coefficient is on the interval (−0.3, 0]. Comparative experiments indicate that the proposed scheme has great potential to identify CCPs in autocorrelated processes.

A novel autonomous machine learning technique for recognizing control chart patterns

A number of strategies have been developed for controlling and monitoring the production process since the quality of the product has emerged as one of the key concerns in today's manufacturing sector. Control charts are the best tools for monitoring and adjusting products and processes. This research proposes a novel automatic method for the recognition of nine control chart patterns (CCPs) based on novel autonomous machine learning. The classification portion and the tuning portion make up the two main components of this procedure. Support Vector Machine (SVM) have demonstrated outstanding performance over the past few years on a variety of applications, including signal dispensation, speech detection, and image processing. SVM is consequently employed as the intelligent classifier for the recognition of CCPs in the classification phase. One key challenge with SVM is that it requires a high level of expertise to choose appropriate parameters, such as the quantity of kernel and their spatial diameters, knowledge rate, etc. It is difficult to fine-tune the SVM parameters because of their domestic dependence. These problems led to the employment of the Harmony Search (HS) Algorithm for the best tuning of SVM parameters in the tuning section of the proposed technique. Instead of depending on any feature engineering procedures, the suggested method, in contrast to the popular CCPs recognition methods, takes raw data and runs it through many hidden layers to obtain the best feature representation. The quantitative and simulation results demonstrate the suggested method's performance advantage over the earlier methods.

An improved features selection approach for control chart patterns recognition

<span lang="EN-US">Control chart patterns (CCPs) are an essential diagnostic tool for process monitoring using statistical process control (SPC). CCPs are widely used to improve production quality in many engineering applications. The principle is to recognize the state of a process, either a stable process or a deterioration to an unstable process. It is used to significantly narrow the set of possible assignable causes by shortening the diagnostic process to improve the quality. Machine learning techniques have been widely used in CCPs. Artificial neural networks with multilayer perceptron (ANN-MLP) are one of the standard tools used for this purpose. This paper proposes an improved features selection method to select the best features as input representation for control chart patterns recognition. The results demonstrate that the proposed approach can effectively recognize CCPs even for small patterns with a mean shift of less than 1.5 sigma. The dimensional reduction was achieved by employing Relief, correlation, and Fisher algorithms (RCF) for feature selection and (ANN-MLP) as a classifier (RCF-ANN). This study provides an experimental result that compares the performance before and after dimensional reduction.</span>

Multivariate Process Control Chart Pattern Classification Using Multi-Channel Deep Convolutional Neural Networks

Statistical process control (SPC) charts are commonly used to monitor quality characteristics in manufacturing processes. When monitoring two or more related quality characteristics simultaneously, multivariate T2 control charts are often employed. Like univariate control charts, control chart pattern recognition (CCPR) plays a crucial role in multivariate SPC. The presence of non-random patterns in T2 control charts indicates that a process is influenced by one or more assignable causes and that corrective actions should be taken. In this study, we developed a deep learning-based classification model for recognizing control chart patterns in multivariate processes. To address the problem of the insufficient representation of one-dimensional (1D) data, we explore the advantages of using two-dimensional (2D) image data obtained from a threshold-free recurrence plot. A multi-channel deep convolutional neural network (MCDCNN) model was developed to incorporate both 1D and 2D representations of control chart data. This model was tested on multivariate processes with different covariance matrices and compared with other traditional algorithms. Moreover, the effects of imbalanced datasets and dataset size on classification performance were analyzed. Simulation studies revealed that the developed MCDCNN model outperforms other techniques in identifying multivariate non-random patterns. For the most significant one, our proposed MCDCNN method achieved a 10% improvement over traditional methods. The overall results suggest that the developed MCDCNN model can be beneficial for intelligent SPC.

Control chart pattern recognition for imbalanced data based on multi-feature fusion using convolutional neural network

As the most practical quality control process monitoring tool, control chart patterns (CCPs) can determine abnormal conditions in the production process. Therefore, automatic and accurate recognition of CCP is critical for improving production efficiency and quality in the manufacturing process. In the actual production process, the CCPs data has mixture patterns as well as single patterns, more importantly, the control chart pattern data proportion is imbalanced. Manual recognition is a costly endeavor, and automatic recognition is simpler and more effective. By applying the deep learning method to control chart pattern recognition (CCPR), abnormal patterns can be recognized in real-time. In this paper, the SMOTE method was used to balance the imbalances in data set, then some features in the control chart patterns were extracted and fused with the data. A method of abnormal control chart pattern recognition based on imbalanced data was proposed by using convolutional neural network. A comparison was made between the performance of the convolutional neural network and that of other methods in recognizing imbalanced patterns in control charts. According to the simulation results, compared with the raw data and other classifiers, the proposed multi-feature fusion and convolutional neural network (MFF-CNN) method has a better recognition performance for abnormal CCPs.

A novel fault early warning method for mechanical equipment based on improved MSET and CCPR

Prompt fault early warning is a significant approach to improve the reliability of mechanical equipment. An abnormal state recognition method based on improved multivariate state estimation technology (MSET) and control chart pattern recognition (CCPR) is proposed. First, the health data band (HDB) of equipment is established through a k-nearest neighbor algorithm. Meanwhile, a dynamic memory matrix construction approach based on a clustering algorithm is presented to improve the calculation efficiency of MSET. The CCPR is integrated to reduce limitations such as artificial interference in the traditional threshold method of fault early warning. The proposed method is applied into two cases of rolling bearing and control moment gyro so as to compare its performance with other MSET modeling methods. According to the results, the proposed method can detect equipment abnormalities hours to weeks in advance, and significantly reduce the false alarm rate and calculation time.

Ensemble Classifier for Recognition of Small Variation in X-Bar Control Chart Patterns

Manufacturing processes have become highly accurate and precise in recent years, particularly in the chemical, aerospace, and electronics industries. This has attracted researchers to investigate improved procedures for monitoring and detection of small process variations to remain in line with such advances. Among these techniques, statistical process controls (SPC), in particular the control chart pattern (CCP), have become a popular choice for monitoring process variance, being utilized in numerous industrial and manufacturing applications. This study provides an improved control chart pattern recognition (CCPR) method focusing on X-bar chart patterns of small process variations using an ensemble classifier comprised of five complementing algorithms: decision tree, artificial neural network, linear support vector machine, Gaussian support vector machine, and k-nearest neighbours. Before advancing to the classification step, Nelson’s Rus Rules were utilized as a monitoring rule to distinguish between stable and unstable processes. The study’s findings indicate that the proposed method improves classification performance for patterns with mean changes of less than 1.5 sigma, and confirm that the performance of the ensemble classifier is superior to that of the individual classifier. The ensemble classifier can distinguish unstable pattern types with a classification accuracy of 99.55% and an ARL1 of 11.94.

PCA-WA Based Approach for Concurrent Control Chart Pattern Recognition

PCA-WA Based Approach for Concurrent Control Chart Pattern Recognition

Control chart pattern recognition using spectral clustering technique and support vector machine under gamma distribution

Most studies on control chart patterns (CCPs) recognition assume that the measured values of process characteristic follow a normal distribution. However, by collecting large-size on-line datasets, it could be found that this assumption may not be tenable. In the present study, a method to combine spectral clustering technique with support vector machine (SVM) for recognizing CCPs under a gamma distribution, which is typically employed to represent various probability distributions, is proposed. Spectral clustering is a useful tool for segmenting the observation window to obtain the correct feature set. Multiple SVM classifiers are applied to recognize CCPs. Recognition accuracy and average run length (ARL) are indicators that are used to validate the CCP recognition ability of our proposed method. Based on the comparative studies, our proposed method in the recognition efficiency of most CCP types generally outperforms the Shewhart X¯ chart and EWMA chart with run rules. It is also noted that a more skewed underlying distribution always leads to a higher false alarm rate for our proposed method. A real-world case study of the tire manufacturing process is given to illustrate how our proposed method can be practically applied to identify and recognize the CCPs.

End-to-End Control Chart Pattern Classification Using a 1D Convolutional Neural Network and Transfer Learning

Control charts are an important tool in statistical process control (SPC). They have been commonly used for monitoring process variation in many industries. Recognition of non-random patterns is an important task in SPC. The presence of non-random patterns implies that a process is affected by certain assignable causes, and some corrective actions should be taken. In recent years, a great deal of research has been devoted to the application of machine learning (ML) based approaches to control chart pattern recognition (CCPR). However, there are some gaps that hinder the application of the CCPR methods in practice. In this study, we applied a control chart pattern recognition method based on an end-to-end one-dimensional convolutional neural network (1D CNN) model. We proposed some methods to generate datasets with high intra-class diversity aiming to create a robust classification model. To address the data scarcity issue, some data augmentation operations suitable for CCPR were proposed. This study also investigated the usefulness of transfer learning techniques for the CCPR task. The pre-trained model using normally distributed data was used as a starting point and fine-tuned on the unknown non-normal data. The performance of the proposed approach was evaluated by real-world data and simulation experiments. Experimental results indicate that our proposed method outperforms the traditional machine learning methods and could be a promising tool to effectively classify control chart patterns. The results and findings of this study are crucial for the further realization of smart statistical process control.

Control chart recognition based on the parallel model of CNN and LSTM with GA optimization

Quality control process has become one of the most critical issues in intelligent manufacturing. As the most practical and prevalent tools for continuously monitoring, control chart patterns (CCPs) can be automatically recognized to judge the fault of production process. However, there are three shortcomings in the previous researches. 1) Insufficient or superfluous features are considered; 2) Few studies simultaneously take the local features and time sequence information into account; 3) With the exception of empiricism, little work has been done on searching for the optimal hyper-parameters for the neural networks. Accordingly, FFR-GACLN, a method for ten CCPs recognition is proposed in this paper. Such a method is comprised of two sections including the fusion feature reduction (FFR) and GACLN model construction. Firstly, the feature extraction comprising of features fusion and feature dimensionality reduction by convolutional auto-encoder is applied to enhance the reliability and performance of the extracted features. Then, such features are input into GACLN, a network combining the convolutional neural network (CNN) and the long-short term memory (LSTM) for achieving the final recognition task. Moreover, genetic algorithm (GA) is applied to search for the optimal hyper-parameters for FFR-GACLN. The quantitative and simulation results demonstrate that the performance of the proposed method is superior over the previous techniques.

On-line concurrent control chart pattern recognition using singular spectrum analysis and random forest

A scheme integrated with engineer process control (EPC) is a useful approach to identify the disturbances for statistical process control (SPC) charting of an auto-correlated process. Although the EPC is able to compensate for the underlying disturbance, the underlying disturbance is embedded in the control chart. Thus, this makes the control chart patterns (CCPs) difficult to be identified. As documented in the literature, much efforts have been put in the recognition of single basic patterns of unnatural variation. However, there could exist a combination of two unnatural patterns simultaneously in a real-world manufacturing process. Also for an automated real-time production line, data are collected automatically and monitored by a computer-based system. Thus, the early detection of abnormality is of high importance for the aforesaid problem. This study aims to establish an on-line detection system used for monitoring the mixture unnatural CCPs for a SPC-EPC process. This paper presents a hybrid approach based on singular spectrum analysis (SSA) and random forest (RF) to identify the concurrent CCPs in an on-line SPC-EPC process. A total of fifteen types of concurrent CCPs were utilized to validate the proposed method. The SSA method was also used to decompose the mixture patterns into single patterns. The RF was employed to identify the types of patterns to which it belongs. The results showed that the proposed method was able to handle most of the concurrent CCPs very successfully with an average accurate identification rate of 91.8%. Also, the proposed method was found to be more accurate and efficient than the use of the hybrid method of SSA and support vector machine (SVM). It is suggested that this proposed system could be possibly applied for monitoring an on-line process to a greater extent.

Monitoring method for gasification process instability using BEE-RBFNN pattern recognition

To identify large fluctuations of parameters in the process of gasification, a control chart pattern recognition method based on optimized radial basis function neural network (RBFNN) is proposed in this paper to conduct pattern recognition of the parameters of gasification process. The proposed method is described in four aspects, feature description, feature extraction, classifier and algorithm learning and training. The initial data is described from shape features and statistical features to reduce the dimension of the data. The optimal characteristic set is selected by the association rule mining algorithm to reduce the complexity of the model. The performance of neural network is affected by the learning method. With RBFNN as a classifier, a learning method based on Bees algorithm is therefore put forward and then training recognition is compared between the proposed method and the traditional method. The results show that the proposed method has higher recognition rate and simpler structure than the traditional method, and it is very effective in monitoring and identifying abnormal fluctuations of gasification process parameters.

An integrated method for variation pattern recognition of BIW OCMM online measurement data

ABSTRACT In order to improve the quality of the body-in-white (BIW), optical coordination measurement machines (OCMM) are used to measure the dimensional variation for BIW. The big OCMM online measurement data with low signal-to-noise ratio makes the variation patterns recognition to be difficult and challenges the traditional statistical process control (SPC) technology and the common variation recognition approaches. In this paper, we propose an automatic and integrated method to recognise the control chart patterns (CCPs), which includes three main modules. The Jarque-Bera test is applied in the wavelet denoising module. The feature extraction module extracts a combination set of shape features and statistical features. In the classifier module, a two-hidden-layer Backpropagation neural network (BPN) is trained and tested. In the experiment, the proposed method is also compared with other CCPs recognition methods. Finally, a practice case is studied to show the application of the integrated method and validate the high recognition accuracy of the integrated system.

Multiscale Control Chart Pattern Recognition Using Histogram-Based Representation of Value and Zero-Crossing Rate

In the research of control chart pattern recognition (CCPR), most previous methods used a classifier to label abnormal CCPs. However, long-term control chart data often contains a large number of small abnormal patterns, with characteristics unlike those seen from a global view of the entire chart. There is also a high probability that local abnormal patterns are worthy of analysis. This article presents a novel multiscale control chart pattern recognition scheme, MS-CCPR, which does not focus on the classification of data from a single chart. Rather, the proposed scheme uses a proposed histogram-based data representation in conjunction with time series subsequence matching to identify abnormal patterns on various scales from a long series of control charts. Experimental results demonstrate the efficacy of the proposed framework in the efficient detection of chart patterns at various scales, outperforming the state-of-the-art time series subsequence matching algorithms.

Fuzzy Heuristics and Decision Tree for Classification of Statistical Feature-Based Control Chart Patterns

Monitoring manufacturing process variation remains challenging, especially within a rapid and automated manufacturing environment. Problematic and unstable processes may produce distinct time series patterns that could be associated with assignable causes for diagnosis purpose. Various machine learning classification techniques such as artificial neural network (ANN), classification and regression tree (CART), and fuzzy inference system have been proposed to enhance the capability of traditional Shewhart control chart for process monitoring and diagnosis. ANN classifiers are often opaque to the user with limited interpretability on the classification procedures. However, fuzzy inference system and CART are more transparent, and the internal steps are more comprehensible to users. There have been limited works comparing these two techniques in the control chart pattern recognition (CCPR) domain. As such, the aim of this paper is to demonstrate the development of fuzzy heuristics and CART technique for CCPR and compare their classification performance. The results show the heuristics Mamdani fuzzy classifier performed well in classification accuracy (95.76%) but slightly lower compared to CART classifier (98.58%). This study opens opportunities for deeper investigation and provides a useful revisit to promote more studies into explainable artificial intelligence (XAI).

A robust data simulation technique to improve early detection performance of a classifier in control chart pattern recognition systems

The quality control process is essential in maintaining the stability of production systems and proactively detecting abnormalities that may result in high mechanical and labor costs. In this study, a new data simulation strategy was devised to improve the early prediction performance (EPP) of an algorithm. Traditional control chart data simulation methods work based on creating abnormal patterns consisting of only abnormal signals. However, a classifier trained with data samples consisting of only abnormal data signals may fail to early detect abnormality in a real-time production line, in which abnormal signals is obscured by volume of normal signals. From this perspective, training a model by imitating real-world cases can improve the performance of an algorithm in terms of early detection of an abnormality. Normal and abnormal patterns were simulated by implementing a new approach called Mix Ratio Data Simulation (MRDS). The proposed methodology MRDS is compared with the customary data simulation method under the predefined scenarios in terms of EPP. The findings indicated that changing the way of simulating dataset increases the EEP of the machine-learning algorithm regardless of abnormality types and parameters.