Simultaneous Fault Detection and Classification for Semiconductor Manufacturing Tools

Detection and classification are two significant tasks for industrial fault diagnosis. However, conventional methods typically treat these tasks as separate and independent problems, and necessitate a retraining process when new fault samples or classes are collected. Therefore, an incremental support vector data description scheme using Gaussian kernel function is pro-posed for industrial process fault diagnosis in a unified frame-work. In this framework, the decision boundary is updated incrementally only based on the specific original support vectors and newly collected samples. An adaptive threshold and a restructured radius are proposed to promote accuracy in the fault detection. In the classification procedure, the hyperspheres for all known classes are constructed by decision tree. The new sample that does not belong to any known class is identified as an unknown class. Without a time-consuming retraining process, the proposed diagnosis method with the incremental learning capability can synchronously achieve the fault detection and classification task. Experimental results demonstrate the effectiveness and superiority in terms of diagnosis performance. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —In practical industrial processes, new fault samples are collected and new fault classes emerge continually. Under such scenarios, fault diagnosis with incremental learning capability is becoming increasingly important. This work proposes a unified incremental framework for fault diagnosis based on support vector data description, which is able to achieve fault detection and classification synchronously. For fault detection, an adaptive threshold and a restructured radius are developed. For fault classification, hypersphere-shaped boundaries of all fault classes are given via the decision tree-based strategy. The proposed method is updated to include new fault samples and fault classes without dimensionality reduction or any distributional assumptions.

Software fault detection is crucial for ensuring reliable and high-quality software systems. However, traditional fault detection methods often rely on manual inspection or rule-based techniques, which are time-consuming and prone to human errors. In this research, the researchers propose an enhanced fault detection approach using an adaptive neural transfer learning algorithm. The goal is to leverage the power of neural networks and adaptability to improve fault detection accuracy and classification performance. The problem addressed in this research is the need for more effective fault detection methods that can handle the complexities of modern software systems. Existing fault detection techniques lack adaptability and struggle to cope with diverse software scenarios. Neural networks have shown promise in pattern recognition and classification tasks, making them suitable for fault detection. However, fixed architectures and training strategy limit their performance in different software contexts. To address this problem, the research proposes an adaptive neural transfer learning algorithm for fault detection. The algorithm dynamically adjusts its neural network architecture and training process based on the characteristics of the software under test. It incorporates adaptive mechanisms, such as adjusting learning rates and regularization techniques, to optimize performance. Real-time feedback and performance evaluation during the training process drive the adaptive mechanisms. To evaluate the proposed approach, the researchers conducted a series of experiments using diverse software systems and fault scenarios. The research compared the performance of the adaptive algorithm with traditional fault detection methods, including rule-based techniques and fixed neural network architectures. Evaluation metrics such as accuracy, precision, recall, and F1 score were used. The results consistently show that the adaptive neural transfer learning algorithm outperforms existing methods, achieving higher fault detection accuracy and improved classification performance.

ABSTRACTFaults in power transmission systems pose significant challenges due to the complexity and length of transmission lines. Effective fault detection, classification and location are essential for preventing further damage to the power grid. While travelling wave‐based algorithms are commonly used for fault location, they often focus on identifying the fault's location without classifying the fault type. Accurate classification is crucial for enabling efficient and timely responses from protection systems. This paper introduces an integrated model for fault detection, classification and location using voltage signals from a single terminal of a series‐compensated transmission line with a static synchronous series compensator (SSSC). The Gabor Transform (GT) is utilised for feature extraction, enabling both fault detection and classification. Travelling wave theory is then applied to identify the faulty segment and estimate the fault location. Additionally, a novel technique adaptively calculates the threshold value during the protection algorithm's execution. The proposed method is validated through a comprehensive analysis of various fault scenarios and sensitivity analysis. Numerical simulations in MATLAB/Simulink show that the model achieves 100% accuracy for fault detection, classification, and faulty segment identification, with 99.7925% accuracy for fault location estimation, demonstrating its effectiveness in fault management.

In this paper, we address the problem of fault classification in PhotoVoltaic (PV) arrays using a semi-supervised graph signal processing approach. Traditional fault detection and classification methods require large amounts of labeled data for training. In utility scale solar arrays, obtaining labeled data for different fault classes is resource intensive. We propose a graph based classification technique that relies on a limited amount of labeled data. We compare our results with the well known supervised machine learning classifiers such as the K-nearest neighbour classifier, random forest classifier, support vector machines, and artificial neural networks. We also show that the graph-based classifiers require lower training computational cost compared to the standard supervised machine learning algorithms. The proposed method also achieves good classification performance with unseen data. We validate our method on a real-time dataset and show significant improvements over existing approaches.

In this paper, the fault detection problem is investigated for a class of wireless networked systems with compressed measurements. A new compressed fault detection framework is proposed. The measurement data is compressed with a compressed sensing technique before being transferred to the fault detection node via wireless communication channel. Besides, the influence of compressed measurements on fault detectability is analyzed and a new fault detection method is provided based on the integration of model-based and statistical methods. The residual generator is designed by the available compressed measurements and priori knowledge about the system, and the statistic testing method is developed to determine whether there is a fault. Maintaining the acceptable fault detection performance, the compressed fault detection framework can greatly reduce the cost of data acquisition, transmission and storage. As the compressed data is utilized directly in the fault detection process, our method can also reduce computational requirements and enhance the efficiency of the whole fault detection system. Finally, a simulation study is carried out to demonstrate the effectiveness and applicability of the proposed method.

The day-by-day increase in digital loads draws attention towards the need for an efficient and compatible distribution network. An LVDC distribution network has the capability to fulfill such digital load demands. However, the major challenge of an LVDC distribution network is its vulnerability during a fault. The need for a high-speed fault detection method is inevitable before it can be widely adopted. This paper proposes a new fault detection method which extracts the features of the current during a fault. The proposed fault detection method uses the merits of overcurrent, the first and second derivative of current, and signal processing techniques. Three different features are extracted from a time domain current signal through a sliding window. The extracted features are based upon the root squared zero, second, and fourth order moments. The features are then set with individual thresholds to discriminate low-, high-, and very high-resistance faults. Furthermore, a fault is located through the superimposed power flow. Moreover, this study proposes a new method based on the vector sum of positive and negative pole currents to identify the faulty pole. The proposed scheme is verified by using a modified IEEE 13 node distribution network, which is implemented in Matlab/Simulink. The simulation results confirm the effectiveness of the proposed fault detection and identification method. The simulation results also confirm that a fault having a resistance of 1 m Ω is detected and interrupted within 250 μ s for the test system used in this study.

The self-attention mechanism has advantages in analyzing the internal characteristics of data and capturing local information. Therefore, it is applied to fault detection. Since complex industrial processes usually contain a mixture of properties, such as nonlinearity, dynamics, and non-Gaussianity, this places higher demands on fault detection methods. To further improve the performance of the self-attention mechanism for fault detection in complex industrial processes, an innovative self-attention mechanism integrating the adaptive double subspace (ISA-ADS) method is proposed for fault detection. First, the query matrix, the key matrix, and the value matrix of the self-attention mechanism are designed so that they can better focus on local information through the adaptive sample weight allocation strategy. At the same time, this design reduces the interference of nonlinear and noisy information and amplifies the impact of important fault information in the sample. Second, considering the problem of Gaussian and non-Gaussian distributions of the data and the different degrees of influence of different subspaces on the final process monitoring results, an adaptive Gaussian and non-Gaussian double subspace fault detection model based on the self-attention output matrix is established, which adaptively assigns different weights to different subspaces by determining their importance. The statistics of different subspaces are fused using the Bayesian inference and a new adaptive monitoring statistic is constructed to obtain a more accurate monitoring of the process state. Finally, the ISA-ADS is validated using the Tennessee Eastman (TE) process and the actual polyethylene production process. The results show that ISA-ADS has a low false alarm rate and a high fault detection rate, verifying the effectiveness of its fault detection performance.

This paper studies the simultaneous fault detection and control problem for linear uncertain discrete-time systems. A design procedure dealing with polytopic uncertainties is proposed for residual generation, where both the fault detection objectives and some control objectives are considered simultaneously through certain performance indexes. To cope with the fault sensitivity performance index in a given frequency range directly, the generalised KYP lemma is introduced in this framework. A two-step procedure is adopted to obtain the solutions through satisfying a set of linear matrix inequalities. A numerical example is given to illustrate the effectiveness of the proposed methods.

This research proposes an application of generative adversarial networks (GANs) to solve the class imbalance problem in the fault detection and classification study of a plasma etching process. Small changes in the equipment part condition of the plasma equipment may cause an equipment fault, resulting in a process anomaly. Thus, fault detection in the semiconductor process is essential for success in advanced process control. Two datasets that assume faults of the mass flow controller (MFC) in equipment components were acquired using optical emission spectroscopy (OES) in the plasma etching process of a silicon trench: The abnormal process changed by the MFC is assumed to be faults, and the minority class of Case 1 is the normal class, and that of Case 2 is the abnormal class. In each case, additional minority class data were generated using GANs to compensate for the degradation of model training due to class-imbalanced data. Comparisons of five existing fault detection algorithms with the augmented datasets showed improved modeling performances. Generating a dataset for the minority group using GANs is beneficial for class imbalance problems of OES datasets in fault detection for the semiconductor plasma equipment.

In response to the challenges posed by the dynamic and multi‐stage characteristics of complex multi‐batch processes in process monitoring, the Baseline Standardization (BS) strategy is proposed, which, together with Multiway Principal Component Analysis (MPCA), forms the BS‐MPCA fault detection method. BS constructs a baseline batch that changes over time based on the mean of samples at the same time point from different batches. Due to inevitable fluctuations in the baseline batch, a Gaussian filter is applied to smooth the baseline batch in order to enhance its robustness. The dataset is standardized based on the standard deviation of samples at the same time point in different batches and the baseline batch, resulting in the standardized dataset. BS scales the data to a unified scale, which eliminates the dynamic and multi‐stage characteristics of the process. The resulting standardized dataset meets the assumption of independent and identically distributed statistics and . Based on this, the MPCA method is used for feature extraction and fault detection. BS‐MPCA is applied to fault detection in numerical processes and the Three‐tank process. Experimental results show that compared to MPCA, DPCA, MDPCA, VBSPCA, KNN, and WKNN methods, the BS‐MPCA fault detection method achieves a higher fault detection rate. Additionally, it eliminates the dynamic and multi‐stage characteristics of the process during the modeling phase, and its detection model outperforms the models of the comparison methods. Theoretical analysis and simulation experiments show that the BS‐MPCA method is suitable for fault detection in multi‐batch processes, whether they have simultaneous or individual dynamic and multi‐stage characteristics. It more effectively ensures the safety of the production process and the high quality of the products.

Accurate fault classification and detection for the microgrid (MG) becomes a concern among the researchers from the state-of-art of fault diagnosis as it increases the chance to increase the transient response. The MG frequently experiences a number of shunt faults during the distribution of power from the generation end to user premises, which affects the system reliability, damages the load, and increases the fault line restoration cost. Therefore, a noise-immune and precise fault diagnosis model is required to perform the fast recovery of the unhealthy phases. This paper presents a review on the MG fault diagnosis techniques with their limitations and proposes a novel discrete-wavelet transform (DWT) based probabilistic generative model to explore the precise solution for fault diagnosis of MG. The proposed model is made of multiple layers with a restricted Boltzmann machine (RBM), which enables the model to make the probability reconstruction over its inputs. The individual RBM layer is trained with an unsupervised learning approach where an artificial neural network (ANN) algorithm tunes the model for minimizing the error between the true and predicted class. The effectiveness of the proposed model is studied by varying the input signal and sampling frequencies. A level of considered noise is added with the sample data to test the robustness of the studied model. Results prove that the proposed fault detection and classification model has the ability to perform the precise diagnosis of MG faults. A comparative study among the proposed, kernel extreme learning machine (KELM), multi KELM, and support vector machine (SVM) approaches is studied to confirm the robust superior performance of the proposed model.

Simultaneous fault detection and control for discrete-time switched systems under relaxed persistent dwell time switching

A Data-Driven Fault Detection and Diagnosis by NSGAII-t-SNE and Clustering Methods in the Chemical Process Industry

Switching LDS-based approach for process fault detection and classification

This study addresses a critical operational gap in smart grid infrastructure management: while hybrid deep learning architectures combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks have been theoretically proposed for simultaneous fault detection and cyber-threat classification in SCADA-based environments, no empirically validated quantitative framework exists to determine which hybrid architecture capability dimensions most strongly predict operational reliability, threat response effectiveness, and detection latency reduction in real-world deployment contexts. The purpose was to empirically validate a hybrid CNN-LSTM deep learning framework that links core architectural capabilities to smart grid operational outcomes within an enterprise-scale, case-based setting. Using a quantitative, cross-sectional, case-study design, data were collected from a purposive sample of N = 218 smart grid security and operations professionals including SCADA engineers, network security analysts, data scientists, grid operations managers, and protection relay engineers working in operational environments where AI-driven monitoring and anomaly detection platforms are actively deployed. Key independent variables were CNN Feature Extraction Capability (C), LSTM Temporal Sequence Modeling Capability (L), and Hybrid Architecture Integration Depth (H), alongside two domain-specific indices: the Hybrid Architecture Performance Index (HAPI) and the Real-Time Threat Detection Alignment (RTDA); key dependent variables were Fault Detection Accuracy (Y1), Cyber-Threat Classification Effectiveness (Y2), and Operational Continuity Performance (Y3). The analysis applied descriptive statistics, reliability and validity testing (Cronbach's alpha, EFA with KMO and Bartlett's test), Pearson correlations, and multiple regression models. Headline findings demonstrate strong measurement quality (alpha = .83-.89; KMO = .91; Bartlett's chi-squared = 2318.7, p &lt; .001) and moderately high capability levels (CNN Capability M = 3.97, SD = 0.61; LSTM Capability M = 4.11, SD = 0.56; Hybrid Integration M = 3.84, SD = 0.67). All core relationships were positive and significant (p &lt; .001), including Fault Detection Accuracy with HAPI (r = .66) and LSTM Capability (r = .63), Cyber-Threat Classification with RTDA (r = .68), and Operational Continuity with Hybrid Integration (r = .64). Regression results indicate substantial explained variance for Fault Detection Accuracy (R2 = .63; F(5,212) = 72.4, p &lt; .001), with HAPI as the strongest predictor (beta = .37, p &lt; .001), followed by LSTM Capability (beta = .25, p = .001).