Article
Automation & Control Systems
Decheng Liu, Jun Shang, Maoyin Chen
Summary: A powerful principal component analysis (PCA)-based ensemble detector (PCAED) is developed for detecting incipient faults in TEP, which cannot be detected by an individual PCA detector. Simulations fully verify the effectiveness of PCAED in detecting faults at the incipient stage.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Computer Science, Artificial Intelligence
Xiaoqiang Wen, Ziang Xu
Summary: A hybrid fault diagnosis method based on ReliefF, PCA, and DNN was developed to solve the problems of big data, inaccurate and untimely fault diagnosis. The method effectively reduced data dimensions and improved accuracy, achieving high accuracies for both single and multi faults. The proposed method outperformed comparison methods in diagnosing wind turbine faults.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Engineering, Electrical & Electronic
Chao Cheng, Weijun Wang, Hongtian Chen, Bangcheng Zhang, Junjie Shao, Wanxiu Teng
Summary: This article focuses on fault detection and diagnosis (FDD) for traction systems in high-speed trains, proposing an enhanced FDD architecture based on data-driven design which achieves fast and accurate fault detection without the need for mathematical models or control mechanisms of high-speed trains.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2021)
Article
Engineering, Civil
Chao Cheng, Yunfei Ju, Shuiqing Xu, Yisheng Lv, Hongtian Chen
Summary: The features of incipient faults in high-speed trains' electrical drive systems are small and can be masked by external environment noise and sensor disturbance. This paper proposes a novel fault detection scheme using a manifold learning method called local linear generalized autoencoder (LLGAE). The LLGAE-based method has three notable characteristics: 1) it can detect faults in electric drive systems even without a physical model or expertise; 2) it performs well for non-Gaussian electrical drives; 3) it considers the locally linear structure of samples. The proposed method is validated through experiments on a high-speed train platform.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Xianrui Sun, Chonghui Song, Yingwei Zhang, Xin Sha, Naizhe Diao
Summary: In this article, a fault diagnosis algorithm for power switch in the motor drive inverter system is proposed. The algorithm utilizes signal normalization preprocessing to handle current fluctuation caused by load change. The algorithm samples the three-phase currents using a sliding window to improve the timeliness of diagnosis. The sampled data is then preprocessed using variable modulus length normalization to eliminate the influence of load change. Additionally, the frequency-domain feature of the data is extracted using discrete Fourier transform (DFT). The PCA method is used to reduce the dimension of the features and locate the fault insulated gate bipolar transistor (IGBT) using its contribution graph. The algorithm effectively overcomes the misjudgment of IGBT failure caused by transient currents when the load changes in the drive system.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Automation & Control Systems
Decheng Liu, Min Wang, Maoyin Chen
Summary: In this article, a novel feature ensemble net (FENet) is proposed for detecting difficult-to-detect faults in the Tennessee Eastman process (TEP). FENet integrates features extracted by basic detectors in the input feature layer, transforms the previous feature matrix using sliding-window patches and principal component analysis (PCA) in the hidden feature transformer layers, and performs sliding technique and uses normalized singular values in the decision layer for detection. Results show that FENet effectively detects Faults 3, 9, and 15 in TEP compared to other methods.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Review
Engineering, Electrical & Electronic
Mohammad Zamani Khaneghah, Mohamad Alzayed, Hicham Chaoui
Summary: Fault detection and diagnosis are crucial for ensuring the safety and reliability of electric vehicles. This paper provides a comprehensive overview of potential faults in EV motor drives and battery systems, while also reviewing the latest research in EV fault detection. The information presented here can serve as a valuable reference for future endeavors in this field.
Article
Computer Science, Information Systems
Yuanyuan Ju, Ziliang Cui, Qingtai Xiao
Summary: Condenser fault diagnosis is crucial for the safe operation of power plants, but existing diagnostic methods lack precision and struggle under complex conditions. This study proposes a novel hybrid model, PCA-DF, that combines Principal Component Analysis with the Deep Forest model to improve diagnosis accuracy. Experimental results demonstrate the feasibility and effectiveness of this method in diagnosing condenser faults using historical data.
Article
Engineering, Industrial
Arthur Henrique de Andrade Melani, Miguel Angelo de Carvalho Michalski, Renan Favarao da Silva, Gilberto Francisco Martha de Souza
Summary: Through Condition-Based Maintenance strategy, a hybrid framework based on Moving Window Principal Component Analysis (MWPCA) and Bayesian Network (BN) was proposed for automated Fault Detection and Diagnosis (FDD) in machinery. The framework was able to detect and diagnose several simulated failures in a simplified model of a hydrogenerator.
RELIABILITY ENGINEERING & SYSTEM SAFETY
(2021)
Article
Automation & Control Systems
Ping Wu, Riccardo M. G. Ferrari, Yichao Liu, Jan-Willem van Wingerden
Summary: The proposed novel hybrid linear-nonlinear statistical modeling approach for data-driven incipient fault detection integrates canonical variate dissimilarity analysis and mixed kernel principal component analysis. Both linear and nonlinear features are considered simultaneously in this approach, leading to improved fault detection performance. Simulations based on a closed-loop continuous stirred-tank reactor process demonstrate the effectiveness of the proposed method compared to other relevant methods.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Engineering, Electrical & Electronic
Jun Hang, Wushuang Sun, Qitao Hu, Xixi Ren, Shichuan Ding
Summary: This article studies the integration of interturn fault diagnosis and fault-tolerant control of PMSM drive system. The interturn fault is diagnosed by zero-sequence voltage component and the torque ripple caused by interturn fault is reduced by current injection-based control strategy. The main novelty of this article is the direct calculation of injection currents for fault-tolerant control based on the information obtained during fault diagnosis, which simplifies the control complexity and enables convenient integration of fault diagnosis and fault-tolerant control. Both simulation and experimental results validate the effectiveness of the proposed integration method for the PMSM drive system.
IEEE TRANSACTIONS ON TRANSPORTATION ELECTRIFICATION
(2022)
Article
Thermodynamics
Fangliang Zhong, John Kaiser Calautit, Yupeng Wu
Summary: This study evaluates the performance differences of proposed convolutional and recurrent neural networks for fault detection and diagnosis (FDD) in limited seasonal fault data scenarios and an ideal scenario covering multiple climatic conditions. The results show that FDD architectures trained on sufficient fault data in a specific season have poor generalization ability to identify faults in unseen seasons. Additionally, the coverage of fault data in different seasons is more important for enhancing FDD performances than the amount of fault data in each season. These findings are crucial for researchers to consider when evaluating data-driven FDD methods.
Article
Chemistry, Analytical
Zahoor Ahmad, Tuan-Khai Nguyen, Sajjad Ahmad, Cong Dai Nguyen, Jong-Myon Kim
Summary: This study proposes a fault diagnosis method for multistage centrifugal pumps using informative ratio principal component analysis. The method selects the fault-specific frequency band to overcome interference and background noise in the vibration signatures of the pump, and extracts statistical features from this frequency band. By introducing a novel informative ratio principal component analysis, discriminant features with low dimensions can be obtained, improving the accuracy of fault classification.
Review
Automation & Control Systems
H. Safaeipour, M. Forouzanfar, A. Casavola
Summary: Incipient faults gradually occur in systems and can have severe effects if not detected in time. Model-based and data-based approaches are essential for incipient fault diagnosis.
JOURNAL OF PROCESS CONTROL
(2021)
Article
Automation & Control Systems
Wanke Yu, Chunhui Zhao
Summary: In this article, a robust canonical variate dissimilarity analysis method is proposed to detect incipient faults for industrial processes with missing value. The method utilizes low-rank matrix decomposition to recover missing elements and reduce noise, and establishes monitoring statistics based on canonical variate analysis to reflect the operation status of online samples. Experimental results show the proposed model can robustly detect incipient faults in industrial applications with incomplete training data.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Automation & Control Systems
Zhuofu Pan, Yalin Wang, Kai Wang, Hongtian Chen, Chunhua Yang, Weihua Gui
Summary: This study proposes an adaptive-learned median-filled deep autoencoder (AM-DAE) for imputing missing values in industrial time-series data. The method replaces missing values with the median of the input and reconstructed data, and adopts an adaptive learning strategy to improve performance. The results show that the proposed method outperforms other advanced techniques in handling missing values.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Energy & Fuels
Yu Lu, Gang Fang, Daoping Huang, Baoping Cai, Hongtian Chen, Yiqi Liu
Summary: With the increasing growth of energy demand and costs, it is important to monitor the operational costs in process industries. A support vector machine recursive feature elimination (SVM-RFE) method was developed to extract important features for constructing a sequential prediction model. Different variants of autoregressive and moving average (ARMA) models were applied to predict the costs in process industries, and the nonlinear model NARXNN achieved better performance for datasets with strong nonlinear characteristics.
FRONTIERS IN ENERGY RESEARCH
(2023)
Article
Engineering, Multidisciplinary
Darong Huang, Min Tang, Shuiqing Xu, Ning Zhao, Yu Zhang, Hongtian Chen
Summary: A digital twin (DT) model of a frequency synthesizer is proposed to perform thermal analysis in this study. The virtual entity of the DT model is established based on the physical entity of the frequency synthesizer. The temperature distribution of the model and cooling channel is analyzed, and a temperature prediction method based on a stochastic configuration network (SCN) algorithm is proposed using the results from the DT virtual entity model. The results show that the SCN algorithm based on the DT frequency synthesizer accurately predicts the temperature of the physical frequency synthesizer, providing accurate solutions and reducing the time required for physical components to adjust to high temperatures.
ENGINEERING ANALYSIS WITH BOUNDARY ELEMENTS
(2023)
Article
Automation & Control Systems
Kai Zhong, Jiayi Wang, Shuiqing Xu, Chao Cheng, Hongtian Chen
Summary: As the heart of high-speed train, traction systems play a crucial role in ensuring safe operation. However, the current level of operation and maintenance is insufficient for modern railway transportation. Fortunately, advanced fault prognosis methods, particularly those based on deep learning, have been developed to address this dilemma. This paper examines the structural characteristics of traction systems, compares and summarizes representative deep learning-based prognosis methods, and discusses the challenges and future trends in this field. It provides valuable insights for researchers interested in fault prognosis and high-speed train traction systems, as well as potential directions for further research.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Automation & Control Systems
Hongtian Chen, Biao Huang
Summary: This study develops three novel data-driven approaches for the development of fault-tolerant soft sensors in automation systems. The approaches, namely MSaS, SSaS, and IMSaS, aim to address the issue of unpredictable faults and their impact on soft sensor performance. MSaS constructs an optimal estimator of faults, SSaS removes influences from unknown sensor faults using a constructed subspace, and IMSaS is an improved version of MSaS that eliminates the effects of past prediction errors. These fault-tolerant soft sensors rely solely on system measurements and are evaluated through performance analysis and case studies.
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
(2023)
Article
Engineering, Civil
Chao Cheng, Yunfei Ju, Shuiqing Xu, Yisheng Lv, Hongtian Chen
Summary: The features of incipient faults in high-speed trains' electrical drive systems are small and can be masked by external environment noise and sensor disturbance. This paper proposes a novel fault detection scheme using a manifold learning method called local linear generalized autoencoder (LLGAE). The LLGAE-based method has three notable characteristics: 1) it can detect faults in electric drive systems even without a physical model or expertise; 2) it performs well for non-Gaussian electrical drives; 3) it considers the locally linear structure of samples. The proposed method is validated through experiments on a high-speed train platform.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Shuiqing Xu, Wenzhan Huang, Darong Huang, Hongtian Chen, Yi Chai, Mingyao Ma, Wei Xing Zheng
Summary: This article presents a reduced-order observer-based simultaneous diagnosis strategy for grid-tied neutral point clamped (NPC) inverters subjected to open-switch and current sensor faults. The strategy involves constructing an augmented descriptor system to transfer the current sensor fault into a generalized state vector, applying matrix transformations to decouple the open-switch fault from the inverter system state and the current sensor fault, developing a reduced-order observer for precise estimation of the phase current and the current sensor fault, and proposing a diagnosis algorithm with an adaptive threshold for distinguishing between different types of faults and locating the faulty power switch. Experimental results and comparisons confirm the effectiveness of the proposed fault diagnosis algorithm.
IEEE TRANSACTIONS ON POWER ELECTRONICS
(2023)
Article
Computer Science, Artificial Intelligence
Guangtao Ran, Hongtian Chen, Chuanjiang Li, Guangfu Ma, Bin Jiang
Summary: This paper introduces a hybrid fault detection design approach for nonlinear dynamic systems whose information is not known beforehand, combining data-driven and model-based methods. The nonlinear system is identified through optimization based on the least-squares method using a T-S fuzzy model. Modeling errors are taken into account when designing residual generators. Statistical learning is adopted to obtain an upper bound of modeling errors, and an optimization problem is formulated to determine a reliable fault detection threshold. An event-triggered strategy is applied in the online fault detection decision to save computational costs and network resources.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Letter
Automation & Control Systems
Engang Tian, Yi Zou, Hongtian Chen
IEEE-CAA JOURNAL OF AUTOMATICA SINICA
(2023)
Article
Engineering, Electrical & Electronic
Shuiqing Xu, Qihang He, Songbing Tao, Hongtian Chen, Yi Chai, Weixing Zheng
Summary: This article proposes a method to ensure the performance of pig face recognition in complex environments. A trapezoid normalized pixel difference (T-NPD) feature is designed for more accurate detection in unconstrained outdoor conditions. The method also utilizes a trimmed mean attention mechanism (TMAM) to assign precise weights to feature channels, which is fused into a ResNet50 backbone network for high accuracy classification. Comprehensive experiments on the JD pig face dataset show that the proposed method outperforms other methods with an overall accuracy of 95.06%.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Civil
Bin Du, Wei Xie, Weidong Zhang, Hongtian Chen
Summary: This paper investigates target tracking guidance for unmanned surface vehicles (USVs) in the presence of obstacles. A bias proportional navigation guidance law with look angle constraints is presented for guiding the USVs to orient and approach the moving target. The experimental results demonstrate that the proposed guidance law is capable of tracking the object, avoiding obstacles, and orienting the USV to the target.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Bin Du, Wei Xie, Yang Li, Qisong Yang, Weidong Zhang, Rudy R. Negenborn, Yusong Pang, Hongtian Chen
Summary: This study proposes a safe adaptive policy transfer RL approach for multiagent cooperative control. The method helps follower agents acquire knowledge and experience from a well-trained pioneer agent and achieves effective policy transfer through adaptive adjustment of learning weight.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Automation & Control Systems
Tao Xie, Weidong Zhang, Yufei Tang, Hongtian Chen
Summary: The problem of imbalanced samples in marine current turbine condition monitoring data poses challenges for training fault diagnosis models and can degrade model performance. In this article, a fault diagnosis approach based on physical-feature interactive expansion (PFIE) and convolutional neural networks (CNN) is proposed to address this issue. Experimental results demonstrate that the proposed method achieves high accuracy compared to other approaches.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2023)
Article
Automation & Control Systems
Shuangxin Zhu, Yu Huang, Engang Tian, Yuqiang Luo, Xiangpeng Xie, Hongtian Chen
Summary: This paper investigates the H-infinity tracking control problem for the series-parallel (SP)-inductive power transfer (IPT) system with coil misalignment. A Takagi-Sugeno (T-S) fuzzy modeling approach is applied to describe the nonlinear characteristic caused by coil misalignment, and an H-infinity parallel distributed compensation (PDC) fuzzy control strategy with an event-triggering scheme (ETS) is proposed to guarantee desired tracking control performance and decrease unnecessary packet transmission. The proposed control strategy ensures the stability of the system and disturbance suppression under large-scale coil misalignment.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Automation & Control Systems
Hao Ma, Yan Wang, Hongtian Chen, Jie Yuan, Zhicheng Ji
Summary: This paper proposes a novel quality-oriented efficient distributed framework for nonlinear plant-wide industrial quality-related process monitoring. The framework divides process variables into quality-related and quality-unrelated parts and selects highly relevant communication variables using the least absolute shrinkage and selection operator technique. It also utilizes orthogonal decomposition and kernel principal component analysis for quality-related and quality-unrelated parts, respectively. The proposed framework improves monitoring efficiency and accuracy of fault detection, facilitating the design of targeted fault-tolerant control schemes and enhancing energy savings. The incorporation of Bayesian fusion enables the generation of global and local fault detection indicators, benefiting fault analysis and system visualization.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
