Article
Automation & Control Systems
He Liu, Xiao-Jian Li
Summary: This paper investigates fault detection and isolation for Takagi-Sugeno fuzzy time-delay systems using a geometric method, constructing unobservability subspaces and designing observers to detect and decouple single faults.
INTERNATIONAL JOURNAL OF FUZZY SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Fanglai Zhu, Yuyan Tang, Zhenhua Wang
Summary: This article deals with fault detection and isolation problems for uncertain discrete-time Takagi-Sugeno fuzzy systems. By combining the H-infinity observer and the zonotope method, a robust observer is designed and a fault detection and isolation scheme is proposed. Numerical simulations are conducted to demonstrate the effectiveness of the proposed method.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Mathematics
Jianing Cao, Hua Chen
Summary: In this paper, a mathematical model based on the T-S fuzzy model is proposed to solve the fault estimation and fault-tolerant control problem for singular nonlinear time-varying delay systems with sensor fault. The TVD problem is solved by using the Laplace transform to build an equivalent system free of TVD. Additionally, the sensor fault is transformed into actuator fault by coordinate transformation. A fuzzy learning fault estimator is built to estimate the sensor fault information, and a PI fault-tolerant control scheme is suggested to minimize the damage caused by the fault. Simulation results show that the proposed algorithms can effectively estimate the fault and ensure system performance.
Article
Automation & Control Systems
Liwei Li, Tu Zhang, Mouquan Shen, Ju H. Park
Summary: An auxiliary filtering system approach is used to design an interval observer for linear systems with state and output disturbances in this paper. The interval observer is built under the unknown input observer framework by augmenting the filtering system and disturbances boundaries. The existence of the interval observer is formulated into strict linear matrix inequalities using the Lyapunov stability method and H infinity technique instead of the combination of linear programming and linear matrix inequalities. Compared to existing results, the proposed method can provide tighter state intervals from simulation studies.
JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS
(2023)
Article
Computer Science, Artificial Intelligence
Linlin Li, Steven X. Ding, Xin Peng
Summary: This article investigates optimal observer-based fault detection and estimation schemes for Takagi-Sugeno fuzzy systems. An optimal fault detection scheme is proposed to enhance sensitivity to faults and increase robustness against unknown inputs, along with a least squares fault estimation scheme. The proposed schemes utilize online recursive observers. A case study on a laboratory three-tank system demonstrates the effectiveness of the proposed approaches.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Automation & Control Systems
Huayun Han, Ying Yang, Linlin Li, Steven X. Ding
Summary: This article addresses the performance-based fault detection and fault-tolerant control for nonlinear systems, utilizing nonlinear factorization and Takagi-Sugeno fuzzy dynamic modeling techniques. It introduces the fault-tolerant margin as an indicator of system fault tolerance and presents FD scheme for stability performance degradation detection and FTC strategy for system performance recovery. The design approach is demonstrated through a case study on the three-tank system.
IEEE TRANSACTIONS ON CYBERNETICS
(2021)
Article
Computer Science, Theory & Methods
Rong Zhao, Lu Liu, Gang Feng
Summary: This paper investigates the problem of asynchronous fault detection filtering design for continuous-time Takagi-Sugeno (T-S) fuzzy affine dynamic systems in the finite-frequency domain. The objective is to design an admissible piecewise affine filter that ensures the asymptotic stability of the filtering error system and meets the prescribed finite-frequency H-/H-infinity performance. The proposed design approach utilizes the S-procedure, the generalized Kalman-Yakubovic-Popov lemma, piecewise quadratic Lyapunov functions, Projection lemma, and matrix inequality linearization techniques.
FUZZY SETS AND SYSTEMS
(2023)
Article
Nuclear Science & Technology
Swetha R. Kumar, Jayaprasanth Devakumar
Summary: The article presents a fault detection and isolation technique that uses a Recurrent Neural Network (RNN) to estimate state variables and build predictive models of a process. A bank of residuals is obtained, which provides exclusive fault signatures for different fault cases. The proposed methodology does not require the fault history of the process or first principle models, making it a more efficient approach.
PROGRESS IN NUCLEAR ENERGY
(2023)
Article
Automation & Control Systems
Zejian Zhu, Tianrui Chen, Yu Zeng, Cong Wang
Summary: This paper presents a systematic deterministic learning-based rapid sensor fault detection, isolation and accommodation (SFDIA) scheme for a class of nonlinear systems with unmodeled output dynamics. The proposed approach achieves locally accurate approximations of sensor faults and unmodeled output dynamics using a deterministic learning-based neural network observer. A sensor fault pattern bank is established to store the knowledge of sensor faults and unmodeled output dynamics. The effectiveness of the proposed method is demonstrated through simulation results.
INTERNATIONAL JOURNAL OF CONTROL
(2023)
Article
Engineering, Electrical & Electronic
Younan Zhao, Yuyan Tang, Fanglai Zhu
Summary: This paper investigates the fault detection problem for uncertain time-delayed systems with nonlinearities that satisfy incremental quadratic constraints. An observer is designed to construct the residual system for fault detection, but since the output disturbance affects the residual dynamics, it cannot be directly used for fault detection. To address this issue, interval estimation of the residual is introduced using the zonotope method. Based on this estimation, a residual-based fault detection scheme is proposed and its effectiveness is verified through numerical simulations and comparisons against other methods.
CIRCUITS SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Engineering, Multidisciplinary
Florent Becker, Ehsan Jamshidpour, Philippe Poure, Shahrokh Saadate
Summary: This paper presents an open-switch fault diagnosis method for a five-level H-Bridge T-type converter, which is capable of detecting and localizing faults based on the switching patterns and the converter output voltage level observation. The effectiveness of the proposed method is validated through simulations and experiments.
IEEE TRANSACTIONS ON INDUSTRY APPLICATIONS
(2023)
Article
Automation & Control Systems
Mohammadhosein Bakhtiaridoust, Meysam Yadegar, Nader Meskin
Summary: This paper proposes a data-driven approach for actuator fault detection and isolation in general nonlinear systems. The method utilizes a deep neural network architecture to obtain a set of basis functions for the Koopman operator, which then forms a linear Koopman predictor for the nonlinear system. The linear model obtained is used for fault detection and isolation without prior knowledge of the system dynamics. The proposed method is entirely data-driven with a recursive approach that has the key feature of global validity for the system's entire operating region, thanks to the global characteristic of the Koopman operator.
Article
Computer Science, Artificial Intelligence
Ruijie Liu, Ying Yang, Linlin Li, Huayun Han
Summary: This article studies a performance-oriented fault detection method for nonlinear control systems in the data-driven framework. The proposed approach evaluates the system performance changes caused by faults, using a regulation performance index and a fuzzy dynamic model. The effectiveness of the approach is demonstrated through case studies on a ship propulsion system and a three-tank system.
IEEE TRANSACTIONS ON FUZZY SYSTEMS
(2022)
Article
Automation & Control Systems
Fanglai Zhu, Yu Shan, Yuyan Tang
Summary: This paper discusses actuator fault and sensor fault detection and isolation problems for switched nonlinear systems with external disturbances. By constructing an augmented descriptor system and designing a sliding mode observer, effective fault detection and isolation methods are proposed to handle actuator and sensor faults.
INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Jinxin Wang, Xiuquan Sun, Chi Zhang, Xiuzhen Ma
Summary: This paper proposes a system-level fault diagnosis methodology based on fault behavior analysis, optimal sensor placement, and intelligent data analytics for multiple fault detection and isolation. By constructing a dynamical model and using set partitioning theory, a condition monitoring system with optimal sensor placement can be designed, and multivariate statistic measures are used to detect potential faults.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
