Convergent Estimation Mechanism Design for Nonlinear Fuzzy Systems With Faults

Sensor fault detection in autonomous mining trucks with unknown varying gross weight

This paper focuses on the problem of the sensor fault detection in Autonomous Mining Trucks (AMTs) with the unknown varying gross weight and measurement noise. The dust and extreme temperatures in strip mines can lead to bias and drift faults in the sensors of AMTs. Besides, due to the bumpy roads in mining areas, the longevity and accuracy of the weight sensors cannot be guaranteed, which makes the weight sensor useless for AMTs. Therefore, the gross weight is treated as an unknown parameter in the lateral dynamics model of AMTs. The emphasis or difficulty lies in obtaining the state estimation and reducing the false alarm rate under the unknown variations in gross weight. This paper proposes an interval observer with the zonotope method to estimate the AMT state under the condition of unknown variations in gross weight. An interval residual generator with the generalized likelihood ratio test and the zonotope method is proposed for the sensor fault detection in AMTs. Finally, the effectiveness of the proposed approach is validated through the simulations of an AMT.

Lower triangle factor based sliding mode observers design for T-S fuzzy systems with faults

The paper mainly focuses on the fault estimation for a class of Takagi-Sugeno (T-S) fuzzy systems with faults. A synthetic estimation observer design method is proposed. The synthetic estimation observer can cover the robust observer, adaptive observer and intermediate estimation observer in the existing study work. Based on the observer design method, an LTF-based sliding mode observer (SMO) is designed for the T-S fuzzy system in consideration. Under the observer, the fault occurring in the system can be well estimated. The obtained LMI-based conditions guarantee the states of the error dynamics to be uniformly ultimately bounded. A numerical example tests the proposed method.

A Novel Active Fault-Tolerant Tracking Control for Robot Manipulators with Finite-Time Stability

Many terminal sliding mode controllers (TSMCs) have been suggested to obtain exact tracking control of robotic manipulators in finite time. The ordinary method is based on TSMCs that secure trajectory tracking under the assumptions such as the known robot dynamic model and the determined upper boundary of uncertain components. Despite tracking errors that tend to zero in finite time, the weakness of TSMCs is chattering, slow convergence speed, and the need for the exact robot dynamic model. Few studies are handling the weakness of TSMCs by using the combination between TSMCs and finite-time observers. In this paper, we present a novel finite-time fault tolerance control (FTC) method for robotic manipulators. A finite-time fault detection observer (FTFDO) is proposed to estimate all uncertainties, external disturbances, and faults accurately and on time. From the estimated information of FTFDO, a novel finite-time FTC method is developed based on a new finite-time terminal sliding surface and a new finite-time reaching control law. Thanks to this approach, the proposed FTC method provides a fast convergence speed for both observation error and control error in finite time. The operation of the robot system is guaranteed with expected performance even in case of faults, including high tracking accuracy, small chattering behavior in control input signals, and fast transient response with the variation of disturbances, uncertainties, or faults. The stability and finite-time convergence of the proposed control system are verified that they are strictly guaranteed by Lyapunov theory and finite-time control theory. The simulation performance for a FARA robotic manipulator proves the proposed control theory’s correctness and effectiveness.

Deep Learning-Based Adaptive Neural-Fuzzy Structure Scheme for Bearing Fault Pattern Recognition and Crack Size Identification

Bearings are complex components with onlinear behavior that are used to mitigate the effects of inertia. These components are used in various systems, including motors. Data analysis and condition monitoring of the systems are important methods for bearing fault diagnosis. Therefore, a deep learning-based adaptive neural-fuzzy structure technique via a support vector autoregressive-Laguerre model is presented in this study. The proposed scheme has three main steps. First, the support vector autoregressive-Laguerre is introduced to approximate the vibration signal under normal conditions and extract the state-space equation. After signal modeling, an adaptive neural-fuzzy structure observer is designed using a combination of high-order variable structure techniques, the support vector autoregressive-Laguerre model, and adaptive neural-fuzzy inference mechanism for normal and abnormal signal estimation. The adaptive neural-fuzzy structure observer is the main part of this work because, based on the difference between signal estimation accuracy, it can be used to identify faults in the bearings. Next, the residual signals are generated, and the signal conditions are detected and identified using a convolution neural network (CNN) algorithm. The effectiveness of the proposed deep learning-based adaptive neural-fuzzy structure technique by support vector autoregressive-Laguerre model was analyzed using the Case Western Reverse University (CWRU) bearing vibration dataset. The proposed scheme is compared to five state-of-the-art techniques. The proposed algorithm improved the average pattern recognition and crack size identification accuracy by 1.99%, 3.84%, 15.75%, 5.87%, 30.14%, and 35.29% compared to the combination of the high-order variable structure technique with the support vector autoregressive-Laguerre model and CNN, the combination of the variable structure technique with the support vector autoregressive-Laguerre model and CNN, the combination of RAW signal and CNN, the combination of the adaptive neural-fuzzy structure technique with the support vector autoregressive-Laguerre model and support vector machine (SVM), the combination of the high-order variable structure technique with the support vector autoregressive-Laguerre model and SVM, and the combination of the variable structure technique with the support vector autoregressive-Laguerre model and SVM, respectively.