Incipient fault prediction for nonlinear stochastic distribution systems

Abstract

In order to avoid the occurrence of major accidents in the industrial production process, incipient fault prediction that can predict fault in advance has attracted more and more attentions. In this article, a generalized correntropy filtering‐based incipient fault prediction strategy is proposed, which is suitable for nonlinear stochastic distribution systems with simultaneous actuator and sensor faults and non‐Gaussian noise. Three filters are designed to predict the incipient fault of nonlinear stochastic distribution systems: the first one is used to predict the time when the incipient fault occurs, and the other two are used to estimate the evolution rate of sensor fault and actuator fault, respectively. When the residual signal exceeds the predetermined threshold, the fault is detected and the time of occurrence can be also estimated. Once the fault is detected, the designed filter can be used to estimate the evolution rate of the unknown fault. After estimating the time of fault appearance and its evolution rate, the change process of incipient fault can be predicted. Finally, an example of a chemical reaction process is given to illustrate the validity of the prediction scheme.

Fault‐tolerant control design for stochastic distribution control systems based on k‐step fault estimation

“…On the contrary, adaptive observer is widely used in fault estimation because of its simple structure and parameter design. However, In existing observer-based FE methods for SDC system, disturbance from the derivative of actuator fault are simply considered as input disturbance to the error dynamic system, which leads to inaccurate results of FE [42,43]. Therefore, improving the accuracy of FE for SDC systems in the adaptive observer framework is the motivation of our work.…”

“…The stabilization of the residual system is proven based on the Lyapunov method. Compared with some previous results [23, 38,43,45,46], the main contributions of this paper are summarized as follows: (1) the mismatched fault is estimated by using the k‐step fault estimation method which the accurate fault information can be provided for fault‐tolerant controller design procedure; (2) based on the information of k‐step fault estimation, a dynamic output feedback fault‐tolerant controller is designed to ensure the output PDF of post‐fault SDC system can still be maintained at a fault‐free level; (3) unlike the reference [45], the mismatched fault is considered in this paper which is common in practical system; note that the observer gain parameters were determined by making the system matrix Hurwitz in reference [46], while the gain parameters are obtained by solving LMIs at the given $$$ {H}_{\infty } $$$ performance in this paper; in some previous results of fault diagnosis for SDC system, for example, previous studies [23, 38,43], the influence from the derivative of the mismatched actuator fault on the error dynamic system was ignored, therefore, the application of k‐step fault estimation method to SDC system has heuristic significance for improving the accuracy of fault diagnosis for SDC system.…”

Application of Patent Right and Trademark Right in Packaging Design Based on Computer Nonlinear Prediction Systems for Virtual Reality Technology

“…e nonlinearity of trademark information data is the basic premise for the effective application of trademark big data technology [9]. Although there are certain differences between digital and digitalization, it will certainly promote the process of trademark information digitalization [10].…”

Layer‐wise contribution‐filtered propagation for deep learning‐based fault isolation

“…A quick response may prevent the spread of the fault in the early stage and reduce significant financial damages. [4][5][6] The difficulty of FI depends on the complexity and transparency of the model, which is usually built based on a fault observer or classifier. It is extremely challenging when the model has low interpretability and complex nonlinearities.…”

“…When the system has detected a fault, the subsequent step is rapidly locating the cause of the abnormality. A quick response may prevent the spread of the fault in the early stage and reduce significant financial damages 4‐6 . The difficulty of FI depends on the complexity and transparency of the model, which is usually built based on a fault observer or classifier.…”