State Estimation And Fault Diagnosis Of Nonlinear Systems Under Unknown Fuzzy Membership Functions

With the advent of the industrial 4.0 era,the scale and complexity of modern control systems continue to increase,and their reliability and security requirements are also increasing.State estimation and fault diagnosis of control systems have always been one of the key issues that need to be addressed and urgently needed in theoretical and engineering applications.However,the existence of nonlinear,modeling errors and unknown input disturbances has brought great limitations to the application in engineering practice.With the rise of fuzzy theory,fuzzy theory has been widely used in control theory,and many theoretical methods have emerged.However,in the commonly used control systems based on TS fuzzy models,we usually rely on empirical assumptions that fuzzy membership functions are known or online Measurable.However,when the membership function of the system is unmeasurable,such assumptions are not so reasonable and accurate.In view of this,this paper carries out state estimation and fault diagnosis of nonlinear systems under the condition of unknown membership function,which fills some shortcomings of existing research results.This paper first expounds the purpose and significance of the state estimation and fault diagnosis of nonlinear systems under the condition of unknown fuzzy membership function.The necessity of its research and the important contribution to the actual engineering system are discussed.At the same time,the advantages and disadvantages of the current research status at home and abroad are analyzed.Then,according to the relevant knowledge involved in this paper,the advantages of T-S fuzzy model and its related mathematical knowledge,including linear matrix inequality method and convex optimization,are introduced respectively.The lemma required for the relevant proof is also listed,which also lays the foundation for the state estimation and fault diagnosis of the next system.For the fault diagnosis of the system,because the previous method is still conservative,this paper proposes a new observer design framework based on the concept of the maximum priority switching law.Then,the truck trailer model is simulated to verify the effectiveness of the method and to reduce the conservativeness of the system.For the state estimation of the system,due to the further consideration of the application of the actual engineering,the stability analysis of the state estimation based on the unknown input is proposed in the system model.The measurable decision variables of continuous and discrete nonlinear systems and the observer design under unmeasured decision variables are fully analyzed.Finally,the feasibility of the method is verified by simulation examples and compared with other literatures,it reduces the conservativeness of the system.