Study On Fuzzy Identification Method And Its Application

Fuzzy identification is often used in system identification methods,which can efficiently make full use of the ability of expert system,the effect of system input and system output data,and give the accountabilityof identification model for the black box model.Fuzzy model is the hardware foundation to establish fuzzy identification theory.It has been proved to play a great role in nonlinear dynamic system modeling,rule-based learning control and pattern recognition.It not only greatly enriches the theoretical method,but also promotes the development of fuzzy control theory and simulation technology.This paper focuses on the study of fuzzy modeling and identification of nonlinear systems,and applies it to the actual industrial production.In this paper,a fuzzy identification method based on fuzzy clustering is proposed,and the method is applied to the prediction of of top-oil temperature for power transformer.The oil temperature prediction model based on T-S model is established.The antecedent parameters of T-S fuzzy model are identified by fuzzy C-means algorithm,And the consequent parameters is determined by the recursive least squares method.Then based on the shortcomings of the existing algorithms,an improved fuzzy identification method is proposed and applied to the of titration process in hair products detection.The establishment of a system process model not only enables automated testing,but also optimizes the efficiency and accuracy of hair product testing.Therefore,this study has a very good application value of engineering practice.The main work of this paper is summarized as follows:(1)The premise structure of fuzzy model is determined by fuzzy clustering method,including the properties and characteristics of fuzzy clustering algorithm.It mainly includes the influence of clustering index and clustering initial value on clustering results.(2)The fuzzy clustering and multi-information recursive least squares are used to identify the structure and parameters of the fuzzy model,and the convergence of the recognition algorithm is studied.An improved fuzzy clustering algorithm is proposed and proved by the theorem.The simulation results show that the proposed method is effective.(3)Fuzzy system identification in the prediction of top-oil temperature for power transformer.Based on the actual monitoring data of the actual running transformer,the current and ambient temperature are used as the input data,the top-oil temperature are used as the output data.A prediction model of top-oil temperature based on T-S fuzzy model has been established.The prediction results achieved a high computational accuracy which agreed well with the test results,It is of great significance to guide the safe operation of the power transformer.(4)Fuzzy system identification in neutralization and titration is studied.According to the input data of the system,the p H value of the effluent in the CSTR is used as the output data,and the fuzzy model is established by fuzzy clustering and multi-information recursive least squares method to establish the fuzzy model of neutral acid titration.Through the fuzzy identification,the proposed identification model can effectively predict the changes in p H,and can be applied in the automatic titrator and other testing equipment.