A Nonlinear System Identification Method Based on Structure Adaptive Filtering

Abstract

The invention provides a nonlinear system identification method based on structure adaptive filtering, which comprises the following steps: obtaining input data and corresponding output data of an identification object as a data set, and dividing the data set into a training set and a test set; constructing a self-adaptive filtering model: randomly selecting a first-order subsystem and a second-order subsystem according to the generation parameters, and randomly cascading the subsystems according to a connection rule to form a complete self-adaptive filtering model; initializing an evolutionary algorithm, and selecting an optimal adaptive filtering model; carrying out loop iteration through an evolutionary algorithm, and finding the optimal structure and parameters of the optimal adaptivefiltering model; and testing by using the trained adaptive filtering model to obtain the actual output of the identification object. According to the method, the nonlinear process of the typical liquid saturated steam heat exchanger is learned by utilizing the nonlinear adaptive filtering model, and the optimal estimation can be found; the method is simple in operation, small in error, high in precision and few in used parameters, and has remarkable nonlinear system recognition capability.

Description

technical field [0001] The invention relates to the technical field of system identification, in particular to a nonlinear system identification method based on structure adaptive filtering. Background technique [0002] System identification design is widely used in process modeling and control in various engineering fields such as chemical engineering, machinery, electrical and civil engineering, which provides a mathematical basis for social production and brings great convenience. [0003] Some system models can be expressed in linear parameterized form, and adaptive control can be applied to realize output tracking, and unknown system parameters can be identified and estimated online. However, in actual social practice and production, many systems have highly nonlinear characteristics, and the models of such systems are relatively complex, making it difficult to model and control them. Traditional system modeling and identification methods are difficult to achieve effe...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the technical problems of complex models, difficult modeling and low recognition rate in nonlinear system identification, the present invention proposes a nonlinear system identification method based on structure adaptive filtering, using evolutionary algorithm to analyze the nonlinear structure adaptive filtering model Conduct training to realize the adaptive iteration of the nonlinear structure adaptive filtering model, so as to search for the globally optimal structure and parameters, and effectively fit the identification objects

Images