Modeling method for noise-uncertainty complicated nonlinear dynamic system

Abstract

The invention discloses a modeling method for a noise-uncertainty complicated nonlinear dynamic system. The method includes the steps: 1), collecting data during industrial process to acquire data (XMN, Y); and 2), calculating noise statistical value of known input data and output data by means of Gamma Test to acquire precise information of system noise. The modeling method for the noise-uncertain complicated nonlinear dynamic system has the advantages that the best ideal point for increasing production and saving energy is searched, and optimal value of technological parameters is determined; and practical production guide is performed according to the optimized optimal value of the technological parameters.

Description

technical field [0001] The invention belongs to the technical field of intelligent information processing. In particular, it relates to a modeling method of a noise uncertain complex nonlinear dynamic system based on an improved kalman filter neural network estimated by Gamma Test noise statistics. Background technique [0002] Neural network statistical modeling method, with its good nonlinear approximation ability, has achieved good industrial process modeling effect. However, when the neural network performs function approximation, although the approximation error can converge to a small neighborhood of zero, the weights of the neural network cannot converge to the optimal value. In other words, the neural network can accurately approximate the real model by learning the existing data, but the information learned by the neural network cannot be further utilized, and the model will not be adjusted once it is determined, which is a static modeling method. However, in the ...

AI Technical Summary

Problems solved by technology

[0004] However, due to the uncertain source of industrial process noise, it is difficult to effectively monitor the noise. In practical applications, the statistical value of the noise is often set to zero, which will inevitably affect the modeling effect of the Kalman neural network.

For industrial systems with uncertain noise, since the observation noise cannot be effectively measured, traditional methods need to estimate the observation noise itself to obtain accurate observation noise statistics, which cannot solve the above problems

Therefore, the usual practice is to set R={R 1 ,...R I ,...,R T }={0,...0,...,0}, that is, the observation noise statistical value R matrix of the Kalman neural network is set to zero, so that the noise estimation value is artificially determined, so that the calculation uses the observation noise statistical value and the actual system Inconsistent statistical values ​​of process noise affect the modeling effect

Images