Control method of non-linear composite PID (Proportion Integration Differentiation) neural network based on triangular basis function

Abstract

The invention discloses a control method of a non-linear composite PID (Proportion Integration Differentiation) NN (Neural Network) based on a triangular basis function. The method comprises the following steps of: firstly, analyzing the action of three PID gain parameters exerting in a control process to acquire rough curves of the three gain parameters varying with error change, and constructing three non-linear composite gain functions based on the triangular basis function according to the three gain curves so as to acquire a non-linear composite PID control rate; using the non-linear composite PID control rate as an NN model; constructing a non-linear composite PID NN controller based on the triangular basis function by respectively merging two linear and non-linear proportional operation units, two linear and non-linear integral operation units and linear and non-linear differential operation units into hidden layer neurons; and generating a non-linear composite PID control signal through the online real-time training of the NN to dynamically control a non-linear controlled object. The invention can rapidly and accurately control a non-linear object and has strong robustness.

Description

technical field [0001] The invention belongs to the field of automatic control, and relates to an intelligent control method for on-line self-stabilization of parameters of hidden layer neurons which integrates a nonlinear composite PID operation unit. Background technique [0002] Proportional, integral, and differential (P, I, D) control according to the deviation is the control method with the longest history and the strongest vitality. Although advanced control strategies are gradually being promoted today, more than 90% of the control loops currently in operation are still PID controllers. However, with the increase of the complexity of the system and the increase of the uncertain factors of the object, the traditional linear PID control is no longer applicable, but the nonlinear PID control can truly reflect the nonlinear relationship between the control amount and the deviation signal. To a certain extent, it overcomes the shortcomings of the linear PID controller, s...

AI Technical Summary

Problems solved by technology

However, on the one hand, this method involves two-layer neural network weight adjustment, and there is a phenomenon of weight coupling, so the algorithm converges slowly and requires a large amount of calculation, which is not conducive to the real-time control of the fast sampling system; on the other hand, this method only uses linear PID The computing unit is integrated into the hidden layer neurons of the neural network after clipping processing, and its nonlinear control ability is not strong

Images