Offset-guiding-information-based parameter self-setting method of SISO partial-format model-free controller

Abstract

The invention discloses an offset-guiding-information-based parameter self-setting method of a single-input-and-single-output (SISO) partial-format model-free controller. Offset guiding information isused as an input of a BP neural network and the BP neural network carries out forward calculation and outputs to-be-set parameters like a penalty factor and a step-size factor of a controller throughan output layer. Calculation is carried out by using a control algorithm of the controller to obtain a control input for a controlled object; a control input is calculated; and for gradient information of all to-be-set parameters, with minimization of a value of a system error function as an objective, system error back-propagation calculation is carried out by combining the gradient informationbased on a gradient descent method. A hidden layer weight coefficient and an output layer weight coefficient of the BP neural network are updated in real time in an online manner; the gradient information is stored as offset guiding information and is used as an input of the BP neural network at a next time. With the offset-guiding-information-based parameter self-setting method, the setting problem of the parameters of the controller can be solved; and the good control effect is realized.

Description

technical field [0001] The invention belongs to the field of automatic control, and in particular relates to a parameter self-tuning method of a SISO partial format model-free controller based on partial derivative information. Background technique [0002] Model-free controller is a new type of data-driven control method, which does not rely on any mathematical model information of the controlled object, but only relies on the real-time measured input and output data of the controlled object for controller analysis and design, and realizes concise, computational The burden is small and the robustness is strong, and the unknown nonlinear time-varying system can also be well controlled, and has a good application prospect. [0003] There are many implementation methods of model-free controllers, among which SISO (Single Input and Single Output, single input and single output) partial format model-free controller is one of the main implementation methods of model-free controll...

AI Technical Summary

Problems solved by technology

The lack of effective parameter tuning methods not only makes the debugging process of the SISO partial format model-free controller time-consuming and laborious, but also sometimes seriously affects the control effect of the SISO partial format model-free controller, restricting the development of the SISO partial format model-free controller. Promote application

Images