PLC-based generalized neural network inverse internal model implementation method for multi-motor speed regulating system

Abstract

The invention discloses a PLC-based generalized neural network inverse internal model implementation method for a multi-motor speed regulating system. The method comprises the steps as follows: three siemens M440 converters drive three three-phase induction motors to drive loads to form the three-motor speed regulating system respectively; a mathematical model of a generalized right inverse system is deduced according to the mathematical model of the three-motor speed regulating system; three PID controllers are designed through an S7-300PLC to stabilize a three-motor drive system and to collect samples; data collected in the previous step is processed through a Wavelet Toolbox; characteristic signals are extracted; a neural network is trained offline; and finally closed-loop internal model controllers are designed for a sub-system for a rotating speed omega<r2> of the 2#motor, the sub-system for tension F<12> between belts and the sub-system for the tension F<23> between the belts in the siemens S7-300PLC respectively. The method disclosed by the invention eliminates a steady-state error of a generalized neural network inverse open-loop system and strengthens the system robustness.

Description

technical field [0001] The invention relates to a method for realizing a neural network generalized inverse internal model controller of a three-motor speed regulating system in a PLC, which is suitable for three three-phase induction motors driven by three Siemens M440 frequency converters with a Siemens S7-300PLC as a controller The invention relates to decoupling control of tension and rotational speed of a speed regulating system, which belongs to the technical field of electric drive. Background technique [0002] In recent years, multi-motor speed control systems have been widely used in textile, metallurgy, rail transit and other fields. However, how to realize the high-precision coordinated control of the multi-motor speed control system and improve the synchronization performance of the system has always been a research hotspot. For the three-motor speed control system, it has the characteristics of high-order, nonlinear, and strong coupling, making each Variables ...

AI Technical Summary

Problems solved by technology

However, how to realize the high-precision coordinated control of the multi-motor speed control system and improve the synchronization performance of the system has always been a research hotspot. For the three-motor speed control system, it has the characteristics of high-order, nonlinear, and strong coupling, making each Variables interact with each other, plus various disturbances in the working environment, sudden load changes, etc., making it difficult for traditional control methods to achieve high-precision control of the three-motor speed control system

However, in industry, three or more motors are often used, the variables and equipment involved are complex, and the existing control strategies have obvious limitations.

[0003] Traditional decoupling methods include feedforward decoupling, adaptive decoupling, self-correcting decoupling, etc., but all rely on the precise mathematical model of the system. Once external disturbance or load disturbance causes system parameters to change, the control effect will change. Poor or even system instability

[0004] At present, most of the intelligent control methods for multi-motor speed control systems are limited to the simulation experiment stage, and there is still a lack of effective and implementable means of operation to transform the simulation experiment into practical applications, which also requires a long accumulation of a large amount of practical experience

Images