Joint Motor Control Of Collaborative Robot Based On Model Predictive Control

This dissertation takes the joint motor of the collaborative robot as the research object,designs a three-vector model predictive control algorithm that considers the switch changes,and proposes a method to suppress the speed pulsation on this basis,thereby improving the low-speed control performance of the motor.Aiming at the problem that the control algorithm is sensitive to parameters,this dissertation designs a parameter adaptive control algorithm and builds an integrated drive control platform for joint motors to verify the feasibility of the algorithmIn terms of control algorithms,this dissertation explores the differences and connections between continuous control set and finite control set model predictive control,designs a three-vector finite control set model predictive control method that considers switch changes,and studies traditional model predictive control methods,namely single vector and double vector model predictive control.Compared with the improved three-vector model prediction algorithm,the results show that the improved three-vector model predictive control has better steady-state and dynamic control performance.In terms of speed ripple suppression,this dissertation considers the control algorithm and torque pulsation from two perspectives,designs a one-step delay algorithm,and proposes a cogging torque feedforward compensation method.Simulation and physical tests verify that the speed ripple suppression effect is good.Especially the improvement effect on the low-speed performance of the motor is obvious.In terms of parameter adaptive algorithm,this dissertation analyzes the necessity of online identification of stator resistance and inductance,improves the current equation in the delay algorithm,turns it into an incremental prediction model,and eliminates the influence of flux linkage on the identification of resistance and inductance.The recursive least square method with forgetting factor is used for online identification,and subsequent experiments are used to verify the control effect of the parameter adaptive algorithm.In terms of physical experiment,this dissertation firstly identifies the joint motor stator resistance,stator inductance and rotor flux linkage in the physical platform through the offline least square algorithm,and designs the software and hardware of the physical platform drive control system.This article tests the dynamic and static performance and anti-disturbance ability of the control algorithm to position,speed and current through experiments such as adding and reducing load and parameter mutation.In summary,the control algorithm proposed in this paper has good dynamic and static performance,and achieves excellent adaptive control effects.The joint motor drive and control integrated platform built in this paper can meet the requirements of high-performance control of collaborative robots,realize flexible expansion functions,and have strong practicability.