Three-motor Synchronous Decoupling Control System Based On Neural Network Adjusting Parameters Of PID Controller

Considering the characteristics of multi-input and multi-output,nonlinearity, time-variation and strong coupling and so on in the three-motor synchronous system, this paper presents a control system centering on neural network arithmetic to realize the decoupling control for three-motor synchronous system.Firstly,on the basis analysing the mathematical model of three-motor synchronous system,according to the control strategy of two-motor synchronous system,we put forward the control strategy of three-motor synchronous system based on master-slave following mode.Namely,one motor is set for the master motor,others are set for the slave motors which run afer the master motor.Secondly,three-motor synchronous control system is designed by applying neural network method and traditional PID control,the system is made up of four parts:three intelligent PID controllers based on BP neural network adjusting the parameters of PID controllers on-line have the stronger ability to adapt and the better real-time and robustness in controlling respectively the speed of the master motor and two tensions; one neuron decoupling compensator fulfills open loop decoupling of multi-variable by training the weighting coefficient of neuron to compensate the coupling.The feasibility and advance of the control system is made sure by decoupling and speed tracking simulation.Thirdly,the experimental platform is constructed,MPI and PROFIBUS-DP network are connected,and transducers' addresses are set;hardware and network of the S7-300 station are configurated in STEP7 to realize the communication between PLC and transducers,the interface monitoring of process is configurated in PC using WinCC to monitor the operating condition of PLC by MPI network,neural network control and PROFIBUS-DP communication programs are compiled and downloaded into PLC,So we realize the long-distance intellectualized control for three-motor synchronous system.Finally,lots of experimental results indicate that the neural network control scheme can realize the decoupling control between the velocity and tension of three-motor synchronous system with the better dynamic and static characteristics relativing to traditional PID control.The control method proposed in the paper is promising and can be applied in many industrial control environments.