Research On Co-simulation And Control System Of Coil-assisted Excitation Brushless DC Motor Dual Controller

Coil assisted excitation brushless DC motor(CAE-BLDCM)is composed of two groups of stator/rotor salient pole structures related to magnetic poles,at the same time,the adjustable magnetic field provided by the central assisted coil is added to avoid the adverse factors such as permanent magnet demagnetization and the inability to control the permanent magnetic quantity in the brushless dc motor,has the advantages of strong structure,high temperature resistance and high reliability.Due to the special structure,complex magnetic circuit and difficult decoupling of CAE-BLDCM,the construction of motor control system is difficult.Firstly,this thesis analyzes the motor torque,flux chain,inductance characteristics and the control effect of the central assisted coil through the static and dynamic simulation of ANSYS,and adopt the intelligent control mode that does not depend on the motor parameters,which is the dual-neuron adaptive intelligent control strategy based on online identification of RBF neural network,achieving the optimal joint control of motor armature winding and central assisted coil.Then,in this thesis,Maxwell,Simplorer and Matlab are used to jointly build a set of simulation platform for CAE-BLDCM,due to the strong non-linearity and special structure of CAE-BLDCM,it is also difficult to establish a mathematical model,therefore,the joint simulation is adopted to realize the simulation analysis of different control strategies of the motor.In the process of motor startup,acceleration and sudden loading,the conventional PID,single neuron adaptive PID and dual neuron adaptive PID are compared and analyzed,the analysis results show that the dual neuron adaptive PID control system has better control accuracy,robustness and response speed.Finally,based on the motor control principle and simulation,the hardware control platform with TMS320F28335 DSP as the core is built and debugged,by programming with CCS6.0 software development platform,the motor conventional PID,single neuron adaptive PID and dual neuron adaptive PID control strategies are respectively realized.Experimental results verified the CAE-BLDCM structure and double neuron adaptive PID control method is feasible and effective.