Direct Torque Control System Based On Intelligent Algorithm

This paper give systematic introduction the situation and development of direct torque control, and give details of its basic theory and principle. The advantage of direct torque control is that it controls motor flux and torque directly, it’s easy for control. The paper gives the relationship between the fluxes of the motor torque with voltage space vector based on the theoretical studies. This article also gives the basic structure for the direct torque control system and a brief description of the various aspects of its working principle and process.Combine the advantages of the intelligent algorithm, introduces smart algorithm control strategy into the additional direct torque control system, Can effectively improve the control performance and enhance the ability of the system such as adaptive. The paper introduces the principle of artificial neural network and particle swarm optimization, According to the particle of global convergence, high precision target, design stator flux observer based on PSO optimization of BP neural network, to achieve the combination of direct torque control system with intelligent algorithms. In order to prove the feasibility of the control strategy, using MATLAB/SEVIULINK, Offline training for the neural network, verify the precision convergence characteristics of the PSO-BP neural network. Prove the reliability of the system by simulation system module.Finally, use of STM32as CPU, combination of algorithmic thinking and realize need, to complete the direct torque control system hardware circuit design. Complete the main monitoring module, tachometer interrupt module and PWM output module software design in the integrated development environment of CCS.