摘要：由于开关磁阻电机的非线性，常规的线性控制方法难以取得良好的效果。针对这一问题，设计了基于单神经元PID的调速系统，利用单神经元的自学习、自适应能力，通过动态调整开关磁阻电机的开通角和关断角以控制电机的转速，同时采用电流斩波控制以防止过流。该设计方法不需要事先采集数据对人工神经网络进行离线训练，实现简单、响应速度快。在Matlab中对电机空载起动、负载起动、设定转速增加、设定转速减少、突加负载、突卸负载等情况进行了仿真，结果表明在上述情况下，电机都能够在设定的转速平稳运行。

关键词：单神经元；PID；开关磁阻电机；转速控制

Abstract: Owing to the nonlinear of switched reluctance motor, a conventional linear control method is hard to perform effectively. To solve this problem, a speed control system based on single neuron PID was designed. Taking advantage of self-learning and adaptive ability of single neuron, controlled the speed of motor by dynamically adjusting the turn-on and turn-off angle. Meanwhile, chopped current control was adopted to prevent over-current. This design method didn′t need to collect data for offline training on artificial neural network, and had the advantages of simple implementation and fast response speed. The processes of non-load starting, load starting, given speed increase, given speed reduction, step load, and sudden unload were simulated with Matlab. Simulation results demonstrate that the motor running stably at given speed.

Key words: single neuron；PID；switched reluctance motor；speed control