【摘要】 使用传统舰船发动机控制方法控制时,因其控制参数均为固定设置,无法随发动机参数变化而变化,导致控制方法的响应与实际控制信号出现较大偏差,影响了控制方法的稳定性和灵活性。针对以上问题,研究神经网络优化PID的舰船发动机自动控制方法。构建发动机闭环增益的PID结构后,设计神经网络自动控制器。利用遗传算法对神经网络PID自动控制器参数进行整定,降低控制器响应控制信号时的超调量,完成对舰船发动机控制方法的设计。通过与传统模糊PID控制方法的对比实验,证明了研究的控制方法能够有效降低24.31%的超调量,并且相比传统方法研究的方法的正弦跟随特性更佳,即神经网络优化后的自动控制方法具有更好的稳定性和灵活性。更多还原

【Abstract】 When using the traditional ship engine control method, the control parameters are fixed and cannot change with the change of engine parameters, resulting in a large deviation between the response of the control method and the actual control signal, which affects the stability and flexibility of the control method. Aiming at the above problems, the automatic control method of ship engine optimized by neural network PID is studied. After the PID structure of engine closed-loop gain is constructed, the neural network automatic controller is designed. The parameters of the neural network PID controller are adjusted by genetic algorithm, and the overshoot of the controller in response to the control signal is reduced. By comparing with the traditional fuzzy PID control method, it is proved that the proposed control method can effectively reduce the overshoot by 24.31%, and has better sinusoidal tracking characteristics than the traditional method, that is, the automatic control method after the neural network optimization has better stability and flexibility.更多还原