基于RBF神经网络的船舶电站PID控制器研究

Research on PID Controller of Ship Power Station Based on RBF Neural Network

针对船舶电站柴油发电机不确定时延性所带来的非线性调速问题,研究一种基于RBF神经网络的船舶电站PID控制器。对RBF-PID调速控制系统在控制结构上进行了改进,实现对船舶电站柴油发电机非线性调速的快速稳定控制。仿真结果表明:在RBF-PID调速控制系统控制结构中,增加排气压力和排气温度两个输入参数,使得PID参数调节更加灵敏,优化了调速系统性能;改进RBF-PID控制器的调速系统控制参数动态调整规律为:当转速变小、排气总管压力和温度变小时,PID控制参数会调高;当转速变大、排气总管压力和温度变大时,PID控制参数会调小。

Aiming at the problem of non-linear speed regulation caused by uncertain time delay of diesel generator in marine power station,a kind of PID controller based on RBF neural network for marine power station is studied. The control structure of RBF-PID speed control system is improved to realize the fast and stable control of non-linear speed regulation of diesel generator in marine power station. The simulation results show that in the control structure of RBF-PID speed control system,two input parameters,exhaust pressure and exhaust temperature,are added to make the adjustment of PID parameters more sensitive and optimize the performance of the speed control system. The dynamic adjustment rule of the control parameters of the speed control system of the improved RBF-PID controller is that when the speed decreases,the pressure and temperature of the exhaust main pipe decreases,the parameters of the PID control will be increased. When the rotational speed increases,the exhaust main pipe pressure and temperature increases, the PID control parameters will be reduced.