BP神经网络在水轮机调节过程中的应用

由于水轮机调节系统的大惯性、"水锤"效应等特点及其结构复杂等问题,采用传统的常规PID控制已很难满足控制要求,控制品质也难以改善,控制过程中易发生超调量大、震荡频次多、收敛时间过长等问题。对此,在常规PID控制基础上设计了基于BP神经网络自适应PID控制,并在Matlab软件中完成相关程序的编写及仿真试验。仿真结果表明,基于BP神经网络自适应PID控制是一种有效的水轮机调速器参数整定方法,相较于常规PID控制能获得更好的动态性能。

Application of BP Neural Network in Water Turbine Regulating System

As for the features of big inertia of water turbine regulating system, the effect of water hammer and its complex structure, the traditional PID control is difficult to meet the control requirements, and it is hard to improve the control quality. Some problems occurred in the control process, such as large overshoot, high-frequency oscillation and long-time convergence. Aiming at these problems, self-adaptive PID control model based BP neural network was designed, and it was completed by writing the related program and simulation experiments in the Matlab software. The simulation results show that the proposed adaptive PID control is an effective method to set the parameters of hydraulic turbine regulation system, and it has a better dynamic performance compared with the conventional PID control.