基于遗传算法BP神经网络的恒压供水系统的研究

Research on constant pressure water supply system based on PID algorithm of optimized BP neural network by genetic algorithm

针对恒压供水系统普遍存在的非线性、大滞后和不确定的特点,设计了一种基于遗传算法BP神经网络的PID控制器,该控制器先通过遗传算法优化BP神经网络的初始权值,再利用BP神经网络的自学习和自适应能力,自动调整PID控制器的参数,达到自适应控制目的,解决了传统PID控制算法难以控制未知复杂系统的问题。软件仿真表明,本系统的恒压性能和动态性能有较大的提高,控制效果比较理想。

Aimed at the characteristics of nonlinearity, large delay and uncertainty for constant pressure water supply system, the paper proposes a PID controller based on optimized BP neural network by genetic algorithm. Firstly, the initial weights of BP neural network are optimized by genetic algorithm. And then the parameters of PID controller are adjusted by BP neural network which has self-learning and adaptive re-use capabilities , which achieves the purpose of adaptive control. The PID control algorithm solves the difficulties of unknown PID control of complex systems. The simulation indicates that the constant performance and dynamic performance of the constant pressure water supply system has greatly improved, and the control effect is satisfying.