过热汽温模糊神经网络预测控制器的设计

A DESIGN OF FUZZY NEURAL NETWORK FORECAST CONTROLLER ON SUPERHEAT TEMPERATURE

针对锅炉过热汽温的特点,设计前馈-反馈串级复合型控制系统。主控制器采用基于神经网络预测模型的模糊神经控制,即该控制器首先是将神经网络与预测控制相结合,采用改进的递阶遗传算法对神经网络的权值和结构同时进行训练,实现了非线性、大时滞系统模型的精确预测;然后将模糊控制与神经网络相结合,实现模糊神经预测控制。副控制器采用二自由度PID控制器。仿真结果表明,该控制显著提高锅炉过热汽温这一非线性、大时滞系统的控制品质,且易于工程实现。

Based on the characteristics of superheated steam temperature of a drum boiler, a new cascade feedback control system with load feed-forward is designed. A fuzzy neural network forecast controller is applied to the outer loop. The controller combines neural network with forecast control, optimizes the neural network by an improved genetic algorithm and realizes accurate forecast to nonlinear and big-lagged system; and then, combines fuzzy control with neural network and realizes fuzzy neural network forecast control. A double-degree PID controller is applied to the inner loop. Simulation result shows that the project is effective in solving nonlinear and big-lagged system, and easy for application.