中文摘要：

针对锅炉汽包系统的强耦合性和非线性及传统的PID控制方法存在控制精度低、调节时间长等问题,提出了利用基于数据的建模方法,对汽包系统进行误差反向传播（BP）神经网络建模,并对神经网络模型进行泛化能力测试,然后利用基于BP神经网络的PID控制方法设计汽包液位优化控制器。实验仿真结果表明,基于BP神经网络建立的汽包模型具有较好的泛化能力,神经网络PID优化控制器在控制精度高、收敛速度快和鲁棒性强等方面都优于传统PID控制器。

英文摘要：

Boiler drum system is a nonlinear and strong coupling system.The traditional PID control method has these problems such as low precision control,long regulation time.A model of boiler drum system based on back propagation（BP） neural network is built,and the generalization ability of neural network model is tested.Then the optimal controller of the drum system is designed by the neural network PID based on BP neural network.Simulation results show that the BP neural network model has a better generalization ability,neural PID controller has high control precision,fast convergence and robustness advantages than traditional PID controller.