基于RBF神经网络自适应整定燃烧串级控制系统

PID cascade control system of power plant combustion with RBF network self-tuning

针对传统的锅炉燃烧串级控制系统采用固定的PID控制参数难以取得满意的控制效果的缺点,引入径向基(RBF)神经网络自适应整定燃烧串级控制系统外回路的PID参数.内回路仍然沿用原来的PI控制方式及控制参数,由RBF网络辨识得到被控对象的Jacobian信息后,根据梯度下降法自适应调整系统外回路的PID控制参数.仿真研究结果表明:新控制算法能够消除控制系统的静态误差;即使被控对象的模型参数发生了很大变化,新控制算法仍然能快速响应蒸汽压力的阶跃扰动,迅速克服燃料量内扰,其控制效果明显优于常规PID串级控制.

The fixed PID control parameters were used in the traditional cascade control system of power plant combustion,but the satisfactory control effects can hardly obtained because the thermal control process of power plants are characterized by non-linear and large inertial time-delay.A new control method of tuning the combustion cascade control system＇s PID parameters using the RBF neural network is put forward.Simulation results show that the new control system can eliminate the system＇s static error,respond the step input of the main steam and overcome the fuel disturbance effectively even if the controlled object＇s parameters varies largely,so it has a better performance than the conventional PID cascade control system.