基于神经网络的解耦方法在直流锅炉中的应用

Application of Decoupling Method Based on Neural Network in the Once-through Boiler

分析了直流锅炉运行时各变量之间的耦合关系,针对直流锅炉参数多变、强耦合的特点,提出了一种改进的误差反向传播算法(BP)的神经网络分散解耦方法;仿真及实验结果表明,神经网络分散解耦算法具有很强的自学习功能和自适应解耦能力,是解决多变量和强耦合问题的一种有效途径。

The article analyses the coupling relations among the variables during the once-through boiler running. According to the features of changing parameters and strong coupling of once-through boiler,an improved error back-propagation algorithm （BP） neural network decoupling method is proposed. The results of the simulation and the experiment show that the neural network decoupling algorithm has a strong self-learning function and adaptive decoupling capacity, which is an effective way to solve the multi-variable and the strong coupling problems.