应用人工神经网络预报电站锅炉的机械不完全燃烧热损失

To Predict Mechanical Incomplete Combustion Heat Loss in Power Plant Boilers with Artificial Neural Network

电站锅炉的热效率是衡量燃煤锅炉和发电机组运行经济性的重要指标,机械不完全燃烧热损失q4是影响锅炉运行热效率的最主要因素之一.要对锅炉热效率进行实时在线监测,必须准确地预报出q4的值.应用BP神经网络,使用MATLAB语言编写程序,对q4进行预报.经实验测得,q4预报值与真实值之间的相对误差在±1.65%之内.

Thermal efficiency in boilers＇ operation is an important index for evaluating economy of coal-fired boilers and generator units. It is heavily affected by mechanical incomplete combustion heat loss （q4）- Real-time and on-line boiler thermal efficiency monitoring depends on precise predicting data of q4. This paper adopts BP NN, and programs with MATLAB language to predict q4. The result of experiment indicates that the relative error between measured data and predicting data is within ± 1.65%.