基于神经网络的电站锅炉辐射受热面污染监测

Fouling Monitoring of Radiant Surface in Station Boiler Based on Artifical Neural Network

对传统的BP算法采取了新的改进方案,建立了基于人工神经网络的电站锅炉辐射受热面污染监测模型,应用于华北某300MW燃煤机组,为实现锅炉受热面优化吹灰建立了基础。与同一时间段内基于热平衡的监测结果的比较证明了该方法的合理性,分析了产生差别的原因,指出了不同监测模型适用于不同性质的受热面。

The model based on artifical neural network and improved BP algorithm is presented to monitor the fouling state of radiant heat absorption surface in power station. Parts of implement results are illustrated and indicate the monitoring model is reasonable and effective. Compared to the results based on heat balance, the reasons of difference are discussed, and application objects of both models are proposed at last. Figs 4, table 1 and refs 5.