基于灰色神经网络的换热器污垢预测研究

由于影响换热器污垢热阻的因素较为复杂,对其预测比较困难.针对这种情况,提出了利用灰色神经网络预测方法对污垢热阻进行预测.本文用一种灰色理论算法改进模型和RBF神经网络分别对换热器污垢热阻值进行预测,并对预测结果进行最优组合,同时给出了预测曲线.结果表明与GM（1,1）模型相比较,灰色神经网络组合模型（GMNN）预测精度更高,可以较准确地预测污垢热阻随时间的变化趋势.

Research on Fouling Prediction of Heat Exchanger Based on Grey Theory and Neural Network

The factors influenced fouling resistance are complicated and it is considerable difficult to forecast the fouling resistance. Whereas, the research on the fouling prediction based on a grey and neural network hybrid model （GMNN） was carried on. This paper used an improved grey prediction model and a neural network prediction model to forecast the fouling resistances of the heat exchanger respectively. The final results came from the optimal combination for the two kinds of prediction results. And the prediction curves were given. The results showed that, compared to GM （1,1）, GMNN could get higher prediction precision and could exactly predict the variation trend with time.