煤灰软化温度建模与预测

基于Elman网络建立了煤灰软化温度预测模型，该模型以煤灰成分为输入向量，以煤灰的软化温度为输出向量。为了选取最优预测模型，分别讨论了输入向量维数、隐含层单元数以及激励函数对模型预测结果的影响。使用获得的最优网络模型对测试样本进行测试，结果表明该模型的预测精度高于常规BP网络。由所建最优预测模型可知，存在一个最优的煤灰成分分析数量。

Modeling and prediction of the softening temperature of coal ash

The model of the softening temperature of coal ash was built based on Elman network.The coal ash compositions were employed as the input vectors and the measured ash softening temperature as the outputs of the neural network.In order to select the optimal prediction model,the relationship between input vector dimensions,the number of hidden layer,the activation functions and the results of the prediction model was discussed respectively.Then，the selected optimal model was used to predict the softening temperature of the test samples.The results show that the prediction accuracy is higher than the conventional BP network.From the optimal prediction model,can draw a conclusion that there is an optimal amount of coal ash composition analysis.