论文部分内容阅读
针对钢铁企业高炉煤气发生量机理模型难以对发生量进行精确预测的问题,通过分析高炉煤气发生量特点,按不同工况利用概率神经网络(PNN)对高炉煤气发生量进行分类,依据分类结果并结合HP滤波、Elman神经网络(ENN)、最小二乘支持向量机(LSSVM)各自的性质,建立了PNN-HP-ENN-LSSVM模型,对高炉煤气的发生量进行分类预测,并用企业实际数据验证.结果表明,随机抽取多组测试结果中的2组,1#高炉80个点、2#高炉60个点的分类准确率分别为95%和93%,模型预测平均相对误差分别为1.0%和1.1%,适合高炉煤气发生量预测.Wilcoxon符号秩检验也验证了所提建模方法的有效性.
Aiming at the problem that it is difficult to accurately forecast the amount of blast furnace gas production mechanism model in iron and steel enterprises, the occurrence of blast furnace gas is classified according to the characteristics of blast furnace gas generation and the probability neural network (PNN) according to different working conditions. According to the classification results Combined with the properties of HP filter, Elman neural network (ENN) and least squares support vector machine (LSSVM), a PNN-HP-ENN-LSSVM model was established to predict the amount of blast furnace gas and verified with the actual data The results show that the accuracy of classification of two groups of 80 sets of 1 # blast furnace and 60 points of 2 # blast furnace is 95% and 93%, respectively, and the average relative error of model prediction is 1.0% and 1.1%, which is suitable for blast furnace gas production forecasting.Wilcoxon signed rank test also verifies the effectiveness of the proposed modeling method.