基于SVC—ENN钢铁企业副产煤气消耗量的预测建模

摘要：针对钢铁企业副产煤气消耗量的机理模型难以对消耗量进行精确预测的问题，通过分析副产煤气消耗量特点，建立SVC—ENN模型对副产煤气的消耗量进行预测．根据企业实际数据应用模型，结果表明，对烧结工序、炼钢工序、连铸工序30个点和60个点进行测试分类准确率分别为90％，96．67％，98．33％；96．67％，95％，100％．根据分类结果建立模型进行预测，预测平均相对误差分别为0．8％，0．5％，0．9％；2．1％，0．8％，1．3％．所建模型分类准确，预测效果良好，适合副产煤气消耗量的预测．

Prediction of By- Product Gas Consumption in Iron and Steel Complex Based on SVC -ENN Model

Aimed at the problem that it is very difficult to accurately predict by-product gas consumption m rote- grated iron and steel works with available mechanism models, by analyzing the characteristics of gas consump- tion, a SVC-ENN model is established in this paper. The simulation results using the practical gas consumption data in an certain iron and steel complex show that for thirty sites and sixty sites in sintering process, steelmaking process and in continuous casting process, the classification accuracies of 90, 96.67 and 98.33 percent and 96.67, 95, 100 percent are tested respectively. Then a forecasting model is founded according to the results of classification to predict gas consumption, and the average relative errors of 0. 8, 0. 5 and 0. 9 percent and 2. 1, 0.8, 1.3 percent respectively are obtained. The model established with precise classification and satisfactory predictive effects is suitable for gas consumption forecast.