基于EMS数据的煤气预测模型建立与应用

Establishment and application of coal gas prediction model based on EMS data

混合煤气不仅能为用户提供所需热值,而且与单一焦煤相比经济性更好,然而当前混煤能源使用多以开环为主,因此导致管网不平衡,能源浪费极大。基于此利用EMS系统提供大量混煤能耗数据,建立BP神经网络混煤能耗预测模型,通过Matlab进行数据运算,依托VB接口,利用Factory Talk View SE进行实际工程界面开发,实际应用显示,系统收敛快、稳定性好、精度高。

The mixed gas could provide the calorific value required by the users. Moreover, its economical efficiency was better than single coking coal. However, the mixed coal energy use at present was mainly in open loop, resulting in the disequilibrium of pipe network and huge waste of energy. Large amounts of data on energy consumption of mixed coal were provided by the EMS system. The prediction model based on BP neural network was established. The data operation was performed by Matlab. The actual engineering interface was developed on the basis of VB interface and Factory Talk View SE. The actual applications indicated that the prediction model system exhibited rapid convergence, good stability and high precision.