热负荷预测中最有影响力参数测定

Heat load forecasting input variable selection

针对集中供热系统,采用自适应模糊神经推理系统分析热负荷的影响因素与系统热负荷之间的相关性和影响程度。分别分析了不同预测时间下的3组输入变量对集中供热系统热负荷预测的影响。仿真结果表明,应用ANFIS选取集中供热系统热负荷预测的输入变量对于简化预测是可行的。

For the central heating system, Adaptive Network-based Fuzzy Inference System （ANFIS） is used to analyze the correlation between the influencing factors of heat load and the heat load of the system. The influence of three sets of input variables on the thermal load forecasting of the central heating system are studied. Simulation results indicate that ANFIS is feasible to be applied to select the input variables of the thermal load forecasting of the central heating system, and simplify the prediction.