基于ANFIS的长期电力负荷预测模型

Long-Term Load Forecasting Model Based on ANFIS

某电力公司的长期负荷预测模型,利用自适应神经网络模糊推理系统(ANFIS)建立。其结构分为计算输入的模糊隶属度、每条规则适用度、适用度归一化、每条规则输出及模糊系统输出5层。系统网络中含14个待定的前件和后件参数。采用Matlab编程,通过Sugeno和evalfis函数训练ANFIS,按指定指标得到这些参数,实现模糊预测。

The ANFIS was adopted by the long-term load forecasting model to establish power company. Its structure was divided into calculating input fuzzy membership degree layer, every rule application degree layer, normalization of application degree layer, every rule output layer and fuzzy system output layer. There are 14 undetermined parameters named predictor and consequent. The Matlab programmer was adopted to train ANFIS through Sugeno and evalfis function then, those parameters were achieved according to appointed indexes and fuzzy forecasting were realized.