Application of Mamdani Fuzzy System Amendment on Load Forecasting Model

A short-term load forecasting model is adopted with a combined method. The model not only summarizes virtues and defects of neural networks and fuzzy system, but also considers that power system load has characteristics of basic load heft and variability load heft. It uses learned capability of neural networks to complete forecasting work of basic heft for power load. Other effect factors that cause variety of load are unconsidered in neural networks. For variability load heft that is affected by many factors, such as weather, data types and holidays, membership functions and fuzzy rules base are constructed in fuzzy logic system, which is used to correct basic load heft. The method simplifies system structure and enhances forecasting precision.