基于区间二型模糊逻辑的短期电力负荷预测

摘要

In order to solve the problem that power load is too stochastic to precisely forecast, an interval type-2 fuzzy logic method was introduced. An interval type-1 non-single-valued type-2 fuzzy logic model was presented for one-hour prediction of time series of power load. And the back-propagation (BP) algorithm was used for tuning model input and stipulating parameters of antecedent and consequent components. Meantime, a non-single-valued type-1 fuzzy logic model was set up on the basis of BP method and its prediction result was taken as a datum for performance checking. Simulation result showed that the interval type-2 fuzzy logic model exhibited a better performance than its type-1 counterpart due to its higher prediction accuracy, justifying the view point that the type-2 fuzzy logic method could deal with uncertainties in a more effective way.%针对电力负荷的波动具有强随机性而难以准确预测的问题,引入区间二型模糊逻辑方法以减小预测误差.提出一个区间一型非单值二型模糊逻辑模型,用于一小时电力负荷的时间序列预测,并利用反向传播算法来调节模型输入及规则前、后件的参数.同时,构建一个基于反向传播算法的非单值一型模糊逻辑模型,将其预测结果作为检验性能的基准.仿真结果表明,区间一型非单值二型模糊逻辑模型提高了预测精度,性能优于相应的非单值一型模糊逻辑模型,验证了二型模糊逻辑方法能更有效地处理不确定性的观点.