基于模式识别的自适应短期负荷预测系统

短期电力负荷预测是电力调度部门制定发电计划的依据，预测系的灵活适应性是实现电网经济运行的重要保证。在分析影响日电力负荷主要因素的基础上，给出了用于日荷预测的负荷模式定义，基于海明距离给出了负荷模式相拟度的计算方法，有效实现了预测负荷所需要的历史负荷模式样本的抽取。利用人工神经网络实现由历史负荷模式到预测负荷的映射。基于C＋＋面向对象的程序设计方法开发了一套灵活的智能自适应短期预测系统。多个用户的应用结果表明，本系统具有很好的实用性和满意的预测结果。

A Pattern-Recognition-Based Adaptive System for Short Term Load Forecasting

Short term load forecasting is the basis of power generation planning for power dispatch department.The flexibility and adaptivity of load forecasting system is the important assurance for the economic operation of power network.Based on the analysis of the major faceors that influence daily load,a load mode for short term load forecasting is defined.The similarity calculation of two load modes is given using the weighted Hanmin distance,by which the necessary historical load mode samples can be extracted efficiently.The artificial neural network is employed to map the forecasting load.A flexible and intelligent adaptive system for short term load forecasting is developed with object oriented program design method based on C++builder.Field application shows that the system is practical with satisfactory results.