摘要
为有效减小短期电力负荷预测的预测误差,提高预测精度、缩短预测时间,应用改进粒子群优化(IPSO)算法建立了1种短期电力负荷预测模型。通过水平方向和垂直方向的平滑修正,对历史数据的异常负荷点进行识别并修正。利用相同日期类型正常负荷,计算缺失数据填充值。采用模糊化处理,计算日期类型、温度、天气隶属度函数,对短期负荷变化因素进行量化处理。将历史数据的负荷值和量化值作为训练数据。为避免粒子群优化(PSO)算法陷入局部最优,采用IPSO算法找到全局最优解,建立了短期负荷预测模型,实现了短期电力负荷预测。试验结果表明,所设计模型预测结果在休息日和工作日的最大相对误差值、平均相对误差值分别为0.97%、0.53%和0.99%、0.65%,能够有效减小预测误差、提高预测精度、缩短预测时间。该研究为电力系统相关人员进行负荷预测提供了参考。
A short-term power load forecasting model is developed by applying the improved particle swarm optimization(IPSO)algorithm to reduce forecasting error effectively,improve forecasting accuracy,and shorten forecasting time of the short-term power load forecasting.The abnormal load points of historical data are identified and corrected by smoothing correction in horizontal and vertical directions.Missing data fill values are calculated by using normal loads of the same date type.Fuzzification is used to calculate the date type,temperature,and weather affiliation functions to quantify the short-term load variation factors.Load values and quantized values of historical data are used as training data.To avoid the particle swarm optimization(PSO)algorithm from falling into local optimum,IPSO algorithm is used to find the global optimal solution,and a short-term load forecasting model is established to realize short-term electric load forecasting.The experimental results show that the maximum relative error value and the average relative error value of the prediction results of the designed model on rest day and work day are 0.97%,0.53%and 0.99%,0.65%,respectively,which can effectively reduce the forecasting error,improve the forecasting accuracy,and shorten the forecasting time.The study provides reference for load forecasting by the relevant personnel in the power system.
作者
王峰
WANG Feng(State Grid Nanjing Power Supply Company,Nanjing 210013,China)
出处
《自动化仪表》
CAS
2023年第4期22-26,共5页
Process Automation Instrumentation
关键词
改进粒子群优化算法
短期电力负荷
负荷预测
电力系统
异常负荷点
模糊化处理
隶属度函数
全局最优解
Improved particle swarm optimization(IPSO)algorithm
Short-term power load
Load forecasting
Power system
Abnormal load points
Fuzzification process
Subordination function
Global optimal solution
