面向配电网下基于大数据技术的电力负荷自动预测研究

Research on automatic power load forecasting based on big data technology in distribution network

随着科技的进步,大数据技术在电力负荷预测中的使用越来越多。为提升在大数据技术基础上的电力负荷预测系统的性能,研究首先设计了一个基于大数据技术的运维管控系统,然后将K均值聚类与长短期记忆网络结合,设计出一种负荷预测模型,最后将运维管控系统与电力负荷预测模块应用于配电网平台。结果表明,在指标计算中,负荷预测模型的决定系数比其他模型分别提升0.051、0.089、0.128,拟合度和精度较高。在实际应用中,负荷预测模型预测值与真实值之间的平均差值为1.0%,准确率较高。以上结论证明了设计模型的性能较好,能够为电力系统的规划和调度提供科学依据。

With the progress of science and technology,big data technology is used more and more in power load forecasting.In order to improve the performance of the power load forecasting system based on big data technology,the research first designed an operation and maintenance management and control system based on big data technology,then combined K-means clustering with long and short term memory network to design a load forecasting model,and finally applied the operation and maintenance management and control system and power load forecasting module to the distribution network platform.The results show that the coefficient of determination of the load forecasting model is increased by 0.051,0.089 and 0.128 compared with other models respectively,and the fitting degree and accuracy are higher.In practical application,the average difference between the predicted value and the real value of the load forecasting model is 1.0%,and the accuracy is high.The above conclusions prove that the design model has good performance and can provide scientific basis for the planning and scheduling of electric power system.