基于SD-SVM的负荷预测研究与应用

Research and Application of Load Forecasting Based SD-SVM

以临沂地区电网负荷预测系统开发为背景,以提高电网负荷预测精确度为目的,提出了基于相似日的支持向量机的(SD-SVM)负荷预测应用方案。介绍其实现的原理、业务功能及其在实例中的系统体系架构和实现技术。重点对系统采用的SD-SVM数据模型算法作了描述。该项应用研发,实现了对地区电网负荷的智能预测、自动分析、气象数据处理等功能,提升了调度人员对电网负荷变化趋势的预见性、掌控的即时性。SD-SVM法的负荷预测结果对气象因素变化较大时有明显的精度优势,具有很好的适应性。实用结果证明这一研发的实时性和预测精度优于基于传统SVM理论的预测方法,能够满足地区电网负荷预测实用性、有效性、智能化的要求。

According to the development of the Linyi power grid load forecasting system,the application program of load forecasting based on similar-day support vector machine （SD-SVM） was proposed for the purpose of improving the accuracy of the power grid load forecasting. The principle, business functions, system architecture and enabling technology are also presented in the paper. The algorithm of the SD-SVM data model is described in detail. The re- search and development of the project is to realize the function which includes intelligent forecasting of the regional power grid,automatic analysis, and meteorological data processing. Improving the dispatcher＇ s ability in the foresight of grid load changing trend and the real-time controlling is another function of the system. While the meteorological factor changes a lot,the accuracy of the result of load forecast using SD-SVM is the best. Therefore,the adaptability of SD-SVM is good. Practical results show that the real-time and accuracy of the load forecasting based on SD-SVM is better than that based on traditional SVM,which can meet the needs of the practicability, validity and the intelligence of the regional power grid load forecasting.