基于多层多源信息的地区级智能电网短期负荷预测研究

项目名称： 基于多层多源信息的地区级智能电网短期负荷预测研究

项目编号： No.51477078

项目类型： 面上项目

立项/批准年度： 2015

项目学科： 电工技术

项目作者： 张智晟

作者单位： 青岛大学

项目金额： 69万元

中文摘要： 短期负荷预测是智能电网关键基础性研究课题，项目以地区级智能电网为研究载体，充分利用多层多源信息，从四个方面研究，第一部分为预测基础研究，包括智能电网负荷序列特性的多层次精细化分析，影响智能电网负荷的显变量因素和潜变量因素研究，基于数据挖掘技术聚类方法和模式识别策略原理结合研究外部随机因素等；第二部分为预测理论机理研究，包括基于多元理论融合、整体混沌动力学行为机理的基本混沌预测负荷分量模型研究，外部随机因素负荷分量的预测模型研究，基于高智能多Agent的群集智能原理的STLF的总体模型等；第三部分为智能电网微观精细化预测研究，包括微网级和用户级短期负荷预测研究等；第四部分为预测评估平台构建，包括构造基于多元理论融合的多层多Agent智能电网STLF评估系统、负荷预测专用数据仓库构建等。通过研究，使预测模型具有好的快速性、准确性和稳定性。预测评估系统建立可加快智能电网STLF科学创新研究速度。

中文关键词： 智能电网；短期负荷预测；预测评估；多元理论

英文摘要： Research on short-term load forecasting is a key basic research project.The research carrier is regional level smart grid, and the multi-layer and multi-source information is used adequately,and the project aims to study the content of four respects.The first respect is the forecasting foundation research part, including detailed analysis with multi-level for load series character in smart grid;and research on significant variable factors and latent variable factors affecting load in smart grid;and research on external random factors based on the fusion of data mining theory and pattern recognition strategy.The second respect is the forecasting mechanism research part, including research on the chaotic load forecasting model based on the fusion of multi-theory and whole chaotic dynamics mechanism; research on forecasting model for load affecting by external random factors; research on load forecasting model based on multi-agent and multi intelligent principle.The third respect is the micro fine forecasting research in smart grid,which include short-term load forecasting in the micro-grid level and in the user level.The fourth respect is the construction of forecast evaluation platform, including constructing the forecast evaluation system with multi layer and multi-agent based on the multi-theory;and constructing special data warehouse for load forecasting. Through the project, the short-load load forecasting model possess good fast performance, accuracy and stability. The construction of forecast evaluation system can accelerate the study speed for short-term load forecasting in smart grid.

英文关键词： Smart Grid;Short-Term Load Forecasting;Forecast Evaluation;Multi-Theory