基于概率盒理论的电网规划决策不确定性建模及风险评估研究

项目名称： 基于概率盒理论的电网规划决策不确定性建模及风险评估研究

项目编号： No.51467007

项目类型： 地区科学基金项目

立项/批准年度： 2015

项目学科： 电工技术

项目作者： 丁家满

作者单位： 昆明理工大学

项目金额： 44万元

中文摘要： 电网规划方案决策中存在大量不确定因素，对不确定性的忽略或简单平均将导致规划方案决策的偏差甚至错误，对原始数据不确定性的正确处理是后续决策成功的基石。概率盒理论在处理不确定性问题上存在强大优势，其应用已在经济、生态、核物理等领域得到证实，本项目拟利用概率盒理论解决电网规划方案中的不确定性问题。以现有电力研究院的电网规划方案为具体研究对象，获取案例的指标体系，根据单指标信息源形式的多样性，选择相应的概率盒建模方法，利用定性和定量约束条件优化模型。分析单指标多源信息的相关性，利用证据理论、包络融合等方法实现单指标多源的融合。获取各单指标权重概率盒，建立决策函数，利用概率盒灵敏度分析方法，获得决策风险评估结果，评价决策的合理性。与忽略指标不确定性决策方法对比，证明概率盒建模方法的有效性。本项目将为概率盒理论与电网决策领域的交叉研究探索提供理论指导和依据，并进一步丰富电网规划方案决策方法。

中文关键词： 智能配电网；不确定性；概率盒理论；风险评估；信息融合

英文摘要： There are a lot of uncertainties in the decision-making of power grid planning. Ignorane or simple averaging of uncertainty will lead to the deviation or even error. Correct processing of raw data uncertainty is the foundation of subsequent decision's success. Probability box (p-box) theory is powerful advantage in dealing with uncertainty. Its application has been confirmed in the areas of economy, ecology, nuclear physics. This project intends to solve the uncertainty problems in the power grid planning by the p-box theory. The existing power grid planning schemes in electric power research institute are the concrete research objects. Index systems of the cases are obtained. According to the diversity of the single index information source, the corresponding p-box modeling methods are selected. The qualitative and quantitative constraints are used to optimize the models. The Correlations of the multi-source information are analyzed. The multi-sources of single index are fused by evidence theory, envelope fusion methods, etc. The single index weight p-boxes are obtained. The decision function is established. The risk assessment results are made by the p-box sensitivity analyzing methods. The rationality of decision-making is evaluated. Contrasted to the decision method ignored uncertainty, the effectiveness of the p-box modeling methods is demonstrated. This project will provide theoretical exploration, guidance and basis for the crossover studies of p-box theory and grid decision making field, and further enrich the methods of grid decision making.

英文关键词： Smart Distribution Grid;uncertainty;probability box theory;Risk Assessment;data fusion