基于粗糙集理论的配电网故障风险模糊综合评价

发布时间：2018-06-22 06:36

本文选题：配电网故障 + 风险评估　；参考：《山东大学》2016年硕士论文

【摘要】：作为复杂的大规模网络动态系统,电力系统基本任务是为用户高效地提供安全、可靠、经济和优质的电力能源。现代电力系统不断向特高压、大容量、自动化、智能化方向发展,系统拓扑结构越来越复杂,尤其是处于发输电与负荷关系密切位置的配电网系统,配电网稳定地运行直接关系着整个电力系统的安全运行与调度控制。配电网系统由于拓扑结构复杂、不确定性与不一致性信息并存等特点给其故障风险评估研究带来挑战。传统的配电网故障风险评估常存在评价指标单一、无法有效地处理配网存在的模糊性与不确定性信息,则如何对配电网运行与故障过程中产生的大量配电网数据信息进行有效的分析,并正确选择风险指标、建立更全面合理的风险评估系统评估风险,是避免或减少由于配电网故障引起的停电损失、提高供电可靠性和电能质量的关键。粗糙集理论与模糊综合评价方法可以有效应对信息的不确定性与模糊性特征,揭示信息数据的潜在规律,是进行风险评估的极其有效的理论依据与方法。本文融合粗糙集理论与模糊综合评价方法在处理包含不确定性以及不一致性特征信息方面的优势,针对配电网系统独特的运行方式与故障属性,通过粗糙与模糊并行的方法进行属性约简与数据处理,提出了一种基于粗糙集理论的模糊综合评价研究方法,并有效地应用于配电网故障风险评估。本文主要从以下几个方面进行了研究：第一,介绍了本文风险评估的理论基础粗糙集理论和模糊综合评价理论。给出了基于粗糙集理论的配电网故障信息的特性统计归类分析与故障风险定量分析方法。为解决故障数据普遍存在的不准确、不完整、不一致性等问题以及评估系统数据无法使用连续数据等问题,在进行配电网故障风险评估前,基于粗糙集理论及相应改进方法对配电网故障数据进行原始数据筛选、补全与离散化,在原有算法基础上结合粗糙集理论与配网故障实际改进算法。第二,深入研究了配电网故障信息,并建立了配电网故障风险评估指标体系。首先利用故障记录,对相关故障属性进行了详细的分析讨论与归类整理。依托于记录的配电网故障信息,从故障相关时间、故障设备、故障发生区域、故障停电原因、故障影响设备、故障级别等多个方面对配电网故障风险源进行了特性统计归类分析,并对故障风险的量化、不确定性及交叉影响进行了定性且定量的研究,基于风险评价理论和可靠性研究选定了故障率等评估指标,并建立了以综合故障概率与综合故障后果为首层、其余风险源影响首层指标为基础的配电网故障停电风险指标体系。第三,提出了基于粗糙集理论的配电网故障风险模糊综合评价方法,建立了由故障属性对应的风险因素影响的配电网故障风险评估模型。对评价风险因素,选择基于粗糙集理论的可辨识矩阵属性频率的属性约简方法对配电网关键风险因素筛选,并求取风险指标权重,结合风险评估理论改进模糊综合评价模型,采用粗糙与模糊并行的方法进行风险评估。第四,应用提出的基于粗糙集与模糊综合评价并行的风险评估方法,对不考虑分布式电源影响的配电网故障原因以及考虑分布式电源影响的多指标配电网故障进行了风险评估计算及对比分析。结果表明,综合多种因素影响的基于粗糙集的配电网风险模糊综合评价模型,可以合理的分析对比配电网停电风险,真实反映不同风险因素影响,验证了方法的合理性与有效性。
[Abstract]:As a complex large-scale network dynamic system, the basic task of the power system is to provide the users with safe, reliable, economic and high quality power. The modern power system is constantly developing to the ultra high voltage, large capacity, automation and intelligent direction, the more complex the system topology is, especially in the transmission and load relation. The distribution network system and the stable operation of distribution network are directly related to the safety operation and scheduling control of the whole power system. The distribution network system brings challenges to the research of the fault risk assessment because of the complex topology structure and the coexistence of uncertainty and inconsistency information. The traditional distribution network fault risk assessment often exists the evaluation index. It can not effectively deal with the fuzziness and uncertainty information of the distribution network, how to effectively analyze the data information of distribution network in the process of distribution network operation and failure, and select the risk index correctly and establish a more comprehensive and reasonable risk assessment system to evaluate the risk, so as to avoid or reduce the distribution network reason. Rough set theory and fuzzy comprehensive evaluation method can effectively deal with the uncertainty and fuzziness of information and reveal the potential law of information data, and it is an extremely effective theoretical basis and method for risk assessment. This paper combines rough set theory and model. The fuzzy comprehensive evaluation method is to deal with the advantages of the information containing uncertainty and inconsistency. In view of the unique operation mode and fault attribute of the distribution network system, the method of attribute reduction and data processing is carried out by rough and fuzzy parallel method. A fuzzy comprehensive evaluation method based on rough set theory is proposed. It is applied to the fault risk assessment of distribution network. This paper mainly studies the following aspects: first, it introduces the theory of rough set and fuzzy comprehensive evaluation theory of the risk assessment in this paper. The classification analysis and quantitative analysis of fault risk based on rough set theory are given. In order to solve the problems of inaccuracy, incomplete, inconsistency and inconsistency of the fault data, and to evaluate the failure of the system data to use continuous data, the original data is screened based on the rough set theory and the corresponding improved method before the fault risk assessment of the distribution network, and the original data is made up and discretized. On the basis of the algorithm, it combines the rough set theory and the actual improvement algorithm of the distribution network fault. Second, the fault information of the distribution network is studied deeply, and the fault risk assessment index system of the distribution network is established. From the fault related time, the fault equipment, the fault occurrence area, the failure cause, the fault influence equipment and the fault level, the characteristics of the fault risk source of the distribution network are classified and analyzed. The qualitative and quantitative research is made on the quantification, uncertainty and cross effect of the fault risk, based on the theory of risk assessment and the feasibility. The failure rate and other evaluation indexes are selected by sex research, and a fault outage risk index system of distribution network based on the first layer of the comprehensive fault probability and the comprehensive fault consequences and the influence of the other risk sources on the first layer is established. Third, a fuzzy comprehensive evaluation method based on Rough Set theory is proposed, and the fault is established. In order to evaluate risk factors, select the attribute reduction method of attribute frequency of discernible matrix based on rough set theory, select the key risk factors of the distribution network, and seek the weight of risk index, and improve the fuzzy comprehensive evaluation model with the risk assessment theory. Fourth, the risk assessment method based on rough set and fuzzy comprehensive evaluation is applied. The risk assessment and comparison analysis of distribution network failure without considering the influence of distributed power supply and the multi index distribution network fault considering the influence of distributed power supply are calculated and compared. It shows that the fuzzy comprehensive evaluation model of distribution network risk based on rough sets, which is influenced by a variety of factors, can reasonably analyze and compare the risk of power outage in distribution network, reflect the influence of different risk factors, and verify the rationality and effectiveness of the method.
【学位授予单位】：山东大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TM73

【相似文献】