基于风险量化分析的电力变压器多层次检修策略研究

发布时间：2018-08-21 08:34

【摘要】：作为电力系统的枢纽设备，电力变压器的运行状态直接关系到电网的安全与稳定，在复杂多变的运行环境下，针对变压器进行故障诊断和预测以掌握变压器运行状态，进而分析设备运行风险以及制定针对性的检修策略，在当前电网设备迅速增加背景下的运维管理具有重要指导意义。电力变压器本身是一个极其复杂的系统，针对当前特征数据存在不确定性和模糊性难以有效分析，以及现有变压器运行状态分析中的问题，提出了基于云理论和隐式半马尔科夫模型的变压器运行状态转移过程研究方法。针对现有风险评估与检修决策关联不足，在基于设备自身运行状态及运行环境下构建了电力变压器动态综合风险评估方法，实现了变压器风险的经济性量化。基于检修决策指标及效能分析，构建了变压器全寿命周期内的固定检修优化策略以及基于动态风险评估的多维度多层面的检修优化策略。论文取得的主要成果如下： ①针对变压器DGA数据存在模糊性、不确定性利用云模型进行值域划分及边界数据的隶属度判定，基于云理论对DGA数据训练，构建了变压器状态观测空间。基于变压器运行过程中状态退化及DGA产生与发展规律，通过引入老化因子并采用多步长加权方法在马氏链中引入历史运行状态信息，构建的加权隐式半马尔可夫退化模型实现了变压器运行状态的准确预测。 ②利用变压器运行状态转移的退化隐式半马尔科夫模型，对变压器寿命周期内的故障率进行分析，分析不同运行工况对变压器风险的影响，并提取热点温升作为综合反映变压器运行工况的参量对变压器故障率修正；基于预想事故的分析，对变压器不同风险模式下导致的本体损失和故障后导致停运而产生的关联系统风险进行经济性量化求解。 ③对设备在全寿命周期内的运行规律进行分析，根据不同检修等级下设备状态的变化、役龄回退以及对应的收益，，以年费用-效能最优为目标函数，制定全寿命周期的固定检修计划。基于变压器自身运行状态及运行环境，针对可靠性、经济性及检修资源进行多维优化，从变电站级、设备级、部件级对变压器的检修顺序、检修等级、检修部位进行决策，构建了多维度多层面的变压器检修优化策略。
[Abstract]:As the hub equipment of power system, the operation state of power transformer is directly related to the security and stability of power network. In the complex and changeable operating environment, fault diagnosis and prediction for transformer is carried out to master the operating state of transformer. Furthermore, it is of great significance to analyze the operational risk of the equipment and to formulate the relevant maintenance strategy, which is of great significance to the operation and maintenance management under the background of the rapid increase of the power network equipment. Power transformer itself is an extremely complex system. It is difficult to effectively analyze the uncertainty and fuzziness of the current characteristic data, as well as the existing problems in transformer operation state analysis. Based on cloud theory and implicit semi-Markov model, a study method of transformer operation state transition process is proposed. In view of the deficiency of the existing risk assessment and maintenance decision, a dynamic comprehensive risk assessment method for power transformers is constructed based on the equipment's own running state and operating environment, and the economic quantification of transformer risk is realized. Based on the maintenance decision index and efficiency analysis, the fixed maintenance optimization strategy and the multi-dimension and multi-level maintenance optimization strategy based on dynamic risk assessment are constructed. The main achievements of this paper are as follows: 1 in view of the fuzziness of transformer DGA data, uncertainty is divided by cloud model and membership degree of boundary data is judged, and DGA data is trained based on cloud theory. The transformer state observation space is constructed. Based on the rules of state degradation and DGA generation and development during transformer operation, the historical operation state information is introduced into Markov chain by introducing aging factor and multi-step weighted method. The weighted implicit semi-Markov degradation model realizes the accurate prediction of transformer operation state. (2) the degenerate implicit semi-Markov model of transformer operation state transition is used. The failure rate in the transformer life cycle is analyzed, and the influence of different operating conditions on the transformer risk is analyzed, and the hot spot temperature rise is extracted as the parameter reflecting the transformer operation condition comprehensively to correct the transformer failure rate. Based on the analysis of the accident, The related system risk caused by transformer's different risk modes and failure caused by outage is solved quantitatively and economically. 3 the operation rule of equipment in the whole life cycle is analyzed. According to the change of equipment condition under different maintenance grades, the retrogression of service age and the corresponding income, the fixed maintenance plan for the whole life cycle is formulated with the optimal annual cost and efficiency as the objective function. Based on the operation state and operation environment of transformer, the multi-dimensional optimization of reliability, economy and maintenance resources is carried out, and the maintenance order, maintenance grade and maintenance position of transformer are determined from substation level, equipment level and component level. A multi-dimensional and multi-level transformer maintenance optimization strategy is constructed.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TM41

【参考文献】