Study On Life Cycle Cost Based Maintanence Decision Making For Power Transformers Considering Condition Assessment And Insulation Life Assessment

An oil filled power transformer is a core component of the power system forenergy transmission and conversion, and its operating status directly affects the securityand stability of the power grid. Once a transformer breaks down, it may cause hugeeconomic losses, and it even causes serious social impact. However, blind repairs willlead to a lot of waste of human and material resources. Therefore, researches oftransformer insulation condition assessment, insulation life prediction, life cycle costanalysis, operational risk of the transformer, and condition based maintenance decisionmaking of the transformer considering the reliability and economy of transformeroperation, which can choose an appropriate condition-based maintenance strategy toreduce accident probability and promote the economical efficiency of electricalcompanies.Based on regulations, this thesis studies transformer fault diagnosis using ICAoptimized SVM model, fault forecasting using cloud model, insulation assessmentmodel using cloud model and evidential reasoning, insulation life assessment modelusing time, temperature, and water information, transformer risk assessment model, andlife cycle cost model. Based on the transformer risk assessment model and life cyclecost model, a maintenance decision is made considering the reliability and economy oftransformer operation for a running power transformer. The achievements are shown asfollows.A decision-making assessment model based on transformer health index isproposed in this dissertation. A cloud and evidence reasoning model of transformerinsulation condition assessment is proposed. A cloud and evidence reasoning model isused to combine the evaluation layer. The assessing results show that the proposedmethod is effective. In addition, considering running time, operating environment,inspection record, and accessories condition, transformer condition assessment modelbased on health index is established in the paper. Cases study shows that the model iseffective.To improve the existing thermal aging oil-paper insulation life assessment model,the influence of initial moisture content and temperature on aging properties ofoil-paper is investigated. Based on the time-temperature superposition (TTSP) method,a time-temperature-moisture superposition (TTSMP) method is proposed to improve the lifetime model. In order to get the water content in paper, a water content of paperassessment method is studied based on FDS and water balance curve. In order to obtainthe DP of the insulation, an insulation aging assessment model based on furfural contentin oil is studied. The transformer hot spot temperature calculation model based on LoadGuidelines is applied to obtain the average temperature of a transformer. Compared withthe previous lifetime models, the benefit derived from the improved method is that notonly temperature, but also initial moisture content in paper is considered, which ishelpful for more rational estimations of insulation lifetime during long-term aging.Aiming at solving the problem that reliability and economy of power transformersare not reasonably considered in the condition based maintenance problem recently, amaintenance decision making model, considering reliability and economy, based ontransformer risk assessment and life cycle cost analysis, is established to select the bestmaintenance strategy of a transformer. The model assess the best ratio of the risk-benefitand the annual life cycle cost in order to obtain the best maintenance decision making.The case study shows that the proposed maintenance decision making model canprovide effective advice to electrical companies.