Research And Application Of Condition Assessment Modeling Of The Power Transmission And Transformation Equipments

Recent years, with the development of science and technology, modern equipments have been applied more and more widely in electric power system. As a result, the significance of fault diagnosis in the display application also becomes more and more important. For a equipment, if its potential failures cannot be found promptly, but in the event of outage maintenance after a failure, it will not only cause the economic loss to people’s production and living, but also even cause the production accidents. In the electric power system, for the purpose of hidden failure investigation, the condition based maintenance could conduct fault diagnosis in the equipments where failures haven’t occurred yet, and issue warning as long as the fault symptoms appear, which therefore has a great significance in daily work and livelihood.Condition Based Maintenance (CBM) is a maintenance technology for the related groups or companies to evaluate the status of the equipment and develop the maintenance decisions to the final status, according to the national standards of equipment status inspection and based on the information obtained from on-line monitoring system. Transformer equipment plays an important role in the power transmission system. So, it is an important component of the maintenance work on the national power equipments to study on the transformer equipment maintenance model, guarantee the healthy status of the equipments, make the corresponding maintenance decision according to the equipments’status, and carry out appropriate maintenance work. Thus, the application prospect of Condition Based Maintenance technique is very broad, and the study of Condition Based Maintenance technique has a very profound significance.Firstly, this dissertation briefly introduces the significance of CBM research overseas and domestic research status and the existing maintenance technologies in electric power system. Secondly, taking the transformer substation equipment as an example, Based on the guidelines of transformer equipment condition maintenance published by State Grid, this dissertation conducts a deep analysis to the evaluating strategies of the status of this kind of equipment, and then proposes some improvement, according to the characteristics of transformer equipments and the original maintenance strategies,.This dissertation presents regression model based single evaluation index of transformer equipment. Regression model has already been widely used in many areas, such as oilfield output forecast, disaster strategy prediction, mathematical data processing and physiological evaluation and so on. There is usually a nonlinear relationship between two variables in nature. Therefore, regression algorithm can be adopted to model the transformer condition evaluation, to simulate the relationships between the possessed evaluation of transformer equipment and various parameters. According to the characteristics of the observations, regression analysis can be divided into linear regression and nonlinear regression where linear regression analysis is the most fundamental study method.On the basis of using the regression to model the single evaluation of the transformer equipments, this dissertation also adopts a widely used BP neural network to model the overall status of a equipment, thereby to predict status of the equipment, find the faults in time and then process. Artificial neural network deals with the data information by collecting, calculating, memorizing and processing according to the characteristics of human neurons. BP model is a multilayer feed forward neural network. It is currently the widest and most important learning algorithm to train the feed forward neural networks. It can perform adaptive learning according to the change of the evaluation guidelines, and then verifies the feasibility of the method through experimental results.Finally, this dissertation practices the research results in real-life application. The maintenance method of power transmission and transformation equipment in substations has been deep studied. By fully understanding the needs of manufacturers of equipment condition maintenance system, with the sampling data from the individual monitoring subsystems, the equipment condition maintenance system has been designed and developed and eventually applied to actual and put into use.