Research On Condition Assessment And Condition Prediction For Power Transformer

Power transformer is one of the most important equipment in electric power system,so it is very important to protect its safety and stable operation on the actual production and people's life.In order to make sure that the transformer could be repaired reasonably,the paper makes a deep study on the state evaluation and prediction about the power transformer.The state evaluation index system is established,and a lot of simulation and example demonstration are made for the evaluation and prediction of power transformer status.Ultimately,the improved assessment and forecasting methods are determined to be true,effective and scientific.Specific works are as follows:(1)The condition evaluation index of power transformer system is studied in this paper.Scientific and effective condition evaluation index system is the basis of condition assessment.This dissertation collects a large number of technical standards,regulations,expert experience and actual state information considering power transformers.Stating with that,the dissertation studies the condition assessment index system.Finally,a completed evaluation index system of power transformer is established from three aspects including oil chromatography test,electrical test and oil test,which ensures the accuracy of transformer state evaluation and prediction.(2)The condition evaluation of power transformers is studied in this paper.This paper further refines the state of the transformer and divides it into five grades.The set pair analysis theory and fuzzy theory are applied to the state evaluation of power transformer.At mean time,a step-by-step analysis of the transformer status assessment is presented,and a theory of subjective and objective integration called integrated weighting is proposed.This theory takes into account the subjective expert experience and objective facts,making the index weight of transformer more reliable and reasonable.The five-level scale method in subjective weighting method is used to reflect the effect of expert experience on transformer state evaluation and the entropy method in objective weighting method is also used to fully response the indicator changes in the role of weight determination.All these are identical with objective facts.Finally,the principle of least squares is applied to integrate the subjective andobjective weight,and the weight of the indicators is scientific theoretically.The final evaluation of the transformer status level is more accurate compared to other methods.This improved evaluation algorithm provides a powerful guarantee for the state maintenance of the transformer.(3)The condition prediction of power transformers is studied in this paper,mainly focusing on data pre-treatment and prediction model.The innovation in data preprocessing is the use of the average weakening operator approach.After the characteristic gas data of the power transformer is averaged and weakened,its smoothness is more in consistent with the requirements of the prediction model,and the accuracy of the prediction is relatively improved.The improvement of the prediction model is that the multivariate gray prediction model MGM(1,N)is used to grouping-predict the characteristic gas of the power transformer.Grouping method is based on the common fault diagnosis method of power transformer,which is based on the three ratio method.This method can avoid the correlation analysis of multiparameter grey prediction model,simplifying the calculation steps of the model and saving computation time.Through the analysis of the experimental results,we can see that the average weakening multi-parameter grey forecasting model basing on the three ratio method is more accurate in forecasting,and the regularity of the system can be better grasped.Finally,the state of the power transformer is predicted by using the ideal point solution theory and the predicted data,which shows that the results were real and reliable.