| Power transformer is one of the critical power system equipments,is the bridgebetween power plants and electric users,and bears the arduous tasks of transforming.Reliable operation of power systems and stability of transformer are inseparable,It’s meaningful to monitor the health of transformers and make a rationalrepairment or scrapping plan,and maximize the economic effect. But variousassessed factors of power transformers often contain great uncentainty during itsrunning,so this paper proposes a residual life prediction model of powertransformers based on the uncertainty theory and the Markov theory,then achieves areasonable assessment of the transformer life.Firstly,this article analyzes the characteristics of power transformers life inservice, summarizes the factors affecting the remaining life of transformers fromthe oil gas analysis, electrical testing, analysis of insulating oil, polymerizationdegree of insulating paper, furfural content analysis and winging mechanicalstrength,then builds the prediction model of the transformer life from theserespects.Next the article deals the randomness,fuzziness and random fuzzy doubleuncertainty of factors with probability theory,credibility theory and random fuzzytheory,then reflects the use of the transformer in the form of health indicators andelaborates evaluation algorithm.Secondly,this paper established residual life prediction model of powertransformers based on random fuzzy simulation and improve Markov algorithm,verifies the rationality and feasibility of the model by a numerical examplesimulation.Then analyzes distinction and applicability of the two models’simulation results.Finally,this article takes the economic life of power transformer as agoal,researches the economic life of the transformer cost component units, handlesthe uncertainty of component units with the theory of uncertainty,constructs a lifemodel of the objective function for the annual minimum life cycle cost with theinterval analysis,so achieves economic life assessment of power transformers,thenverifies by a numerical example simulation. |