Research On Life-cycle State Assessment Of AC Contactor Based On Data-driven

AC contactor is a kind of control appliance widely used in various low-voltage electrical control lines.It is mainly responsible for frequently switching on and off the AC main circuit and high-current control circuit from a long distance.As the number of breaking of the AC contactor increases,the electrical contacts gradually wear and the breaking performance gradually decreases.The working state of the AC contactor directly affects the safety and reliability of the operation of the power system,so it is important to accurately evaluate its operating status.The failure mechanism of AC contactor is analyzed in this thesis.On this basis,based on the data-driven method,the fuzzy comprehensive evaluation model and the M-RVM model are established to evaluate the state of the AC contactor.First,the basic structure,working process and failure mechanism of AC contactor are analyzed,and the extraction method of characteristic parameters affecting its operation state is studied.Second,in order to obtain the operation data of AC contactor,the whole life test system is designed.The virtual instrument technology based on Labview is used to realize the automatic control of AC contactor interruption.The life test of CJX2-8011 AC contactor is carried out.The three-phase main contact voltage and current signals,coil voltage and current signals of AC contactor are collected during the on-off process.A large number of degraded data are obtained from the experiment,which provides a data base for the later work.Third,the state evaluation method of AC contactor based on variable weight coefficient is studied.The characteristic parameters affecting the state of AC contactor are selected and the state evaluation index system is established.Based on the constant weight coefficient,the variable weight coefficient method with equilibrium function is introduced to obtain the variable weight coefficients of each characteristic parameter.Based on the variable weight coefficient,the comprehensive evaluation model of AC contactor is studied.Finally,the state evaluation method of AC contactor based on M-RVM is studied.In order to improve the accuracy of state model evaluation,this thesis adopts the relief-F feature selection algorithm to reduce the dimensionality of original feature parameters.The state recognition of M-RVM AC contactor under two input conditions before and after dimension reduction of characteristic parameters is studied.