Research On Life-cycle Condition Monitoring And Degradation State Recognition Of AC Contactor

AC contactor is a kind of low-voltage electrical apparatus used to switch on and off AC circuits, which is widely used in various fields of national economy. AC contactor is usually used to control a motor and also used in other electrical power loads, such as electric heater, power units and electric welder. AC contactor has the advantages of great capacity control, remote operation and so on. The fault of AC contactor will cause enormous losses because it works under frequently breaking conditions. It is of great significance to forecast the working conditon and reliability level of AC contactor for safe operation of power system, condition monitoring and degradation state recognition is of equal importance.The degradation mechanism and evolution rules of AC contactor is systematically analyzed in this thesis, degenerative process is demonstrated by using mathematic model. The feature information extraction based on life-cycle condition monitoring data and HSMM degradation state recognition of AC contactor is studied on the basis of preceding part of this paragragh.First, main degradation mechanism mode and evolvement rules of AC contactor is systematically analyzed in this dissertation as a basis. The HSMM degradation model is constructed making use of the common ground between HMM and the degradation process of AC contactor that sense changes of state via observed values. The characteristic that HSMM can logically describe degradation state resident time in the evolution process is another reason to construct HSMM model, which lay a foundation for the next research.Second, test platform is designed and built to perforam an life-cycle experiment for CJX2-8011 AC contactor as the object of study. A great deal of effective degradation data is gained during the process of the test, which provides preliminary data support to AC contactor state monitoring and degradation state recognition based on data driven.Third, the feature information extraction technology based on wavelet transform and comprehensive evaluation of improved principal component analysis is studied in order to resolve the feature information extraction problem of the degradation state recognition. Feature information extraction has direct relation to the accuracy of the degradation state recognition, yet the noise is the main obstacle. In this thesis, comprehensive evaluation of improved principal component is used by condition monitoring data after wavelet transforming, quantitative evaluation is given for each signal of decomposition and reconstruction, providing the accurate feature information.Finally, to determine current degraded condition of the AC contactor, HSMM degraded state recognition techniques have been obtained. In order to improve the accuracy of forecast and state recognition, KPCA is proposed from the perspective of multi-source information fusion, the degradation state recognition method of AC contactor is studied based on KPCA-HSMM.