Study On The On-line Dielectric Loss Tangent Monitoring Technique Of Capacitive Equipment

Dielectric Loss Tangent (tang) is a key parameter in capacitive equipment insulationon-line monitoring, which can reflect the insulation station of the capacitive equipments.However, it is a challengeable task to measure tanδprecisely because many factors introduceconsiderable error to the measurement of tanδ. Therefore, tanδon-line measurement methodshave been a hot research task for many scholars in the field of power equipment on-linemonitoring. It is very useful and meaningful to study and apply advanced monitoring anddiagnosis techniques. In this dissertation, the high accuracy measurement of tanδ,preprocessing of the on-line data, analysis of the data regularity and application of thediagnosis approaches for capacitive equipments are investigated. The main researches are asfollows: The influential regularity of each factor on tanδseries is researched. The on-linemonitoring data series of capacitive equipments are analyzed in frequency domain. Based onthe frequency spectrum characteristic of the series, the model of the tanδdata series isconstructed. Then a approach to preprocess the on-line data based on the adaptive generalizedmorphological filter is designed. The results of the Simulation study show that the generalizedmorphological filter can suppress different kinds of the white noise and pulse noise among thedata. The principle of the trend-extracting approach through relative comparison is introducedand its problems are pointed out too. Based on the filtering idea, the morphological filterused to extract the trend of tanδdata are designed. A virtual instrument based onmorphological filter is presented in this paper. By applying the construction of virtualinstrument, simulation test can be completed. The analysis results of the simulation dataverify the effectiveness of the above approaches. Finally, the novel diagnosis approaches areintroduced and the control mode of morphological filter is discussed.