Study For On-line Monitoring Method And Device Of Polluted Insulator Electrical Insulation Strength Based On Acoustic Emission

Due to the long-distance transmission, the insulators in the transmission line are covered with dust and powder contamination of various industrial emissions. With time goes on, the contamination will be more serious. In some meteorological conditions, such as fog, drizzle, condensation and other wet weather, the dirt surface of the insulator are affected with damp moist which lead to the decline of electrical insulation.The transmission lines are apt to causing pollution flashover which is a serious threat to the security and stability of the power system.On the basis of referring to a large number of literature, the on-line monitoring method of the polluted-insulators discharge of the transmission line was studied. Actually, the discharge intensity of polluted-insulator reflects the polluted-insulators external insulation status. Insulators are going through the normal insulation condition to the flashover, corresponding to the polluted-insulators discharge enduring three stages which are corona discharge, partial discharge and arc discharge. The more serious the polluted-insulators discharge is, the more pessimistic polluted external insulation status is. Based on the acoustic emission(AE) technology, the relationship of the acoustic emission signals and the different stages of polluted insulator discharge was studied.The study carried out in both theoretical and experimental aspects is described as follows.From the phenomenon of increasing polluted insulator discharge accidents, this study carries out the research work of the relationship of acoustic emission characteristics, acoustic emission signal feature and different stages of polluted insulator discharge based on acoustic emission technique and the method and equipment of on-line monitoring.The mechanism of polluted insulatorformation, the phenomenon of acoustic emission of polluted insulator is analyzed in this paper. It is also studied that the emission source characteristics of polluted insulator discharge which refer to the feature of acoustic source, the distribution of sound field and the propagation characteristics of polluted insulator discharge. These analyses provide theoretical bases of acoustic sensor design, location of its installation and the criteria to the division of the different stages.The acoustic emission signals’ characteristic of three discharge patterns is discussed, which are corona discharge, partial discharge and are discharge. The feature, which can reflect the different stages of polluted insulator discharge effectively, accurately and reliably, includes amplitude, average power which means sound pressure amplitude of the integrated area of per unit time and frequency of the acoustic emission signal, which can provide the bases of dividing the intensity of polluted insulator external insulation and the status of insulation precisely.Different material and model insulators with different ESDD were used to conduct polluted-insulators discharge tests so as to prove the correctness and feasibility of polluted-insulators discharge online-monitoring method. Polluted-insulators discharge process consists corona discharge, partial discharge and arc discharge, in which the partial discharge was divided into mild, moderate and severe partial discharge. Tests on the relationship of different discharge strength and the AE signal characteristics were carried out. Then AE signal emitted in different discharge stages of polluted-insulators in operation were analyzed to confirm the relationship between different discharge-intensities and the AE signal characteristics. Arc is the important phenomenon flashover is adjoining, so the characteristics of arc discharge are one of the key research contents.Characteristic quantities, such as amplitude, average power and frequency., were researched, and the extraction methods were proposed. Frequency characteristic is not affected by the detection distance, so frequency characteristic and its extraction method were the focused research content. The empirical mode decomposition(EMD) was used to decompose polluted-insulators discharge AE signal, combining fast Fourier transform(FFT). Frequency characteristics of AE signal in different polluted-insulators discharge stages are extracted, combining the amplitude and power characteristics, is realized.To enhance the discharge pattern recognition validity, algorithm based on kernel principle component analysis(KPCA) and random forests was researched. Based on the experimental polluted-insulators discharge AE signal data, sufficient sample data were selected, and main characteristic quantities, such as amplitude, energy and frequency, were extracted. With least squares support vector machine (SVM) as the basic classifier, the polluted-insulators discharge-intensity and external-insulation status recognition model was established. The recognition validity of the model are verified and it can be used to recognize the external-insulation status correctly.Parabolic focused acoustic sensor and polluted-insulators online monitoring device prototype were developed. Experiments on polluted-insulators external-insulation intensity were carried out.When the polluted-insulators discharge are close to severe partial discharge, the device can give the warning message and when it reaches arc discharge,it can give early warning message. The results show that the correct auto-recognition of different polluted-insulators discharge stages is realized.The polluted-insulators online monitoring device based on AE technology and its related methods have been put into operation.