Search

This paper introduces a computerized on-line partial discharge (PD) monitoring and diagnostic system for transformers. The system, which is already in use in a power station, uses wide-band active transducers and a data acquisition unit with modularized and exchangeable components. The system software is a power equipment monitoring and diagnostic system, which is based on the component object model, and was developed for monitoring multiple parameters in multiple power supply systems. The statistical characteristics of PDs in power transformers were studied using 7 experimental models for simulating PDs in transformers and 3 models for simulating interfering discharges in air. The discharge features were analyzed using a 3-D pattern chart with a three-layer back-propagation artificial neural network used to recognize the patterns. The results show that PDs in air and oil can be distinguished. The model can be used for interference rejection on-line monitoring of partial discharge in transformers.

To suppress interference is a critical technique of online partial discharge monitoring for power equipment insulation. They can be divided into two types: periodical and impulse interference according to their signal characteristic. Different anti-interference techniques can then be chosen to suppress them separately. This paper describes various anti-interference techniques which have been developed recently and their actual effect on online partial discharge monitoring. The above techniques include a combined filter, a digital filter, a differential system and a pulse discrimination system. This paper presents some other anti-interference ideas which are satisfied for online partial discharge monitoring. >

A computer-aided online detection system of partial discharges (PDs) in power transformers is described. The impulse electrical signals of PD are detected with a current transducing unit and the acoustic emission signals of the PD are detected with acoustic transducing units. Various detected analog signals are transmitted with an optical fiber transmission system to the data-acquisition circuit and microcomputer. The frequency-division multiplexing technique is applied to transmit four channels of analog signals in one optical fiber. The amplitudes of detected electrical and acoustic signals are acquired and transmitted to the microcomputer with DMA. The microcomputer is used to store and to process the acquired signals. Two kinds of online PD calibration methods were developed, i.e. the transformer bushing method and the impulse charge couple method. With these methods various functions of this detection system are realized, such as offline and online PD detection, online PD calibration, recognition of electromagnetic interference, and location of the PD source. Onsite experiments with this system were performed in several power substations, and the test results were satisfactory. >

To analyze the statistical characteristics of partial discharges (PDs) in power transformers, 7 types of experimental models simulating PD in transformers and 3 types of models simulating interfering PD in air are designed and model experiments are conducted in a screened room. Using a digital measuring device with the sampling rate of 50 kHz, the quantity phase information of PD pulse current in models is obtained after the signal is processed by a peak-holding hardware. The PD features are extracted using the 3D pattern chart and then a group of three-layer back-propagation ANNs (artificial neural networks) is used to recognize the PD patterns. The investigation shows that ANN has enough ability to recognize different PD in oil-paper insulation of power transformers and PD interference from air can be recognized and eliminated.

This paper discusses the development of a simulation model used to study the propagation of partial discharge (PD) pulses of transformers in a 500 kV substation. The results are used to design an on-line PD monitoring system for transformers. The impedance matrix of a 500 kV single-phase transformer is computed with a special program based on geometry data and its external coupling network is calculated with EMTP. The factors that affect the coupling network (often referred to as coupling impedance) are also analyzed. It was found that the frequency characteristic of the coupling network is complex and an equivalent lumped coupling capacitance representation is only feasible below 60 kHz. The impedance varies greatly as frequency increases. The simulated PD current responses are obtained in tank grounding and 500 kV bushing tapping grounding and the optimal monitoring point is discussed. Applying resonant frequency of PD responses as monitoring frequency is suggested; this can increase the monitoring sensitivity effectively. A practical off-line PD calibration model for transformers in operation is designed according to the simulation results.