Fuqi Li

PD pattern recognition for stator bar models with six kinds of characteristic vectors using BP network

The application of six different kinds of characteristic vectors to recognize PD sources is studied. Four kinds of model bars are used to simulate typical partial discharges in generator stator winding. The PD signals were measured by using a computer aided digital sampling system. The sampling results are processed by six kinds of feature extraction methods and different characteristic vectors are obtained. Then these vectors are used as input patterns for BP network. Recognition results using all six kinds of vectors are reasonable. Further analysis shows that vectors formed by moment features or fractal dimensions possess fairly good abilities of pattern identification and data compression.

Internet based virtual hospital architecture for power equipment

Equipment diagnosis and maintenance plays a key role in reliable power system operations. Information technology opens new doors in this area with powerful and convenient information sharing capabilities. The concepts and prototype development of an Internet based virtual hospital (VH) for power equipment are introduced in this paper. The VH will serve as an information center for the power equipment diagnosis and maintenance community. The virtual hospital will feature a diagnosis/maintenance knowledge and education center, a diagnostic tools center, a worldwide maintenance expertise pool, forums and user groups, links to equipment manufacturers, links to standards, and many other relevant information. VH provides a convenient platform for experience/information sharing, discussions, and worldwide collaborations.

PD monitor system for power generators

This paper introduces a computerized PD monitoring system for generators and presents the experimental and numerical study of discharge pattern recognition methods. The system has specially designed transducers, data acquisition unit and software, and can obtain statistical as well as individual discharge information. In order to validate the performance of the system, experiments were done in the laboratory, using elaborately designed models that can generate various types of discharges. Feature extraction of the gathered data and neural network (NN) classification of the acquired discharge patterns were studied. The results showed that the surface fitting method is able to extract features from statistical data of discharges, and that NN is a potential classifier in practical applications.

Partial discharge acoustic emission detector based on computer for power capacitor

A computer-based digital measuring system for detecting partial discharge (PD) within a power capacitor is introduced in this paper. The principle and construction of this instrument and some results obtained from typical discharge experiments are given. The electrical signal caused by PD in a capacitor with large capacity is very weak; therefore this system is based on acoustic emission (AE) method and detects the acoustic pulse signal caused by PD. Mass capacity, high-speed synchronous A/D (analog/digital) converter cards are used to record the acoustic signals and this makes the subsequence processing possible. Through a series of processing, the frequency spectrum of PD signals, the discharge waveform information and the PD characteristic reflected by /spl phi/-q-n chart (/spl phi/: occurring phase, q: discharge amplitude, n: repetitive rate) could be extracted. Some feature extracting methods such as waveform feature analysis, frequency spectrum analysis and artificial neural network used as a classification tool are applied to recognize the discharge pattern. The study results could provide reference for a factory to improve manufacturing technologies.

PD pattern recognition of power capacitors model based on Combinational neural network

Five types of partial discharge (PD) models are designed to represent typical PD phenomena in power capacitors. A computer-based acoustic emission signal detecting instrument that is developed for PD is used to collect a lot of acoustic signals of model discharge. The acoustic signal duration of gas cavity discharge is very short, therefore, it is easy to be differentiated from other discharge patterns. According to the time domain and frequency domain graph of acoustic signals, the characters of oil gap discharge is close to that of discharge along oil impregnated paper surface. However, the character of discharge of metallic impurity in oil impregnated paper insulation is close to that of surface discharge of bushing. Combinational neural network (CNN) is used to recognize the five kinds of representative discharge patterns. The result shows that CNN is effective and the characteristics of acoustic signals could be used to recognize PD patterns in power capacitors.

A novel on-line PD monitoring and diagnostic system for power transformers

This paper introduces a computerized online PD (Partial Discharge) monitoring and diagnostic system for power transformers, which has been installed in a 500 kV substation in China. It consists of wide-band active transducers and DAU (Data Acquisition Unit) with modularized and exchangeable components. This system has been developed to monitor multiple parameters of multiple power transformers, which is based on COM (Component Object Model). In addition, the system can also be used to monitor other power equipment. Some results of on-line monitoring are discussed.

Interference analysis and rejection of partial discharge (PD) monitoring signal on-site

According to the waveform and spectrum characteristics, various interferences during on-line PD measurement could be divided into three types such as the sinusoidal continuous noises, pulse shaped interference (including the periodical and stochastical) and background noises (similar to the white noise). Generally speaking, various digital filters could be used to suppress the sinusoidal noise. Some rejection methods for the pulse shaped interference have been developed according to some waveform characteristics and statistical features in the time domain. And the wavelet analysis is a feasible solution to reduce background noises. This paper makes an overall introduction to the anti-interference architecture of a PD monitor and diagnosis system installed in a transformer-generator unit of a power plant in China. Results were also provided in detail of the analysis and rejection for various interferences during on-site PD measurement.

The UHF method for measurement of partial discharge in oil-impregnated insulation

In recent years, the ultra high frequency (UHF) method is employed to suppress the interferences during the partial discharge (PD) measurements with the detecting frequency range from 300 MHz to 1.5 GHz. However, with such width of frequency range, the interferences, sometimes, can not be avoided. In order to suppress the interferences for on-line PD detection of a power transformer effectively, a narrow band UHF method is put forward in this paper. The fundamental works have been carried out in high voltage laboratory to confirm the frequency range for PD detection with UHF method. To suppress the continuous and random UHF interferences, the envelope detection technique is introduced in the paper too. The experimental results show that the frequency of the UHF signals radiated from PD sources in oil-impregnated insulation is mainly concentrated on the range from 0.4 GHz to 0.8 GHz. The envelope detection can not only suppress the continuous high frequency interferences, but also reduce the sampling rate of PD measuring systems.