New approach of sensor positioning for optimizing source location of partial discharge

Transformers are one of the most important and costly components of any power system and thus proper monitoring of the health of a transformer is indispensable. Online monitoring of partial discharges (PD) is one of the ways by which the risk of catastrophic failure of a power apparatus can be reduced. In order to assess severity of the PD, location of the PD needs to be determined. Study of available literature shows that in case of acoustic methods of PD location, various source location algorithms have been formulated, but for their optimal implementation, it is of utmost importance to have the exact time difference of acoustically emitted (AE) signals. It has been observed that none of the experimental methods have been able to conclusively determine the exact time difference due to the complex nature of the peaks present in the experimental signals obtained. Thus it can be concluded that the experimentally obtained time difference may be only a close approximation of the exact value. Taking cognizance of this problem, usually a time range, within which the exact time difference is expected to lie, is assigned in the source location algorithm. By increasing the time range, the probability of the exact time difference lying in it increases. But increasing the time range also increases the detection of a number of points, which are not faulty, as probable PD points hence require a different approach towards PD source location. This paper proposes a method in which the acoustic sensors are placed in a helical manner, surrounding the transformer tank. The paper also establishes the fact that the proposed method is more efficient than the conventional method. Hence the errors induced in the time difference by the digital filters can be compensated by using the techniques proposed in the paper.