Intelligent background noise reduction technology in cable fault locator using the magneto-acoustic synchronous method

摘要

The magneto-acoustic synchronous method has found wide application in accurate positioning of power cable fault due to its advantages of high accuracy and strong ability to reject interference. In the view of principle, the magneto-acoustic synchronous method needs to detect the discharge sound signal and electromagnetic signal emitted from the fault point, but the discharge sound signal is easy to be interfered by the ambient noise around and the magnetic sound synchronization. Therefore, it is challenging to quickly and accurately detect the fault location of cable especially in strong background noise environment. On the other hand, the spectral subtraction is a relatively traditional and effective method in many intelligent background noise reduction technologies, which is characterized by a relatively small computational cost and strong real-time performance. However, its application is limited because the algorithm displays poor performance in low Signal to Noise Ratio (SNR) environment. Aiming at the shortcoming of the spectral subtraction that de-noising effect is weak in low SNR environment, this paper proposes an improved spectral subtraction combining the magnetic sound synchronous principle and analyzing the properties of discharging sound. This method can accurately estimate noise in real time and optimize the performance of the basic spectral subtraction thus solving the problem that the magneto-acoustic synchronous method is unsatisfactory for positioning cable fault in the strong background noise environment.