Application of independent component analysis on state inspection for transmission lines

According to the serious theft of transmission tower materials, a new function in the state maintenance for transmission lines is presented to monitor towers and examine the vibration signal produced from the action of stealing the tower materials, in order to ensure the reliability and safety of the power grid operation. This paper presents the FastICA using negentropy to effectively extract and separate the vibration signals caused by human activity. A new method combined Empirical Mode Decomposition (EMD) method with adaptive threshold pulse distraction method is used to extract the integrity vibration pulses, and suppress interference signals. The practical tests demonstrate the method proposed in the paper is effective in separating and extract the vibration signals successfully.