Identification of defects in overhead ground lines based on GA-BP neural network

The harsh operating environment of overhead ground lines inevitably produces structural damage. Accurate measurement of the size of defects on the surface of the overhead ground line can effectively determine the safety of the structural condition of the overhead ground line. However, it is difficult to quantitatively recognize the size of defects on the surface of the overhead ground line by the current structural condition detection methods, or the detection accuracy and efficiency can hardly meet the actual requirements. In this paper, we firstly constructed a finite element model for magnetic leakage detection of overhead ground line, analyzed the change rule of defect size and magnetic leakage field in the space around the defects and determined the magnetic leakage characteristic quantity, and then used the finite element calculation to construct the magnetic leakage sample database of different sizes of defects, and then utilized matlab to construct the GA-BP neural network to realize the accurate prediction of the width and depth of defects on the surface of the overhead ground line. This method can efficiently and reliably detect the defect size of overhead ground line to ensure the structural safety of overhead ground line, which is of great significance for the safe and stable operation of power system.