TEM apparent resistivity imaging for grounding grid detection using artificial neural network. Issue 17 (19th August 2019)

• Record Type: Journal Article
• Authors: Qin, Shanqiang; Wang, Yao; Tai, Heng‐Ming; Wang, Haowen; Liao, Xian; Fu, Zhihong
• Is Part Of: IET generation, transmission & distribution. Volume 13:Issue 17(2019)
• Page Start: 3932
• Page End: 3940
• Publication Date: 2019-08-19
• Subjects: neural nets -- substations -- nondestructive testing -- fault diagnosis -- earthing; TEM apparent resistivity imaging -- grounding grid detection -- artificial neural network -- transient electromagnetic method -- substation -- TEM system -- TEM response characteristics -- recorded TEM data -- apparent resistivity image -- TEM imaging method; Electric power production -- Periodicals; Electric power transmission -- Periodicals; Electric power distribution -- Periodicals; 621.3105
• DOI: 10.1049/iet-gtd.2018.6450
• Languages: English
• ISSNs: 1751-8687

Abstract

Abstract : Transient electromagnetic (TEM) method is a reliable means of non‐destructive testing for fault diagnosis of grounding grid in substation. In view of the small transmitter loop of the TEM system and many measuring points and survey lines of grounding grid detection in substation. This study presents a fast solution to TEM apparent resistivity imaging for grounding grid detection using artificial neural networks. The input ‐output mapping relation of neural network is established based on TEM response characteristics of the grounding grid. The built network could map the recorded TEM data of grounding grid detection and quickly obtain the resistivity image. The proposed method offers accuracy and fast computation for the resistivity imaging of grounding grid. Feasibility and technical attractiveness of the proposed method in fast imaging of apparent resistivity is investigated with the measurement of grounding grids in the actual substations. It can be used in real time so that the recorded TEM data in substation can be calculated without re‐training, which avoids time‐consuming inversion computation. The rapid processing of massive data and submitting the detection results of the grounding grid to customers rapidly and in real time, which is expectation for a modern power industry.