Work

Toward Online Frequency Response Analysis of In-Service Power Transformers: A Single-Coil Three-Phase Parallel-Connected Sensing System

Year: 2026

Type: article

Abstract: Winding deformation is an important contributor to power transformer faults and forced outages. Effective and mature online detection methods are still lacking. The online frequency response analysis (FRA) method is a promising candidate for further development. A major obstacle in online FRA is the strong interference posed by the high-amplitude power-frequency load current (hundreds of amperes) to the extremely weak FRA test signal at the sub-milliampere level. This paper proposes a single-coil, three-phase parallel-connected sensing system, which employs three identical coil-type current sensors. Its capability to suppress power-frequency interference while measuring high-frequency signals has been theoretically analyzed and experimentally validated. An online measurement system based on it was applicated on an operating 110 kV power transformer. Field measurements verified that the proposed single-coil sensing system suppresses power-frequency interference by a factor of 100. Furthermore, the new frequency response functions derived from the online current signals exhibited resonance features closely aligned with the conventional offline FRA method, confirming the method's stability and suitability for diagnosing winding deformation in in-service power transformers.

Source: IEEE Access

DOI: 10.1109/access.2026.3660082

Authors: Sen Yang, Yuhan Wang, Guangmao Li, Yangchun Cheng, Shirui Gao, Yuanshu Wei

Institutions: Guangdong Power Grid Company (China), China Southern Power Grid (China), North China Electric Power University

Language: English

Topic: Power Transformer Diagnostics and Insulation

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

SDG: Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $2,075