Abstract:The type and concentration of dissolved gases in transformer insulating oil are used to assess transformer conditions. In this paper, an online detection setup for measuring the concentration of multicomponent gases dissolved in transformer insulating oil is developed, which consists of an oil-gas separation system and an optical system for acquiring the transformer status in real time. The oil-gas separation system uses low pressure, constant temperature, and low-frequency stirring as working conditions for degassing large-volume oil samples based on modified headspace degassing. The optical system uses tunable diode laser absorption spectroscopy (TDLAS) to determine the gas concentration. Six target gases (methane, ethylene, ethane, acetylene, carbon monoxide, and carbon dioxide) were detected by three near-infrared lasers (1569, 1684, and 1532 nm). The stability of the optical system was improved by the common optical path formed by time-division multiplexing (TDM) technology. The calibration experiments show that the second harmonics and the concentrations of the six gases are linear. A comparison experiment with gas chromatography (GC) demonstrates that the error of acetylene reaches the nL/L level, while the other gases reach the μL/L level. The data conforms to the power industry testing standards, and the state of the transformer is analyzed by the detected six characteristic gases. The setup provides an effective basis for the online detection of dissolved gas in transformer insulating oil.

An On-Line Condition Assessment and Diagnosis System for Traction Transformers Based on Dissolved Hydrogen Analysis

On-line DGA Monitoring System for Traction Transformers

Power Transformer Fault Diagnosis Based on Extension Theory

Study on On-Line Monitoring Technique of Traction Transformer Insulation

Wireless Monitoring And Analysing System Of Dissolved Gases In The Transformer Oil

Online Dissolved Gas Analysis (DGA) Monitoring System

A New Condition Assessment Method for Distribution Transformers Based on Operation Data and Record Text Mining Technique

An intelligent monitoring and diagnosis system for 330kV oil-immersed power transformer

A Methodology for Transformer Fault Diagnosis Based on the Feature Extraction from DGA Data

A Methodology to Diagnose Transformer Faults Based on Principal Components Analysis and Artificial Neural Network

Deep Machine Learning-Based Asset Management Approach for Oil- Immersed Power Transformers Using Dissolved Gas Analysis

Convolutional neural networks applied to dissolved gas analysis for power transformers condition monitoring

Development of an Online Detection Setup for Dissolved Gas in Transformer Insulating Oil

Fault Diagnosis Method for Power Transformers Based on Patterns of Dissolved Gases in the Oil

Online State Diagnosis of Transformer Windings Based on Time-Frequency Analysis

Dissolved Gas Analysis of Insulating Oil in Electric Power Transformers: A Case Study Using SDAE-LSTM

Transformer Health Monitoring Using Dissolved Gas Analysis

Assessment of computational intelligence and conventional dissolved gas analysis methods for transformer fault diagnosis

Study on the on-line insulation monitoring technique for transformer oil

Power Transformers Insulation Faults Identification with DGA: A Molecular Dynamics-Assisted Method

Application of Dissolved Gas Analysis for Transformer Condition Assessment Using Computational Intelligence