Online monitoring device for dissolved gas in transformer oil

As the core asset of the power grid, the health of the transformer is directly related to the stability and safety of the entire power system. As an advanced means of predictive maintenance, DGA (Dissolved Gas in Transformer Oil) technology is able to provide a real-time, continuous "perspective" of the internal operating status of the transformer, thus providing early warning of failures before they occur, and effectively preventing major power outages.

From offline to online: the evolution of monitoring technology

Traditional transformer health assessment relies on taking oil samples from the equipment at regular intervals and sending them to a specialized laboratory for analysis. There is a significant lag in this offline testing method. As the sampling period is usually long (e.g. once a year), sudden failures that may occur between tests cannot be detected in time. Statistics show that a significant proportion of transformer failures occur within months of the annual test, which exposes the limitations of intermittent testing.

The on-line DGA monitoring system realizes a revolutionary breakthrough by installing the monitoring device directly on the transformer body. It is able to automatically complete gas extraction and analysis around the clock without interruption, and feedback real-time data to O&M personnel. This proactive, continuous monitoring mode captures the weakest early signals of faults, buying valuable time for preventive maintenance and timely intervention, thus preventing faults from worsening and costly equipment damage and grid paralysis.

Core value of online DGA monitoring

Online DGA monitoring offers multiple core advantages over traditional offline testing:

• Early Failure Warning: It can instantly detect trace amounts of faulty gases due to overheating, arcing or partial discharges, etc., and warn of problems at their earliest stages.

• Improve system reliability: By effectively preventing unplanned outages, it greatly enhances the reliability of the grid's power supply and ensures the safety of personnel and equipment at the site.

• Optimize asset life and cost: Timely maintenance response reduces cumulative damage to a transformer's insulation system, thereby extending its effective service life. In the long run, the economic losses saved by avoiding a single major failure far outweigh the initial investment in an online monitoring system.

• Data-driven decision support: The continuous monitoring data stream provides a solid scientific basis for transformer condition assessment, load management and maintenance strategy formulation, and promotes the development of operation and maintenance mode in a more intelligent direction.

Principles of operation and types of technology

The operating principle of an on-line DGA unit is divided into two main steps: first the dissolved gases are separated from the transformer oil and then the composition and concentration of these gases are precisely analyzed. Extraction techniques for the gases are usually based on headspace circulation or highly selective permeation membranes.

Depending on the monitoring range and analyzing technique, online DGA monitoring devices are mainly classified into the following categories:

1. Key Gas Monitor: This is an economical solution that focuses on the core metrics, primarily monitoring hydrogen (H₂) levels. Since the vast majority of internal transformer faults produce hydrogen, monitoring this gas can provide an effective early warning of a wide range of potential problems. However, it is difficult to make an accurate diagnosis of the root cause of a fault based on hydrogen data alone.

2. Multi-Gas Monitor: Such devices are capable of measuring multiple characteristic gases at the same time, thus enabling accurate diagnosis of the type of fault.

   • Five gas monitoring: Five flammable gases, hydrogen, methane, ethane, ethylene and acetylene, are typically monitored for the presence of overheating or discharge faults.

   • Seven gas monitoring: The monitoring of carbon monoxide (CO) and carbon dioxide (CO₂) has been added to the above. These two gases are key indicators of whether the solid insulating materials (e.g., insulation paper) inside the transformer are aging or decomposing.

   • Nine gas monitoring: It can provide the most comprehensive gas component data, and through comprehensive analysis, it can accurately distinguish between different failure modes such as arc, high and low temperature overheating or partial discharge.

At the level of core analytic techniques, the current mainstream approaches include:

• Gas Chromatography (GC): As the gold standard in laboratory analysis, gas chromatography is capable of accurately separating and quantifying a wide range of gas components, providing highly reliable data. Modern on-line GC systems are highly automated and equipped with automatic calibration to ensure long-term operational stability.

• Infrared Spectroscopy: The technique includes non-dispersive infrared (NDIR) and photoacoustic spectroscopy (PAS), among others. The significant advantage is that it eliminates the need for carrier gases and consumables required for chromatography, thus simplifying maintenance and reducing long-term operating costs.

future outlook

With the deepening of smart grid construction, transformer online monitoring technology is developing in the direction of more integration and intelligence. The future trend is to integrate DGA with other monitoring technologies (e.g., bushing monitoring, partial discharge monitoring, vibration analysis, etc.) to build a comprehensive transformer condition sensing system. At the same time, combining artificial intelligence and big data analysis algorithms to deeply mine the massive monitoring data will realize more accurate fault diagnosis, health status assessment and remaining life prediction, and provide powerful support for the intelligent operation and maintenance of the next-generation power grid.