Global committement towards sustainability and environmental conservation as a part of Agenda 2030 is pushing for new stragies for a complete Life Cycle Analysis (LCA) including noise impact assessment of products and processes. Transformers, being vital assets in the power sector with significant and often less-discussed environmental impacts, are also under radar.
What causes transformer noise?
Typically, vibration and noise from a transformer are excellent markers of its performance and operational quality. As mentioned earlier, some of this noise exists due to the flow of alternating current across the windings. However, unchecked contamination of transformer oils can also cause buzzing beyound permissible limits.
Polychlorinated Biphenyls (PCBs)
Over the years, electrical transformers filled with insulating fluids often face harsh checks for long-term contamination of Polychlorinated Biphenyls or PCBs. These are persistent and pernicious pollutants with severe environmental impacts. Specific details are available by bodies like Stockholm Convention which regulates the deadlines for decontamination and disposal worldwide.
Can tools like health indexing help?
Health indexing is a popular strategy within the power industry to rank transformers based on condition monitoring data, visual inspection, historical profile, and field information. It ensures a quick overview on the risk and reliability aspects of a transformer using compound quantitative data and translating that in to qualitative information. The health condition of transformers have a strong impact on the environment and is a function of its carbon footprint. Tools such as Ronin allows one to take the necessary actions to improve them and reduce energy consumption and environmental impact.
Challenges in PCBs sound management
To provide an effective strategy on environmental sound management in favor the Global Environmental Facility (GEF), supported by the World Bank, finances disposal projects. They consider that the preliminary inventory step is to identity contaminated transformer among fleets. The decision-making criteria is to choose the best available technique and prioritize actions.
However, the main obstacle during project debriefings was to define a univocal criterion to prioritize assets treatment. Experts often debate on the best model to use, particularly due to lack of data. The final aim is to decide whether a contaminated transformer should be scrapped in authorized centers, often in distant countries from where it is located, or to be decontaminated and treated on site.
Depending on the choices made, the costs and the technical and logistical implications differ and can cause significant gaps between the allocated and actual expenses.
There are very few technical tools available to experts right now. These are based on the individual expert’s assessment of the data available on the level of contamination. Moreover, as the data is not cross-referenced with the functional state of the transformer, defining cost-effectiveness of decontamination actions becomes challenging.
How can Ronin and AI offer effective PCB management?
An Original Equipment Manufacturer (OEM) will always evaluates carbon emission and its potential impact along the production steps to certify carbon footprints on the frame of corporates Environmental Sustainability and Governance (ESG) matrix. This is where Artificial Intelligence (AI) helps by providing a key to interpreting intervention priorities.
For example, Ronin AI Dashboard has a range of application from 500 kva up to 800 MVA for any type of oil filled transformer which are commonly susceptible to PCB contamination. We can offer a computational matrix by including PCB contamination in oil versus noise levels to match the transformer health index.
This will ensure the best response for the asset in the form of intervention, disposal or decontamination. It is possible to screen thousands of assets by ranking their health index and expect compliance with estimated lifespan in few second. Even the lack of data while testing Ronin’s efficacy on this will not be a hindrance thanks to its atypical inference pattern. Once the remaining lifespan is identified, it willbe possible to create a decision tree based on contamination level and economic impact of disposal/decontamination.
