Digital Condition monitoring for smart transformers
Real-time diagnosis of condition-monitoring data can maximise the benefits of a digital transformer
bySruti Chakraborty and Marius Grisaru
Intelligent frameworks for transformer diagnostics
- Introduction
The decentralisation of power generation sources, strict environmental laws, and various techno-economic regulations are driving the power sector towards optimisation of the existing business model. As a key asset, power and distribution transformers endure severe operational stresses that gradually deteriorate their lifespan and often lead to irreversible damages. Conversely, meticulous condition monitoring of operational transformers can maximise their availability and reliability by optimising the maintenance efforts. It is also the key to stabilise the ageing infrastructure and transformer population by reinforcing effective asset management decisions.
In this context, digitisation of transformer and energy grids can make the business agile, cost-effective, and environment friendly. In fact, a digital transformer enables self-measurement, monitoring, diagnosis, and two-way communication of its condition using various electronic devices in real-time. The interoperability of smart grids using digital transformers can effectively avoid failure occurrence, unwanted downtime, and consequential market backlash.
Real-time diagnosis of condition-monitoring data can maximise the benefits of a digital transformer
A primary challenge for energy utilities right now is to control the operational costs of energy transmission without compromising on asset availability. In this context, a digital transformer is an optimum response to such challenges making the whole process agile, compact, and even intelligent enough to accelerate asset management decisions. Modern-day transformers, bushings, and tap changers contain various built-in electronic devices for self-monitoring purposes. However, interpretation of the data from such devices to achieve a conclusive decision or action plan requires vast knowledge and experience on diagnostic matters. Furthermore, a limited number of human experts and lack of knowledge preservation makes the process tedious, costly, and partially inaccessible. A truly digital transformer that can mitigate the said limitations is yet to emerge. Hence, upgrading the analytical platforms by integrating data collection and storage channels, and maximising diagnostic efforts using artificial intelligence and machine learning can be the optimum solution.
This paper brings a discussion on the existing state and evolution of transformer condition monitoring and systems with respect to digital technologies.
Log in or subscribe to read the whole article.