Power substations are advancing with sustainable energy

Emerging technologies such as artificial intelligence (AI) and machine learning (ML) have transformative potential in optimising energy systems and substations.

Technological advancements such as artificial intelligence (AI) and machine learning (ML) have transformative potential in optimising energy systems and substations. These technologies enable utilities to harness data collected from grid devices such as transformers, meters, and feeders, leading to predictive maintenance and enhanced outage management. By analysing data in real time, AI and ML can identify patterns and potential issues before they become critical, thus minimising downtime and improving operational efficiency.

Data analytics enable predictive maintenance, which allows for servicing equipment based on its condition rather than following a fixed schedule. This targeted approach reduces unnecessary maintenance and associated costs, prolonging the equipment’s lifespan. For example, companies can use data analytics and machine learning to predict when a 400 kV transformer requires maintenance based on performance data rather than arbitrary time intervals.

In addition, the concept of a digital twin, a virtual replica of a transformer or other equipment, is gaining traction in the industry. Digital twins provide real-time monitoring and analysis, improving maintenance schedules and extending equipment lifespan.

Integrating information from various grid devices into a centralised system offers a holistic view of the network’s status, allowing for coordinated efforts between healthy and compromised feeders. This approach streamlines supply chain operations, reduces supply outages, and enhances customer communication. However, the successful integration of AI and ML with distributed control systems requires robust cybersecurity measures to protect against data breaches and maintain the reliability of power systems. Adopting these emerging technologies can lead to a more efficient, reliable, and sustainable energy supply, benefiting the industry and consumers. Industry experts share their opinions on the subject with the EPR Magazine. Let us understand what they have to say.

Emerging technologies

In today’s rapidly digitising world, one of the major challenges in the energy sector is the need to use data acquired at the relay level more. Despite significant investments in relay digitisation, much of the potential data must be explored. This creates missed opportunities for predictive maintenance, system optimisation, and enhanced customer service.

Effective software integration is necessary for the proper utilisation of available data. When data from devices like transformers, meters, and feeders is collected and processed accurately, it can reveal valuable insights into current levels, system logic, and potential failures. For instance, observing a 6-crew ampere current at a tipping point could signal a specific outage type, enabling targeted responses and minimising disruptions.

Rajiv Goyal, CEO and full-time director of EKI Power Trading, says that energy companies can transform raw data into actionable intelligence by employing AI and machine learning. For example, AI can predict and identify issues in real-time, allowing for preventive measures and better outage management. Combined with customer service systems, this can improve transparency and communication with customers regarding outages and expected restoration times.

He further opines, “Integrating information from various grid devices into a centralised system can provide a holistic view of the network’s status. This centralised approach can help coordinate efforts between healthy and compromised feeders, streamline supply chain operations, and reduce supply outages. To optimise the energy sector’s operations and improve customer satisfaction, companies must invest in software solutions that integrate data from different grid devices.”

“AI and machine learning can enhance predictive maintenance and outage management, leading to a more efficient and reliable energy supply. This, in turn, will benefit not only the industry but also the end consumers,” he adds.

Sanjiv Prasad, Vice President and head of Power & Utilities, RIL-DMD, comments that artificial intelligence (AI) and machine learning (ML) have significant potential for advancing electrical systems, including substations. AI can aid troubleshooting by analysing data to identify potential issues before they become critical, thus improving maintenance and reliability. ML can optimise operational efficiency and safety through predictive maintenance, anomaly detection, and load forecasting.

He further says, “The concept of a digital twin, a virtual replica of a transformer, is gaining popularity in the industry. It allows real-time monitoring and analysis, providing insights into performance and potential issues. This digital representation can improve maintenance schedules and extend the lifespan of equipment. However, the widespread adoption of digital twins in power plants and electrical systems is still developing. Integrating AI and ML with a distributed control system (DCS) can centralise data management and provide a holistic view of the electrical system, enhancing operational control and efficiency.”

Suvendra Kumar Senapati, Head of Sales & Commercial, L&T Digital Energy Solutions, shares his views. He says that AI can play a crucial role in modernising and optimising energy systems, whether upgrading existing infrastructure or retrofitting systems. It aids in operational management, release planning, and decision-making across energy sectors such as generation, transmission, distribution, and substations.

“AI can also support renewable energy planning and the integration of electric vehicles by optimising charging infrastructure. By analysing data and predicting patterns, AI helps improve efficiency and reliability. Proper utilisation of AI can enhance the overall effectiveness of energy systems, from the top management level down to individual operations, leading to smarter and more sustainable energy solutions,” he comments.

Data analytics and cybersecurity measures

Data analytics is integral in various fields, particularly machine operations and maintenance. Companies can achieve greater reliability and minimise downtime by analysing data collected from machinery and systems. Currently, many substations follow a preventive maintenance schedule based on past experiences. However, transitioning to data-driven, predictive maintenance can optimise operations and improve efficiency.

Sanjeev Prasad comments, “Predictive maintenance allows for equipment to be serviced only when necessary, based on observed data patterns and potential issues. This reduces unnecessary downtime and the costs associated with it. For example, a 400 kV transformer might traditionally undergo maintenance after several years, even if it is functioning well. Data analytics allows companies to predict potential issues and perform maintenance only when required.”

He further says that incorporating machine learning and real-time data monitoring, such as using cameras to inspect equipment, can provide insights and early warnings of possible failures. This data-driven approach supports targeted interventions, ensuring the longevity and performance of equipment. Overall, effective data analytics helps optimise maintenance schedules and improve operational reliability.

Sharing his views, Mahendra Singh says that migrating from time-based maintenance to condition-based or predictive maintenance is crucial for modern utilities. Instead of following a strict schedule (e.g., servicing a breaker every two years), utilities can use real-time data to determine when maintenance is needed. This shift allows for proactive interventions and extends equipment lifespan.

In addition to his views, he says, “In substations with hundreds of pieces of equipment, data analytics plays a key role. Manually analysing reports for all equipment is impractical. Instead, automated systems can process large data sets to assess equipment health. This information can be presented in a web-based dashboard for easy monitoring. By utilising data analytics, utilities can improve efficiency and reduce downtime.”

Key challenges and modernising power substation

The current challenge in power distribution systems is the need for real-time data transfer capabilities. Engineers face issues connecting renewable energy sources (RE) due to capacity constraints and fluctuating grid conditions. The absence of real-time data hampers the management of power purchase, demand, and supply, leading to errors in power system operations.

Addressing the issue, Rajiv Goyal says, “It is crucial to establish a system that provides real-time data from 11 kV feeders to the control room. This would facilitate better integration of renewable energy and improve the accuracy of power system management.”

The upcoming installation of 25 million smart meters over the next 3-4 years will generate valuable data points, but transferring data remains an issue. Transitioning to optical fibre networks for data transfer in 11 kV and LT networks could offer seamless data transmission, as reliance on telecom networks has proven unreliable.

He further shares that utilities must find sustainable solutions that can accommodate technological advancements like 4G and 5G without constantly upgrading devices. A stable optical fibre network may provide a viable long-term solution for real-time data management in power distribution systems.

Speaking on the issue, Mahendra Singh comments that grid integration of inverter-based resources such as solar, wind, and battery storage requires thorough system studies to ensure stability and reliability. Unlike traditional substations, integrating renewable sources demands precise modelling and realistic simulations. Inverter-based sources can introduce grid oscillations and disturbances, as seen in Western Rajasthan, India, where severe grid oscillations lasted for hours.

He opines that “Challenges in grid integration are not unique to India; similar issues are observed in Australia and the UK. One key difficulty is the lack of detailed inverter models available to utilities and grid operators, limiting the accuracy of system studies. Manufacturers possess these models, but sharing them with operators would enable more effective grid studies and better integration.”

Mahendra Singh says that conducting comprehensive system studies with accurate models can help minimise grid challenges. While some level of challenge is inevitable due to the nature of inverter-based resources, realistic simulations and proper planning can lead to smoother integration and enhanced grid stability. A collaborative approach involving manufacturers, grid operators, and utilities is essential for successfully integrating inverter-based resources.

Suvendra Kumar Senapati has something more to say, “To address challenges related to ageing infrastructure and equipment failure, various strategies can be implemented. Firstly, adopting condition-based maintenance using AI and IoT can help predict equipment failures and extend the lifetime of assets. Real-time monitoring allows for early detection of issues, enabling proactive repairs and reduced downtime.”

He again says upgrading infrastructure with modern, compact technology can improve efficiency while conserving space. This approach integrates the latest advancements, providing higher reliability and performance. Asset performance management can also guide decisions on when to upgrade or replace equipment to effectively defer investments.

Suvendra Kumar Senapati further states, “Considering customer feedback and specific needs can inform the development of tailored solutions that prolong the lifespan of infrastructure by at least 3-5 years. Combined with regular assessment and maintenance, such strategies contribute to the overall improvement and longevity of ageing infrastructure.”

As the power industry evolves, utilities must address data integration, cybersecurity, and grid stability challenges. By effectively leveraging AI and ML, companies can optimise energy operations, improve customer satisfaction, and drive a more sustainable and resilient energy supply for both industry and consumers.

Quotes:

Sanjiv Prasad, Vice President, Head Power & Utilities, RIL-DMD

“Widespread adoption of digital twins in power plants and electrical systems is still developing. Integrating AI and ML with a distributed control system (DCS) can centralise data management and provide a holistic view of the electrical system, enhancing operational control and efficiency.”

Mahendra Singh Hada, Head – Substations, Indigrid

“In substations with hundreds of pieces of equipment, data analytics plays a key role. Manually analysing reports for all equipment is impractical. Instead, automated systems can process large data sets to assess equipment health.”

Rajiv Goyal, CEO & Whole-Time Director, EKI Power Trading

“Utilities must find sustainable solutions that accommodate technological advancements like 4G and 5G without constant device upgrades. A stable optical fibre network may provide a viable long-term solution for real-time data management in power distribution systems.”

Suvendra Kumar Senapati, Head of Sales & Commercial, L&T Digital Energy Solutions

“Upgrading infrastructure with modern, compact technology can improve efficiency while conserving space. This approach integrates the latest advancements, providing higher reliability and performance.”