Artificial intelligence (AI) is revolutionising the world, and the electricity transmission sector is no exception. Flexibility needs of the electricity grid are growing with the increasing penetration of variable renewable energy, distributed generation, and shifting demand patterns. This also means that the grid now depends more on ancillary services such as frequency regulation and reserves to keep operations stable. Hence, operational resilience in the face of contingencies is now a critical requirement. Renewable energy, distributed energy sources, electric vehicles, advanced metering and communication infrastructure, management algorithms, energy efficiency programmes, and new digital solutions are driving change in the power sector. They are altering energy supply chains while revising energy landscapes. To cater to these diverse set of complex requirements and fulfil load generation balance in all scenarios, AI is increasingly seen as a necessary tool to support forecasting, optimisation, control and security functions at scale. With the share of renewable energy increasing and digitalisation getting integrated into each layer of power sector operations, the adoption of AI is no longer optional.
The perceived advantages of AI in grid operations are multi-fold, including the optimisation of grid performance, better management of renewable energy variability, and a reduction in operator workloads through automation. Innovative solutions, enabled by AI and machine learning (ML) techniques, and blockchain will also facilitate the unlocking and exploitation of system flexibility, for example, by enabling the development of solutions enhancing horizontal and vertical system integration. AI is a critical enabler for improving operational performance, predictive maintenance, and smart grid planning, as traditional methods for optimising transmission line efficiency often struggle with the increasing complexity and the dynamic nature of contemporary grids. By leveraging machine learning algorithms, real-time data analytics, and predictive modelling, AI can optimise power flow, minimise losses, and enhance grid reliability.
Relationship between AI, ML and statistics
The International Renewable Energy Agency (IRENA) defines AI as an area of computer science that focuses on creating intelligent machines that follow human behaviour, based on the data collected. However, this concept can sometimes be misunderstood or considered too broad, as some techniques from statistics and data mining (DM) are also components of AI and ML.
ML is a sub-field of AI that evolved from pattern recognition to analyse the data structure and to fit it into models that can be understood and replicated by users. It is segmented into four categories: supervised learning, unsupervised learning, reinforcement learning (RL), and deep learning (DL). It aims to predict or describe the existing relationships within the data set. RL is a computational approach that learns from interaction with the environment. This means defining how system agents can take actions in their environment to maximise the cumulative reward. RL is implemented mainly to solve energy dispatch problems and for enabling energy management scheduling. DL belongs to the artificial neural network (ANN) field. DL techniques can be applied to power systems in different scenarios such as fault detection in transformers or day-ahead (DA) electricity market price forecasting. And, last but not the least, statistics is a science dealing with analysis and data modelling. DM aims to extract patterns and knowledge from large datasets.
Figure 1: Relationship between AI, ML and statistics
Note: AI–Artificial intelligence; MI–Machine learning SVM–Support vector machines; NB–Naive Bayes; DT–Decision tree; KNN–k-nearest neighbours; RF–Random forest; MLP–Multi-layer perceptron; DBSCAN–Density-based spatial clustering of applications with noise; SOM–Self-organising map; PCA–Principal component analysis; TSNE–t-distributed stochastic neighbour embedding; ELM–Extreme learning machine; DNN–Deep neural network: CNN–Convolutional neural network; LSTM–Long short-term memory; GRU–Gated recurrent unit;
Source: ENTSO-E
Applications of AI in power transmission
AI-based technologies are expected to provide breakthrough innovation and will be leveraged to make control room operations more efficient, starting from routine activities, such as producing continuous and periodic reports, in addition to enabling a detailed analysis of network status. Moreover, from a longer term perspective, AI will be introduced in decision support systems, thus providing direct support to the online system operation.
AI’s application for the electricity transmission system includes power flow optimisation; power grid loss prediction; the optimisation of DA predictions; weather forecasts; renewable energy generation; asset management (visual inspection, vegetation management, environmental threats, predictive maintenance); data quality assessment (real-time detection, auto validation processes, estimation and correction); large language models for technical documentation and also for assisting system operation and market activities. Additionally, AI can also support transmission system operator (TSO)/distribution system operator (DSO) cooperation by gathering securely energy from DER sources or storage; help in grid health monitoring (real-time monitoring system of transmission lines, transformer and breaker monitoring in a substation); detect faults based on a deep learning intelligent approach in detection and classification of transmission line faults; devise optimisation codes for automated energy trading; and use DLs in drones for maintaining overhead lines (OHLs). Utilising AI recommendations to support dynamic load control through the management of smart appliances or EV charging infrastructure has started in several countries. These interventions help reduce peak demand and enhance grid stability. Further, with respect to outage management of grid elements, a few predictive maintenance applications that use AI to detect anomalies and predict equipment failures have been developed and are in use at some utilities. For example, ML algorithms can analyse operational data to flag emerging issues in transformer health before they lead to outages. Additionally, AI supports grid optimisation functions such as facilitating real-time balancing and congestion management. AI can also be applied in energy storage dispatch to determine the optimal charging strategy.
Table 1: AI applications in key TSOs across the globe
Source: Global Transmission Report
AI for reliable renewable grid integration
Integrating wind and solar power into power grids remains one of the biggest challenges for the power industry. Renewable energy is inherently variable, making real-time coordination essential for stability, and AI is proving to be of critical importance in this regard by enhancing efficiency, improving forecasting, and enabling smarter grid management. Accurate power forecasting is critical for renewable energy integration, and AI leverages real-time weather data, historical trends, and market conditions to refine predictions. Unlike traditional methods, ML continuously adapts, reducing uncertainty and improving energy trading and grid planning. AI-driven energy management systems now analyse weather forecasts, market dynamics, and consumption patterns to optimise grid operations continuously. These intelligent agents balance supply and demand with unprecedented precision, helping prevent blackouts, minimising energy waste, and enhancing overall grid resilience — especially as the share of renewables accelerates.
In terms of grid flexibility and balancing, AI forecasts variability and recommends grid reconfiguration or reserves dispatch while also supporting frequency and voltage stability under variable generation. Another benefit of AI is that it can carry out predictive curtailment management wherein it anticipates when generation will exceed demand or grid capacity, and helps in planning curtailments proactively.
As weather and generation relationships are non-linear (such as partial clouding impacts photovoltaic, and turbulence impacts wind turbines), AI learns these hidden patterns better than traditional statistical methods, which results in improved prediction of ramp-up/ramp-down events, essential for grid balancing.
AI also optimises battery charging/discharging to smooth fluctuations and shifts excess renewable generation to demand peaks. In collaboration with the Indian Institute of Technology (IIT) Kanpur, Grid Controller of India Limited (Grid-India), the country’s TSO that controls the National Load Despatch Centre and the five Regional Load Despatch Centres, has developed an AI/ML engine to clean real-time renewable energy data for better forecasting outcomes. The pilot has been implemented at a 250 MW solar power plant in the southern region. As part of this effort, three models have been developed using approximately 200 features, including direct point values, historical point values and engineered mean values. These models are designed to address various operational anomalies, such as non-updating data from multiple inverters, partial data availability, missing weather data and station-wide real-time data lapses. This cleaned and validated data is now being used to improve forecasting algorithms and enhance grid predictability.
Recent developments
Some TSOs have entered into collaborations, including with manufacturers and vendors to test AI applications through pilots. For instance, in July 2025, Open Access Technology International (OATI) announced the launch of the world’s first generative and agentic AI platform purpose-built for the energy industry, OATI Genie™, through a pilot programme with the California Independent System Operator (CAISO). Headquartered in Minnesota, US, OATI is a global leading innovator of solutions for the energy industry and the power grid. CAISO, the balancing authority for 80 per cent of the state of California and parts of Nevada, and operator of the Western Energy Imbalance Market, is collaborating with OATI to understand the use cases of its outage management system, design the solution architecture, evaluate and select AI models, and deploy OATI Genie™ in an operations pilot.
In June 2025, State Grid Brazil Holding S.A. (SGBH), a Brazilian subsidiary of China’s largest grid developer State Grid Corporation of China (SGCC), deployed an AI platform for inspecting transmission lines in Brazil. SGBH launched the platform during the BRICS (Brazil, Russia, India, China, and South Africa) high-level forum on AI in Brasilia. The platform uses an image recognition system with AI, which increases the efficiency of inspections, the rate of flaw detection in images, and risk control during manual inspections. It is seamlessly integrated with local systems, which streamlines workflows and unifies the inspection process. An offline mode has been introduced, to mitigate issues caused by limited network coverage in remote locations. This feature allows users to collect data, log faults, and execute work orders without internet access. SGBH is the first transmission company in Brazil to implement an AI-proctored solution for transmission line inspection.
In the same month, the Southwest Power Pool (SPP), a regional transmission organisation (RTO) that operates the grid for all or parts of fourteen US states in the Midwest and West, announced a partnership with energy technology company Hitachi to develop an AI-based solution for supporting power transmission reliability and overcoming flexibility challenges. SPP and Hitachi announced that the technology enables “end-to-end use of industrial AI and advanced computing infrastructure to help significantly speed up safe integration and use of additional energy sources supporting central US power grids. The partnership will draw on expertise from Method’s design services; GlobalLogic’s software engineering services; Hitachi Energy’s energy portfolio management asset modelling solutions; Hitachi R&D’s AI-based energy grid algorithm; and Hitachi Vantara’s integrated storage and compute platform Hitachi iQ, built on NVIDIA accelerated computing, networking, and AI software.
Earlier, in April 2025, another US-based RTO, PJM Interconnection, announced a multi-year collaboration with Google and Tapestry — a Google-backed initiative described as a “moonshot for the electrical grid” — to deploy AI tools meant to reduce the time it takes to process new generation interconnection requests, aiming to streamline grid planning and improve efficiency. Tapestry’s AI solutions aim to increase decision-making speed and accuracy in PJM’s planning department — making it easier to assess project feasibility, manage transmission constraints, and anticipate impacts on grid stability.
In Africa, the month of May 2025 saw Zambia’s state-owned power company ZESCO Limited announcing the upgrade of its National Control Centre (NCC) using AI, the Internet of Things (IoT), and real-time data analytics through a partnership with Swedish SwedFund Project Accelerator, which extended grants worth SEK12 million, to achieve digital transformation. This upgrade will enable the NCC to leverage cutting-edge technology, improving grid management and facilitating proactive issue prediction and resolution, ultimately ensuring a more reliable power supply for users.
Some vendors are making acquisitions or entering into collaborations to gain further advantage and expertise in the field. For instance, in July 2025, GE Vernova Inc. announced an agreement to acquire Alteia SAS, a France-based software firm known for its experience in AI, computer vision, and ML. The move aims to enhance GE Vernova’s visual data and AI capabilities, particularly in support of utility grid operations and situational awareness. According to the company, by linking visual data with operational systems, such as advanced distribution management software (ADMS), utilities will be better equipped to anticipate and respond to disruptions, such as extreme weather events, while improving grid resilience and reducing downtime.
In another development, in April 2025, Mexican engineering firm Mexmot partnered with GE Vernova to launch its software, which comprises sensors for existing as well as future substations. This software, utilising AI, will be capable of predicting maintenance needs, failures, and other factors affecting optimal network usage remotely from a PC, along with information stored in the Cloud. The two partners plan to tap into the opportunity presented by the Mexican state-owned power company Comisión Federal de Electricidad (CFE), which intends to begin digitalising over 110,000 km of transmission lines by 2030.
Some of the other key vendors in the industry include ABB (which acquired a minority stake in North Carolina company DG Matrix in March 2025 to support the commercialisation of solid-state power electronics for generative AI data centres and renewable microgrids), IBM, Honeywell, and Schneider Electric. In March 2025, Schneider Electric announced the launch of the One Digital Grid Platform, an integrated and AI-powered platform designed to enhance grid resiliency, reliability, and efficiency.
Challenges and the way forward
With continued advancements in AI technologies, the future of energy transmission is set to become more intelligent, efficient, and sustainable. Security situation has also improved through proactive intrusion detection. Most importantly, decision-making is now based on real-time data and machine intelligence. The increased adoption of AI, cloud computing, and other digital technologies is enabling utilities to respond and adjust to changing market dynamics, stemming from huge influx of renewable energy and the volatility that comes with it.
However, challenges remain. For instance, the quality and consistency of input data need continuous improvement. Model development requires domain expertise and technical depth, and integration with legacy infrastructure can be complex. Addressing these gaps will be essential to scale AI, from niche applications to a universal layer in grid management.
Another major concern with AI is cybersecurity. AI relies on vast amounts of data, which poses a risk of exposing consumer information and operational data for the electrical grid, thus increasing data privacy risks. Moreover, AI models can be manipulated, impacting grid reliability. Increased connectivity (through IoT, smart meters, etc.) also expands potential attack surfaces. For instance, in 2021, the RedEcho attack on India’s power companies exploited weak links in the system. In developing countries like India, there is still a lack of expertise to manage AI-specific threats.
As utilities seek to advance further in their digital transformation journey, addressing these gaps will be essential for scaling AI from niche applications to a universal layer in grid management.
