Operationalizing Artificial Intelligence and Machine Learning in Electricity Value Chain, Dr. Joseph Nyangon

To meet the 2015 Paris Agreement on climate change and avoid the catastrophic impacts of a changing climate, governments need to reduce emissions by 45% by 2030 and achieve a net zero greenhouse gas/CO2 emissions target by 2050. The urgency to minimize and reduce emissions while boosting energy deployment is picking up fast due to game-changing innovations in the energy sector. Public and private institutions are evaluating numerous alternatives to switch power generation from fossil fuels (coal, oil, and gas) to renewable options (solar and wind). The rising investment in grid modernization compels electric utilities and the industry to rethink and reevaluate transitioning models. However, a profound and fast structural shift to renewable energy sources from the fossil-fuel economy exposes existing energy infrastructure investments to the risk of stranded assets.

The pressure to orchestrate and deliver more distributed, reliable, affordable, and sustainable power drives electric utilities to invest in grid modernization and advanced automation to increase revenue and improve performance. However, making the switch is more challenging because it involves significant restructuring and redevelopment. Companies risk their energy assets becoming stranded because the redeveloped infrastructure might render many plants and machinery assets unprofitable relative to initial expectations. Transitioning to a clean energy production and distribution system also comprises reengineering a fossil-fuel-powered enterprise to alternative modes of energy consumption. For example, switching from a coal-fired electricity plant to a natural gas-fired generating system is more challenging than a consumer switching from a gasoline-powered car to an electric vehicle. Energy companies, therefore, are compelled to consider the sunk costs of existing assets because substituting fossil fuel with renewable sources entails replacing assets and infrastructure. 

Stranded energy assets refer to infrastructure and equipment that could potentially become obsolete in the absence of fossil fuels or the utilization of alternative energy sources. International governments are rapidly pledging carbon neutrality in the next few years, and the risk of stranded assets looms for many energy producers and distributors. Companies will face significant write-downs and losses due to the sunk costs in conventional fuel equipment, machinery, and infrastructure. However, some investors and firms are introducing innovative ideas and concepts to help mitigate the risks of stranded electricity and energy assets. Mitigating the risks of stranded assets using artificial intelligence (AI)-powered models and machine learning (ML)-enabled solutions are progressing rapidly due to the energy sector’s perceived long-term benefits.

Dr. Joseph Nyangon is a leading energy economist with an engineering education working to confront asset-stranding risks arising due to the realignment of expectations, disruptive policy, and technological innovations in the energy sector, including the application of ML and AI strategies. Dr. Nyangon holds Ph.D. and postdoctoral degrees in energy and environmental policy with a concentration in energy economics and engineering systems. His experience in clean energy transition and energy economics has helped electric utilities to tackle intractable challenges stemming from disruptive energy transition scenarios and the transition risk associated with stranded fossil-fuel assets. Dr. Nyangon also has an M.Sc. degree in computing systems which offers him a unique combination of computer systems, energy economics, and policy background critical to addressing the noble trifecta of sustainable development—economic growth, environmental integrity, and social inclusion. Dr. Nyangon’s expertise and knowledge of power systems, computing (ML and AI applications), and energy economics and policy have enabled him to tackle multiple transformational issues of the energy transition currently bedeviling electricity markets, such as changing government policy and regulatory environment, technological advancement, the scale of financing needed and the impact on consumers and business model innovation. He utilizes his unique academic and industry experience to help utilities uncover unmet customer needs and influence decision-making in a continuously changing environment.

Digitalization and migration of utility systems and applications to the cloud can help the industry reduce carbon emissions and accelerate its socio-economic impact without rendering existing infrastructure redundant. Dr. Nyangon developed a model to monitor and reduce carbon emissions in buildings and improve energy efficiency adoption through ML- and AI-powered techniques. His expertise in streamlining energy asset management by adopting deep and fast structural efficiency measures offers opportunities for electricity generation providers, energy trading firms, and utility distribution companies to minimize the risk of stranding energy assets. ML and AI are innovative techniques for improving the performance of energy assets and infrastructure investment, requiring a unique blend of interdisciplinary perspectives. Experts like Dr. Nyangon can play a prominent role in advising energy institutions on strategies for improving the operational efficiency of electricity systems, leveraging knowledge spillovers through R&D innovation, and managing infrastructure investment.

The pressure to reduce global emissions and transition to a low-carbon economy significantly increases the risks of stranded fossil-fuel assets. However, Dr. Nyangon’s primary focus is low-carbon electricity systems functioning on variable or controllable mechanisms for tackling climate change issues. He emphasizes using ML and AI techniques to mitigate the fast phase-out of fossil-fuel production and the likely stranded energy assets challenges based on the concept of “leaving no one behind” to safeguard a responsible, people-centered, and just energy transition. As the utility industry transforms, delivering customer-centric strategies and high-touch customer engagement to balance the energy transition would gain more prominence. Applying AI and ML-powered techniques across several use cases to uncover unmet customer needs and improve customer experience would position utilities as the true orchestrators of change. The digitalization solutions for improving energy management and efficiency utilize various mechanisms, including sensors, meters, and energy auditing devices. AI and ML can also help monitor and improve the performance of electricity networks and advance cooperation across the utility ecosystem.