SCOPE

ABSTRACT

Energy generation, distribution, and management is central to humanity. At the scale of a single home, AI can be used to help minimize energy use through smart sensors and off-peak use of machines. At the scale of cities and states, AI can be useful in power distribution, load balancing relative to predicted and actual demands, and to change these relative to system dynamics. Even at national and global levels, energy production and consumption play pivotal political roles, with AI helping to revolutionize the way in which global energy markets trade and interact. Autonomous systems can assist with the control of energy plants leading to reductions in emissions. Such systems are being used currently to accelerate a transition to renewable forms of energy in intelligent ways with market stakeholders using tremendous volumes of data to model alternative futures. AI has also a wide applicability in the oil and gas energy sector.

This vertical will address, amongst others:

• AI in Autonomous Control, for wind and solar farms and for the discovery of novel energy resources.
• AI in Carbon Emissions Reduction, by developing cleaner production processes, and improving monitoring and compliance standards for emissions.
• AI in Consumer Products, helping to provide devices that assist consumers in the direct management of energy consumption, helping to decrease demand and placing less strain on power networks.
• AI in Demand Forecasting, to improve load balancing and management, optimize dynamic distribution and use of different energy resources to maximize customer benefit and grid utility.
• AI for Digital Twins, real-time virtual models of physical grid assets that can be used to study wind turbines and power-generation facilities. Digital twins can improve servicing, experimenting, maintaining, and optimizing the energy network.
• AI for Energy Communities enabling fair energy trading and distribution of benefits, efficient energy resource management (generation, storage, demand flexibility), efficient participation in wholesale and local energy markets 
• AI for Energy Consumers to support decision making regarding efficient energy use and active consumers’ participation namely in demand response programs, and in energy transactions with neighbors 
• AI for Energy Efficient Industrial Plants to enable energy efficient industrial tasks and processes and more sustainable factories by promoting the increased use of renewable energy sources
• AI for Energy Efficient Transportation, addressing electric vehicles’ charging and routing management and their efficient integration with the electric grid, including V2X (vehicle-to-grid, vehicle-to-home, vehicle-to-building, vehicle-to-vehicle, etc.)
• AI for Energy Markets enabling realistic energy market simulation, decision-support to market players, market models suitable for intensive use of renewables, and coordination with local energy markets.
• AI in Oil and Gas. In upstream operations (exploration and production), AI can assess the value of specific reservoirs, customize drilling and completion plans based to the geology of the area, and assess risks of each individual well. In midstream and refining, AI can optimize pipeline and refining scheduling, forecast commodity and product market prices, and optimize commodity trading and hedging. In downstream operations, AI can minimize costs and maximize spreads.
• AI in Plant Management, forecasting the longevity of systems, predictive maintenance scheduling, and true production capabilities, reducing operation costs and carbon emissions.
• AI in Security for infrastructure protection of critical energy systems from network intrusion and malware.
• AI for Sensor Fusion, for the transformation of information from thousands to millions of sensors simultaneously for improved tracking and decision-making in cities, states, and nations.
• AI in Smart Grid Operation, Control, and Management, to forecast likely outages before they occur, to improve automated switching by predicting grid imbalance and optimize power yield towards improved demand-side management, reducing peaks in energy demand. Smart Grid enables a two-way flow of power and data among suppliers and consumers to optimize the power flow in terms of economic efficiency, reliability and sustainability. AI facilitates collaboration between stakeholders, control of network imbalances (e.g., frequency and voltage regulation), decentralized network management and operation, and security and privacy.
• AI in Transition to Renewables, by providing real-time monitoring of power grids, accurate predictions of power fluctuations, and the development of new strategies to work with geothermal energy sources.
• Environmental sustainability aspects of AI systems. While most of the previous applications of this vertical are focused on more standard applications of AI to energy, we want to remind the audience that it is possible to reduce the energy consumption posed by AI. We will explore impact of training of large language models (AI energy consumption during training), data center energy usage, etc.

ORGANIZING TEAM 

Ganesh Kumar Venayagamoorthy (Clemson University)

Ahad Esmaeilian (Audubon Companies)

Piero Bonissone (Piero P Bonissone Analytics, LLC)

Francesco Grimaccia (Politechnico di Milano, Italy)

Horste Schulte (Hochschule für Technik und Wirtschaft Berlin)

Joao Soares (Polytechnic of Porto, School of Engineering – ISEP, Portugal)

Fernando Lezama (Instituto Superior de Engenharia do Porto, Portugal)

Zita Vale (GECAD, Portugal)

Brian K. Spears (Lawrence Livermore National Laboratory)

Jose Celaya Galvan (Schlumberger)