University of Southampton researchers and IBM solve energy problems

To help reduce the world’s dependence on fossil fuels, researchers across the globe are looking at ways to generate, distribute, and use electricity in a more efficient, reliable and environmentally responsible manner. One such effort is taking place in the United Kingdom at the University of Southampton’s School of Electronics and Computer Science, where researchers led by Professor Nicholas Jennings and Dr. Alex Rogers have been working on the iDEaS Project (Intelligent Decentralized Energy-Aware Systems) since March 2009. A key aspect of the project is to study how advanced smart meters can be used to control energy storage devices (i.e., batteries) that could be purchased by residents for their homes.

As envisioned by the research team, software agents in the smart meter would monitor energy use in the house, learning the energy-consumption profile of the inhabitants. The meter would also receive and track real-time electricity rates from the energy supplier that reflect the differences in cost to generate electricity during peak and off-peak periods. Peak periods occur when demand for electricity is highest—for example, on very hot summer days when everyone is running his or her air conditioning units.

To supply the electricity needed during these times, power companies often resort to carbon-intensive and expensive “peaking plant” generators that significantly increase both energy costs and greenhouse gas emissions. Moreover, unmet demand for electricity can lead to brownouts, infrastructure damage, and power outages across the grid.

Knowing the home’s energy profile and the supplier’s energy rates, the meter would then optimize the use of the battery to both recharge and run home devices at the “right” times. In effect, the meter would seek to minimize each homeowner’s energy costs by using the battery to meet demand at peak times and recharge at off-peak.

Testing the solution with IBM ILOG optimization
What the researchers needed next was a way to test the feasibility of their approach, so they turned to IBM ILOG CPLEX for the answers. IBM ILOG CPLEX is a high-performance optimization engine that allows organizations to transform complex problems into mathematical models to provide precise and logical answers for decision making. “IBM ILOG CPLEX, with its rich set of optimization algorithms, allows us to quickly research and analyze strategies and solutions that can meet the world’s energy needs with more cost-effective, reliable and cleaner technologies,” says Sarvapali Ramchurn, one of the researchers on the team.

Using data characterizing the domestic electricity market and usage patterns of homes in the United Kingdom, the research team built
the model and let IBM ILOG optimization determine the results for various scenarios.

• Could the meter manage the electricity requirements of the user with minimum impact on the homeowner’s habits and comfort?
• Would the savings in energy costs for the homeowner provide enough financial incentive to purchase an energy storage device?
• Could the meters, when storing electricity “selfishly” based on their individual needs, learn to interleave their charging times, thereby preventing a demand peak and more costly electricity and higher carbon emissions?
• Would the collective use of the meters flatten demand peaks and create a more stable, environmentally friendly system overall?

What the researchers discovered confirmed their expectations and even provided a surprising result.

Not only did the simulations show a reduction in energy costs by 13 percent on average, more than enough for the homeowner to purchase the 4kWh storage device that was indicated by the simulations, but they also showed a significant reduction in peak demand, a consequent reduction in carbon generation and an evening out of market prices. Moreover, energy consumers without storage devices indirectly benefited from the lower prices. If rolled out across the entire country, this would point to annual savings of nearly £1.5 billion.

However, what surprised the researchers was that only 38% of the population needed to have storage devices to maximize the social welfare of the overall system. In other words, at the equilibrium point, a consumer without a storage device would have the same energy costs as a consumer with a storage device.

Implementation on the horizon
Is the iDEaS Project just research or is the scenario presented possible in the very near future? “We’re five to ten years away from having the storage device technology that we need, as well as the pricing mechanisms that would allow suppliers to send current electricity prices to the smart meters,” says Ramchurn. “We already have the software to run the smart meter and have experimented on real data.”

In fact, as battery technologies improve, the team expects that the equilibrium point for the system will be reached with less than the 38% adoption rate noted above, and the smart meters may even use the batteries of the owner’s electric or plug-in hybrid vehicle to store energy for the home. Ramchurn and his colleagues recently published their research in a paper entitled Agent-based Micro-Storage Management for the Smart Grid, which won Best Paper Award at the International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS) in Toronto in May 2010. They’ve also used IBM ILOG CPLEX to examine other energy issues. Another paper, Trading Agents for the Smart Electricity Grid, shows how agents controlling micro-generation capacity or consuming energy located at different points in the grid could use congestion pricing to trade energy more efficiently to maximize the social welfare
of the system.

Why IBM
IBM is a leader is the field of operations research and offers some of the world’s most advanced optimization technologies for solving tough business and research problems. More than 1,000 universities use IBM ILOG optimization for research and teaching, and more than 1,300 corporations and government agencies, including over 160 of the Global 500, use IBM ILOG optimization in some of their most important planning and scheduling applications.

Today’s energy- and climate-related issues are at the top of IBM’s strategic agenda. Along with IBM ILOG CPLEX, IBM solutions can help customers reduce costs, improve operational efficiency and systemically minimize energy, water, carbon emissions and waste. IBM is helping customers become more energy efficient, implement new ways to source, manufacture and distribute goods and services in a more sustainable manner, enable safe and renewable sources of energy and manage resources at a macro level, transforming entire industries. IBM takes a holistic approach to our planet’s challenges that combines innovative technology, deep business insight, and industry expertise.

For more information
Contact your IBM sales representative or IBM Business Partner.
Visit us at:
ibm.com/smarterplanet/energy