Award-winning paper on storing solar energy in residential power networks

Two electrical engineers have won an award for their paper on an economically feasible way to store solar energy in existing residential power networks.

Reza Arghandeh, a doctoral candidate in the Bradley Department of Electrical and Computer Engineering at Virginia Tech, and his advisor, Robert Broadwater, professor of electrical and computer engineering, wrote the paper which was awarded Best Student Paper at the 20th International Conference on Nuclear Engineering, held in conjunction with the American Society of Mechanical Engineers Power 2012 Conference.

“Selling the household-generated electricity into the electric energy market and the storage of electricity in storage systems and demand control systems provide a variety of economic opportunities for customers and utility companies to use more renewable resources,” Arghandeh said. Some residential houses already do this by selling power back to an electrical distribution industry.

The engineers’ work provides an optimisation algorithm for a distributed energy storage (DES) system on a broad scale. The system they developed presents a fleet of batteries connected to distribution transformers. The storage system can then be used for withholding distributed photovoltaic power before it is bid to market, Arghandeh said.

“Withholding distributed photovoltaic power, probably gained from rooftop panels, represents a gaming method to realise higher revenues due to the time varying cost of electricity,” he said. “The distributed photovoltaic power adoption can be controlled with the help of real-time electricity price and load profile.”

Modern power systems are moving towards a smart grid concept to improve efficiency, reliability, cost and sustainability. Arghandeh and Broadwater say solar technologies should be integrated with existing technologies like energy storage and control systems.

The DES system computation they designed is called a “discrete ascent optimal programming approach”, which ensures convergence of the various power systems after a finite number of computational iterations. A solution determined by using their approach depends on the forecast of load variation, market prices and photovoltaic generation.

The output of their optimisation algorithm is a distributed energy storage charging and discharging schedule with maximised operation benefits.