Empower’s Genesys 8K™ Modular Smart Home Energy Platform
    By 2030, 95 percent of U.S. passenger miles traveled by autonomous electric vehicles (AEVs)

    This is a mouthfull: Optimal constrained self-learning battery sequential management in microgrid via adaptive dynamic programming

    Solution to optimize power consumption

    Smart homes need smart batteries. Current systems overindulge on power, which can shorten the life of batteries and the devices they power. Future batteries may get an intelligence boost, though.

    A collaborative research team based in Beijing, China, has proposed a novel programming solution to optimize power consumption in batteries. The scientists, from the Institute of Automation, the Chinese Academy of Sciences, and the School of Automation and Electrical Engineering at the University of Science and Technology Beijing, published their results in IEEE/CAA Journal of Automatica Sinica (JAS), a joint publication of the IEEE and the Chinese Association of Automation.

    “In smart home energy management systems, the intelligent optimal control of [the] battery is a key technology for saving power consumption,” Prof. Qinglai Wei, with the State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences, wrote in the paper.

    To develop a system in which batteries can learn and optimize their power consumption, Wei and his team turned to adaptive dynamic programming. This method breaks down one big problem – how to best use batteries in smart home systems – into smaller problems. The answer to each small problem builds into the answer to the big problem, and, as the circumstances of the question change, the system can examine all the small answers to see if and how the big answer adapts.

    Wei and his team are the first to use this method while also considering the physical charging and discharging constraints of the battery. The algorithm learns which inputs, such as the demand for power from a device, lead to which outputs, such as providing power. By continually questioning the link between input and output, the algorithm learns more about the best times to charge and to discharge to limit power consumed from the grid. To extend the battery life, every iteration of learning is constrained by the understanding that the battery can only charge and discharge to certain limits. Anything more, and the battery could experience excessive wear.

    “The battery [makes] decisions to meet the demand of the home load according to the real-time electricity rate,” Wei wrote, noting that the objective of optimal control is to find the ideal balance for each battery state (charging, discharging, and idle) within the battery’s constraints, while still minimizing the power needed from the grid.

    To further extend the lifetime of batteries in smart home systems, Wei and his team will next examine how the damage caused by frequently switching between charging and discharging modes may be avoided.

    Reference

    Fulltext of the paper is available: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7783983 http://html.rhhz.net/ieee-jas/html/20170203.htm

    Empower’s Genesys 8K™ Modular Smart Home Energy Platform
    By 2030, 95 percent of U.S. passenger miles traveled by autonomous electric vehicles (AEVs)