Taiwan’s Industrial Technology Research Institute (ITRI) has unveiled research into the use of artificial intelligence to maximise the use of battery arrays in vehicles and energy storage applications.
ITRI’s Reconfigurable Array of Inexpensive Batteries Architecture (RAIBA), a system featuring AI “self-learning” and optimisation technology, was showcased at the annual CES 2018 technology innovation show in Las Vegas.
According to ITRI, RAIBA can be used in “heterogeneous” battery modules, including arrays that combine both lithium-ion and lead-acid technologies, to handle specific load requirements.
In addition, RAIBA is “capable of dynamically reconfiguring the array to bypass nearly exhausted modules— avoiding the shutdown of battery arrays when one module stops performing”, researchers said.
“Even when multiple modules are depleted, the system will still function at full potential as RAIBA transfers the required load to other performing modules.
Researchers say RAIBA can also reduce system-level decay by 51% and extend system-running time by 90% for a set of heterogeneous 20-battery modules with capacity ranging from 1.5 to 4Ah.
ITRI said the use of AI allows the system to simulate feasible configurations based on cell characteristics and system load requirements, after its initial diagnostic analysis, and to ensure optimum utilisation before requiring a recharge, which in turn helps to extend and optimise charge/discharge cycles.
The research is the latest to emerge from ITRI, which last year announced details of studies into the manufacture of rapid charge, long-life batteries from low-cost and abundant aluminium.