The single biggest factor hindering the convenience, and therefore the adoption, of electric vehicles is the batteries used to power them. While filling up an ICE vehicle takes just a few minutes at the pump, electric vehicle recharge times are measured in hours. Engineers at the University of California, San Diego, have developed new algorithms that improve the efficiency of existing lithium-ion batteries and could allow them to be charged twice as fast than is currently possible.

Toyota is so concerned about the limitations of EVs it recently announced it would severely limit the availability of its Scion iQ EV microcar in favor of hybrid vehicles. While most research efforts to overcome the limitations of lithium-ion batteries focus on using new materials, such as graphene, carbon nanotubes or even sugar, the sophisticated estimation algorithms developed by Professor Miroslav Krstic and UC President's Postdoctoral Fellow Scott Moura at the Jacobs School of Engineering at UCSD promise to make existing lithium-ion battery technology more effective.

Lithium-ion batteries comprise three layers – an anode, a cathode, and a separator layer – all of which are rolled together to form a cylinder. When fully charged, the lithium ions are stored at the anode before moving through the separator layer to the cathode when powering the device to which the battery is connected. Knowing where the ions are in the anode provides an indication of whether the battery is functioning properly, but since the ions are usually lodged deep inside irregularly-shaped particles within the anode, this is very hard to measure.

As a result, manufacturers currently monitor a battery’s behavior and health by measuring the voltage and current. Krstic says this is a crude measure and results in batteries that are oversized, weigh and cost more, and take a long tome to charge.

The estimation and control algorithms developed by Krstic and Moura allow them to estimate where the particles are so the anode can be filled to capacity safely and efficiently. The algorithms can also estimate how the health of the battery changes over time. The researchers claim their approach has the potential to reduce the production costs of lithium-ion batteries by 25 percent, while also allowing them to run more powerful electric motors and slashing charge times in half.

The researchers have received a US$460,000 share of a $9.6 million grant from the Department of Energy’s ARPA-E research agency. They are sharing the grant with automotive product supplier Bosch and battery manufacturer Cobasys, who will supply testbeds on which Krstic and Moura will refine and test their algorithms using actual batteries. The first step will see them estimating the charge distribution within the battery before estimating its state of health. The final step will be to find the optimal charge and discharge rates of the batteries.

Source: UCSD Jacobs School of Engineering