Data-driven charging strategies for grid-beneficial, customer-oriented and battery-preserving electric mobility
Authors:
Karl Schwenk,
Tim Harr,
René Großmann,
Riccardo Remo Appino,
Veit Hagenmeyer,
Ralf Mikut
Abstract:
Electric Vehicle (EV) penetration and renewable energies enables synergies between energy supply, vehicle users, and the mobility sector. However, also new issues arise for car manufacturers: During charging and discharging of EV batteries a degradation (battery aging) occurs that correlates with a value depreciation of the entire EV. As EV users' satisfaction depends on reliable and value-stable…
▽ More
Electric Vehicle (EV) penetration and renewable energies enables synergies between energy supply, vehicle users, and the mobility sector. However, also new issues arise for car manufacturers: During charging and discharging of EV batteries a degradation (battery aging) occurs that correlates with a value depreciation of the entire EV. As EV users' satisfaction depends on reliable and value-stable products, car manufacturers offer charging assistants for simplified and sustainable EV usage by considering individual customer needs and battery aging. Hitherto models to quantify battery aging have limited practicability due to a complex execution. Data-driven methods hold feasible alternatives for SOH estimation. However, the existing approaches barely use user-related data. By means of a linear and a neural network regression model, we first estimate the energy consumption for driving considering individual driving styles and environmental conditions. In following work, the consumption model trained on data from batteries without degradation can be used to estimate the energy consumption for EVs with aged batteries. A discrepancy between the estimation and the real consumption indicates a battery aging caused by increased internal losses. We then target to evaluate the influence of charging strategies on battery degradation.
△ Less
Submitted 14 October, 2019;
originally announced October 2019.
