An enhanced multi-constraint state of power estimation algorithm for lithium-ion batteries in electric vehicles

In this paper, an enhanced multi-constraint (MC) state of power (SOP) estimation algorithm in a prediction window up to 120 s is developed for lithium-ion batteries in electric vehicles (EVs). First, a novel regressionbased algorithm (RBA) incorporating a model parameter forward prediction is devised for voltage-constraint SOP estimation to reduce battery linearization error and improve model accuracy by estimating model parameters to the end of a prediction window. The convergence condition is analytically formulated and can be satisfied over a whole battery operating range. Second, an improved state of energy (SOE)-constraint SOP estimation algorithm is proposed from a more practical perspective by considering battery terminal voltage variation in a prediction window. Together with the other constraints, the enhanced MC SOP estimation algorithm is achieved, which specifies a dynamic safe operating area for power regulation in EVs. Moreover, an incremental pulse test is designed to obtain reference SOPs in high fidelity under static conditions and dynamic load profiles. Simulations and experimental results demonstrate the improved accuracy of the enhanced MC SOP estimation algorithm over the conventional MC online SOP estimation algorithm.

Automated summary

In this paper, an enhanced multi-constraint (MC) state of power (SOP) estimation algorithm is developed for lithium-ion batteries in electric vehicles (EVs). The algorithm improves accuracy by incorporating a regression-based algorithm and considering battery terminal voltage variation. Simulation and experimental results demonstrate its superiority over conventional algorithms.