An electrochemical lithium-ion battery model is well known to be suited for effectively describing the microstructure evolution in charging and discharging processes of a lithium-ion battery with practically realizable complexity. This paper presents a neural network-based parameter estimation scheme to identify the parameters of an electrochemical lithium-ion battery model in a near-optimal and real-time manner in order to consistently observe the electrochemical states of batteries. The network is first trained to learn the dynamics of the electrochemical lithium-ion battery model, and then, it is applied to estimate the parameters with available finite-time measurements of voltage, current, temperature, and state of charge. In order to efficiently learn the dynamic characteristics of a lithium-ion battery, a well-known recurrent neural network, called a long short-term memory model, is employed with other techniques such as batch normalization, dropout, and stochastic gradient descent with warm restarts for learning speed enhancement and regularization. Using synthetic and experimental data, we show that the proposed estimation scheme works well, finding parameters and recovering the voltage profiles within the root-mean-square error of 0.43% and 26 mV, respectively, even with measurements obtained within a sufficiently short interval of time. INDEX TERMS Electrochemical battery model, lithium-ion battery, long short-term memory, real-time parameter estimation, recurrent neural network, synthetic data generation.
Electrochemical models of lithium-ion batteries are derived according to the laws of physics; therefore, the parameters represent specific physical quantities such as lithium diffusivities, particle volume fractions, and ion intercalation rates. It is important to estimate these parameters to identify the internal states of a lithium-ion battery for efficient and safe management. Until now, parameter estimation algorithms for electrochemical lithium-ion battery models have been developed without considering the unequal identifiability among the target parameters. Thus, it is highly likely that existing algorithms exhibit inefficient exploration and lead to a slow convergence rate and even large parameter estimation error. For more accurate parameter estimation of an electrochemical lithium-ion battery model, we propose a new adaptive exploration harmony search (AEHS) scheme that provides a wide search space for a longer period of time when estimating parameters with low identifiability. The proposed algorithm is based on improved harmony search; its bandwidth parameters for determining the level of exploration are adjusted according to the individual and joint variabilities computed from the distributions of previously estimated parameters. Such adaptive bandwidth parameters can reduce inefficient exploration and enable fast convergence, allowing exploration that achieves global optimality. Simulation results show that the proposed parameter estimation algorithm produces the highest convergence rate and the smallest parameter estimation error compared with existing schemes. The performance of the proposed scheme is also validated using real data generated from experiments. INDEX TERMS Adaptive exploration harmony search, electrochemical model, lithium-ion battery, meta-heuristic algorithm, parameter estimation, parameter identifiability.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.