The Lithium-ion batteries are certainly the battery of choice today because of their high voltage platform, low maintenance, high energy and power density, good cycle performance, and low self-discharge. Developments in lithium-ion battery technology make them inexpensive, small and extremely light-weight. These batteries are capable of providing the great power needs of today’s high-tech industries, ranging from small portable devices, e.g. smart phones, to full/hybrid electric vehicles and also alternative energy storage requirements of renewable power generation. The lithium-ion battery performance plays a significant role in the system energy management. As a consequence, battery real-time monitoring and accurate state-of-charge (SOC) estimation are necessary. However, full order electrochemical models are not suitable for this purpose due to high computational cost. Reduced order model (ROM) is one of the promising models which is computationally cost effective while ensuring accurate results. Using data clustering are one of attractive ways to make reduced order model. The capability of cluster-based reduced order modeling (CROM) has been proven in fluid mechanics engineering fields. This approach, starts with a snapshot data set. The observations are clustered based on their similarity and then the means of these clusters (centroids) are used to constitute the reduced order basis.

   In the present study, cluster-based reduced order modeling is applied to electrochemical transport governing equations in order to simulate a lithium-ion battery. For this purpose, the governing equations including conservation of species and charge are solved all together using finite volume method (FVM) by which the snapshot data are provided. The data are clustered by k-means clustering algorithm in order to obtain the centroids which are optimal low-dimensional basis for representing an ensemble of high-dimensional simulation data. This low-dimensional basis are used to formulate reduced order models of complex fields of battery. The obtained numerical results show that not only the CROM of lithium-ion battery significantly decreases the computational time but also there is an excellent agreement with the results of computational fluid dynamics (CFD) models, consequently is suitable for deployment in real-time monitoring and battery management systems.