Impact of battery degradation models on energy management of a grid-connected DC microgrid Shuoqi Wang a , Dongxu Guo a , Xuebing Han a , Languang Lu a , Kai Sun b , Weihan Li c , Dirk Uwe Sauer c , Minggao Ouyang a, * a State Key Laboratory of Automotive Safety and Energy, Tsinghua University, Beijing,100084, China b State Key Laboratory of Power Systems, Department of Electrical Engineering, Tsinghua University, Beijing, 100084, China c Chair for Electrochemical Energy Conversion and Storage Systems, Institute for Power Electronics and Electrical Drives (ISEA), RWTH Aachen University, Jaegerstrasse 17/19, 52066, Aachen, Germany article info Article history: Received 19 March 2020 Received in revised form 28 May 2020 Accepted 23 June 2020 Available online 4 July 2020 Keywords: DC microgrid Battery degradation model Energy management Particle swarm optimization Combined factor-based model abstract Battery degradation cost is one of the major concerns when designing energy management strategies of DC microgrids. However, many battery degradation models used in the previous works are over- simplified and the effectiveness of which has not been verified. As a result, this paper presents a comparative study of the impact of battery aging models on energy management of the microgrid. Four popular single factor-based semi-empirical models are investigated while a combined factor-based Combined Arrhenius-Peukert-NREL (CAPN) model is proposed with the best fitting performance compared with the experimental data. The five degradation models are considered as part of the objective function in the particle swarm optimization-based energy management structure of a grid- connect microgrid. The optimized power scheduling and state of charge trajectory of the battery un- der different single factor-based models exhibit enormous deviations, so as the calculated total costs, which have the maximum error of 63.9% compared with the CAPN model. The application of the studied single factor-based models will also result in 3.5%e12.5% additional actual operating cost under non- optimal conditions. This paper first reveals the significant and unneglectable influence of the simpli- fied degradation models on the microgrid energy management, the abandon of the single factor-based models is also recommended. © 2020 Published by Elsevier Ltd. 1. Introduction Under the increasing demand of distributed renewable energy sources integration and the growing burden of electric vehicle (EV) fast charging against the utility grid [1], DC microgrid is understood to be the feasible solution and key technology for photovoltaics (PV) penetration, EV adoption and energy transmission efficiency improvement [2]. One of the major obstacles of promoting DC microgrids results from their large economic cost during con- struction and operation processes [3]. To overcome this challenge, numerous microgrid energy management strategies [4] have been put forward to achieve economic operation of the system in the literature. The proposed energy management techniques could be diversified into rule-based energy management [5] and optimization-based energy management [6], while the latter approach is more popular among the researchers as the minimum operating cost could be obtained base on the forecasted condition [7]. Major optimization objectives include the operation cost of distributed generators, the electricity cost purchased from the grid, the load shedding penalty cost and the energy storage system (ESS) replacement cost due to battery degradation [8]. From the point of view of the objective functions, the modeling of the battery degradation cost is the most diversified part and is often ignored in some studies due to its complexity [9]. The aging mechanism of the battery can be divided into three modes: loss of lithium ion inventory, loss of anode/cathode active materials and internal resistance increase [10]. The degradation modes are caused by various coupled internal side reactions, such as the solid elec- trolyte interface (SEI) film formation and continuous thickening, metal dissolution and electrolyte decomposition [11]. The side * Corresponding author. State Key Laboratory of Automotive Safety and Energy, Tsinghua University, Beijing, 100084, China. E-mail addresses: wang-sq2013@qq.com (S. Wang), ouymg@tsinghua.edu.cn (M. Ouyang). Contents lists available at ScienceDirect Energy journal homepage: www.elsevier.com/locate/energy https://doi.org/10.1016/j.energy.2020.118228 0360-5442/© 2020 Published by Elsevier Ltd. Energy 207 (2020) 118228