Traditional battery energy storage systems (BESS) are based on the series/parallel connections of big amounts of cells. However, as the cell to cell imbalances tend to rise over time, the cycle life of the battery-pack is shorter than the life of individual cells. New design proposals focused on modular systems could help to overcome this problem, increasing the access to each cell measurements and management. During the design of a modular. Traditional battery energy storage systems (BESS) are based on the series/parallel connections of big amounts of cells. However, as the cell to cell imbalances tend to rise over time, the cycle life of the battery-pack is shorter than the life of individual cells. New design proposals focused on modular systems could help to overcome this problem, increasing the access to each cell measurements and management. During the design of a modular battery system many factors influence the lifespan calculation. This work is centred on carrying out a factor importance analysis to identify the most relevant variables and their interactions. The analysis models used to calculate the reliability of the batteries are the state of health (SoH) and the Multi-State System (MSS) analysis with the Universal Generating Function (UGF), while electronic devices reliability is approximated using constant failure rate achieved with FIDES guide. Thus, it is determined numerically that module redundancy, cell capacity, module voltage and their interactions are the most determinant design characteristics.••••Modular and traditional battery systems' reliability analysis••Lifetime improvement of battery systems through modular solutions••Relevance analysis of the variable battery design factorsBattery-packLithium-ion batteryLow voltageModular batteriesThe penetration of renewable energy sources into the main electrical grid has dramatically increased in the last two decades. Fluctuations in electricity generation due to the stochastic nature of solar and wind power, together with the need for higher efficiency in the electrical system, make the use of energy storage systems increasingly necessary. To address this challenge, battery energy storage systems (BESS) are considered to be one of the main technologies.Every traditional BESS is based on three main components: the power converter, the battery management system (BMS) and the assembly of cells required to create the battery-pack. When designing the BESS for a specific application, there are certain degrees of freedom regarding the way the cells are connected, which rely upon the designer's criterion. Taking the energy of the battery-pack as a design specification and assuming that a DC/DC converter will adapt the voltage level required by the application, the number of cells connected in series and in parallel is a decision that will need to be addressed.Most of the BESSs that have been developed until now were designed specifically for one application and in most cases scalability and reusability crit. 2.1. Reliability model of a BESSIn order to evaluate the BESSs' reliability, it is necessary to deeply analyse the failure rate of each of the components. All these items are considered to be independent and not reparable. Once the individual result of both, the cells and the electronic hardware is obtained, the next step is to establish a reliability calculation method for the whole BESS. Fig. 1 presents an overall point of view of the steps followed during the reliability analysis.2.2. Factor importance analysis methodologyWith the main components of the reliability estimation method defined, the next step is to focus on the whole BESS analysis. To that end, different variable factors have to be defined. Nevertheless, it should be noticed that there is no previous knowledge of which of these factors are going to improve the final design. Having that said, this article aims to describe how each variable influences the reliability result in order to establish a criterion for the design process of battery-packs.Based on the numerical reliability analysis method of Section 2.1, an iterative process able to estimate the MTTF for each design characteristic combination is developed. Among the different importance analy.
