Development of a management system for the active control of the main battery parameters

In the last years electric vehicles (EV) have gained a large share of the market, due to the problems of environment pollution and energy saving. Thanks to their high-performance properties and good knowledge, the majority of EVs employ as energy source lithium-ion batteries which require a reliable control system to guarantee an optimal management and ensure proper operation inside the battery, avoiding dangerous situations which could seriously damage their operational capacity and drastically reduce their lives. This thesis is addressed to achieve these goals through the creation of our own battery management system (BMS), able to monitor the key parameters and activate whenever one of them is captured with values out of the range. The battery system considered is made up of nine modules in parallel each consisting of twenty cells while the parameters are the current, voltage, temperature and state of charge (SOC). Related to the first one an over-charging (over-discharging) situation could appear causing possible electrolytic decomposition and short-circuit. Associated with the second parameter problem such overheating could emerge, damaging the battery sometimes irreversibly. With high (low) temperature the chemical reactions efficient can be greatly reduced inside the battery causing a fast degradation and possible explosion since unstable chemical reaction can be involved as well as capacity problems. Last but not least, issues could occur with the presence of error in the SOC balance between the different modules, leading to unexpected and unwanted low performances such as in terms of power as well as unpredicted behaviour due to dangerous level of state of charge, if not correctly tracked. The model created in the thesis involves the use of SIMULINK/SIMSCAPE within which it is subdivided in two main parts, the BMS itself and the battery model. The latter is controlled by the BMS through the implementation of switches for each module, which are activated whenever is required (fault detection of one or more parameters). Several tests and analyzes were conducted on the model to verify its functionality, starting from single error cases up to simulations involving the presence of multiple simultaneous errors as well as charging phases, in order to test the reliability of the BMS.