IMPRESS

Innovative and multiparametric monitoring system, with AI micronodes, for Electric Battery 

This project represented a first step towards the realization of a system capable of diagnosing a battery module in real-time based on multiparametric measurements. In particular, an experimental prototype was designed, implemented, and tested on a four-battery module to measure voltage, current, impedance spectroscopy, and temperature for each battery in the module. In parallel, some AI algorithms were tested, using databases available in the literature, in order to identify the best models that can be implemented in the future on the realized prototype.
Finally, through a dedicated experimental setup, a testing campaign was conducted under various operating conditions to obtain an innovative dataset related to the estimation of the state of charge of the adopted batteries. In order to create a complete system for the diagnostics of battery modules, numerous developments can be implemented on the proposed prototype. First of all, it will be necessary to use the proposed prototype in subsequent experimental campaigns to increase the amount of measurement data obtained.
In fact, only through the exploration of different operating conditions can algorithms of artificial intelligence aimed at battery diagnostic operations be trained and tested.
Subsequently, it will be necessary to implement the useful AI algorithms on the prototype to perform the battery diagnosis on the prototype. In this sense, a careful analysis of the microcontrollers to be used in the prototype will be necessary to manage the computational complexity required by the identified algorithms.
Finally, an engineering process of the prototype will be necessary to meet certain functional requirements dictated by the specific application regarding the placement of the sensors needed for monitoring within the battery modules.