علي شفيعي فيني

The ability of lithium-ion batteries to provide high energy density and long life has made them an essential component in electric vehicles. The battery thermal management system will inevitably be used to control these batteries' maximum temperature and temperature difference. Thermal management methods for batteries encompass a range of techniques, including air cooling systems, liquid cooling, refrigeration cycles, heat pipes, and phase change materials. Compared to other methods, battery thermal management based on phase change material does not consume energy. Therefore, thermal management based on phase change material is the best compromise between cost, efficiency, and life cycle. This research aims to introduce pressure as a tool to increase energy storage capability in polyethylene glycol 1000 (as a phase change material), to achieve more efficient thermal management of lithium batteries. The current research includes battery cell and module experiments in the first step. In the second step, two machine learning loops were used to predict the results in the desired scenario. The cell test has been performed considering the input parameters of initial temperature (30-37°C), pressure (100-500 kPa), and discharge rate (1-7C) and their effects on maximum temperature, depth of discharge, and the extracted energy. During the discharge process, an 18650 cylindrical cell was investigated. The results showed that, for the discharge rate of 7C, the pressure of 500 kPa increases the discharge time almost twice that of atmospheric pressure. In addition, increasing the pressure from 100 to 500 kPa approximately increases the depth of discharge by nearly 10% and doubles the energy extracted from the cell. For the battery module, the effect of applying pressure on the maximum temperature and the maximum temperature difference has been eva‎luated in the discharge rate of 3 to 7C, and in the initial temperature range of 22 to 30 °C. The module test results showed that applying a pressure of 500 kPa has the greatest effect on the maximum temperature and the maximum temperature difference at the discharge rate of 7C. So the maximum temperature decreases from 64.8 °C at 100 kPa to less than 54.6 °C at 500 kPa and the maximum temperature difference also decreases by about one °C. Then, regression analyses were performed to obtain the best response surface according to all recorded tests. In the last step, machine learning techniques were used to predict the battery cell discharge time. In the second loop of the machine learning algorithm, the machine could draw the temporal diagram of cell temperature changes, ampere-hour capacity changes, and energy discharged from the cell for the desired scenario.

فشار , سيستم مديريت حرارتي باتري , يادگيري ماشين , باتري ليتيوم يوني , پلي اتيلن گلايكول , ماده تغيير فاز دهنده

pressure , Battery thermal management system , Machine Learning , Lithium-ion battery , Polyethylene glycol , phase change material

Ali Shafiei

Dr. Ayat Gharaghani