Since the beginning of the automobile era, the internal combustion engine (ICE) has been used for vehicular propulsion. The thermal vehicles powered by the ICE are significant contributors to air pollutants and greenhouse gases linked to global climate change. As the global economy begins to strain under the pressure of rising petroleum prices and environmental concerns, research have been spurred into the development of various types of clean energy transportation systems using electrically propelled vehicles such as Hybrid Electric Vehicles (HEVs), Battery Electric Vehicles (BEVs) and Plug–in Hybrid Electric Vehicles (PHEVs). Especially PHEVs acquire the most attention due to the combination of the electrical source and conventional engine. The PHEVs can provide high fuel economy and less emission, and can provide an extended range compared to other vehicles. However, the major obstacles of the development of the PHEVs are the sizing, characterization, battery behaviour and modeling of the energy storage system. Therefore, in the scientific area, a lot of research work has been performed to provide solutions for these challenges. In this PhD research work, an unique extended characterization methodology has been performed on different rechargeable energy storage systems for PHEVs based on novel developed test procedures, which never have been addressed in the literature. The developed test procedures can reduce the test time with 80% compared to the existing test procedures. The analysis showed the high-energy performances of nickel manganese cobalt oxide based (NMC) battery cells against lithium iron phosphate (LFP) and lithium titanate oxide (LTO). However, from the power point of view, the capabilities of LFP are more powerful than NMC an LTO based battery cells. Furthermore, this technology exhibits superior fast charge capabilities compared to NMC battery technology. Then, the performed investigation showed the variation between the cells of the studied battery types and the need for the development of the appropriate balancing system in PHEVs. Due to the importance of the developed characterization methodology, the detailed behaviour of the various rechargeable energy storage systems can be figured out. This analysis is the basis for the development of an accurate electrical battery model and a thermal battery model in a battery system in PHEVs. Then, this analysis has demonstrated the impact of the different parameters on the battery behaviour. The main parameters that have been considered were state of charge, working temperature, cycle life, current rates, and self–discharge. Based on the battery behaviour and the above-mentioned performance parameters, a novel electrical battery model has been proposed to predict the battery behaviour during transient and steady state durations accurately. In the framework of this PhD thesis, a novel extended experimental investigation has been performed for analysis of the well–known Peukert relationship on lithium-ion batteries. The analysis has shown that the Peukert number is not an unambiguous value. It changes in function of working temperature, current rate ratio, cycle life and depth of discharge. Furthermore, the relationship does not take into account the capacity recovery occurring after stopping the discharge process. Therefore, the novel Noshins relationship has been proposed which takes into account all the mentioned performance parameters based on partial differential equations. This analysis is very unique and has not been used for lithium-ion battery technologies. Comprehensive unique cycle life tests have been carried out in order to investigate the main ageing phenomena in lithium-ion battery technology. This new analysis is of high importance to assess the parameters that influence the battery performances, evaluating the impact of the working temperature, current rates and depth of discharge on the battery cycle life. Furthermore, this analysis is essential for an accurate estimation of the state of health of a battery and development of an ageing battery model. This performed extended analysis is in the scientific world very unique and never has been carried out on lithium–ion batteries. The results of this study indicated that operating a battery at elevated working temperatures (40°C) can accelerate the corrosion phenomena in the battery through the increase of the internal resistance and capacity degradation. Moreover, at low working temperatures (-18°C) the cycle life performances of the batteries are poor (200 cycles) compared to 2600 cycles at 25°C. Furthermore, the analysis exhibits that subjecting a battery to high current rates is harmful for the battery cycle life performances. The cycle life of the investigated battery is 2900, 2060, 1100 and 560 cycles at 1 It, 5 It, 10 It and 15 It, respectively. On the other hand, the cycle life test at different depth of discharge levels indicates that a battery is able to perform 3221 cycles (till 80% DoD) compared to 34957 shallow cycles (till 20% DoD). In order to investigate the cycle life capabilities of lithium iron phosphate based battery cells during fast charging, a number of cycle life tests have been carried out at different charge current rates. The experimental analysis indicated that the cycle life of the battery degrades the more the charge current rate increases. The cycle life of the battery is 2950 cycles at 1.25 It, 1607 cycles at 2.5 It, 660 cycles at 4 It, 567 cycles at 7 It and 414 at 10 It. The analysis on fast charging is unique and never has been performed earlier in the research community. Battery model parameter analysis has been performed using an advanced, high accurate (99% fitting) and non-linear minimization algorithm called ”Levenberg–Marquardt”. In this PhD thesis, the proposed innovative technique has been applied at different battery models such as Thévenin, and FreedomCar, and never has been accomplished before on batteries. The developed minimization technique is much faster (50%) and accurate than the existing battery model parameter estimation techniques. Based on the extended performances of the studied battery models, an innovative electrical battery model has been developed, which is able to achieve higher accuracy than other battery models. The results demonstrate that the error percentage between the simulations and experimental results is 3% compared to >10% for the Thévenin and FreedomCar models. This model is novel and has high performances. During this PhD dissertation, the battery thermal behavior has been described using an innovative developed electro–thermal battery model. The model has been validated at different working conditions and the errors between the simulated and experimental surface temperature of the battery were 1.5°C compared to >4°C for the thermal model. This model is essential for development of an appropriate battery system and thermal management system in PHEVs and can be used for evaluation of the heat development and heat distribution inside a battery pack, which is of high importance in the selection of the appropriate battery package topology and dimensioning of the cooling system. Finally, the impact of the electrical–double layer capacitors (EDLCs
) on the behaviour of high-optimized energy batteries has been investigated, considering different hybridization topologies. The active and the passive hybrid configurations seem not attractive due to the high cost, volume and weight in PHEVs. To overcome these obstacles, a new hybrid topology has been developed based on parallel combination of battery cells and EDLC cells on cell level compared to the active and passive systems at pack level. Therefore, the cost, weight and volume can be significantly reduced by factors 9.23, 2.75 and 5.67, respectively. Such topology is innovative and can accelerate the use of EDLCs with batteries in PHEVs. Cycle life testing has demonstrated the positive impact of the new hybrid configuration on the battery lifetime (30 up to 40% depending of the used EDLC cells).
Promotor: Prof. Dr. Ir. Peter Van den Bossche, Prof. Dr. Ir. Joeri Van Mierlo
Committee members: Prof. Hamid Gualous (Université de Caen Basse Normandie), Prof. Elena Lomonova (Technische Universiteit Eindhoven), Dr. Jean–Marc Timmermans (Unilin)