Development of Intermediate Controllers for Inductive Charging Systems in Dynamic Roadway Systems

Abstract

Electric Vehicles (EVs) are becoming a popular alternative to gasoline fuelled vehicles. Presently most electric vehicles have very large batteries in order to have comparable driving ranges with traditional vehicles and need to be regularly plugged in by the user to charge. A perceived downside however is that EV charging stations besides roads and footpaths are unsightly and prone to weathering and vandalism. Inductive Power Transfer (IPT) is a technology which can allow energy to be transferred wirelessly and presents an attractive means to charge EVs without the need for cables and plugs. Power is transferred via electromagnetic induction from the primary charging coil in the ground to the secondary coil that is mounted underneath the vehicle, allowing EVs to be charged at power levels and efficiencies comparable to traditional plug in chargers. IPT charging systems can also be integrated into highways to allow vehicles to be powered while they are in motion. With widespread adoption of IPT charging systems users can seamlessly charge their vehicle throughout the day and even potentially travel an extended or unlimited distance along highways while being dynamically charged. By taking advantage of this opportunistic charging, smaller batteries can be used in EVs, making them more affordable which should further drive the adoption of EVs. This thesis focuses specifically on dynamic charging issues given most current literature related to IPT is focused on stationary charging. Dynamic charging systems tend to have low coupling factors so a means to improve the coupling using an independently tuned coplanar intermediate coil embedded into the primary pad was investigated. This was compared to a traditional two coil system with the same amount of copper in the primary pad to fairly and fully evaluate the benefits of such a coplanar intermediate coil. It was found that the independently tuned coplanar coil improves the efficiency of a series-series tuned system since it reduces the source losses. However, there appears to be no benefit to having an intermediate coupler with parallel-parallel tuned systems. It was shown that a traditional two coil IPT system was also simpler to tune. Following this, the demands on the power supplies for dynamic charging systems were investigated. These power supplies need to tolerate operating with different air gaps, power levels and secondary electronic topologies and as a result have the potential to become mistuned during operation. The push-pull converter was chosen because of its voltage boosting characteristics. It is also straightforward to drive since all of its switches are referenced to ground. A switchable capacitor bank was added to the push-pull converter, allowing it to operate at a fixed frequency while maintaining ZVS operation with a wide range of mistuning. The proposed system was experimentally validated and it was found to tolerate large coupling variations but lacked some desired controllability. A variation of this power supply known as the unidirectional switch push-pull (USPP) with a reduced buck converter was then investigated. It was found that this circuit was able to tolerate mistuning caused by the vehicle movement and had desirable features such as constant current control and controlled start up and shut down. The impact that the mistuning tolerance had on the efficiency was also discussed. In order to ensure lumped dynamic IPT systems operate at the highest efficiency, a robust detection scheme to recognise upcoming secondary pads for highway charging systems was investigated. The proposed detection scheme uses only the current sensors already present in most power supplies to detect variations in the free resonant current induced within the power supply. The detection scheme was shown to work with most popular power supply and pad topologies and ensures that the primary pad is only on for the period over which it can couple useful power to the secondary pad. Dynamic simulations showing the secondary pad moving at 100 km/h were used to evaluate the performance of the detection scheme at high speeds and experimental validation of the detection scheme with a 3.3 kW dynamic charging system were presented. Finally, a general mathematical model for different power supply and secondary topologies was developed. It utilises the harmonic summation model to allow all of the harmonics in the system to be fully accounted for. Techniques to linearise the secondary rectifier and output filter are developed and an iterative solving procedure is presented to generate all of the steady state waveforms of the IPT systems. Waveforms from the developed model were found to in excellent agreement with commercially available simulation tools. The model can be useful to optimise different power supply and secondary topologies and compare them to each other in terms of effciency, mistuned performance and harmonics generation. This thesis covers a variety of work mostly focused on modelling and developing the electronic systems required for IPT systems. It adds to the state of wireless power transfer as it exists for dynamically powered electric vehicles and lays down the pathway for further development in these areas.