Poly-phase Inductive Power Transfer Systems for Electric Vehicles

Abstract

Inductive power transfer (IPT) was first discovered over a century ago as a method of electromagnetically transferring power. The absence of physical contact enables IPT systems to achieve advantages over conventional conductive methods, allowing greater freedom of movement in the secondary, electrical isolation and resistance to abrasion, corrosion and environmental factors. Due to these advantages, IPT systems have become increasingly ubiquitous during the past two decades in a wide spectrum of applications ranging from materials handling to biomedical devices and battery charging. One application of IPT systems that has shown promise recently is wireless charging of electric vehicles (EVs). As the number of EVs on the road continues to increase in the future due to lower life-time costs of EVs and environmental concerns, wireless charging of EVs is a convenient alternative that could complement the relatively short range of EVs. As IPT systems for EV charging is an active topic of research, numerous systems have been suggested to date. However, a disjoint is present between many of the proposed systems since each system is designed for a specific range of power levels for either smaller passenger vehicles or larger vehicles for materials and public transportation. The interoperability problem is further exacerbated by the systems using different magnetic structures topologies with sizes and tolerances to misalignments that are specific to their application. Additionally, these IPT systems need to minimise the generation of leakage magnetic flux, which is the unwanted stray magnetic flux that may be harmful to people. Constraining the leakage magnetic flux below the regulatory guidelines potentially impedes IPT systems from transferring high levels of power. This thesis investigates the use of multiple independent coils in IPT systems to solve the challenges in power transfer capability and leakage magnetic flux. A collation of mathematical models are derived in order to understand the functionality of systems with a single coil or multiple independent coils. A controller is designed based on the mathematical models to optimise the energisation of the systems with multi-ple independent coils to generate magnetic field shapes that minimise effort exerted from the primary electronics to transfer power. The controller is implemented on a new magnetic structure proposed in this thesis called the Tripolar pad (TPP), which consists of three independent coils that can be energised individually, to evaluate the performance of the proposed pad. A TPP under optimal control conditions for maximising coupling to the secondary is shown to be interoperable with existing magnetic structure topologies and highly tolerant to misalignments of the secondary by generating magnetic field shapes that best suit power transfer. IPT systems using such a TPP to transfer power at 3.3 kW and 20 kW are examined in the thesis to show that the TPP is suitable for transferring both lower and higher power. The TPP can also be built using a similar volume of material to conventional pads and driven with power electronics with lower VA ratings, which substantially offsets some of the cost from the additional switches, diodes and tuning networks. The TPP is also found to generate less leakage magnetic flux than other magnetic structure topologies in most cases, especially when the secondary is misaligned. The findings in the thesis indicate that IPT systems using multiple independent coils, such as the TPP, are a viable alternative to other conventional IPT systems for EV charging.