Monitoring system parameters and loads in wireless power transfer systems without radio-frequency communication system

Abstract

Wireless power transfer technology has become a preferable solution for supplying contactless energy over a certain air-gap to various applications, ranging from microwatt biomedical implantable devices to kilowatt battery charging systems for electric vehicles, due to its exclusive features of safety, flexibility and convenience. In such applications several sensors mounted on both the input and output sides are typically necessary to monitor the load variations or other system parameter changes. The communication system of radio-frequency is installed to link the two separate circuits between the power supplying and sinking terminals as the feedback loop for the control purpose. The concept of getting rid of such the complicated installations to monitor systems parameters and loads only based on the input side measurements is presented in this thesis.

An offline method to identify the system parameters based on the input voltage and current measurements as well as the use of a Genetic Algorithm is proposed. It is effective in finding all the unknown parameters even in large wireless power transfer systems which consist of multiple resonant coils. This method has been successfully illustrated in a three-coil experiment with good agreements between calculated and measured parameter values.

If the system parameters are known, a method is presented to show that, based only on the measurements of the input voltage and current, the load impedance of a wireless power transfer system can be instantaneously monitored without using any direct measurement from the load. Subsequently the power loss optimization and output power control can be achieved by tuning the input frequency and adjusting the input voltage amplitude. The principles are favorably verified with practical measurements obtained from an eight-coil system.

In a series-series compensated two-coil system, both the mutual inductance and load resistance can be determined uniquely with measurement of the input voltage and current at only one specific operating frequency. This discovery is useful in applications where the load is movable. In addition it can be adopted as a simple effective method to calculate the mutual inductance between any pair of coupled coils if it is difficult to get access to either coil.

There is a need for powering multiple loads simultaneously from the power source in some applications. A computational method for monitoring double loads from the front-end of a wireless power transfer system without using any wireless communication link is reported. Measurements of the system responses under more operating frequencies are required to derive the load conditions with good accuracy. A four-coil system has been set up to confirm the effectiveness of the proposed approach under both static and dynamic load conditions.