Electromagnetic Model-Based Measurement, Sensing, and Detection for Wireless Power Transfer

Abstract

Advances in measurement, sensing, and detection are the basis of technological development, which drives or hampers innovation, technology adoption, confidence in the value services, and security. As electric vehicles (EVs) become one of the dominant means of transportation in the next few decades with wireless power transfer for charging, the accuracy and fairness of energy metering and charging safety become prominent.

This dissertation aims to develop smart electromagnetic measurement and detection systems integrated for wireless charging. A small number of sensors sample the electromagnetic field to reconstruct the information needed for power measurement for fair metering and foreign object detection for safe wireless charging. This electromagnetic model-based measurement, sensing, and detection provide accurate solutions for receiver coil misalignment and power level variations.

First, Faraday coil transfer-power measurement (FC-TPM) is presented for fair metering and transactions of wireless charging in electric vehicles. The transfer-power is defined from the Poynting vector, which is the directed power density. The winding losses in the transmitter and receiver coils are derived and decomposed based on heat dissipation to show how the measurement of transfer-power demarcates the losses and imposes the costs for power losses to each coil based on physical power dissipation, resulting in fair metering. FC-TPM employs non-contact, open-circuited sense coils to calculate the transfer-power. The information obtained from the sense coils (e.g., sense coil voltages) is combined uniquely for the power reconstruction, which is accurate despite receiver coil misalignment without explicitly measuring the misalignment. The coupling coefficient variations to the misalignment are approximated by quadratic functions, explaining why a linear combination of multiple sense coil voltages results in accurate power reconstruction across the variation. Furthermore, this method is accurate over other types of variations (e.g., operating frequencies, different types of wires) since polynomials can generally approximate variations. FC-TPM was demonstrated in hardware accurately within 0.1% errors despite a receiver coil misalignment of up to 10 cm using a 1kW wireless power transfer system.

Second, Electromagnetic Model-Based Foreign Object Detection (EM-FOD) is presented for safe wireless power transfer, where foreign objects neighboring wireless power transfer systems are fire hazards. The same electromagnetic physics model, constructed by the transmitter, receiver, and sense coils, is the normal model and can be used to detect the hazardous objects by excluding them from the original normal model. A target information (e.g., the transmitter coil current) reconstruction by the normal model becomes inaccurate when there is a foreign object. The detection metric is a sequent error in the information reconstruction compared to the true information obtained by an independent measurement simultaneously. The detection metric is invariant to receiver coil misalignment and power level, allowing less risky pre-startup low-power detection. Hardware demonstrations show that a 2 cm diameter U.S. nickel coin can be detected using only 9 W regardless of a receiver coil misalignment of up to 10 cm.

This dissertation is concluded by presenting a calibration-transfer strategy to consider the practical deployment of the wireless charging models for FC-TPM to energy service stations. Open-circuited sense coils are chosen as transfer standards that convey accurate data obtained from the certified standard in standards laboratories to the transmitter and sense coils in energy service stations.