Artificial Neural Network Based Wireless Power Transfer Behavior Estimation

The proper design of Wireless Power Transfer (WPT) devices requires prior estimation of behaviors such as load voltage, current and power. The value of these behaviors depend on the physical construction of the coils and variable factors such as position between the coils. Moreover, the accurate estimation of such functions is very computational time consuming. This cost distresses situations such as optimization process where you require multiple estimations of those functions. Using pattern recognition to model such parameters is expected to reduce the time and the simulation cost. In this paper, Artificial Neural Network (ANN) is implemented to identify the patterns in a biomedical application of WPT to transcutaneously transfer energy to a device inside human's body. The outcome of the ANN is expected to simplify future optimization processes.