Achieving Fast Electric Vehicle Charging on a Limited Domestic Supply

Vajira Dhanapala, Gayan Wickramasinghe, David Gurwicz, David Hughes

Achieving fast Electric Vehicle charging on a limited domestic supply “Power Harvester” PCIM Europe 2023; International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, Nuremburg Germany 09-11 May, 2023

An increasing number of households are purchasing electric vehicles due to their well-known advantages and environmental friendliness. However, the time required to charge these vehicles remains a major issue. Although technologies exist to charge vehicles at rates of up to 50kW or even 120kW for certain models, access to fast charging is limited to dedicated community charging locations where sufficient power is available. Most residential properties have a total power availability in the range of 25kW, with a maximum power rating of 7.2 kW for domestic EV chargers after allowing for household electric usage. As a result, charging a long-range car at home can take over 10 hours, requiring careful planning. This problem is exacerbated if a single residence has more than one electric vehicle. To address this issue, a patented solution that dynamically provides the maximum available power for charging electric vehicles is presented here, along with details of a proof-of-concept sample.

Transformation of Transformer Design with Artificial Intelligence

Vajira Dhanapala, Gayan Wickramasinghe, David Gurwicz, David Hughes

Transformation of Transformer Design with Artificial Intelligence PCIM Europe 2023; International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, Nuremburg Germany 09-11 May, 2023.

In the process of designing magnetic components, more than forty different parameters can impact the final design outcome. These parameters can each have over two hundred possible values, resulting in a nearly infinite number of possible design permutations for a given set of input parameters, which can make it challenging to determine an optimal design solution. A human designer with significant expertise and experience can reduce the number of potential design permutations, but different experts may have varying opinions on the significance of each parameter, leading to subjective design outputs. Exhaustively evaluating all parameters using traditional computational approaches is impossible due to the vast number of possible permutations exceeding the number of atoms in the world. However, an advanced autonomous approach that employs artificial intelligence algorithms has been developed to surpass even the capabilities of a superhuman designer. This paper discusses this AI based process for magnetic designs.