Porsche Engineering is testing an artificial-intelligence-supported “soft switching” approach designed to significantly reduce power losses in electric vehicle inverters, a development the company says could lift driving range by a high single-digit percentage in certain operating ranges.
According to the German automaker’s engineering arm, simulations show that intelligent control of power transistors could cut switching losses by as much as 95%. In electric vehicles, the inverter is a major source of inefficiency, with losses typically divided into line losses—caused by the residual resistance of transistors—and switching losses that occur as transistors move between on and off states.
See also: Porsche to Close Self-Operated Charging Stations in China Amid Strategic Shift
“Line losses are a physical property of the transistors, and one that cannot be influenced by the inverter’s circuit design,” Porsche Engineering said, adding that switching losses increase as transistors are activated more frequently, even though higher switching frequencies can improve the quality of the generated alternating current.
The proposed solution relies on so-called soft switching, in which transistors are not switched abruptly but instead at optimized moments to minimize energy loss. Porsche Engineering is focusing on zero-voltage switching (ZVS), where the transistor is activated when voltage is close to zero, a method the company believes is more efficient than zero-current switching.
See also: Porsche Designs EV Batteries for 15-Year Lifespan via Advanced Management and Crash Safety
“The more often the transistors are switched over, the greater the problem,” said Volker Reber, senior manager for function and software development at Porsche Engineering. “Then again, high switching frequencies are welcome in the inverter, because this can improve the quality of the alternating current that is generated.”
Porsche Engineering said the main challenge has been the highly dynamic operating conditions of electric vehicles, which until now have prevented widespread adoption of soft switching. By applying artificial intelligence, the company aims to overcome this limitation. The algorithm predicts optimal switching moments in real time, even as load conditions change.
See also: Porsche Plans Virtual Gearshift Feature for Future Electric Models, Report Says
“An electric vehicle operates under constantly changing loads, which is why soft switching has not yet been applied in any other way,” Reber said. “We were able to change that by using artificial intelligence to optimize transistor switching… This allows us to achieve full soft switching with minimal losses and, as a result, higher ranges.”
The company is currently validating the concept in simulations, with a focus on reducing losses in silicon-carbide and gallium-nitride power transistors, materials increasingly used in modern EV power electronics.
