This article is part of our exclusive IEEE Journal Watch series in partnership with IEEE Xplore.
Researchers from Odisha University of Technology and Research, in India, have developed a model for a direct-current electric motor powered by a photovoltaic array. The system relies on AI to optimize the solar array’s output and operate the motor at 88 percent efficiency; real-world DC electric motors have efficiencies of 75 to 80 percent. Such solar-powered motors could someday be used in industrial machines, household appliances, and even electric cars.
Bismit Mohanty, the lead author on the study, says the focus of the model was on boosting the overall efficiency of the system, to obtain the highest output of the motor for the solar power available. The efficiency gains come from the AI algorithm, which optimizes the power output from the solar array, as well as the motor’s regenerative braking system and a battery that can be charged from both the solar array and the braking system.
Solar cells have a maximum-power point, which is the maximum electrical power they’ll deliver for a given amount of irradiation. The maximum-power point fluctuates with both temperature and sunlight, and so the solar cells don’t always output the maximum amount of power. The way to get as close to the maximum output as possible is by changing the resistance of the solar cells, which changes the amount of power that’s extracted.
That’s where the AI model comes in. In their MATLAB/Simulink model, Mohanty and his colleagues trained a neural network to calculate the solar cell resistance that would yield the maximum output, based on thousands of daily temperature and irradiance measurements. The technique takes advantage of existing AI techniques for maximum-power-point tracking. Because the model is trained using a neural network, it can make predictions using complex criteria but can’t convey the exact criteria for those predictions, operating more as a predictive black box.
According to the model, when it’s sunny, the solar array generates enough power to operate the motor, storing excess energy in the battery. When it’s overcast, the motor runs off the battery. The motor’s regenerative braking system charges the battery whenever the brakes are applied, turning kinetic energy into electrical energy. The team created only a virtual model, but building a working physical model could be a future step.
This model for a solar-powered electric motor could be used in an industrial setting or for household appliances, such as refrigerators and fans. Mohanty says he hopes to see such a system someday used in electric vehicles, which would eliminate the need to plug the EV into the main power grid.
“Now we have to charge the electric vehicle at a station or from the home,” Mohanty says. “I want a chargeless electric vehicle in which power is taken directly from the solar array [on the vehicle].”
The findings were presented in July at the 2023 International Conference on Smart Systems for Applications in Electrical Sciences.
- This Axial-Flux Motor With a PCB Stator Is Ripe for an Electrified World ›
- This Inside-Out Motor for EVs Is Power Dense and (Finally) Practical ›
- New Electric Motor Could Boost Efficiency of EVs, Scooters, and Wind Turbines ›
- Will Nanophotonics Save Solar Power Tech? - IEEE Spectrum ›
- Wind, Water, and Solar Power for the World - IEEE Spectrum ›