article · Energies
Controlling motor speed accurately under varying loads and changing system parameters is essential for electric motor systems powered by fuel cells. Different control strategies were evaluated for a high-performance brushless direct current motor, including a classical proportional integral controller, an adaptive neuro-fuzzy inference system, and an approach combining that system with particle swarm optimisation. Testing was conducted on a physical prototype comprising a 1.2 kilowatt proton exchange membrane fuel cell generator, a 1 kilowatt motor, and a dedicated control board. The control goal was to maintain desired rotor speed despite load torque disturbances and parameter variations. Experimental results demonstrated that the particle swarm optimised adaptive neuro-fuzzy controller achieved superior performance compared to the alternatives. Optimising the parameters yielded higher tracking accuracy along with reduced overshoot and shorter settling times during operation.
Electric vehicles relying on fuel cells require precise and reliable motor control to cope with sudden load changes and shifting operating conditions. Demonstrating that optimised neuro-fuzzy controllers deliver faster settling times and reduced overshoot on physical fuel cell hardware helps advance practical motor drive management, supporting smoother and more stable vehicle performance.
This work relates to electric powertrain systems and motor drive management in fuel cell electric vehicles. Automotive engineers and motor controller manufacturers could adopt this approach to improve drive stability and precision. Having been validated on a physical 1.2 kilowatt fuel cell and motor test bench, the technology is at an applied and tested stage, though integration into full vehicle systems would be required.
AI-generated from the published abstract. Always read the original work before citing.
This paper compares the performance of different control techniques applied to a high-performance brushless DC (BLDC) motor. The first controller is a classical proportional integral (PI) controller. In contrast, the second one is based on adaptive neuro-fuzzy inference systems (proportional integral-adaptive neuro-fuzzy inference system (PI-ANFIS) and particle swarm optimization-proportional integral-adaptive neuro-fuzzy inference system (PSO-PI-ANFIS)). The control objective is to regulate the rotor speed to its desired reference value in the presence of load torque disturbance and parameter variations. The proposed controller uses a dSPACE platform (MicroLabBox controller board). The experimental prototype comprises a PEMFC system (the Nexa Ballard FC power generator: 1.2 kW, 52 A) and a brushless DC motor BLDC of 1 kW 1000 rpm. The PSO-PI-ANFIS controller presents better performance than the PI-ANFIS and classical PI controllers due to its ability to optimize the PI-ANFIS controller’s parameters using the particle swarm optimization (PSO) algorithm. This optimization results in improved tracking accuracy and reduced overshoot and settling time.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.3390/en16114395
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
