首页|Comparative analysis of PI, ANFIS, and SQP-GD controllers for load variation mitigation in off-board EV DC-DC charging stations
Comparative analysis of PI, ANFIS, and SQP-GD controllers for load variation mitigation in off-board EV DC-DC charging stations
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Elsevier
This study examines the effectiveness of three controllers-Proportional-Integral (PI), Sequential quadratic programming-gradient Descent (SQP-GD), and Adaptive Neuro-Fuzzy Inference System (ANFIS)-in reducing the effects of abrupt load changes in an off-board electric vehicle (EV) charging station using an isolated dual active bridge converter (DAB) topology. The charging station uses single-phase shift modulation to transfer power from the DC grid to an electric vehicle. It works with an input of 24 V and an output of 12 V and 10 A current, following the laboratory model. The controllers, PI, SQP-GD, and ANFIS, are evaluated in the same operating conditions, and their transient settling times are measured and studied. The study shows that despite ANFIS being a contemporary controller, it is not superior to SQP-GD. PI is a mathematically proven controller, however it has a longer settling period when dealing with load disturbances. Settling times for PI are 302.89 ms and 569.03 ms, SQP-GD are 33.92 ms and 51.846 ms, and for ANFIS are 69.99 ms and 187.95 ms. The results highlight the excellent performance of SQP-GD in attaining quick settling times, making it a strong option for handling abrupt load fluctuations in off-board EV charging stations.
ANFISPI controllerResponse optimized PI controllerSequential quadratic programming usinggradient descentEV chargingOPTIMIZATION
NETL
NSTL
Elsevier
