Numerical harmonic modelling of low wattage LED lamps based on parameter estimation algorithms.

Abstract

This paper presents a numerical harmonic modeling approach for 18 low-wattage Light-Emitting Diode (LED) lamps. The proposed methodology utilizes meta-heuristic algorithms, namely Multi-Stage Ant Colony Optimization (MSACO) and Harris Hawk Optimization (HHO), to estimate optimal parameters for the equivalent circuits. Additionally, the study examines eight different configurations of Electromagnetic Interference (EMI) Filter circuits, including their respective parameters, which function as front-end circuits for the Full Bridge Rectifier (FBR) of the LED drivers. Through extensive investigations, the primary objective is to establish generalized and accurate models for various EMI filter circuits that can be universally applied, with validation through comparison against experimental results. Under both MSACO and HHO, the lowest average total error ranges were observed in the LC-Filter and CL-Filter configurations, with errors ranging from 21.302 % and 22.769 %, respectively. On the other hand, the L-Filter and R-only configurations exhibited the highest average total error ranges, with errors of 42.376 % and 44.648 %, respectively. Notably, this paper introduces a non-intrusive, non-invasive method for estimating optimal parameters based solely on measurements obtained from the input terminals of the LED lamps.