Fault diagnosis algorithm for LED lamp driven by segmented linear solution in indoor environment via illumination waveform fluctuation similarity calculation

Abstract

In this paper, fault diagnosis (FD) algorithm for LED lamp driven by segmented linear solution via illumination waveform fluctuation similarity calculation is proposed for indoor environment application. The proposed FD algorithm can be divided into four steps: illumination signal pre-processing, illumination waveform fluctuation similarity calculation, similarity assessment, and fault recognition. In illumination signal pre-processing, smoothing the original illumination signal using average filter algorithm, and illumination data normalization with min-max normalization method. Calculating illumination waveform fluctuation similarity with euclidean distance between the illumination signal of detected-lamp and standard illumination templates. Diagnosing the detected-lamp whether occurs fault or not and confirming the fault-type with illumination waveform fluctuation similarity value. To validate the proposed FD algorithm, an experimental setup is established in indoor environment and illumination signal of LED lamps are captured using Light Flicker Analyzer. The experimental results demonstrate the proposed FD algorithm can effectively diagnose faulty LED lamps driven by segment linear solution in indoor environment. The proposed algorithm developed here can benefit the fault diagnosis technology of LED lamp driven by segmented linear solution.