DC Arc Fault Model Superimposing Multiple Random Arc Noise States on an Average Model

This paper presents a new series DC arc fault model to provide accurate simulations in systems such as photovoltaic (PV) arrays. The model is derived by observing the different physical arcing states in air which correspond to specific frequency spectrums. The foundation of the model is pink noise, which is used to derive a current noise power spectral density inversely proportional to the square root of the frequency at different slopes depending on the physical state of the arc. The simulated noise is superimposed on an existing DC arc fault model in the time domain. This new modified pink noise model can: 1) increase the accuracy of frequency-dependent attenuation simulations and filtering effects of inverters, converters, and PV panels on arc faults; 2) facilitate the development of detection algorithms that use the frequency domain by reducing the number of real experiments and enabling simulations of arc faults anywhere in the system.