A Single Voltage Sensor Bypass Switch-Based Photovoltaic Fault Localization

Photovoltaic (PV) energy systems are becoming an important source of sustainable energy. However, undiscovered faults within these systems may cause significant efficiency reduction. Localizing these faults to the module level is important for a quick fault diagnosis and maintaining the overall system efficiency. This article presents a novel method to localize intrastring, line-ground, cross-string, and partial shading faults in an <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$N$</tex-math></inline-formula> × <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$M$</tex-math></inline-formula> PV system down to the module level. The approach utilizes a single voltage sensor in the combiner box of the PV system and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\lceil N/2 \rceil$</tex-math></inline-formula> bypass switches per string to bypass the connected PV modules during faults. The technique initially relies on identifying the faulty string. Once this string is determined, the voltage associated with each module in that string is found. Each module's voltage in that string is obtained by measuring the string voltage after bypassing each module corresponding to an activated switch. Subsequently, the resulting linear equations are solved to obtain the voltage of each module in the faulty string. The technique is verified using simulation and an experimental setup for a 5 x 4 small-size PV system. Experimental and simulation results demonstrate that the technique can accurately localize faulty modules with only <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$N$</tex-math></inline-formula> voltage samples of the faulty string. The proposed method is robust to variations in the maximum power point tracking algorithm, ensuring faults are localized effectively in real-time.