RESEARCH ARTICLE Kalman filter–based approach for detection of series arc fault in photovoltaic systems Mohammad Ahmadi | Haidar Samet | Teymoor Ghanbari School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran Correspondence Haidar Samet, School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran. Email: samet@shirazu.ac.ir Summary Series arc fault is one of the main failures of photovoltaic (PV) systems that should be detected as soon as possible after occurrence. Electrical shock hazard and fire in PV systems are the main consequences of delayed series arc fault detection. In this paper, an approach based on analyzing high‐frequency compo- nents of voltage signal derived by Kalman filter is presented to detect series arc fault occurrence. A criterion is defined using energy variations of the estimated harmonics for this purpose. The performance of the proposed method is evalu- ated by plenty of simulations and experiments in different conditions. These cases include different arc lengths, electrodes movements, temporary shadings, and different types of noise. The results confirm that the proposed technique is efficient from speed response, reliability, and accuracy points of view. KEYWORDS fault identification, photovoltaics, power system protection 1 | INTRODUCTION In the new world, fossil fuels have been known as energy sources with pollution that are reaching their end, so solar energy usage as a clean and renewable energy is more popular. Photovoltaic (PV) systems are among the prevalent solar systems. There are many possible fault types in PV systems, from which the systems should be protected. 1 Series arc fault is one of the common failures in these systems that may happen in any electrical contacts and electrical circuitry. It may occur in the PV inverter circuitry or in modules of a string. In case of delayed fault detection, series arc may lead to harmful and serious risks in PV systems. Fire in the system and electrical shock hazard are among these risks. Generally, voltage of PV strings or PV busses is more than 100 V, which can make them highly susceptible to dif- ferent electrical arc faults. 2,3 Thus, DC arc faults detection is essential to have a safe and reliable system operation. There are three major groups of arc faults that are series, parallel, and ground arc faults. 1 The main origins of series arc fault are corrosion and loose connection of each pole, while parallel fault type appears in case of connection between negative and positive poles of a system and have more broadly origins. Parallel and ground arc faults lead to significant increase in fault current. Therefore, they can be detected by an overcurrent protection, straightforwardly. 4,5 However, List of symbols and abbreviations: PV photovoltaic; KF Kalman Filter; V PV terminal DC voltage; N Maximum available harmonic order in V;n Harmonic order; v n Voltage magnitude of n th harmonic; ω n Angular frequency of n th harmonic; φ n Phase of n th harmonic; v dc Magnitude of exponential term of V; α dc Inverse of time constant of exponential term of V; μ t Zero mean random noise; k Sample number; V k k th sample of the V; T s Sampling period; μ k k th sample of zero mean random noise; S nk k th sample of n th component of V; E nk k th sample of energy of n th estimated component; i Data‐window number; L nk k th sample of a line which is fitted to energy signal of n th frequency component; a n Intercept of L nk ; b n Ratio of L nk ;V arc Arc voltage; I arc Arc current; g Arc conductivity; R gap Arc resistance; e gap Arc electromotive force pulse; T average Mean detection time; M Number of test cases; MDT Mean detection time; TmaxDT Maximum detection time Received: 17 May 2018 Revised: 9 September 2018 Accepted: 6 December 2018 DOI: 10.1002/2050-7038.2823 Int Trans Electr Energ Syst. 2019;e2823. https://doi.org/10.1002/2050-7038.2823 © 2019 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/etep 1 of 21