Abstract
Due to the rise in power consumption in recent years, the rated capacity of the power system has increased, resulting in an increase in the presence of Distributed Generators (DGs) in electrical networks. As a result, short-circuit currents surge when shunt faults occur. Fault Current Limiters (FCLs) are an effective way to suppress fault currents in the power systems. On the other hand, FCLs have an impact on the response speed of the protective devices, such as over-current relays (OCRs), which increase the relay operating time, raising the electrical and mechanical stresses on the system equipment. This paper presents an adaptive OCR algorithm considering the FCLs effect without any delay time. The proposed algorithm includes two modules: (1) a Z-score algorithm based on both the mean and the standard deviation values of the input current data, which is used to detect fault conditions, and (2) tripping characteristic curves based on the current Mean Ratio, which are applied to estimate the appropriate operating time of the adaptive OCR. To verify the method performance, a power system with real parameters is simulated on the Alternative Transient Program platform, and the algorithm procedure is implemented in the MATLAB program. Extensive simulation studies of load changes and various fault types are conducted, encompassing a wide range of fault initiation angles, fault resistances, and fault zones. The quantitative findings of these studies are analyzed in the presence and absence of FCLs/DGs. The simulation results indicate that the proposed algorithm can operate online and adjust its operating time settings automatically. As a consequence, it is able to detect fault instances upon which the relay sends a tripping flag, yet remains inactive under normal operating conditions. The algorithm speed and sensitivity are controllable using a moving data window size. Moreover, it is characterized by being easy to use, reliable, and accurate. Furthermore, the Z-score of the phase current can be used to identify the faulty phase and classify the fault type. In addition, the algorithm can be integrated with other digital protection and automation systems to be applied in conventional and smart grids.