Sustainable Energy, Grids and Networks 22 (2020) 100331 Contents lists available at ScienceDirect Sustainable Energy, Grids and Networks journal homepage: www.elsevier.com/locate/segan Centralized radial feeder protection in electric power distribution using artificial neural networks Marko Išlić a,* , Stjepan Sučić a , Juraj Havelka b , Ante Marušić b a Končar-Power Plant and Electric Traction Engineering Inc., Zagreb, Croatia b University of Zagreb Faculty of Electrical Engineering and Computing, Zagreb, Croatia article info Article history: Received 14 December 2019 Received in revised form 8 February 2020 Accepted 7 March 2020 Available online 20 March 2020 Keywords: Artificial neural network Power distribution grid Power system protection IEEE radial test feeder Matlab simulation Simulink model Radial feeder abstract The paper deals with a new way of determining a fault location in an electric power distribution radial feeder using an artificial neural network (ANN) and presents a new centralized protection method based on ANN. The aim of the devised algorithm is to detect the presence of a fault in a radial feeder and to disconnect faulted laterals from the distribution network. The protection schemes that are currently in use cannot leave the whole feeder energized by disconnecting faulted laterals only without even short de-energization of the healthy part of the feeder. A simulation model is also made for training the ANN and for testing the results. Current values for faulty and healthy states are generated by the simulation model and are used as training and testing data for the algorithm. The purpose of the algorithm is to make a decision which feeder circuit breakers should trip. The simulation model and the ANN are modeled by using MATLAB tools. The results show that the tripping of circuit breakers initiated by the algorithm is correct for all states. © 2020 Elsevier Ltd. All rights reserved. 1. Introduction The extensive development of computers and computational science has resulted in the application of computers in the area of electric power system protection in unprecedented ways. In- dustrial computers have been continuously improved so that they can process increasingly demanding tasks. Without such improvements, the application of machine learning algorithms such as ANNs would not be possible. An ANN mimics the hu- man nervous system and solves the most complex operations by doing many simple operations in parallel. If we assume that the development will be equally rapid in the near future, it will be possible to take a new, centralized protection scheme in the electric power distribution into consideration. The centralized protection scheme consists of a centralized intelligent electronic device (CIED) which collects and processes current measurement values obtained from current transformers on feeder laterals and makes a decision about which circuit breaker has to trip so that the most of the feeder remains energized. The proposed algorithm for decision-making uses an ANN, which means that it is necessary to provide a large dataset for training and testing. Such a dataset is obtained from a simulation model made just for this purpose. The simulation model consists * Corresponding author. E-mail addresses: marko.islic@koncar-ket.hr (M. Išlić), stjepan.sucic@koncar-ket.hr (S. Sučić), juraj.havelka@fer.hr (J. Havelka), ante.marusic@fer.hr (A. Marušić). of all feeder elements that have an impact on short circuit cur- rents. It is possible to simulate a large number of various states that can affect the feeder (regular operation or faulty state) and have an impact on the current values. The current values obtained from the simulation are used as input vectors in the dataset. The outputs of the dataset are actions that affect circuit breakers (trip or no trip). The complete method of the presented approach to the radial feeder protection consists of a simulation model for the dataset creation, the preparation of dataset values for the training and testing stage of the algorithm as well as an algorithm based on an ANN that makes a decision on whether the fault is present in the feeder and which circuit breaker must trip. The main objective of this paper is to show that it is possible to create an algorithm that can identify a fault in a feeder and determine in which lateral the fault occurred. In reality, eleven types of faults occur: • short circuit between three phases ABC, • short circuit between three phases and the ground ABCG, • short circuit between two phases; AB or AC or BC, • short circuit between two phases and the ground; ABG or ACG or BCG and • ground faults; AG or BG or CG. Protection schemes for radial feeders that are currently in use are described in [2] and [3]. A typical protection scheme [2] uses the coordination of a recloser and fuses, but it cannot locate and https://doi.org/10.1016/j.segan.2020.100331 2352-4677/© 2020 Elsevier Ltd. All rights reserved.