OPTIMIZED FAULT LOCATION FORMULATION FOR UNBALANCED DISTRIBUTION FEEDERS CONSIDERING LOAD VARIATION Mariana Resener 1 mariana@ece.ufrgs.br Rodrigo H. Salim 1 salim@ece.ufrgs.br André D. Filomena 1,2 afilomena@ece.ufrgs.br Arturo S. Bretas 1 abretas@ece.ufrgs.br 1 Federal University of Rio Grande do Sul (UFRGS) Electrical Engineering Department Porto Alegre, RS, Brazil 2 Companhia Estadual de Geração e Transmissão de Energia Elétrica (CEEE-GT) Porto Alegre, RS, Brazil Abstract – In this paper an extended impedance based fault location formulation for unbalanced distribution systems is presented. The majority of distribution feeders are characterized by the existence of several laterals, non- symmetrical lines, highly unbalanced operation and time- varying loads. These specific characteristics compromise traditional fault location methods performance. The one- terminal proposed formulation uses as input data local voltages and currents, system’s topology and pre-fault loads. Test results are obtained from numerical simula- tions using real distribution system data from a Brazilian energy company. Comparative results show the technique robustness in respect to traditional fault location limita- tions, such as fault distance, resistance, inception angle, as well as for load variation. Keywords: Fault Location, Fault Detection, Unbal- anced Distribution Feeders, Power Distribution Pro- tection 1 INTRODUCTION Electric power systems are constantly exposed to faults, which affect the system’s reliability, security and energy quality. Different stochastic events may cause systems faults, such as lightning, insulation breakdown and trees falling across lines. Protection schemes are important to maintain system stability and minimize the damages to the consumer and the electrical network, detecting and isolating the fault as fast as possible. In case of permanent faults, fault location techniques should be used for faster system restoration. In trans- mission systems, the most accepted one-terminal data- based fault location methods are based on the positive sequence apparent impedance [1]. Fault location in Electric Power Distribution Systems (EPDS) however, due to their specific topological and operational characteristics, still present challenges [2],[15]. Still today, the fault location in EPDS is done by the great majority of electric power distribution companies through visual inspection, technique not feasible in underground systems and very time consum- ing in large distribution networks. Therefore, a specific fault diagnosis method for EPDS becomes very impor- tant and useful. In recent years, several fault location methods for transmission [3],[4],[15] and distribution systems [5],[7],[15] have been proposed by the scientific com- munity. These methods, however, do not totally con- sider the unique characteristics of distribution systems, which affect considerably the methods performance. These characteristics have already been considered before, but not in a single unified fault location formu- lation. The fault location methods described in [5],[7] consider the existence of different laterals and sublater- als in the distribution system, but only a single loading characteristic is considered. The load compensation applied in [6] considers measurements in each one of the load points. However, these measurements are not always available in distribution systems. In this paper, an extended fault location formulation for unbalanced distribution feeders is presented and discussed. The proposed fault location method uses as input data local voltages and currents and also pre-fault system’s loads and topology. The method is based on the positive sequence apparent impedance calculation and fundamental phasor quantities. Also, the formula- tion considers the load time-varying characteristics of EPDS [8]. In order to validate the proposed method, the formulation was implemented in MATLAB platform [9] and had its performance tested on a real data under- ground distribution feeder of the Electrical Energy Dis- tribution State Company of Rio Grande do Sul (CEEE- D), southern Brazil, simulated with the use of ATP- EMTP [10]. The remaining of this paper is organized as follows. Section 2 describes and discusses the proposed fault location formulation. Section 3 presents the study sys- tem description. The results and conclusions are pre- sented in the sections 4 and 5, respectively. 2 FAULT LOCATION FORMULATION Consider the faulted system illustrated in figure 1.The terminal S voltages are given by (1): ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ + ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⋅ ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ ⋅ = ⎥ ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎢ ⎣ ⎡ Fc Fb Fa Sc Sb Sa cc cb ca bc bb ba ac ab aa Sc Sb Sa V V V I I I Zl Zl Zl Zl Zl Zl Zl Zl Zl x V V V (1) where 16th PSCC, Glasgow, Scotland, July 14-18, 2008 Page 1