Reliability Improvement of Distribution System: A Hybrid Approach Based on GA and NN
S Chandrashekhar Reddy1, P.V.N.Prasad2, A. Jaya Laxmi3
1S.Chandrashekhar Reddy, Assoc.Professor, Electrical & Electronics Engineering , Christu Jyoti Institute of Technology & Science, Jangaon, Warangal, A.P-India.
2P.V.N.Prasad, Professor, Electrical Engineering, Osmania University College of Engineering, Osmania University, Hyderabad, A.P-India
3A.Jaya Laxmi, Assoc.Professor, Electrical & Electronics Engineering, J.N.T.U.College of Engineering, J.N.T.University, Hyderabad, A.P-India
Manuscript received on January 01, 2013. | Revised Manuscript received on January 02, 2013. | Manuscript published on January 05, 2013. | PP: 1-3 | Volume-2, Issue-6, January 2013. | Retrieval Number: F1069112612/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Due to high power demand, modern utilities are continuously planning the expansion of the electrical networks. One of the methods used for the expansion of electrical networks is connecting distributed generator (DG) in the distribution system. The main function of DG is to generate power based on the load condition or any fault occurs in the electrical network. By connecting DG in the distribution system, the power demand of the system can be satisfied and also it improves the reliability of the electrical network. The major problem in DG is, identifying the optimal location for fixing DG in the system and also computing the optimal number of DG to be connected in the system. By considering the abovementioned problem, here a hybrid technique is proposed, which includes genetic algorithm and neural network to identify the optimal number & location of DG to be connected in the system. The proposed method also computes the amount of power to be generated by each DG for various load conditions. By connecting DGs, the number of generators in the network increases and so that different generator states are possible for a particular load condition. From the possible generator states, the best state is selected based on some reliability parameters. Here, the reliability parameters that are considered for identifying the best generator states are loss of load probability (LOLP), loss of load expectation (LOLE), expected energy not supplied (EENS) and system expected outage cost (ECOST). The above reliability parameters are computed for different load conditions and also for the optimal number of DG identified using the proposed method. By using this method, the best generator state for different load conditions and also for different number of generators is computed. The result obtained shows the development in system reliability due to connecting optimal number of DG in the system.
Keywords: Reliability, ECOST, EENS, LOLP, LOLE, DG, Distribution system.