On optimal multistage electric power distribution networks expansion planning S. Najafi Ravadanegh ⇑ , R. Gholizadeh Roshanagh Department of Electrical Engineering, Azarbayjan Shahid Madani University, Tabriz, Iran article info Article history: Received 27 September 2012 Received in revised form 4 July 2013 Accepted 13 July 2013 Keywords: Optimal Distribution System Planning (ODSP) Optimal feeder routing Optimal HV substation placement Imperialist Competitive Algorithm (ICA) Multistage planning Greedy algorithm abstract The optimal expansion planning of electric power distribution network to meet system load growth and overcome to pseudo dynamic behavior of network parameters considering a large number of constraints is a hard satisfactory multiobjective optimization problem. This paper implements new developed Impe- rialist Competitive Algorithm (ICA) for the optimal expansion planning of distribution network. The topology of medium voltage (MV) distribution network as backbone of electric power distribution sys- tems is designed by optimal sizing, siting and timing of medium voltage network components such as HV substation and MV feeders’ routes. A multistage expansion planning is proposed to consider dynamic behavior of the system parameters asset management and geographical constraints. In order to reach the global solution an efficient coding is developed for ICA parameters. The Greedy algorithm is used to solve the minimum spanning tree problem to construct a radial configuration of the mesh network. At each stage of the problem the results are fully illustrated either by figures or by tables. A sensitivity analysis is used to show the robustness of the results with respect to ICA parameters variation. The obtained results are compared with GA as well known heuristic optimization tool. The efficiency and capability of the methodology has been tested on an under developed relatively large-scale distribution network. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction Electric power distribution system is connection point between power transmission network and energy supply to the consumers. Recently because of the importance of the distribution systems in deregulated and smart grid environments from both electric utili- ties and customers’ viewpoint, the operation, expansion and plan- ning of electric power distribution systems has been in more attention. One of the main problems of long-term planning in dis- tribution system is its sensitivity to system parameters such fore- casted load, asset management and economical constraints. These uncertainties enforce the planning procedures to encounter with many unwanted problems. To overcome for deficiency of the men- tioned problems it is necessary that the planned network is evalu- ated and resolved at a given periods. A multistage planning procedure can help to solve the non-dynamic expansion planning of the network at the previously work. This paper uses the Imperi- alist Competitive Algorithm as recently developed heuristic algo- rithms for the optimal design of large-scale electric distribution systems in order to provide optimal sizing and locating of high voltage (HV) substations and medium voltage (MV) feeder routing simultaneously. In this situation a hard satisfactory optimization problem with different continuous and discrete constraints in large-scale distribution networks should be solved. Because of pseudo-dynamic behavior of mid and long-term planning nature of the problem, a multistage solution is proposed to overcome the system problem arising from parameters variation during plan- ning periods. The aim of this paper is to propose a general frame- work for large-scale distribution system expansion and planning. After mid and long-term spatial load forecasting and medium voltage substation placement, the major and important topics of optimal distribution network planning are MV and HV system planning which includes optimal HV substation planning and opti- mal feeder routing. These two topics may be solve separately or combined with together in a general problem. In this paper a gen- eralized formulation for both optimal HV substation placement and optimal feeder routing problems is developed which directly inserts the constraints of two sub problem in a multiobjective mul- ti constraints optimization problem. The new algorithm aims to minimize capital investment and operating costs of expanded and new developed installation con- sidering electrical, geographical and other constraints in ODSP (Optimal Distribution System Placement) [1]. The proposed model can modify the existing network and determined the new HV sub- stations as well as new MV feeders’ location, size and installation time regarding future load growth. 0142-0615/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijepes.2013.07.008 ⇑ Corresponding author. Tel.: +98 9125597097; fax: +98 4125327500. E-mail address: s.najafi@azaruniv.edu (S.N. Ravadanegh). Electrical Power and Energy Systems 54 (2014) 487–497 Contents lists available at ScienceDirect Electrical Power and Energy Systems journal homepage: www.elsevier.com/locate/ijepes