Abstract—The placement and sizing of Distributed Generators (DGs) can be formulated as a nonlinear optimization problem to maximize the benefits from its placement, while minimizing its size. This paper proposes a voltage stability index based method for the DG placement and a meta-heuristic technique based on Multi-Objective Particle Swarm Optimization (MOPSO) to provide Pareto optimal solutions. The MOPSO minimizes the DG size as well as the weighted sum of multiple system indices, reflecting the impact of DGs on the improvement of the system performance. The Pareto-optimal solutions present the possible trade-off between the multi-objective index and the DG size. The studies have been carried out on a 30-bus test system and a 41- bus Indian distribution system. The impacts of the DG placement on the system voltage profile, line loss, environment and cost of generation have also been investigated. Index Terms—Distributed generation, multi-objective particle swarm optimization, radial distribution system. NOMENCLATURE C alcc Annualized life cycle cost (Rs.). C 0 , C f Initial capital cost (Rs.), Annual fuel cost (Rs.) C O&M Annual operating and maintenance cost (Rs.) d, D Discount rate, Capital recovery factor. GE Gas emission (Kg/MW). m, P dg DG life in years, DG size (MW). P L(x)dg Real power line loss with DG (MW). P L(x)0 Base case real power line loss (MW). Q L(x)dg Reactive power line loss with DG (MVAr). Q L(x)0 Base case reactive power line loss (MVAr). SLIP System real power line loss index. SLIQ System reactive power line loss index. SVPI System voltage profile improvement index. SGEI System gas emission index. Z ij ,S ij Line impedance and transferred apparent power. I. INTRODUCTION istributed Generation (DG), which is known as embedded generation in Anglo-Saxon countries, dispersed generation in North American countries and decentralized generation in Europe and Asian countries, is an electric power source Naveen Jain, (e-mail: njain@iitk.ac.in), S.N. Singh (e-mail: snsingh@iitk.ac.in) and S. C. Srivastava (e-mail: scs@iitk.ac.in) are with the Department of Electrical Engineering, Indian Institute of Technology Kanpur, Kanpur – 208016, India. connected directly to the distribution network or on the customer side of the meter. DGs can be (a) renewable energy sources such as: wind turbine, Solar Photo Voltaic (SPV), bagasse cogeneration in sugar factories and biomass gasifier, or (b) non-renewable sources such as: internal combustion engine, fuel cell and micro turbine, etc. The DG is a feasible alternative for new capacity addition and can be built in small- capacity modules in short lead time. Hence, it can track load variation more closely, especially in the competitive electricity market environment. DG offers several advantages such as increased system reliability, loss reduction, voltage profile improvement, reducing the level of pollutants by use of green energy sources, transmission and distribution capacity relief and low investment risk [1-2, 10]. The problem of DG planning has recently received much attention by power system researchers. In recent years, several studies have considered techniques for locating DG units in distribution feeders. Methods and procedures of the DG placement vary according to the objectives and solution techniques. Lot of work have already been carried out in the DG planning, which show the growing interest of the researchers in this area. Chiradeja et al. [17] proposed a set of indices to assess some of the technical benefits in a quantitative manner with concentrated load. In this analysis, the DG size as well as its location was considered to be fixed. An analytical approach based on the exact loss formula for optimal size and location considering loss minimization as an objective was presented in [3]. Optimization methods such as Golden section search and Grid search algorithm, using successive load flows for DG siting and sizing and considering the loss minimization criterion, were presented in [4]. Hasan et al. [14] proposed a method based on continuation power flow. The placement of DG was considered at the buses most sensitive to the voltage collapse. Gözel et al. [5] proposed an analytical method for DG sizing and siting, considering the loss minimization in which the topological characteristics of a distribution system was exploited to avoid the use of Jacobian matrix. The effect of various types of loads on the DG planning was discussed in [7]. Kennedy and Eberhart [11-12] presented a Particle Swarm Optimization (PSO) approach which has already been used by many reseachers in variety of applications. Pindoriya et al.[13] presented MOPSO based approach for day-ahead optimal self scheduling of generators under electricity price forecast uncertainty. Reference [15] has compared various options of energy to understand the viable option of DG in Indian Planning and Impact Evaluation of Distributed Generators in Indian Context using Multi- Objective Particle Swarm Optimization Naveen Jain, Student Member, IEEE, S.N. Singh, Senior Member, IEEE, and S.C. Srivastava, Senior Member, IEEE D 978-1-4577-1002-5/11/$26.00 ©2011 IEEE