PSO technique introduced originally by Kennedy and Eberhart in 1995 75. it involves simulating social behavior among individuals(particles) flying through a multidimensional search space, in which eachparticle represents a single intersection of all of the search dimensions. Theparticles evaluate their positions relative to a goal at every iteration, andthe particles in a local neighborhood share memories of their best positionsand use those memories to modify their own velocities and subsequent positions.PSO has the advantages of parallel computation and robustness, and it can findthe global optimal solution with a higher probability and efficiency thantraditional methods.
PSO is easy to realize, fast converging, and intelligent.W. Yang and Q. Li 76 proposed an approach based on a PSO algorithm to solve the capacitorplacement problem in radial distribution systems. Under consideration of thepotential harmonic effects, different load levels, and practical aspects offixed or switched capacitor banks, the target problem was reformulated by acomprehensive objective function and a set of equality and inequalityconstraints. The proposed solution method employed PSO to search for theoptimal location, type, and size of capacitors to be placed and the optimalnumbers of switched capacitor banks at different load levels.
X. Yu et al. 77 proposed a PSO-based parallel search technique to estimate therequired level of shunt capacitive compensation to improve the voltage profileof the system and reduce active power loss.
K. Prakash, M. Sydulu 78 present a novel approach that determines the optimal location andsize of capacitors on radial distribution systems to improve voltage profileand reduce the active power loss. Capacitor placement and sizing are done byloss sensitivity factors and particle swarm optimization respectively.M.
Al-Hajri et al. 79 proposed a novel approach to optimally solve the problem ofdetermining the location and size of shunt capacitors in distribution systems.the proposed method solves the problems of finding the optimal capacitor sizeand location simultaneously. Throughout the optimization process, both thecapacitor injected reactive power and its location are being treated asdiscrete variables. The objective function considered is to minimize thetotal feeder losses. M.
Khalil et al.7 presented a PSO for optimal capacitor placement considering voltagestability enhancement. A.
Eajal and M. El-Hawary 80 present a discrete version of PSO combined with a radial distributionpower flow algorithm to form a hybrid PSO algorithm. The former is employed asa global optimizer to find the globally optimal solution, while the latter isused to calculate the objective function and to verify bus voltage limits. P.
Sonwane and B. Kushare 81used a particle swarm optimization technique to evaluate objective function for capacitor placement andsizing in IEEE bus system. Result analysis shows that optimal capacitorconfiguration finds proper places and size of the capacitor. This placementimproves power factor, reduces active and reactive losses, maintain voltageprofile and KVA release.
T.Selim 82applied Selective Particle Swarm Optimization in a real large distributionsystem to find the optimal capacitor placement, optimal conductor selection,and network reconfiguration, the main objectives are to increase the networkefficiency by increasing the annual benefits, reducing power and energy lossesand improving the voltage profile. H. Lotfi, M.
Samadi and A. Dadpour 3 proposeda novel method, Improved Particle Swarm Optimization, to placement capacitor inradial distribution systems. This method used a combination of shuffledFrog-leaping algorithm and particle swarm algorithm. the objective function is acombination of the power losses cost and the cost of installing capacitors, todetermine the location and optimal size of capacitors and constraint includingminimum and maximum voltage limitations.