A combined approach to multi-objective optimization of capacitor placement in radial distribution networks

A problem of capacitor placement in nodes of radial distribution networks is multi-objective because it has been formulated by using several objectives. The main criteria for capacitor placement and options selection are usually considered as the following functions: minimization of active losses and minimization of investment cost. The problem solution does not always give a positive result in terms of engineering logic, the need to consider imprecision of technical information, specific features of the networks requirements, inadequate pricing in conditions of market economics in transition periods etc. Therefore, an additional third criterion, which is based on the Technical-Coenose approach, is proposed in order to obtain a specific solution for radial distribution network. This criterion requires ensuring maximum compliance of capacitor parameters (such as their conductivities, capacities or power) to the specifically normalized hyperbolic rank-parameter dependence. The proposed combined algorithm uses three mentioned criteria in order to find an acceptable solution in the space solutions formed by method of concessions by making the permissible deviations from the optimal value of each criterion. The proposed method allows to find the vector of ranked capacitor conductivities with specific nonlinear dependence, such as hyperbole. The comparison is made between two approaches: usage of two main criteria and proposed combined algorithm. Calculation procedures are based on the modified version of Particle Swarm Optimization approach. The medium voltage distribution networks with balanced loading and radial configuration with possibility of installation of parallel compensation are considered for analyses. A radial distribution feeder with 27 nodes is used as an example to show the proposed method of capacitor placement. The results show that obtained solutions of shunt capacitor banks placement satisfy the formulated problem and show effectiveness of the prospec- of practical purposes applying.