Abstract In this paper , a Reactive Power Planning (RPP) approach based on a proposed voltage stability index is introduced to allocate and size reactive power compensators. The proposed index is employed to identify the weakest bus for compensator placement, while the compensator rating is determined using a sensitivity-based approach to satisfy reactive power requirements under normal operating conditions, ( N − 1) contingency scenarios, and light-load conditions. Unlike many existing voltage stability indices that rely on solving vector or differential equations and computing Jacobian matrices, the proposed index is formulated as a simple algebraic expression and is efficiently evaluated using a single load-flow solution implemented in tools such as MATLAB script. Based on the proposed index, a sequential RPP framework is developed to systematically allocate and size Static VAr Compensators (SVCs) in order to enhance the system voltage profile and maintain all bus voltages within permissible limits, while achieving the minimum required SVC capacity and the corresponding minimum overall investment cost. The performance of the proposed RPP methodology is benchmarked against metaheuristic optimization techniques, namely Particle Swarm Optimization (PSO) and Grey Wolf Optimization (GWO). A comprehensive performance assessment is conducted using quantitative indices related to voltage profile improvement and economic performance. The proposed voltage stability index is validated using the standard IEEE 9, 14, and 39-bus test systems. The sequential RPP strategy is applied to modified IEEE 14 and 39-bus systems incorporating renewable energy sources such as hydro, wind, and photovoltaic (PV) solar generation. The comparative results, including voltage profile enhancement and cost analysis, demonstrate that the proposed RPP approach effectively compensates reactive power requirements, achieves improved voltage profile and minimum investment costs. All simulation studies are carried out using DIgSILENT PowerFactory and MATLAB.
Sonbol et al. (Thu,) studied this question.