A computer package for optimal multi-objective VAr planning in large scale power systems

This paper presents a simulated annealing based computer package for multi-objective, VAr planning in large scale power systems-SAMVAR. This computer package has three distinct features. First, the optimal VAr planning is reformulated as a constrained, multi-objective, nondifferentiable optimization problem. The new formulation considers four different objective functions related to system investment, system operational efficiency, system security and system service quality. The new formulation also takes into consideration load, operation and contingency constraints. Second, it allows both the objective functions and equality and inequality constraints to be nondifferentiable; making the problem formulation more realistic. Third, the package employs a two-stage solution algorithm based on an extended simulated annealing technique and the ε-constraint method. The first-stage of the solution algorithm uses an extended simulated annealing technique to find a global, noninferior solution. The second-stage uses the simulated annealing technique to find the global optimal solution. A salient feature of SAMVAR is that it allows planners to find an acceptable, global noninferior solution for the VAr problem. Simulation results indicate that SAMVAR has the ability to handle the multi-objective VAr planning problem and meet with the planner´s requirements