Passive filters' placement considering parameters' variations

Summary The presence of power system harmonics due to the increasing number of nonlinear loads in modern industry has become a growing concern for the engineers. Among the harmonic mitigation methods available, the use of single tuned passive filters is 1 of the techniques most widely employed due to their simplicity and low cost. The formulation here presented considers the probabilistic nature of parameters' deviations of the filters and of the system by using the Monte Carlo statistical simulation method. The optimization determines a set of single tuned harmonic filters to obtain the maximum saving of annual cost as well as the best power quality indices of the circuit. The obtained solutions comply with the constraints of power quality and of stress of capacitors. The multiobjective optimization problem is solved by the nondominated sorting genetic algorithm. The presented methodology is tested with an example circuit of literature.