A multi-objective optimized design of LCL filters for grid-connected voltage source inverters considering discrete components

This work proposes a multi-objective optimized design for an LCL filter taking into account discrete components. Volume and power losses are objective functions, while harmonic attenuation, total inductance, and control capability are handled as constraints. For power loss calculation, the currents at fundamental and switching frequency in the LCL filter are calculated. Here, currents at switching frequency are calculated using corresponding simplified transfer functions. On the other hand, decision variables (inductances, capacitance, and damping resistance) are taken from a database of commercial components (discrete decision variables), getting a filter whose final performance results directly from the design process. Under this scenario, the optimization problem is solved using the Non-Dominated Sorting Genetic Algorithm II. The methodology provides a Pareto Front of solutions with both objective functions. Every solution (filter design) has specific commercial components, and designer can select the one, which better fits with his requirements. As example, an LCL filter for a 1 kVA three-phase inverter is designed, where two of the resulting filters are simulated together with the inverter, using PSIM software, in order to evaluate their performance.