Particle Swarm Weighting Factor Optimization for Predictive Control of Three Level Inverter with Balanced Voltages

The paper presents a design procedure of weighting factor optimisation for finite states model predictive direct torque and flux control of an induction machine (IM). The multilevel converter feeding the machine is a three level neutral point clamped voltage source inverter (3LNPC-VSI). The flexibility of the predictive method permits to control simultaneously the torque, the flux and the DC link input voltage of the source. The cost function uses at least one weighting factor adjusted to satisfy desired performance. The chosen value of this factor may not be suited for all ranges of operating points and finding these values are very time consuming and complex. The main aim of this paper is to propose an online optimisation of the weighting factor by the particle swarm optimisation (PSO) approach well known by its robustness and fast convergence to the global optimum. Simulation results show that PSO strategy is very efficient to design accurately and quickly these weighting factors.

• weighting factor optimization
• particle swarm optimization
• predictive direct torque control
• three-level neutral point clamped inverter
• voltage source inverter
• induction motor
• multilevel converters
• online optimization
• DC bus voltage regulation

• 7 — Affordable and clean energy
• 9 — Industry, innovation and infrastructure

• Multilevel Inverters and Converters
• Sensorless Control of Electric Motors
• Electric Motor Design and Analysis
• weighting factor optimization
• particle swarm optimization
• predictive direct torque control
• three-level neutral point clamped inverter
• voltage source inverter
• induction motor
• multilevel converters
• online optimization
• Electrical and Electronic Engineering
• Engineering