Experimental Validation of the Performances of an Inverter Sized with Optimization Methods

This work shows the experimental performances of a PWM inverter designed with a fast design automation tool. The chosen optimization method uses a deterministic algorithm with continuous equations. Analytical models are used to design active and passive components under loss, harmonic, thermal and efficiency constraints. Precise calorimetric test benches are used to measure the losses, and an original method is proposed to measure capacitor currents. Then, the results given by the optimization tool, presented as Pareto front, are discussed. A 10kW prototype was built to compare the constraints between the optimization results and the experimental values.

1. Introduction

The concept of More Electrical Aircraft consists in improving the energetic efficiency of aeronautic systems by increasing the electrical embedded vectors. It is now one of the key drivers of technological developments in aerospace. Power converters represent a significant part of the new technologies, and their power density still needs to be improved in order to save weight and therefore fuel consumption. Design automation using optimization method is an interesting way to reach this goal, defining the converter weight as the objective function, and considering the numerous aeronautical constraints. Sequential approach in the design may result in sub-optimal results, because several compromises are necessary during the choice of some parameters [1]. For example, the switching frequency has an opposite impact on the sizing of the semi-conductors and on the design of a low-frequency filter. Solving all tradeoffs simultaneously is the role of the optimization tool.