Design by Optimization of Electromechanical Actuators - Applications and Industrial Examples

In the first part of our article, we presented the methodological and theoretical aspects of our approach to the design of electromechanical actuators. By posing these design problems as inverse problems and reformulating them as continuous or mixed optimization problems with constraints, we showed how the use of optimization codes could potentially lead to optimized designs. We ended this first part of our article by giving our expertise in how best to combine direct problem-solving methods for calculating cost or constraint functions with high-performance optimization methods.

To illustrate our point, in the second part of this article we will apply this methodology to concrete examples of electromechanical actuator design. As a first step, we have chosen a representative design example that is not too complex, in order to be as didactic as possible. The problem is the design of a magnetic coupling. This illustrative example will be treated in as much depth as possible, in order to highlight the strengths and weaknesses of the various techniques available for solving the associated optimization problems. Different types of models will be proposed:

• analytical models ;

• 2D numerical models based on finite element solving ;

• 3D numerical models also using the finite element method.

Several simplifications of these models will also be discussed, and various optimization codes will be tested and compared.

Before concluding, we will present in a final section three industrial design problems that we have dealt with in contractual settings. Our activities as researchers in electromechanical actuator design have enabled us to acquire a certain expertise in associating the best possible models with the best possible optimization methods, in order to solve these industrial cases. For these 3 industrial examples, we will outline the resolution methods used to obtain optimized designs for these actuators. This demonstrates the effectiveness of the design methodologies discussed and presented in this article.