Abstract This paper presents a Python framework for the preliminary design of embedded mechatronic systems. Mechatronic systems have the particularity to involve several levels of design and several technologies. In addition, embedded systems introduce to the design problem specific constraints like energy consumption, resistance/impact to/on environment, geometrical integration and reliability. In order to support the designer in satisfying these constraints, a Model-Based Design methodology and the corresponding framework is proposed. Models used during preliminary design of such multi-domain systems come from several disciplines and have different scales: distributed parameters (3D FEM, 3D CFD) for local level, lumped parameters (1D/0D, ODE/ADE) and state machine for global level. The dynamic simulation through 0D-1D models of the system to be designed is commonly used to validate architectural choices and preliminary sizing that requires multi-disciplinary optimization. Unfortunately, optimization can not only apply to the system level and resort to 3D models cannot be avoided, even during the early design phases. The proposed framework aims at implementing the optimization loops at both local and global levels in an open source environment. The proposed framework’s features will be underlined through a case study - a rotary electromechanical actuator for primary flight control.
[1]
Jean-Charles Maré,et al.
Requirement-based system-level simulation of mechanical transmissions with special consideration of friction, backlash and preload
,
2016,
Simul. Model. Pract. Theory.
[2]
F. Thielecke,et al.
A Knowledge-Based Approach for Design and Modelling of High Lift Actuation Systems
,
2011
.
[3]
Michel Lebrun.
Simulation et CAO en automatique et mécatronique
,
2003
.
[4]
Jean-Charles Maré,et al.
Optimal preliminary design of electromechanical actuators
,
2014
.
[5]
Marc Budinger,et al.
Dimensional analysis and surrogate models for the thermal modeling of Multiphysics systems
,
2017
.
[6]
H. Elmqvist,et al.
Automated formula manipulation supports object-oriented continuous-system modeling
,
1993,
IEEE Control Systems.