This paper presents work towards a formal framework to support model-based integrated design and optimization of mechatronic products for early-phase conceptual design. This paper describes an integrated design framework through the introduction of its software implementation and a specific use case. The contribution is to introduce mathematical formalism to define the concepts, semantics, computation rules and system architectures of the formal framework. The advantage of the formal definitions is to clearly expose functionality and the limitations of the design framework and facilitate the software implementation. The modelling capability of the framework is enhanced to include non-linear mechatronic components, such as a two degrees-of-freedom arm. Further, an optimal proportional-integral-derivative control component is added to the software library supporting the framework.

Mechatronic system design includes a combination of different engineering disciplines. A common approach in design of mechatronic systems is based on a sequential method, where different disciplines are treated and designed separately. This paper extends earlier work on integrated physical and control design optimization with integrating an additional aspect of the corresponding embedded control system implementation. Our previous publications describe integrated design optimization through a few specific use cases but the impact of embedded control implementation on the structural design of the systems is neglected. In this paper, the approach is extended to cover discussions on control implementation and its effect on the physical dimensioning and vice versa. A multi-objective optimization approach is implemented and tested on a mechatronic system case study consisting of a DC-motor, a planetary gear, a flexible shaft, an embedded controller and a load. The couplings between the properties of different engineering domains are studied and highlighted. The presented approach which is aimed for early phases of design, considers the integration of three engineering disciplines in one design framework which so far has been missing.

Integrated design optimization of mechatronic systems necessitates a concrete definition of requirements, design parameters and variables, optimization objectives and constraints. A multidisciplinary mechatronic design requires engineers from different domains to work together on the design approach. Different engineering disciplines impose different constraints on the system. To consider the impact of these disciplines and the interactions and coupling between different parameters, there is a need for integrated design method. In this paper, an integrated design optimization approach is presented that takes into account three engineering domains, physical dimensioning, control design and embedded control implementation impact, simultaneously. The optimization problem is a multi-objective method that considers the size of system, sampling frequency and sensor resolution as main objectives. Regarding each objective, there are a few constraints to be satisfied by the final optimum system. The approach is applied to a non-linear mechatronic case including a DC-motor, a timing belt drive and a load. Non-linear dynamics of a timing belt are considered as a new physical component in the method and the system is linearized, controlled and optimized at defined operational points. Sensor resolution and location impact on the optimal system are analysed using a quantization model. It is shown that the approach allows engineering designers to integrate multidisciplinary optimization method to achieve a logical optimal system without degrading the system performance in each domain.

Methodologies on design optimization of mechatronic systems are usually based on consecutive methods, i.e., the procedure of physical design, control and optimization of a system is performed step by step to achieve the final goal. This paper is built upon previous works on developing a toolbox to integrate several engineering backgrounds in early design phase to avoid time and cost consuming iterations in later deign steps. The previous methodology was mainly applicable for linear one-degree of freedom systems without time-variant dynamics. In this paper, the method is upgraded towards covering concepts on nonlinear systems where extra degrees of freedom are added to the system. Additionally, the library of the mentioned toolbox is extended to include ball-screw drive and rotational rigid beam components in terms of physical design, dynamic and static models to examine the feasibility of the design. A conceptual nonlinear multi-degree design case is presented and linearized at specified operational points in the supported software framework and the implemented models are verified in both SimMechanics and Simulink.