Automated Conceptual Design of Mechanisms Using Improved Morphological Matrix

Abstract

Conceptual design of mechanisms has attracted a number of research efforts in recent years due to its significance in product development. However, existing approaches for automated conceptual design of mechanisms are either prone to a loss of optimal solutions or inextensible to achieve conceptual design of complex mechanisms. This paper is devoted to developing a comprehensive and extensible methodology for automated conceptual design of mechanisms utilizing a design prototype synthesis methodology. To support automated mechanism synthesis effectively, the traditional morphological matrix is improved as a motional function matrix (MFM). In addition, a mechanism prototype knowledge base is developed to provide systematic knowledge support for conceptual design decision-making. Based on the integrated MFM, an exhaustive mechanism synthesis algorithm is developed to yield as many solutions as possible to desired functions to facilitate the discovery of novel and optimal combinatorial solutions. To curb the possible combinatorial explosion from the exhaustive search, a performance constraint verification approach is proposed to help designers filter out combinatorial solutions violating performance constraints, followed by a satisfaction degree-based approach for evaluating the total performances of combinatorial solutions according to the performances of their subsolutions. An automated mechanism conceptual design prototype system is developed and a design case is presented to illustrate the feasibility and practicality of the proposed methodology.