Automation of thermal and structural design using artificial intelligence techniques

Abstract This paper describes an application of artificial intelligence techniques to design automation of structures to mechamical as well as heat loading. In order to design structures considering many coupled phenomena such as structural deformation, fracture, heat conduction and so on, an ordinary design procedure based on the generation and test strategy seems practical, rather than mathematical techniques such as a synthesis approach because of the broad applicability. To achieve the design automation of structures related to many coupled phenomena with maintaining the flexibility and extensibility, artificial intelligence techniques are introduced in this study. An object-oriented knowledge representation technique is adopted to store knowledge modules related to the design of structures subjected to mechanical as well as heat loading. A data-flow processing technique is also utilized as an inference mechanism among the knowledge modules, that is, objects. In addition, experts' empirical knowledge on design modification is utilized to attain the efficient convergence feature of theiterative design process based on the generation and test strategy. The developed system is applied to the thermal and structural design of a two-dimensional model of a fusion first wall with a cooling channel, and the fundamental preformance of the system is clearly demonstrated.