Case-based Reasoning for the Creative Design of Electromagnetic Devices.

摘要

Design is concerned with the question of how to construct an artifact or system to meet the desired specification and is normally categorized as either routine, non-routine or creative. Up until now, the study of knowledge-based design has concentrated on routine or non-routine design. In order to explore creativity in design, a new computational model based on the combination of case based reasoning (an approach to solving new problems based on the solutions of similar past problems) and soft computing techniques (such as machine learning, data mining and optimization), is proposed in this thesis. The new model is able to deal with the four challenging issues related to creativity: generation of a design prototype from incomplete requirements, judgment and improvement of system performance given a sparse initial case base library, extraction of pivotal features from a given feature space, adaptation of retrieved previous solutions to similar problems for deriving a new innovative solution to a given design task. The core principle underlying this model is that different knowledge from various level cases can be explicitly explored and integrated into a practical design process.;In addition, in order to demonstrate the practical significance of our proposed computational model, a design system for electromagnetic devices which is capable of deriving a new design prototype from a real-world device case base with high dimensionality has been developed.