Research On The Technology Of Computer-aided Conceptual Design Based On Product Style Recognition Model

Product style conveys customers' emotional needs on one hand and represents designers' intention on the other hand, which has become an important communication media between customers and designers. Also it is an important way for designers to understand correctly what customers' actual feel, emotional recognition and expectation to products. In traditional way, collecting customers' suggestion about products is made by a detailed investigation. The cost and time of investigation cause overburden in the process of product development, and decrease competitive strength of products. In order to avoiding them, we can use an automatically simulative system, study the model of the product style recognition and its related computer-aided conceptual design technology, summarize potential information from the history database of products, and then mine the potential emotional needs of customers. It certainly will stimulate designers' inventive inspiration, lead to product design process, accelerate product developing process and highly improve the whole benefit of product development.It is an intersection study of computer application, industrial design, artificial intelligence and cognitive psychology. The following works have been researched on product style recognition, computer-aided conceptual design and style computing model.In the 1st part, the review of computer-aided conceptual design is presented, and the requirement analysis phase and the conceptual design phase are syncretized, then the conceptual design is divided into three phases: concept orientation, concept creating and concept choosing. The technology and system of computer-aided conceptual design for product innovation is proposed and the ideal CACD system is presented. Furthermore, the aim, purport, content and status in quo of this research are analyzed and the key research problems and hotspots are brought forward.In the 2nd part, the key research is on the process and method of style-oriented product conceptual design. Firstly the definition, cause and attribute of product style are presented. Secondly, based on the analysis of its connotation, the formerly dissevered designers' coding process and customers' decoding process is put together and a loop is formed, then one general model of the process and method of style-oriented product conceptual design is presented. The model illustrates the twofeatures clusters which compose the product style: physical shape features cluster and mental imago features cluster. The reflection of the two constructions really is studying the product style recurrence and innovation process in conceptual design. Finally, the two ways of product style recurrence and innovation are put forward, also with the corresponding examples.In the 3rd part, the key research is on the computing model of product style recognition, which is illustrated in the two chapters. In chapter 3, it is the first time to use the method of multi-experiment to create the model of product style recognition model based on feature-matching, such as experts visit, eye track observing experiments of influencing the car's appearance features, the features extraction experiments influencing product style recognition, the quantitative relation experiments between key formal features and product style recognition. In chapter 4, aim at the following questions:(l)the method of fuzzy sets must define membership functions, while it is very hard to define membership functions in style space. (2)Artificial neural network represented as the basic indicative unit of product style is too particular.(3)The existing product style recognition model using fuzzy sets is lack of knowledge expressing ability, reasoning ability, uniforming ability and learning ability. This research creatively takes the fuzzy relations as the most basic starting point, instead of membership functions. It adopts multi-matrix to describe the fuzzy relations qualitatively and quantificationally, and adopts random sets theory especially nested random sets to establish uncertain inference method. It has established the product style recognition model based on nested random sets. This automatically simulating mechanism is provided with strict logic basis and does not rely on membership functions.In the 4th part, the key research is on the imprecise concept creating technology base on product style feature lines. Take cars for example, at first the possibility distribution is converted to unprejudiced probability distribution, carrying out the integration of style feature shapes. The method of product style feature lines creating technology based on possibility distribution is put forward. Ulteriorly, the constraints between im-style feature lines and style feature lines are defined and the method of the imprecise concept creating technology base on constraints is presented.In the 5th part, the computer-aided conceptual design system based on product style recognition model is implemented, including product style investigation subsystem and product style computing subsystem, which offer the decision makingtools to designers and customers. Through this system; customers (or designers) can input the style words to get the related conceptual design schemes. Designers can make a sketch of simple formal features in the computers and the system can automatically bring forward the style corresponding to the sketch. Also take cars for instance to verify the practicability and correctness of the theories; methods and technologies proposed in this dissertation.In the end; all of the contributions is summed up and the conclusions about the works are drawn. Moreover; it discusses some unsolved problems and future research works.