Automation of Design Innovation Process Based on CAD Technology and Reinforcement Learning

Abstract

Design innovation is the core power that promotes the sustainable development of modern industry, architecture, science, technology, and many other fields. Computer-aided design (CAD) technology has developed from initial 2D drawings to powerful tools such as 2D modelling and simulation analysis, which provide designers with convenient design means. This article will discuss how to combine reinforcement learning (RL) algorithms with CAD technology to realize automatic image enhancement and Optimization. The experiment uses a large number of product modelling data to train the model and analyzes the convergence of the algorithm in the iterative process in detail. The results show that the new method has a high error in the initial iteration, but after about 20 iterations, the error gradually decreases and tends to be stable. In addition, through comparative experiments, it is found that this method has achieved a higher score in design performance, and its comprehensive performance is better than the traditional method. It is worth mentioning that this method has also achieved a significant improvement in user satisfaction, thanks to its Optimization in user experience, accuracy and reliability, and personalized and customized services. These advantages together reflect the practical application value and potential of the new method in the design field and provide strong support for future research and application.