Research on topology optimization algorithm in additive manufacturing of continuous fiber reinforced composites based on deep learning

Abstract With the rapid development of additive manufacturing technology, continuous fiber reinforced composites (CFRCs) have gradually become the preferred material for high-performance structures due to their excellent mechanical properties. However, in the additive manufacturing process, how to reasonably optimize the fiber distribution of composite materials to improve structural performance and material utilization remains a difficult problem. Although traditional topology optimization methods can optimize the distribution of materials, it is difficult to fully consider the nonlinear characteristics of the fiber direction in complex composite structures. To this end, this paper proposes a hybrid topology optimization algorithm based on deep learning, combining CNN with traditional topology optimization algorithms to optimize the design of fiber reinforced composites in additive manufacturing. The algorithm automatically learns and predicts the optimal fiber layout through deep learning model training, and combines it with traditional topology optimization algorithms to form an efficient optimization framework. Through a large amount of experimental data and simulation verification, this paper demonstrates the optimization effect of this method under different loading conditions. Compared with the conventional method, the proposed method can greatly increase the rigidity and strength of the composite structure and reduce the waste of materials. The optimal efficiency is about 30%. Moreover, the computation time of this method is reduced by 20%, which shows that it has great potential in additive manufacturing.