Parametric Structural Design for Mycelium and Seaweed Fiber: Sustainable Furniture Component Optimization Method Based on Physical Simulation and Reinforcement Learning

This study presents a parametric structural design framework for sustainable furniture components fabricated from mycelium and seaweed fibers by integrating Proximal Policy Optimization (PPO) with continuum-mechanics-based physical simulation. Fiber orientation angle, porosity distribution, and layer thickness are encoded as continuous state variables, enabling the policy network to generate adaptive parameter adjustment strategies during the optimization process. Finite element analysis is employed to evaluate stress distribution, deformation behavior, and failure probability, while reward functions are formulated to improve structural stiffness and stress uniformity. Under a compressive load of 2.5 kN, the optimized structure achieves a stress standard deviation of 2.98 MPa, a displacement norm integral of 5.93×10 mm/N, and a failure probability of 12.17%, indicating enhanced mechanical stability and microstructural controllability. Transferability analysis further demonstrates improved elastic recovery and breathability regulation in smart textile applications across different porosity levels. The proposed framework establishes a quantitative relationship between microstructural parameters and macroscopic performance, providing an efficient design methodology for porous bio-based materials and functional material architectures. The approach also offers potential value for the optimization of structurally adaptive components used in intelligent material systems and electromagnetic-responsive engineered structures.