Adaptive Infill Method Based on Cross-Section Features and Transition-Layer Design

This paper proposes an adaptive infill strategy for FDM that addresses two practical issues: weak interlayer bonding at the minimum cross-section and the loss of efficiency caused by globally increasing infill density. The method constructs an adaptive coefficient from the slice-wise area sequence and maps it to the layer density; transition layers are inserted immediately before density changes to smooth the gradient and enlarge the effective interlayer contact. A manufacturable toolpath is obtained by linking the density–line-spacing relation to Zigzag and Hilbert patterns and clipping them within the contour via Boolean operations. Two compression groups and two tensile groups were tested. At identical nominal density, Hilbert patterns achieved higher peak loads and energy-absorption rates than Zigzag. With the adaptive strategy, compression curves evolved from single-peak to multi-peak with milder load drops, increasing total and specific energy absorption while incurring only a small rise in build time (approximately 1.2%–4.12%). In tensile tests, interlayer bonding strength improved by about 12.6%–40.5% and 29.3%–41.2% across the two groups, with negligible or slightly reduced mass (about 1%–5%); Hilbert again performed best. These results show that cross-section-driven variable density combined with transition layers can markedly improve load carrying and energy absorption of FDM parts at modest manufacturing cost.