Answers are grounded in this paper and cite their sources.
Simulation study demonstrates rapid visual drape prediction in weft-plain knitted fabrics, highlighting new efficiencies for digital garment manufacturing.
Visual drape simulation of fabrics enables a more intuitive demonstration of their comprehensive mechanical properties. However, the current process relies heavily on time-consuming physical prototyping and repeated experiments, resulting in poor timeliness. This paper aims to study the rapid prediction method of fabric drape visualization. This paper proposes a fast and reliable visual prediction method of fabric drape based on coil structure and physical particle simulation, using weft-plain knitted fabric as example. First, mesoscale decoupled finite element simulations were conducted based on the structural and yarn characteristics of 70 weft-plain knitted fabric. The obtained fabric morphological mechanical parameters were specifically used to match the virtual mechanical parameters of the Mass-Spring Model (MSM). Through MSM-based drape visualization simulations and experimental validation, a more reliable and comprehensive drape parameter and visual image database was established. Then, a random forest regression algorithm suitable for high-dimensional and nonlinear fitting was employed to construct a drape prediction model using coil structural parameters as independent variables. Based on the prediction results, the corresponding visual simulation effects could be rapidly matched and displayed. Finally, validation on the test set showed a high average R2 value of 0.8124, with 95% of the projected area errors confined within a small range of ±4.8 cm2. Moreover, compared to the time-consuming physical prototyping and experiments, the efficiency of visual prediction was significantly improved. Therefore, this method can rapidly and reliably predict drape visualization based on predefined structural parameters, contributing to the advancement of fabric digitization, intelligent garment manufacturing, and related industries.
No takes yet. Share an insight, caveat, or question.
Li et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: