Early spring low temperatures and plastic film use constrain sustainable cotton production in Xinjiang. This study compared five integrated cultivation systems: conventional tillage with plastic mulch (CK), no-tillage with non-woven fabric arch shed (NTN), no-tillage with blue plastic film arch shed (NTB), conventional tillage with plastic film mulching and non-woven fabric arch shed (CTPN), and conventional tillage with plastic film mulching and blue plastic film arch shed (CTPB). Soil hydrothermal conditions, plant growth, root traits, yield components, and seed cotton yield were evaluated. CTPN and CTPB improved seedling-stage hydrothermal conditions and early growth relative to CK. Across the monitored 0–450 mm soil profile, mean soil temperature under CTPN and CTPB was 21.13% and 17.75% higher than under CK, respectively, while mean soil water content was 17.12% and 10.68% higher, respectively. However, these advantages did not result in the highest yield. NTN achieved the highest seed cotton yield, 15.19% higher than CK, followed by NTB with a 9.02% increase. NTN’s yield advantage was mainly associated with higher seed cotton weight per boll. Correlation analysis showed that boll number per plant was negatively associated with yield, whereas seed cotton weight per boll was positively associated with yield. Overall, favorable seedling-stage conditions alone did not determine final productivity, and NTN showed potential for practical application under comparable arid Xinjiang conditions while reducing reliance on conventional surface plastic mulch.

Current small arch shed machine designs rely on manual pole placements, resulting in low construction efficiency and mechanized levels. These machines were not designed with key components tailored to the agronomic requirements of Xinjiang’s small arch shed cotton cultivation model. An automated rod-feeding mechanism for a small arch shed was designed using SolidWorks 2023 to bridge this gap. Its major components include rod separation and conveying units, enabling the separation and orderly transportation of tunnel rods. A kinematic simulation of the conveyor rod during the transport process using ADAMS 2024.1 software was performed to examine the effects of motor speed, synchronous belt stop block height, and horizontal distance on the conveyor rod. Using MATLAB 2023a to fit the center-of-mass distance curve yields the optimal values for the parameters (motor speed = 17.57 rpm, stop block height = 16.79 mm, and horizontal distance = 103.95 mm). Bench test results confirmed the simulation performance of the device with a motor speed of 17 rpm, a synchronous belt stop block height of 15 mm, and a horizontal distance of 100 mm. The automated rod-feeding device exhibited an 80.8% feeding rate. The prototype operates stably, and this design can serve as a reference for developing automated equipment for small arch sheds.