Considering the failure of fully exploring the transferability of samples, domain adaptation (DA) is still troubled by the problem of the disturbed distribution of the target domain. In this article, a conditional domain confusion network (CDCN) is proposed, which commits to exploit the minimum class confusion (MCC) to achieve unsupervised cross-domain detection. To utilize the transferability of samples to better respond to domain shift, we first propose a transfer disturbance factor under the guidance of class confusion to constrain the classifier, designed to eliminate the impact of hard-to-transfer samples on the model. Moreover, an adaptive auxiliary component is embedded before the classifier, aiming at improving the generalization performance of the model, which adjusts the intra- and interclass distances of samples by uniting the classification loss to better fit the classifier. Extensive experiments are performed on multiple transfer tasks, which demonstrate the superiority of our proposed CDCN over the state-of-the-art DA strategies.

Understanding the microscopic interaction between agricultural tillage tools and soil is essential for improving wear resistance. In this study, molecular dynamics (MD) simulations are employed to investigate the tribological behavior of the Fe–SiO₂ interface under varying loads and sliding velocities. The results demonstrate that the coefficient of friction increases with both normal load and sliding velocity, accompanied by a clear running-in stage. Under high loads, significant plastic deformation occurs, characterized by asymmetric atomic pile-up, expansion of the strain field, and heterogeneous von Mises strain distribution. Energy analysis reveals intensified kinetic and potential energy variations, indicating enhanced defect accumulation and interfacial non-equilibrium states. Temperature distributions are highly localized at the interface, with thermal saturation observed under high-velocity conditions. Mean square displacement (MSD) results confirm that higher loads and velocities promote atomic migration and plastic flow. This study provides atomic-scale insights into wear mechanisms under extreme mechanical conditions, offering theoretical support for the design of durable soil-engaging components in agricultural machinery.

Silica-rich dust intrusion is a persistent challenge for lubrication systems in agricultural machinery, where abrasive third-body particles can accelerate wear and shorten component service life. Here, molecular dynamics simulations are employed to elucidate how SiO₂ nanoparticle contamination degrades polyalphaolefin (PAO) boundary lubrication at the atomic scale. Two confined sliding models are compared: a pure PAO film and a contaminated PAO film containing 7 wt% SiO₂ nanoparticles between crystalline Fe substrates under a constant normal load and sliding velocity. The contaminated system exhibits a higher steady-state friction force, faster lubricant film disruption and migration, and consistently higher interfacial temperatures, indicating intensified energy dissipation. Substrate analyses reveal deeper and stronger von Mises stress penetration, increased severe plastic shear strain, elevated Fe potential energy associated with defect accumulation, and reduced structural order. Meanwhile, PAO molecules store more intramolecular deformation energy (bond, angle, and dihedral terms), reflecting stress concentration and disturbed shear alignment induced by nanoparticles. These results clarify the multi-pathway mechanisms by which abrasive SiO₂ contaminants transform PAO from a protective boundary film into an agent promoting abrasive wear, providing insights for designing wear-resistant lubricants and improved filtration strategies for particle-laden applications.

Graphene coatings have considerable potential to improve the tribological performance of polyetheretherketone (PEEK) under dry friction conditions. For PEEK components operating in marine environments, direct exposure to seawater introduces water molecules and dissolved ions into the sliding interface, which may alter interfacial interactions and lubrication behavior. However, how seawater alters the lubrication mechanism and protective effect of graphene at the PEEK interface remains poorly understood. To address this issue, molecular dynamics simulations were performed to compare PEEK with and without graphene under dry and seawater conditions, thereby revealing the interfacial deformation and molecular response. The results show that graphene preserved the structural integrity of PEEK, reduced its tangential displacement by more than 99%, and decreased the shear stress transmitted into the polymer matrix by approximately 85% to 90%. In the absence of graphene, seawater reduced the tangential mean square displacement of PEEK by approximately 20%, while increasing the normal mean square displacement by about 17%, indicating that seawater redistributed part of the polymer motion from tangential dragging toward normal and local molecular rearrangement. The combined analysis shows that graphene mainly limits the penetration of sliding loads into PEEK, whereas seawater regulates the direction and spatial distribution of the remaining deformation. This work provides a molecular basis for the design of graphene-protected PEEK interfaces in marine friction applications.