Investigation of Wear Behavior and LSTM-Based Friction Prediction in Cr/Nanodiamond-Coated Al10Cu Alloys

Cr-based composite coatings with superior wear resistance are in growing demand for high-performance applications in the automotive, aerospace, and general manufacturing sectors. In this study, an Al10Cu alloy produced via powder metallurgy was coated with a chromium/nanodiamond (Cr/ND) composite layer using an electrodeposition process to enhance its tribological performance. The coatings were characterized using scanning electron microscopy, energy-dispersive X-ray spectroscopy, and X-ray diffraction. The resulting Cr/ND layer exhibited a uniform thickness of 73.5–76.2 μm and markedly improved surface hardness (809.4 HV), representing a 15-fold increase over the uncoated alloy (53.6 HV). Pin-on-disk tribological testing under dry sliding conditions showed complete elimination of detectable mass loss (0.00 mg vs. 0.55 mg for uncoated) within the measurement system resolution, indicating excellent resistance to both abrasive and adhesive wear. XRD analysis revealed the formation of a hexagonal close-packed Cr2H phase with incorporated nanodiamond particles. To capture and predict the temporal evolution of the friction coefficient, a customized dual-layer long short-term memory neural network—optimized with a look-back window of 3 timesteps and ReLU-activated dense layers—was implemented. The model achieved superior predictive performance on the coated system, with validation and test R2 values of 0.9973 and 0.9965, respectively, demonstrating enhanced modeling accuracy for surface-engineered materials. These findings demonstrate a significant advancement in wear protection for aluminum alloys and introduce a robust data-driven approach for real-time friction prediction in engineered surfaces.