Direct energy deposition (DED) technology shows promising applications in the production of roller die cutters. The optimization of process parameters, scanning strategies, and analyses of compressive properties and wear behavior are required prior to application. Therefore, this work investigated the influence of scanning strategy and overlap ratio on the microstructure, microhardness, compressive properties, and wear resistance of M2 high-speed steel (HSS) with DED on a 316 L cylindrical surface. The results reveal that along the deposition direction of the sample, the grain size gradually decreases, with hardness increasing from 187 HV in the matrix to 708 HV. As the overlap ratio increases, the grain size initially rises and then decreases, while hardness first declines and subsequently increases. The cross-scanning strategy effectively enhances the compressive strength by reducing porosity defects. Furthermore, the compressive strength of the samples initially increases with the overlap ratio before experiencing a slight decrease. The M-3 sample with a 50% overlap ratio exhibits the best compressive strength (3904 MPa). The wear rate decreases and then increases with the rising overlap ratio. Therefore, the M-3 sample, prepared using cross-scanning strategies with an overlap ratio of 50%, demonstrates a uniform and dense microstructure, resulting in superior wear resistance, and the wear rate is as low as 8 × 10−6 mm3·N−1·m−1. The current experimental results provide valuable references for the DED of die-cut knives.
Laser-cladding directed energy deposition enables both the repair and fabrication of complex metallic components with curved surfaces. However, during multi-axis deposition on curved substrates, sharp transient feed-rate fluctuations at corner segments—together with an approximately constant powder feed rate—readily cause local over-deposition and geometric defects (e.g., nodules and humps). These defects compromise surface-profile fidelity, thereby creating a major barrier to practical deployment. To overcome this limitation, we propose a corner-oriented path-optimization strategy based on geometric code parsing. By operating directly on the toolpath without modifying the Computer-Aided Design model or slicing workflow, the proposed method suppresses corner overbuild and associated morphological distortion in curved-surface directed energy deposition, substantially improving dimensional consistency and surface quality. Overall, this strategy provides a scalable and broadly applicable route toward high-precision, high-reliability, industrial-scale curved-surface additive manufacturing.