A time-sequence optimization model for single-layer path in LPBF additive manufacturing based on discrete scanning units

Abstract

Laser powder bed melting (LPBF) additive manufacturing technology serves as the core process for producing complex metal components. The sequence of single-layer scanning paths directly impacts printing efficiency, production throughput, and equipment utilization. The traditional row-first scanning strategy, due to its fixed scanning order, inevitably generates extensive long-distance idle movements across rows, resulting in excessive idle time and limited improvements in printing efficiency, thereby failing to meet the practical demands of high-efficiency additive manufacturing. Addressing this challenge, this paper focuses on three key objectives: temporal decoupling of single-layer printing, design of timing optimization algorithms, and validation of optimization results. We have established four core modules: “underlying physical modeling – timing optimization – advantage comparison－result validation”. Additionally, we propose an LPBF single-layer path timing optimization model based on discrete scanning units. Specifically, we developed component-based modeling for single-layer printing durations using temporal decoupling theory, constructed a reference path model incorporating geometric relationships of discrete units, and introduced an enhanced genetic algorithm for timing optimization that achieves precise separation between reference and optimized paths while adapting to diverse component geometries. An error-checking method was employed to automatically validate path effectiveness, alongside an explicit calculation model for optimization efficiency, incorporating temporal components and multi-parameter consistency constraints to enhance model stability. To validate the proposed methods and models, we conducted experiments on a single-layer part composed of 38 discrete units using a dedicated MATLAB simulation platform, generating four sets of independent visualizations. Experimental results demonstrate that the proposed model maintains high computational efficiency even for large-scale discrete scanning units, with a single optimization task for 1,000 units completed within 12 seconds. This method effectively reduces long-distance idle movement, decreasing idle time by over 13.8% and improving overall printing efficiency by 5.4%. It also supports optimization for discrete units of varying scales, demonstrating significantly superior performance compared to traditional line-first scanning methods. The model balances theoretical rigor with strong engineering applicability, featuring high computational efficiency that directly meets the real-time optimization requirements of additive manufacturing. It provides a new technical approach and theoretical foundation for enhancing LPBF additive manufacturing efficiency, holding significant engineering and academic value for improving production productivity and advancing the high-quality development of additive manufacturing.