目的 齿轮系统作为印刷装备核心子系统之一，其优化设计直接影响着印刷装备的动态特性，进而对印品质量产生重要影响，因此亟须对其研究进展进行综述。方法 通过文献综述法，梳理总结印刷装备齿轮系统动力学建模方法、优化设计及动态特性，以及与印品质量之间的关系等方面的研究成果，归纳概括各研究方法及其适应性，并展望未来的研究方向。结果 目前，针对印刷装备复杂齿轮系统精细化动力学建模方法及运行稳定性分析的研究相对匮乏，针对印刷装备齿轮传动系统、执行系统、结构系统动力学建模与优化设计中的耦合作用的研究相对较少，针对印刷装备复杂齿轮系统的动态特性与印品质量指标之间的关系规律的研究相对缺乏，这些将成为未来的研究重点。结论 采用高精度多物理场仿真技术有望进一步对印刷装备齿轮系统进行精细化、复杂性建模，采用拓扑优化算法可以满足印刷装备齿轮系统高稳定性、高耦合性优化设计的需求，采用深度学习方法能够挖掘动态特性与印品质量评价指标的关联规律，可为高端印刷装备关键自主可控技术的研发攻关提供支撑。

Gear systems, as one of the key subsystems in printing equipment, have a direct impact on the dynamic characteristics of printing equipment and further affect print quality. Therefore, it is urgent to provide a review of their research progress. Through a comprehensive literature review, research findings pertaining to dynamic modeling techniques, optimization design strategies, dynamic characteristics of gear systems in printing equipment, and their correlation with print quality were summarized. Methodologies employed in various research endeavors were synthesized and summarized along with their applicability. Insights into prospective avenues for further investigation were provided. Currently, there is a relative lack of research on refined dynamic modeling methods and operational stability analysis of complex gear systems in printing equipment. Research on the coupling effects in the dynamic modeling and optimization design of gear transmission systems, execution systems, and structural systems in printing equipment is relatively limited in the field. There is a relative lack of research on the relationship between dynamic characteristics of complex gear systems in printing equipment and quality evaluation indicators of printing products. These will be the focal points of future research endeavors. High-precision multi-physics simulation technology holds promise for further refining and complex modeling of gear systems in printing equipment. Topology optimization algorithms can meet the demands of high stability and high coupling optimization design for gear systems in printing equipment. Deep learning methods can explore the correlation patterns between dynamic characteristics and print quality evaluation indicators. This research can provide support for the development and breakthroughs in key autonomous and controllable technologies for high-end printing equipment.