数控加工系统在线优化持续学习方法综述

Overview of Online Optimization Methods for NC Machining Systems

航空航天零件具有品种多、批量小、形状复杂和精度高的特点,针对上述加工场景,数控加工技术逐渐发展并有效推动了制造过程的高效化和自动化,为飞行器的更新换代提供强有力的支持。飞行器性能的不断提升对零件加工质量提出了更为严苛的要求,传统数控加工技术面临严峻挑战。近年来,人工智能技术的发展为进一步提高加工质量提供了有效途径,本文对人工智能领域的持续学习方法进行分析,进而结合持续学习研究进展,提出一种面向数控加工系统的在线优化研究方案。

Aerospace parts have the characteristics of varied varieties,small batch,complex shape and high precision.For the above scenarios,numerical control technology is proposed and gradually developed.The continuous development of numerical control technology effectively promotes the high efficiency and automation of the manufacturing process,and provides a strong support for the upgrading of aircraft.Due to the continuous improvement of the performance of aircraft,the machining quality of parts has been put forward more stringent requirements,the traditional numerical control technology is weak in the face of the above problems.In recent years,with the development of artificial intelligence technology,it has provided an effective way to further improve the machining quality.Since it is difficult to accumulate a large amount of machining data for a single part in the aerospace manufacturing field,effective learning algorithms under the condition of small samples are needed.As different types of data are accumulated in the process,online learning can effectively improve the representation ability of the model.