基于BP神经网络的某型飞机部件装配工时估算方法  

Assembly Time Estimation Method of a Certain Aircraft Based on BP Neural Network

作　　者：刘宝明 刘鑫荣 宋国秋 LIU Baoming;LIU Xinrong;SONG Guoqiu(Engineering Training Center,Shenyang Aerospace University,Shenyang 110136,China;Faculty Aerospace Engineering,Shenyang Aerospace University,Shenyang 110136,China;AVIC Shenyang Aircraft Industrial(Group)Co.,Shenyang 110850,China)

机构地区：[1]沈阳航空航天大学工程训练中心,辽宁沈阳110136 [2]沈阳航空航天大学航空宇航学院,辽宁沈阳110136 [3]中航工业沈阳飞机工业(集团)有限公司,辽宁沈阳110850

出　　处：《郑州航空工业管理学院学报》2024年第4期72-78,共7页Journal of Zhengzhou University of Aeronautics

摘　　要：传统飞机部件装配工时定额依靠人工经验确定,导致装配工时预测准确性低、稳定性差。为解决该问题,提出一种基于BP神经网络的飞机部件装配工时估算方法。首先,以某型飞机机身装配为例,分析其装配工艺流程,划分典型装配工艺环节并归纳装配工时主要集中工艺环节,从装配对象、装配要求两方面分析装配工时的影响因素。然后,构建不同工艺环节理论装配工时预测神经网络模型,并提出操作技能水平、操作姿态修正系数,实现了某型飞机机身装配工时定额准确估算。最后,设计开发了基于CAA技术的装配工时影响因素提取系统,将所得信息输入工时定额计算模型得到预测值并与实际装配工时对比,验证了所提方法的可行性与实用性。Traditional aircraft component assembly time quotas rely on manual experience,resulting in low accuracy and poor stability in assembly time prediction.A method for estimating aircraft component assembly time based on the BP neural network is proposed to address this issue.Firstly,taking the assembly of a specific type of aircraft fuselage as an example,the assembly process is analysed,typical assembly process segments are identified,and the main assembly process segments are summarised.The factors influencing assembly time are analysed from the aspects of assembly object and assembly requirements.Then,theoretical assembly time prediction neural network models for different process segments are constructed,and the coefficients for manual skill and operation posture correction are proposed,achieving an accurate estimation of assembly time quotas for a specific type of aircraft fuselage.Finally,a system for extracting the factors affecting assembly time based on CAA technology is designed and developed.The obtained information is input into the time quota calculation model to obtain the predicted value,which is then compared with the actual assembly time to verify the feasibility and practicality of the proposed method.

关 键 词：飞机部件 典型装配工艺 BP神经网络 工时定额估算 CAA 

分 类 号：V262.3[航空宇航科学与技术—航空宇航制造工程]