«上一篇/Previous Article|本期目录/Table of Contents|下一篇/Next Article»

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:

34

期数:

2024年10期

页码:

192-196

栏目:

新型计算应用系统

出版日期:

2024-10-10

文章信息/Info

Title:

Research on Machining Quality Prediction Method of Aerospace Precision Parts Based on BAS-RF

文章编号:

1673-629X(2024)10-0192-05

作者:

张茜1; 吕倩1; 孔宪光2*

1. 中国电子科技集团公司第十研究所,四川 成都 610036;2. 西安电子科技大学,陕西 西安 710071

Author(s):

ZHANG Qian1; LYU Qian1; KONG Xian-guang2*

1. The 10th Research Institute of China Electronics Technology Group Corporation,Chengdu 610036,China;2. Xidian University,Xi’an 710071,China

关键词:

零件加工; 随机森林; 天牛须算法; 质量预测; 优化算法

Keywords:

parts processing; random forest; beetle antennae search; quality prediction; optimization algorithm

分类号:

TP39

DOI:

10.20165/j.cnki.ISSN1673-629X.2024.0195

摘要:

在传统航空零件加工过程中,零件质量的波动问题一直是制造业面临的挑战之一。 为了解决这一问题,提出了一种基于 BAS-RF(天牛须优化的随机森林算法)的零件加工质量预测方法。 该方法结合了启发式优化算法(BAS)和集成学习算法(RF)的优势,旨在提高零件质量预测的准确性和稳定性。 首先,采用 RF 算法对零件特征进行预测。 RF 算法通过构建多个决策树,并通过投票机制综合它们的预测结果,能够有效地处理复杂的非线性关系,具有较强的泛化能力。 为优化 RF 算法的参数,引入 BAS 算法进行模型参数的优化。 BAS 算法是一种基于天牛须行为的启发式优化算法,通过模拟天牛在寻找食物过程中的行为,能够高效地搜索复杂的参数空间。 文中 BAS 算法用于调整 RF 算法中的关键参数,以实现最佳的零件质量预测效果。 通过 BAS 算法的优化,模型能够更好地适应不同的零件加工场景,提高了预测的精度和泛化能力。为了验证该方法的有效性,进行了一系列实验,结果表明,基于 BAS-RF 的零件加工质量预测方法相比传统方法具有更高的预测精度,有效提高了航空精密零件的质量预测精度,在降低零件质量波动、提高生产过程可控性方面具有显著的应用实践意义。

Abstract:

In the process of traditional aviation parts processing,the challenge of fluctuating component quality has long been a prominent issue in the manufacturing industry. To address this challenge, we propose a method for predicting the quality of components in manufacturing based on the combination of the Behavioral Attraction Search (BAS) algorithm and Random Forest (RF),referred to as BAS-RF. This method aims to enhance the accuracy and stability of component quality predictions by leveraging the strengths of the heuristic optimization algorithm BAS,and the ensemble learning algorithm RF. Primarily,the RF algorithm is employed for predicting component features. By constructing multiple decision trees and aggregating their predictions through a voting mechanism, the RF algorithm effectively handles complex non-linear relationships and exhibits strong generalization capabilities. To optimize the parameters of the RF algorithm,we introduce the BAS algorithm. BAS,inspired by the foraging behavior of beetles, is a heuristic optimization algorithm that efficiently explores complex parameter spaces. In this study,BAS is utilized to adjust key parameters within the RF algorithm,aiming to achieve optimal predictions of component quality. Through the optimization facilitated by the BAS algorithm,the model demonstrates improved adaptability to diverse manufacturing scenarios,resulting in enhanced prediction accuracy and generalization capabilities. To validate the effectiveness of the proposed method,a series of experiments were conducted. The results indicate that the BAS- RF method outperforms traditional approaches in terms of prediction accuracy, particularly showcasing significant practical implications for reducing component quality fluctuations and enhancing the controllability of the manufacturing process. This research contributes to advancing the accuracy of quality predictions for aerospace precision components,underscoring its noteworthy practical sig-nificance.

常用功能

更新日期/Last Update: 2024-10-10