Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins

Traditional feed system mechanism models can partially capture the characteristics of feed systems. However, to simplify the modeling process for simulation and calculation, certain factors such as component interactions and environmental influences are often neglected. These omissions lead to discr...

Full description

Saved in:
Bibliographic Details
Main Authors: HUANG Hua, MEI Le, ZHI Xiaobo, ZHANG Huiwang
Format: Article
Language:zho
Published: Harbin University of Science and Technology Publications 2025-02-01
Series:Journal of Harbin University of Science and Technology
Subjects:
Online Access:https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2401
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849703261153525760
author HUANG Hua
MEI Le
ZHI Xiaobo
ZHANG Huiwang
author_facet HUANG Hua
MEI Le
ZHI Xiaobo
ZHANG Huiwang
author_sort HUANG Hua
collection DOAJ
description Traditional feed system mechanism models can partially capture the characteristics of feed systems. However, to simplify the modeling process for simulation and calculation, certain factors such as component interactions and environmental influences are often neglected. These omissions lead to discrepancies between the mechanism model ’ s predictions and actual performance, ultimately affecting prediction accuracy. To address this issue, this study proposes a digital twin error prediction method for feed systems that integrates a mechanism model with a CNN-BiLSTM neural network. First, the feed system is abstracted as a mass spring-damping system model, taking into account the nonlinear effects of friction. Friction forces and mechanism model parameters are identified through experimental analysis. Subsequently, a neural network is employed to correct residuals between the mechanism model’s predictions and the physical system’ s actual behavior, constructing a comprehensive digital twin error prediction model. Finally, the proposed method is validated using spiral spatial trajectories. Experimental results demonstrate that the error twin model improves prediction accuracy by 76. 04% compared to traditional mechanism models and achieves superior accuracy compared to similar neural network models. This underscores the method’ s effectiveness in predicting feed system following errors, enhancing model generalization and stability, and providing a robust modeling strategy for complex systems.
format Article
id doaj-art-5a355336cd374403bee70dac3dce93c9
institution DOAJ
issn 1007-2683
language zho
publishDate 2025-02-01
publisher Harbin University of Science and Technology Publications
record_format Article
series Journal of Harbin University of Science and Technology
spelling doaj-art-5a355336cd374403bee70dac3dce93c92025-08-20T03:17:19ZzhoHarbin University of Science and Technology PublicationsJournal of Harbin University of Science and Technology1007-26832025-02-013001829710.15938/j.jhust.2025.01.009Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital TwinsHUANG Hua0MEI Le1ZHI Xiaobo2ZHANG Huiwang3College of Mechanical and Electrical Engineering, Lanzhou University of Technology,Lanzhou 730050College of Mechanical and Electrical Engineering, Lanzhou University of Technology,Lanzhou 730050College of Mechanical and Electrical Engineering, Lanzhou University of Technology,Lanzhou 730050College of Mechanical and Electrical Engineering, Lanzhou University of Technology,Lanzhou 730050Traditional feed system mechanism models can partially capture the characteristics of feed systems. However, to simplify the modeling process for simulation and calculation, certain factors such as component interactions and environmental influences are often neglected. These omissions lead to discrepancies between the mechanism model ’ s predictions and actual performance, ultimately affecting prediction accuracy. To address this issue, this study proposes a digital twin error prediction method for feed systems that integrates a mechanism model with a CNN-BiLSTM neural network. First, the feed system is abstracted as a mass spring-damping system model, taking into account the nonlinear effects of friction. Friction forces and mechanism model parameters are identified through experimental analysis. Subsequently, a neural network is employed to correct residuals between the mechanism model’s predictions and the physical system’ s actual behavior, constructing a comprehensive digital twin error prediction model. Finally, the proposed method is validated using spiral spatial trajectories. Experimental results demonstrate that the error twin model improves prediction accuracy by 76. 04% compared to traditional mechanism models and achieves superior accuracy compared to similar neural network models. This underscores the method’ s effectiveness in predicting feed system following errors, enhancing model generalization and stability, and providing a robust modeling strategy for complex systems.https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2401digital twinnumerical control machinefollowing errorfrictionfeed system
spellingShingle HUANG Hua
MEI Le
ZHI Xiaobo
ZHANG Huiwang
Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins
Journal of Harbin University of Science and Technology
digital twin
numerical control machine
following error
friction
feed system
title Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins
title_full Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins
title_fullStr Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins
title_full_unstemmed Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins
title_short Modeling and Prediction of Mixed Errors in Feed Systems Based on Digital Twins
title_sort modeling and prediction of mixed errors in feed systems based on digital twins
topic digital twin
numerical control machine
following error
friction
feed system
url https://hlgxb.hrbust.edu.cn/#/digest?ArticleID=2401
work_keys_str_mv AT huanghua modelingandpredictionofmixederrorsinfeedsystemsbasedondigitaltwins
AT meile modelingandpredictionofmixederrorsinfeedsystemsbasedondigitaltwins
AT zhixiaobo modelingandpredictionofmixederrorsinfeedsystemsbasedondigitaltwins
AT zhanghuiwang modelingandpredictionofmixederrorsinfeedsystemsbasedondigitaltwins