木勺毛坏捆抓取柔性夹持器结构优化

打开文本图片集

关键词：柔性夹持器；木勺毛坏捆；有限元分析；BO-RF随机森林；SHAP分析；遗传算法；多目标优化；应变能 中图分类号：TP242.6；S776.3 文献标识码：A DOI： 10.7525/j.issn.1006-8023.2026.01.013

Structural Optimization of Wooden Spoon Blank Bundle Gripping Flexible Gripper

ZHANGJiacheng，ZHANG Sen，LIU Yutong，HUANG Hui，ZHAO Hui* （College of Mechanical and Electrical Engineering，NortheastForestry University，Harbin 150040，China）

Abstract：Toaddressthelow efficiencyof manual waterimmersionand thesusceptibilitytodamage of woodenspoon blanks inthe molding process，this paper presents the design of an adaptive flexible gripper for grasping wooden spoon blank bundles.Anoptimized experimentaldesign was developed using finite element analysis software.Aregression prediction modelforthe gripper's strain energywas constructed using a Random Forest（RF） methodoptimized bya Bayesianoptimization（BO）algorithm，termed BO-RF.Furthermore，the explainable machine learning method，SHAP （SHapleyAdditiveexPlanations），was employed toprovideboth global andlocal interpretabilityanalysisof the model. Based onthis predictive model，a genetic algorithm（GA）was utilized to perform anoptimal design of theflexible gripper，with theconflicting objectives of maximizing the topsurface strain energyand minimizing the overallstrain energy. The multi-objective Pareto front was computed usingthe BO-RF model.Simulationresults validatethe effectiveness and feasibility of the proposed modeling and optimization methodology.

Keywords：Flexible gripper；wooden spoon blank bundle；finite elementanalysis；Bayesian optimization-random for est；shapley additive explanations；genetic algorithm；multi-objective optimization； strain energy

0引言

在木勺模压工艺中需要将木勺头部浸水3～6s，以备第二天的模压成型。（剩余11159字）