基于机器学习的柱塞式橡胶集流器优化方法  

Optimization method of plunger rubber collector based on machine learning

作　　者：孔维航[1,2] 李绍华 郝虎 邵玉广 孔令富[1,2] KONG Weihang;LI Shaohua;HAO Hu;SHAO Yuguang;KONG Lingfu(School of Information Science and Engineering,Yanshan University,Qinhuangdao,Hebei 066004,China;Hebei Key Laboratory of Computer Virtual Technology and System Integration,Qinhuangdao,Hebei 066004,China;Aerospace Science&Industry Shenzhen(Group)Co.Ltd.,Shenzhen,Guangdong 518000,China)

机构地区：[1]燕山大学信息科学与工程学院,河北秦皇岛066004 [2]燕山大学河北省计算机虚拟技术与系统集成重点实验室,河北秦皇岛066004 [3]航天科工深圳(集团)有限公司,广东深圳518000

出　　处：《燕山大学学报》2023年第5期458-470,共13页Journal of Yanshan University

基　　金：国家自然科学基金资助项目(62173290,62306264);中央引导地方科技发展资金项目(236Z0303G);河北省自然科学基金资助项目(F2022203008);河北省高等学校科学技术研究资助项目(ZD2020118,QN2023189);河北省创新能力提升计划资助项目(22567626H)。

摘　　要：为有效提升测井用柱塞式橡胶集流器的集流性能,开展柱塞式橡胶集流器结构参数优化方法研究。首先构建橡胶弹性件受力膨胀的有限形变机理模型,理论分析橡胶弹性件受力膨胀内压与主伸长的内在关系,并确定适用于橡胶弹性件的Mooney-Rivlin超弹模型;然后基于橡胶弹性件有限形变机理构建其流固耦合仿真模型,确定橡胶弹性件受力形变影响因素:橡胶弹性件的厚度、轴向长度和硬度;进一步以橡胶弹性件的性能和内液体体积为优化目标,建立双隐层BP网络模型对橡胶弹性件结构参数进行寻优求解,得到橡胶弹性件的最优结构参数组合:硬度43 HA、厚度4 mm和轴向长度25 cm,优选后的橡胶弹性件使得橡胶集流器整体性能得到大幅提升,其集流性能提升57.69%、注液口压强减少20%、橡胶弹性件内液体体积增加25%。该双隐层BP参数优化模型最大预测误差仅为7%,误差率低,对数据的拟合程度更好,为后续优化设计提供理论支撑。In order to effectively improve the flow collection performance of the plunger rubber collector,an optimization method of the structural parameters of the plunger rubber collector is proposed.Firstly,the feasibility of Mooney-Rivlin model for rubber elastic part is determined.The finite deformation mechanism model of the expansion of the rubber elastic part is established.The relationship between the internal pressure and the principal elongation of force expansion of the rubber elastic part is analyzed theoretically.Then,based on the finite deformation mechanism model of the rubber elastic part,the fluid-structure interaction model is constructed.The factors affecting the force deformation of the rubber elastic member is determined as the thickness,the axial length and the hardness of the rubber elastic part.Further,with the performance and internal liquid volume of the rubber elastic part as optimization objectives,the double-hidden-layer BP network model is established.The optimal structural parameters of the rubber elastomer are obtained:hardness of 43HA,thickness of 4mm and axial length of 25cm.The optimized rubber elastomer improved the overall performance of the rubber flow collector by 57.69%,reduced the pressure at the injection port by 20%,and increased the volume of liquid in the rubber elastomer by 25%.The maximum prediction error of the double hidden layer BP parameter optimization model is only 7%,which is a low error rate and a better fit to the data,providing theoretical support for the subsequent optimization design.

关 键 词：橡胶集流器 Mooney-Rivlin模型 集流性能 参数优化 机器学习 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]