基于代理模型和遗传算法的热芯缠绕芯模结构优化

发布时间：2018-07-11 17:33

本文选题：热芯缠绕 + 响应曲面法　；参考：《哈尔滨理工大学》2017年硕士论文

【摘要】：热芯缠绕成型方法是在传统外固化成型方法的基础上发展而来新型复合材料制品生产制造方法,该方法可实现缠绕固化同步进行,省去了脱模、转运工序,减少了设备占地空间,提高了生产效率,降低了生产成本。其工作流程是在进行纤维缠绕的同时,芯模内部通入高温蒸汽,蒸汽流动对芯模金属外表面进行加热,缠绕与固化同时进行。采用热芯缠绕成型方法生产的复合材料制品,成型效率更高,废品、次品率低,目前此种工艺广泛应用在高端复合材料制品生产线上。热芯缠绕成型方法的出现使得复合材料成型技术得到了很大的改进,但在实际生产中依然存在问题:芯模是一根圆柱状金属管,高温蒸汽在芯模内流动时,极易造成芯模轴向产生较大的温度梯度,造成其外表面温度不均。复合材料制品易出现部分管段固化不完全,进而影响制品品质。芯模轴向温度场的分布在很大程度上受其内部结构的影响,针对以上问题,本文以芯模结构优化为研究核心,对芯模进行结构优化设计以达到芯模外表面温度场均匀的目的。本文主要针对热芯缠绕成型方法进行研究,研究使用高温蒸汽作为加热介质的热芯缠绕成型方法原理,构建芯模及其内部流动的蒸汽流体的物理模型、蒸汽流动与传热过程的数学模型,利用计算流体力学软件Fluent进行芯模蒸汽加热过程数值模拟。依据响应曲面法进行试验方案设计,确定试验因素及因素水平,依据Box-Behnken设计的试验方案进行建模、划分网格、数值模拟仿真。数值模拟仿真结束后,在芯模外表面设置温度跟踪点,读取温度数据,计算得到芯模外表面温度的标准差以此值作为试验响应值,进而拟合出温度均匀性与其内部结构参数之间关系的二阶响应曲面模型。最后利用Matlab遗传算法工具箱,寻找出最优的结构参数组合。依据最优结构参数组合再次进行数值模拟,验证利用遗传算法寻优方案的正确性,验证结果表明,试验值与拟合值相差很小。在利用遗传算法寻出的最优结构参数方案下,芯模表面温度均匀性良好,实现了芯模结构优化的目的,为实际生产提供了理论指导。
[Abstract]:The hot core winding molding method is based on the traditional outside solidification molding method to develop a new production and manufacturing method of composite products. This method can realize the winding solidification synchronously, thus eliminating the demoulding and transferring process. The invention reduces the space occupied by the equipment, improves the production efficiency and reduces the production cost. At the same time of filament winding, high temperature steam is introduced into the core mold, and the outer surface of the core metal is heated by steam flow, and the winding and solidification are carried out at the same time. The composite products produced by hot core winding method have higher molding efficiency, lower rate of waste and inferior products. At present, this technology is widely used in the production line of high-end composite products. The appearance of the hot core winding method has greatly improved the composite molding technology, but there are still some problems in practical production: the core mold is a cylindrical metal tube, and the high temperature steam flows in the core mold. It is easy to cause a large temperature gradient in the axial direction of the core die, resulting in uneven temperature on the outer surface. Composite products are prone to partial tube curing incomplete, and then affect the quality of products. The distribution of the axial temperature field of the core die is largely influenced by its internal structure. In view of the above problems, the core die structure optimization is taken as the research core, and the core die structure optimization design is carried out in order to achieve the goal of uniform temperature field on the outer surface of the core die. This paper mainly focuses on the research of the forming method of hot core winding, and studies the principle of the hot core winding forming method using high temperature steam as the heating medium, and constructs the physical model of the core mold and the steam fluid flowing inside it. The mathematical model of steam flow and heat transfer process is numerically simulated by using the CFD software fluent. According to the response surface method, the experimental scheme is designed, and the test factors and the level of factors are determined. According to the test scheme designed by Box-Behnken, the model is built, meshed, and simulated numerically. At the end of the simulation, the temperature tracking point is set on the outer surface of the core mold, the temperature data are read, and the standard deviation of the outer surface temperature of the core mold is calculated as the experimental response value. Then the second order response surface model of the relationship between the temperature uniformity and its internal structure parameters is fitted. Finally, the optimal combination of structural parameters is found by using Matlab genetic algorithm toolbox. According to the optimal structure parameter combination, the numerical simulation is carried out again to verify the correctness of the optimization scheme using genetic algorithm. The results show that the difference between the experimental value and the fitting value is very small. Under the optimal structural parameter scheme found by genetic algorithm, the surface temperature uniformity of the core die is good, which realizes the optimization of the core die structure and provides theoretical guidance for practical production.
【学位授予单位】：哈尔滨理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TB33;TP18

【相似文献】