导　　师： 张湘伟

学科专业： 0802

授予学位： 硕士

作　　者： ;

机构地区： 广东工业大学

摘　　要： 随着家用空调的广泛使用,各空调企业的竞争力在不断发展,消费者对于空调产品的性能要求也越来越高。伴随着各大空调制造商对市场占有率的争夺,原材料价格的上涨以及运输费用的增加等因素,空调行业的利润增速放缓。作为空调主要支撑件的钣金件在空调中占有相当大的比重,怎样在进一步的降低钣金件厚度的情况下使其刚强度同样能够满足使用要求成为各大空调制造商急需解决的难题。 优化设计能够更加合理的利用材料的性能,使得内部结构得到最好的协调。本文针对某款空调产品,通过优化空调底板的内部结构来减薄底板的厚度,并使其满足使用性能。首先,在Pro//E中建立空调底板的三维模型,然后在ANSYS Workbench平台对减薄后的底板进行有限元分析,然后采用多级优化的思想对底板进行结构多目标优化,使底板的刚强度满足要求。 对于结构优化后的空调底板,需要对其进行冲压成形仿真以验证其可制造性。在DYNAFORM平台多次对底板进行冲压成形仿真,并调整其冲压工艺参数,使得优化后的底板取得较好的成形效果。 因为冲压参数的设置严重依赖于试验者的相关经验,DYNAFORM仿真计算非常费时,本文对MATLAB平台底板冲压成形质量的BP神经网络预测进行了研究。通过正交试验获得神经网络的训练样本,并编程实现了破裂、起皱衡量指标的计算,利用MATLAB神经网络工具箱函数创建了空调底板的BP神经网络成形预测模型,经测试取得了良好的效果。 With the wide use of household air conditioner, and continuous development of air conditioner manufacturers competitiveness, the consumer has higher requirement for the performance of air conditioner products. With each big air conditioner manufacturers competition for market occupancy, and the rise in raw material prices and the rise in the cost of transportation, air conditioner industry profit growth is slowing. As air conditioner's main support, the sheet metal parts occupies a considerable proportion in air condition, how to further reduce the thickness of the sheet metal parts to make it also meet the stiffness and strength requirements have become each big air conditioner manufacturers need to solve problems. Optimization design can use of material performance more reasonable and make the internal structure best coordination. This paper based on a certain type of air conditioner products, through optimize the internal structure of air conditioner chassis, Under thinning the thickness of the chassis circumstances, and make it meet use requirement. First, building up air conditioner chassis's three-dimensional model in Pro//E, and then do finite element analysis for the thinning chassis in the ANSYS Workbench platform, and then using the idea of multi-stage optimization to multi-objective optimize the chassis, make it also meet the stiffness and strength requirements. Stamping simulation to verify its manufacturability is needed for structure optimized air conditioner chassis. Stamping simulation was carried out to chassis for many times in DYNAFORM platform, by means of adjust the stamping process parameters, makes the chassis to achieve good forming effect. Because of stamping parameters setting relies heavily on the experience of experimentalist, DYNAFORM simulation calculation takes a lot of time, this paper studied air conditioner chassis forming quality BP neural network prediction in MATLAB platform. Through the orthogonal experiment to obtain the neural network training samples, and. achieve fracture and wrinkling index's computation through the programming, Build air conditioner chassis's BP neural network prediction model using the MATLAB neural network toolbox function, and the test achieved good effect.

关 键 词： 空调底板 优化设计 冲压成形仿真 正交试验 神经网络

分 类 号： [TG386]

领　　域： [金属学及工艺]