期刊
JOURNAL OF MANUFACTURING SYSTEMS
卷 59, 期 -, 页码 607-616出版社
ELSEVIER SCI LTD
DOI: 10.1016/j.jmsy.2021.04.012
关键词
Multi-task learning; Gaussian process; Response surface modeling; Ultrasonic metal welding; Smart manufacturing; Data-efficient learning; Process optimization
资金
- National Science Foundation [1944345]
- Div Of Civil, Mechanical, & Manufact Inn
- Directorate For Engineering [1944345] Funding Source: National Science Foundation
Response surface modeling is essential for identifying optimal input parameters in processes. This paper introduces a new approach based on hybrid multi-task learning to address challenges such as the high cost and time-consuming nature of traditional models and the inadequacy of existing methods to account for nonlinear relationships. The effectiveness of this method is demonstrated through a case study on ultrasonic metal welding processes, highlighting its superior performance without increasing experimental costs.
Response surface modeling is an essential technique for identifying the optimal input parameters in a process, especially when the physical knowledge about the process is limited. It explores the relationships between the process input variables and the response variables through a sequence of designed experiments. Conventional response surface models typically rely on a large number of experiments to achieve reliable modeling performance, which can be cost prohibitive and time-consuming. Furthermore, nonlinear input-output relationships in some processes may not be sufficiently accounted for by existing modeling methods. To address these challenges, this paper develops a new response surface modeling approach based on hybrid multi-task learning (H-MTL). This approach decomposes the variability in process responses into two components-a global trend and a residual term, which are estimated through self-learning and MTL of Gaussian process (GP), respectively. MTL leverages the similarities between multiple similar-but-not-identical GPs, thus achieving superior modeling performance without increasing experimental cost. The effectiveness of the proposed method is demonstrated by a case study using experimental data collected from real-world ultrasonic metal welding processes with different material combinations. In addition, the hyperparameter selection, the effects of the number of tasks, and the determination of the stopping criterion are discussed in detail.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据