Physical Logic Enhanced Network for Small-Sample Bi-layer Metallic Tubes Bending Springback Prediction

نویسندگان

چکیده

Bi-layer metallic tube (BMT) plays an extremely crucial role in engineering applications, with rotary draw bending (RDB) the high-precision processing can be achieved, however, product will further springback. Due to complex structure of BMT and high cost dataset acquisition, existing methods based on mechanism research machine learning cannot meet requirements springback prediction. Based preliminary analysis, a physical logic enhanced network (PE-NET) is proposed. The architecture includes ES-NET which equivalent single-layer tube, SP-NET for final prediction sufficient samples. Specifically, first stage, theory-driven pre-exploration data-driven pretraining, are constructed, respectively. In second under logic, PE-NET assembled by then fine-tuned small sample composite loss function. validity stability proposed method verified FE simulation dataset, small-sample angle potential interpretability applications demonstrated.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Laser Bending of Tubes : Mechanism , Analysis , and Prediction

Laser bending of tubes is a process in which laser-induced thermal distortion is used to bend tubes without hard tooling or external forces. Mechanisms of the process are examined to better understand the deformation characteristics such as wall thickness variation, cross-section ovalization, bending radius, and asymmetry. Factors important to these characteristics are experimentally and numeri...

متن کامل

Springback Effect Prediction in Wipe Bending Process of Sheet Metal: a Ga-ann Approach

Sheet metal is one of the most important semi-finished products used in automobiles, domestic appliances, aircraft and other familiar products. Therefore sheet metal forming technology is an important engineering discipline in the area of mechanical engineering. In sheet metal forming, wipe bending process plays a major role in which the sheet metal tries to return to its original shape after r...

متن کامل

Analytical prediction of springback based on residual differential strain during sheet metal bending

As the springback of sheet metal during unloadingmay cause deviation from a desired shape, accurately predicting springback is essential for the design of sheet stamping operations. Finite-element models have not been successful in predicting springback; hence there is a need for analytical models to make such predictions. In this study, a model based on differential strains after relief from t...

متن کامل

Prediction of Bending Angle for Laser Forming of Tailor Machined Blanks by Neural Network

Tailor-made blanks are sheet metal assemblies with different thicknesses and/or materials and/or surface coatings. A monolithic sheet can be machined to make the required thickness variations that is referred as tailor machined blanks. Due to the thickness variation in tailor machined blanks, laser bending of these blanks is more complicated than monolithic plates. In this article, laser formin...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Lecture Notes in Computer Science

سال: 2022

ISSN: ['1611-3349', '0302-9743']

DOI: https://doi.org/10.1007/978-3-031-20500-2_10