Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

A hybrid model compression approach via knowledge distillation for predicting energy consumption in additive manufacturing

Li, Yixin, Hu, Fu, Liu, Ying ORCID:, Ryan, Michael ORCID: and Wang, Ray 2023. A hybrid model compression approach via knowledge distillation for predicting energy consumption in additive manufacturing. International Journal of Production Research 61 (13) , pp. 4525-4547. 10.1080/00207543.2022.2160501
Item availability restricted.

[thumbnail of Liu Y.pdf] PDF - Accepted Post-Print Version
Restricted to Repository staff only until 26 February 2024 due to copyright restrictions.

Download (605kB)


Recently, additive manufacturing (AM) has received increased attention due to its high energy consumption. By extracting hidden information or highly representative features from energy-relevant data, knowledge distillation (KD) reduces predictive model complexity and computational load. By using almost predetermined and fixed models, the distillation process restricts students and teachers from learning from one model to another. To reduce computational costs while maintaining acceptable performance, a teacher assistant (TA) was added to the teacher-student architecture. Firstly, a teacher ensemble was combined with three baseline models to enhance accuracy. In the second step, a teacher ensemble (TA) was formed to bridge the capacity gap between the ensemble and the simplified model. As a result, the complexity of the student model was reduced. Using geometry-based features derived from layer-wise image data, a KD-based predictive model was developed to evaluate the feasibility and effectiveness of two independently trained student models. In comparison with independently trained student models, the performance of the proposed method has the lowest RMSE, MAE, and training time.

Item Type: Article
Date Type: Publication
Status: Published
Schools: Engineering
Publisher: Taylor & Francis
ISSN: 1366-588X
Date of First Compliant Deposit: 20 December 2022
Date of Acceptance: 15 November 2022
Last Modified: 19 Nov 2023 18:24

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics