Rotor Profile Design for Internally Geared Screw Machines Using Deep Neural Network
Lacevic, H., Kovacevic, A.
ORCID: 0000-0002-8732-2242, Read, M.
ORCID: 0000-0002-7753-2457 & Ponnusami, S. A. (2026).
Rotor Profile Design for Internally Geared Screw Machines Using Deep Neural Network.
In:
14th International Conference on Compressors and Their Systems.
14th International Conference on Compressors and Their Systems, 8th - 15th September, City St George's, University of London, United Kingdom.
doi: 10.1007/978-3-032-04102-9_26
Abstract
The internally geared screw machine is a rotary positive displacement machine with two helical rotors rotating in the same direction on offset parallel axes. Working chambers are formed by continuous contact points, with their volume varying cyclically. By controlling the timing of fluid entry and exit, the machine achieves compression. Rotor profile design is a critical phase in compressor development, as it influences working chamber volumes, contact forces, and overall performance. Various established methods, such as the rack and pin-generation methods, use precise mathematical formulations to define rotor profiles. A more advanced approach involves deep learning. Artificial intelligence (AI) is transforming many fields, including compressor design. Recent research has demonstrated its potential in generating rack profiles for conventional screw machine rotors. However, internally geared screw machines impose additional design constraints. This paper presents a preliminary study on training a Wasserstein Generative Adversarial Network (WGAN) to generate rotor profiles that ensure continuous contact. The feasibility of this approach is demonstrated through generated profiles, and future research will explore integrating efficiency-related constraints to enhance rotor design.
| Publication Type: | Conference or Workshop Item (Paper) |
|---|---|
| Additional Information: | This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/978-3-032-04102-9_26 |
| Publisher Keywords: | Screw compressor, Rotor profiling, Gan |
| Subjects: | T Technology > TA Engineering (General). Civil engineering (General) T Technology > TJ Mechanical engineering and machinery |
| Departments: | School of Science & Technology School of Science & Technology > Department of Engineering |
| SWORD Depositor: |
This document is not freely accessible until 8 February 2027 due to copyright restrictions.
To request a copy, please use the button below.
Request a copyExport
Downloads
Downloads per month over past year
Metadata
Metadata