A study into the potential of GPUs for the efficient construction & evaluation of kriging models
- Submitting institution
-
University of Southampton
- Unit of assessment
- 12 - Engineering
- Output identifier
- 20751627
- Type
- D - Journal article
- DOI
-
10.1007/s00366-015-0421-2
- Title of journal
- Engineering With Computers
- Article number
- -
- First page
- 377
- Volume
- 32
- Issue
- 3
- ISSN
- 0177-0667
- Open access status
- Out of scope for open access requirements
- Month of publication
- September
- Year of publication
- 2015
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
0
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Surrogate model construction can introduce a significant burden within a design optimisation. The demonstration of GPUs as a method to significantly reduce this burden an enable larger datasets with high dimensions to be employed is therefore quite important. Enabling accelerations of an order of magnitude in some instances, the algorithms presented within this paper have been implemented within the proprietary Rolls-Royce optimisation suite, OPTIMATv2, and are routinely used in the development of their engines. Applications include whole combustor (Trent XWB) and combustor tile (Trent 1000, Trent 900) optimisation (contact marco.zedda@rolls-royce.com) and turbine blade robust design (contact Marcus.Meyer@rolls-royce.com).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -