On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions
Experiments in engineering are typically conducted in controlled environments where parameters can be set to any desired value. This assumes that the same applies in a real-world setting, which is often incorrect as many experiments are influenced by uncontrollable environmental conditions such as t...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Cambridge University Press
2024-01-01
|
| Series: | Data-Centric Engineering |
| Subjects: | |
| Online Access: | https://www.cambridge.org/core/product/identifier/S2632673624000406/type/journal_article |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841526435511009280 |
|---|---|
| author | Mike Diessner Kevin J. Wilson Richard D. Whalley |
| author_facet | Mike Diessner Kevin J. Wilson Richard D. Whalley |
| author_sort | Mike Diessner |
| collection | DOAJ |
| description | Experiments in engineering are typically conducted in controlled environments where parameters can be set to any desired value. This assumes that the same applies in a real-world setting, which is often incorrect as many experiments are influenced by uncontrollable environmental conditions such as temperature, humidity, and wind speed. When optimizing such experiments, the focus should be on finding optimal values conditionally on these uncontrollable variables. This article extends Bayesian optimization to the optimization of systems in changing environments that include controllable and uncontrollable parameters. The extension fits a global surrogate model over all controllable and environmental variables but optimizes only the controllable parameters conditional on measurements of the uncontrollable variables. The method is validated on two synthetic test functions, and the effects of the noise level, the number of environmental parameters, the parameter fluctuation, the variability of the uncontrollable parameters, and the effective domain size are investigated. ENVBO, the proposed algorithm from this investigation, is applied to a wind farm simulator with eight controllable and one environmental parameter. ENVBO finds solutions for the entire domain of the environmental variable that outperform results from optimization algorithms that only focus on a fixed environmental value in all but one case while using a fraction of their evaluation budget. This makes the proposed approach very sample-efficient and cost-effective. An off-the-shelf open-source version of ENVBO is available via the NUBO Python package. |
| format | Article |
| id | doaj-art-848bbed2dd3246599662bdd728e8a2bd |
| institution | Kabale University |
| issn | 2632-6736 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | Cambridge University Press |
| record_format | Article |
| series | Data-Centric Engineering |
| spelling | doaj-art-848bbed2dd3246599662bdd728e8a2bd2025-01-16T21:48:44ZengCambridge University PressData-Centric Engineering2632-67362024-01-01510.1017/dce.2024.40On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditionsMike Diessner0https://orcid.org/0000-0001-9838-0862Kevin J. Wilson1Richard D. Whalley2School of Computing, Newcastle University, Urban Science Building, Newcastle upon Tyne, United KingdomSchool of Mathematics, Statistics and Physics, Newcastle University, Herschel Building, Newcastle upon Tyne, United KingdomSchool of Engineering, Newcastle University, Stephenson Building, Newcastle upon Tyne, United KingdomExperiments in engineering are typically conducted in controlled environments where parameters can be set to any desired value. This assumes that the same applies in a real-world setting, which is often incorrect as many experiments are influenced by uncontrollable environmental conditions such as temperature, humidity, and wind speed. When optimizing such experiments, the focus should be on finding optimal values conditionally on these uncontrollable variables. This article extends Bayesian optimization to the optimization of systems in changing environments that include controllable and uncontrollable parameters. The extension fits a global surrogate model over all controllable and environmental variables but optimizes only the controllable parameters conditional on measurements of the uncontrollable variables. The method is validated on two synthetic test functions, and the effects of the noise level, the number of environmental parameters, the parameter fluctuation, the variability of the uncontrollable parameters, and the effective domain size are investigated. ENVBO, the proposed algorithm from this investigation, is applied to a wind farm simulator with eight controllable and one environmental parameter. ENVBO finds solutions for the entire domain of the environmental variable that outperform results from optimization algorithms that only focus on a fixed environmental value in all but one case while using a fraction of their evaluation budget. This makes the proposed approach very sample-efficient and cost-effective. An off-the-shelf open-source version of ENVBO is available via the NUBO Python package.https://www.cambridge.org/core/product/identifier/S2632673624000406/type/journal_articleBayesian optimizationblack-box optimizationcomputer emulatorGaussian processeswind farm optimization |
| spellingShingle | Mike Diessner Kevin J. Wilson Richard D. Whalley On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions Data-Centric Engineering Bayesian optimization black-box optimization computer emulator Gaussian processes wind farm optimization |
| title | On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions |
| title_full | On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions |
| title_fullStr | On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions |
| title_full_unstemmed | On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions |
| title_short | On the development of a practical Bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions |
| title_sort | on the development of a practical bayesian optimization algorithm for expensive experiments and simulations with changing environmental conditions |
| topic | Bayesian optimization black-box optimization computer emulator Gaussian processes wind farm optimization |
| url | https://www.cambridge.org/core/product/identifier/S2632673624000406/type/journal_article |
| work_keys_str_mv | AT mikediessner onthedevelopmentofapracticalbayesianoptimizationalgorithmforexpensiveexperimentsandsimulationswithchangingenvironmentalconditions AT kevinjwilson onthedevelopmentofapracticalbayesianoptimizationalgorithmforexpensiveexperimentsandsimulationswithchangingenvironmentalconditions AT richarddwhalley onthedevelopmentofapracticalbayesianoptimizationalgorithmforexpensiveexperimentsandsimulationswithchangingenvironmentalconditions |