Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network
As compared with the computational fluid dynamics(CFD), the airfoil optimization based on deep learning significantly reduces the computational cost. In the airfoil optimization based on deep learning, due to the uncertainty in the neural network, the optimization results deviate from the true value...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-09-01
|
| Series: | Frontiers in Bioengineering and Biotechnology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fbioe.2022.927064/full |
| _version_ | 1798002337851113472 |
|---|---|
| author | Xiaoyu Zhao Weiguo Wu Wei Chen Yongshui Lin Jiangcen Ke |
| author_facet | Xiaoyu Zhao Weiguo Wu Wei Chen Yongshui Lin Jiangcen Ke |
| author_sort | Xiaoyu Zhao |
| collection | DOAJ |
| description | As compared with the computational fluid dynamics(CFD), the airfoil optimization based on deep learning significantly reduces the computational cost. In the airfoil optimization based on deep learning, due to the uncertainty in the neural network, the optimization results deviate from the true value. In this work, a multi-network collaborative lift-to-drag ratio prediction model is constructed based on ResNet and penalty functions. Latin supersampling is used to select four angles of attack in the range of 2°–10° with significant uncertainty to limit the prediction error. Moreover, the random drift particle swarm optimization (RDPSO) algorithm is used to control the prediction error. The experimental results show that multi-network collaboration significantly reduces the error in the optimization results. As compared with the optimization based on a single network, the maximum error of multi-network coordination in single angle of attack optimization reduces by 16.0%. Consequently, this improves the reliability of airfoil optimization based on deep learning. |
| first_indexed | 2024-04-11T11:50:32Z |
| format | Article |
| id | doaj.art-0dcc8f7098e4469eb26c072cab943eb2 |
| institution | Directory Open Access Journal |
| issn | 2296-4185 |
| language | English |
| last_indexed | 2024-04-11T11:50:32Z |
| publishDate | 2022-09-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Bioengineering and Biotechnology |
| spelling | doaj.art-0dcc8f7098e4469eb26c072cab943eb22022-12-22T04:25:24ZengFrontiers Media S.A.Frontiers in Bioengineering and Biotechnology2296-41852022-09-011010.3389/fbioe.2022.927064927064Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial networkXiaoyu Zhao0Weiguo Wu1Wei Chen2Yongshui Lin3Jiangcen Ke4Hubei Key Laboratory of Theory and Application of Advanced Materials Mechanics, Department of Engineering Structure and Mechanics, Wuhan University of Technology, Wuhan, ChinaGreen and Smart River-Sea-Going Ship, Cruise and Yacht Research Center, Wuhan, ChinaHubei Key Laboratory of Theory and Application of Advanced Materials Mechanics, Department of Engineering Structure and Mechanics, Wuhan University of Technology, Wuhan, ChinaHubei Key Laboratory of Theory and Application of Advanced Materials Mechanics, Department of Engineering Structure and Mechanics, Wuhan University of Technology, Wuhan, ChinaSchool of Transportation and Logistics Engineering, Wuhan University of Technology, Wuhan, ChinaAs compared with the computational fluid dynamics(CFD), the airfoil optimization based on deep learning significantly reduces the computational cost. In the airfoil optimization based on deep learning, due to the uncertainty in the neural network, the optimization results deviate from the true value. In this work, a multi-network collaborative lift-to-drag ratio prediction model is constructed based on ResNet and penalty functions. Latin supersampling is used to select four angles of attack in the range of 2°–10° with significant uncertainty to limit the prediction error. Moreover, the random drift particle swarm optimization (RDPSO) algorithm is used to control the prediction error. The experimental results show that multi-network collaboration significantly reduces the error in the optimization results. As compared with the optimization based on a single network, the maximum error of multi-network coordination in single angle of attack optimization reduces by 16.0%. Consequently, this improves the reliability of airfoil optimization based on deep learning.https://www.frontiersin.org/articles/10.3389/fbioe.2022.927064/fulldeep learningairfoilparticle swam optimisationrandomparameter optimizatioin |
| spellingShingle | Xiaoyu Zhao Weiguo Wu Wei Chen Yongshui Lin Jiangcen Ke Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network Frontiers in Bioengineering and Biotechnology deep learning airfoil particle swam optimisation random parameter optimizatioin |
| title | Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network |
| title_full | Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network |
| title_fullStr | Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network |
| title_full_unstemmed | Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network |
| title_short | Multi-network collaborative lift-drag ratio prediction and airfoil optimization based on residual network and generative adversarial network |
| title_sort | multi network collaborative lift drag ratio prediction and airfoil optimization based on residual network and generative adversarial network |
| topic | deep learning airfoil particle swam optimisation random parameter optimizatioin |
| url | https://www.frontiersin.org/articles/10.3389/fbioe.2022.927064/full |
| work_keys_str_mv | AT xiaoyuzhao multinetworkcollaborativeliftdragratiopredictionandairfoiloptimizationbasedonresidualnetworkandgenerativeadversarialnetwork AT weiguowu multinetworkcollaborativeliftdragratiopredictionandairfoiloptimizationbasedonresidualnetworkandgenerativeadversarialnetwork AT weichen multinetworkcollaborativeliftdragratiopredictionandairfoiloptimizationbasedonresidualnetworkandgenerativeadversarialnetwork AT yongshuilin multinetworkcollaborativeliftdragratiopredictionandairfoiloptimizationbasedonresidualnetworkandgenerativeadversarialnetwork AT jiangcenke multinetworkcollaborativeliftdragratiopredictionandairfoiloptimizationbasedonresidualnetworkandgenerativeadversarialnetwork |