Pub Date : 2024-11-12DOI: 10.1016/j.jweia.2024.105954
Shizeng Liu , Wentong Zhang , Qiang Li , Shicheng Yan , Shihong Zhang , Chao Li , Lixiao Li
For transmission tower-line (TL) systems, the coupling effect between line cables and towers under strong winds is significant. This paper presents a method to quantify the coupling effect. Assuming that effective separation of line cables and towers is attainable, this work transforms the coupling effect into the transferred load from the line cable to the target tower, the coupling participation mass of the line cable, and the additional damping. The effective separation conditions are defined through an optimization method minimizing the residual errors of the wind-induced response and dynamic characteristics between the separated bodies and the TL system. A typical TL system is considered and analyzed for its structural dynamic characteristics and wind-induced response. Particularly, the quantities associated with the coupling effect of the TL system are estimated. It reveals that the transferred dynamic load component parallel to the line cable which is overlooked in current codes is significant and highly sensitive to the separation boundary conditions of line cables. Furthermore, the coupling participation mass of the conductor is more prominent than that of the ground wire. The proposed method is feasible for quantifying the TL coupling effect and incorporating it into the wind-induced response analysis of transmission line structures.
{"title":"Engineering method for quantifying the coupling effect of transmission tower-line system under strong winds","authors":"Shizeng Liu , Wentong Zhang , Qiang Li , Shicheng Yan , Shihong Zhang , Chao Li , Lixiao Li","doi":"10.1016/j.jweia.2024.105954","DOIUrl":"10.1016/j.jweia.2024.105954","url":null,"abstract":"<div><div>For transmission tower-line (TL) systems, the coupling effect between line cables and towers under strong winds is significant. This paper presents a method to quantify the coupling effect. Assuming that effective separation of line cables and towers is attainable, this work transforms the coupling effect into the transferred load from the line cable to the target tower, the coupling participation mass of the line cable, and the additional damping. The effective separation conditions are defined through an optimization method minimizing the residual errors of the wind-induced response and dynamic characteristics between the separated bodies and the TL system. A typical TL system is considered and analyzed for its structural dynamic characteristics and wind-induced response. Particularly, the quantities associated with the coupling effect of the TL system are estimated. It reveals that the transferred dynamic load component parallel to the line cable which is overlooked in current codes is significant and highly sensitive to the separation boundary conditions of line cables. Furthermore, the coupling participation mass of the conductor is more prominent than that of the ground wire. The proposed method is feasible for quantifying the TL coupling effect and incorporating it into the wind-induced response analysis of transmission line structures.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105954"},"PeriodicalIF":4.2,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142660006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A precise command of railway operations according to the measured instantaneous wind speed on an anemometer tower along a railway line is the development trend, whose challenges lie in the unknown transfer relation(s) between wind speed fluctuations on a moving train and an anemometer tower, i.e. the turbulence correlation between them. To address this issue, in the current work, the cross-correlation functions of wind speed fluctuations at the moving train and anemometer tower are derived, and an empirical formula of the coherence functions is obtained. The turbulence correlation is inversely related to the separation distance from the anemometer tower to the line, and there is little correlation when this distance is longer than double the longitudinal turbulence length scale. Field measurements of wind characteristics were carried out on an anemometer tower and a moving vehicle, and the turbulence correlation and its expression were validated. Three methods are proposed and compared to evaluate the instantaneous wind speed at the anemometer tower and moving train with this correlation. The methods based on the cross-spectral density and coherence function can accurately simulate the correlation, and the latter performance is slightly better (its simulation of the frequency domain correlation is 52.9% better than the former); the method based on solely independent and identically distributed random phases cannot fully simulate the correlation. From this, the effects of the correlation on train operations are studied and analysed in detail. Our analysis shows that neglecting this correlation leads to conservative estimates: wind speed differences between the anemometer tower and the moving train are at least 18.1% greater, and the safety and economic assessments of train operations in crosswinds are underestimated by at least 32.0%. Considering the correlation can reduce the (excess) safety risk/margin and is an inevitable development of adapting to the detailed assessment of the crosswind stability of vehicles. The quantitative description and simulation of the correlation presented in this work point to the critical importance of wind speed monitoring systems for the detailed crosswind assessment, and provide a theoretical basis for further research work on the crosswind stability of vehicles under true/realistic turbulent flow wind.
{"title":"Turbulence correlation between moving trains and anemometer towers: Theoretical analysis, field measurements and simulation","authors":"Hongrui Gao , Tanghong Liu , Xiaodong Chen , Haoyang Zeng , Jiyun Jiang , Xinran Wang , Boo Cheong Khoo","doi":"10.1016/j.jweia.2024.105949","DOIUrl":"10.1016/j.jweia.2024.105949","url":null,"abstract":"<div><div>A precise command of railway operations according to the measured instantaneous wind speed on an anemometer tower along a railway line is the development trend, whose challenges lie in the unknown transfer relation(s) between wind speed fluctuations on a moving train and an anemometer tower, i.e. the turbulence correlation between them. To address this issue, in the current work, the cross-correlation functions of wind speed fluctuations at the moving train and anemometer tower are derived, and an empirical formula of the coherence functions is obtained. The turbulence correlation is inversely related to the separation distance from the anemometer tower to the line, and there is little correlation when this distance is longer than double the longitudinal turbulence length scale. Field measurements of wind characteristics were carried out on an anemometer tower and a moving vehicle, and the turbulence correlation and its expression were validated. Three methods are proposed and compared to evaluate the instantaneous wind speed at the anemometer tower and moving train with this correlation. The methods based on the cross-spectral density and coherence function can accurately simulate the correlation, and the latter performance is slightly better (its simulation of the frequency domain correlation is 52.9% better than the former); the method based on solely independent and identically distributed random phases cannot fully simulate the correlation. From this, the effects of the correlation on train operations are studied and analysed in detail. Our analysis shows that neglecting this correlation leads to conservative estimates: wind speed differences between the anemometer tower and the moving train are at least 18.1% greater, and the safety and economic assessments of train operations in crosswinds are underestimated by at least 32.0%. Considering the correlation can reduce the (excess) safety risk/margin and is an inevitable development of adapting to the detailed assessment of the crosswind stability of vehicles. The quantitative description and simulation of the correlation presented in this work point to the critical importance of wind speed monitoring systems for the detailed crosswind assessment, and provide a theoretical basis for further research work on the crosswind stability of vehicles under true/realistic turbulent flow wind.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105949"},"PeriodicalIF":4.2,"publicationDate":"2024-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142660548","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-09DOI: 10.1016/j.jweia.2024.105933
Agostino Cembalo , Jacques Borée , Patrick Coirault , Clément Dumand
Under disturbed upstream conditions, numerous wind tunnel studies have shown that the near wake region of a vehicle loses its average symmetry, resulting in an increase of drag. The aim of this research work is to analyze the large scale response of a vehicle wake to on-road perturbations by using an instrumented vehicle and by comparing scale one wind tunnel tests, track trials and on road experiments. More precisely, in all these tests, we focus on the analysis of the asymmetry of the pressure distribution at the base. Proper Orthogonal Decomposition (POD) is used. For all cases considered, POD analysis reveals two dominant modes, respectively associated with vertical and horizontal wake large scale reorganization. More than 50% of the total energy is carried by these two modes and this value increases significantly for on-road tests. Noteworthy, the low-frequency energy content of the temporal coefficients of these modes is significantly higher on-road. Low frequencies (even very low ones) then play a major role, corresponding to a quasi-static perturbation domain of the velocity seen by the vehicle. We show that a quasi-steady exploration of the on-road yaw angle statistical distribution during a wind tunnel test captures phenomena similar to those observed on the road and is therefore interesting to evaluate on-road aerodynamic performances. This also opens perspectives for developing closed loop control strategies aiming to maintain a prescribed wake balance in order to reduce drag experienced on the road.
{"title":"Large scale response of a vehicle wake to on-road perturbations","authors":"Agostino Cembalo , Jacques Borée , Patrick Coirault , Clément Dumand","doi":"10.1016/j.jweia.2024.105933","DOIUrl":"10.1016/j.jweia.2024.105933","url":null,"abstract":"<div><div>Under disturbed upstream conditions, numerous wind tunnel studies have shown that the near wake region of a vehicle loses its average symmetry, resulting in an increase of drag. The aim of this research work is to analyze the large scale response of a vehicle wake to on-road perturbations by using an instrumented vehicle and by comparing scale one wind tunnel tests, track trials and on road experiments. More precisely, in all these tests, we focus on the analysis of the asymmetry of the pressure distribution at the base. Proper Orthogonal Decomposition (POD) is used. For all cases considered, POD analysis reveals two dominant modes, respectively associated with vertical and horizontal wake large scale reorganization. More than 50% of the total energy is carried by these two modes and this value increases significantly for on-road tests. Noteworthy, the low-frequency energy content of the temporal coefficients of these modes is significantly higher on-road. Low frequencies (even very low ones) then play a major role, corresponding to a quasi-static perturbation domain of the velocity seen by the vehicle. We show that a quasi-steady exploration of the on-road yaw angle statistical distribution during a wind tunnel test captures phenomena similar to those observed on the road and is therefore interesting to evaluate on-road aerodynamic performances. This also opens perspectives for developing closed loop control strategies aiming to maintain a prescribed wake balance in order to reduce drag experienced on the road.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105933"},"PeriodicalIF":4.2,"publicationDate":"2024-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142660008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-09DOI: 10.1016/j.jweia.2024.105937
Simian Lei , Luca Patruno , Claudio Mannini , Stefano de Miranda , Yaojun Ge
Motion-induced aerodynamic forces play a fundamental role in the stability and buffeting analysis of long-span bridges, which are traditionally performed in the frequency domain adopting the well-known approach based on flutter derivatives and aerodynamic admittance functions. However, the increase in span of newly designed bridges currently raises concerns regarding the role of nonlinear aerodynamic effects, the response to non-stationary winds and the aerodynamic coupling in multi-modal vibrations. Addressing these issues requires to calculate aerodynamic forces induced by arbitrary motions and, possibly, consider large variations in the incoming flow orientation, a task better suited for time-domain approaches. In this study, we introduce a time-domain state-space model formulation for motion-induced aerodynamic forces, which systematizes and generalizes previous models, while keeping a simple structure and ease of calibration. We tailor the model formulation to allow for a clear distinction between quasi-static and purely transient aerodynamic contributions and investigate the relations between the proposed model and other available models, highlighting their common underlying framework. The model is finally calibrated for a selection of bridge decks, showing a very good ability to reproduce motion-induced aerodynamic forces.
{"title":"Time-domain state-space model formulation of motion-induced aerodynamic forces on bridge decks","authors":"Simian Lei , Luca Patruno , Claudio Mannini , Stefano de Miranda , Yaojun Ge","doi":"10.1016/j.jweia.2024.105937","DOIUrl":"10.1016/j.jweia.2024.105937","url":null,"abstract":"<div><div>Motion-induced aerodynamic forces play a fundamental role in the stability and buffeting analysis of long-span bridges, which are traditionally performed in the frequency domain adopting the well-known approach based on flutter derivatives and aerodynamic admittance functions. However, the increase in span of newly designed bridges currently raises concerns regarding the role of nonlinear aerodynamic effects, the response to non-stationary winds and the aerodynamic coupling in multi-modal vibrations. Addressing these issues requires to calculate aerodynamic forces induced by arbitrary motions and, possibly, consider large variations in the incoming flow orientation, a task better suited for time-domain approaches. In this study, we introduce a time-domain state-space model formulation for motion-induced aerodynamic forces, which systematizes and generalizes previous models, while keeping a simple structure and ease of calibration. We tailor the model formulation to allow for a clear distinction between quasi-static and purely transient aerodynamic contributions and investigate the relations between the proposed model and other available models, highlighting their common underlying framework. The model is finally calibrated for a selection of bridge decks, showing a very good ability to reproduce motion-induced aerodynamic forces.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105937"},"PeriodicalIF":4.2,"publicationDate":"2024-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142660005","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-08DOI: 10.1016/j.jweia.2024.105946
Congjie Shang , Huoyue Xiang , Yulong Bao , Yongle Li , Kou Luo
The long-span bridge has a flexible structure and low damping. And the truss girder section is relatively blunt, which is prone to vortex-induced vibration(VIV). Initially, the wind tunnel tests are carried out at two scales to test the two lock-in regions of VIV for the truss girder. Then, the fluid-structure coupling analysis numerical model of the simplified two-dimensional(2-D) section of the truss girder is established, and the analysis results are compared with the experiment. Finally, the time-frequency characteristics of the aerodynamic lift and the wind speed at the movable monitoring points, further the meso-mechanism of the vortex evolution in two lock-in regions are analyzed. The results indicate that there are both twice key changes in the lift frequency components during the whole process of VIV in the two lock-in regions with the increase in amplitude, which are related to the change in flow pattern. VIV in the first lock-in region is self-excited by the coupling of the wake vortex of the bridge deck and the motion of the model. VIV in the second lock-in of VIV is self-excited by the impinging shear layer instability at the leading edge of the bridge deck and the motion of the model.
{"title":"Mechanism of vortex-induced vibration in two lock-in regions for truss girder sections","authors":"Congjie Shang , Huoyue Xiang , Yulong Bao , Yongle Li , Kou Luo","doi":"10.1016/j.jweia.2024.105946","DOIUrl":"10.1016/j.jweia.2024.105946","url":null,"abstract":"<div><div>The long-span bridge has a flexible structure and low damping. And the truss girder section is relatively blunt, which is prone to vortex-induced vibration(VIV). Initially, the wind tunnel tests are carried out at two scales to test the two lock-in regions of VIV for the truss girder. Then, the fluid-structure coupling analysis numerical model of the simplified two-dimensional(2-D) section of the truss girder is established, and the analysis results are compared with the experiment. Finally, the time-frequency characteristics of the aerodynamic lift and the wind speed at the movable monitoring points, further the meso-mechanism of the vortex evolution in two lock-in regions are analyzed. The results indicate that there are both twice key changes in the lift frequency components during the whole process of VIV in the two lock-in regions with the increase in amplitude, which are related to the change in flow pattern. VIV in the first lock-in region is self-excited by the coupling of the wake vortex of the bridge deck and the motion of the model. VIV in the second lock-in of VIV is self-excited by the impinging shear layer instability at the leading edge of the bridge deck and the motion of the model.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105946"},"PeriodicalIF":4.2,"publicationDate":"2024-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142660456","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-11-07DOI: 10.1016/j.jweia.2024.105932
Juan A. Cárdenas-Rondón , Omar Gómez-Ortega , Carlos Rodríguez-Casado , Mikel Ogueta-Gutiérrez , Sebastián Franchini
Based on the experimental data of the aerodynamic derivative presented in the literature, evidence was found that supports the existence of a similarity solution between , the effective mean angle of attack, , the tracker height-to-width ratio, , and the reduced velocity. . With this similarity solution, it is possible to estimate for any and with a significantly reduced amount of experimental data. This represents a notable advancement compared to the current state of the art, as it could allow for a more detailed analysis of aeroelastic instability in flat solar trackers with fewer experimental requirements. This article presents the developed formulation and the process followed to obtain the discovered similarity solution. Relying on the similarity solution, a simplified model of has been proposed as a function of the reduced velocity and for .
{"title":"Similarity solution for sectional A2∗ aerodynamic derivative for single axis solar trackers at various angles of attack and ground distances","authors":"Juan A. Cárdenas-Rondón , Omar Gómez-Ortega , Carlos Rodríguez-Casado , Mikel Ogueta-Gutiérrez , Sebastián Franchini","doi":"10.1016/j.jweia.2024.105932","DOIUrl":"10.1016/j.jweia.2024.105932","url":null,"abstract":"<div><div>Based on the experimental data of the aerodynamic derivative <span><math><msubsup><mrow><mi>A</mi></mrow><mrow><mn>2</mn></mrow><mrow><mo>∗</mo></mrow></msubsup></math></span> presented in the literature, evidence was found that supports the existence of a similarity solution between <span><math><msubsup><mrow><mi>A</mi></mrow><mrow><mn>2</mn></mrow><mrow><mo>∗</mo></mrow></msubsup></math></span>, the effective mean angle of attack, <span><math><msubsup><mrow><mi>α</mi></mrow><mrow><mi>m</mi><mi>e</mi><mi>a</mi><mi>n</mi></mrow><mrow><mi>e</mi><mi>f</mi><mi>f</mi></mrow></msubsup></math></span>, the tracker height-to-width ratio, <span><math><mrow><mi>H</mi><mo>/</mo><mi>B</mi></mrow></math></span>, and the reduced velocity. <span><math><msub><mrow><mi>U</mi></mrow><mrow><mi>r</mi></mrow></msub></math></span>. With this similarity solution, it is possible to estimate <span><math><msubsup><mrow><mi>A</mi></mrow><mrow><mn>2</mn></mrow><mrow><mo>∗</mo></mrow></msubsup></math></span> for any <span><math><mrow><msubsup><mrow><mi>α</mi></mrow><mrow><mi>m</mi><mi>e</mi><mi>a</mi><mi>n</mi></mrow><mrow><mi>e</mi><mi>f</mi><mi>f</mi></mrow></msubsup><mo>∈</mo><mfenced><mrow><mo>−</mo><mn>40</mn><mo>°</mo><mo>,</mo><mo>+</mo><mn>40</mn><mo>°</mo></mrow></mfenced></mrow></math></span> and <span><math><mrow><mi>H</mi><mo>/</mo><mi>B</mi><mo>∈</mo><mfenced><mrow><mn>0</mn><mo>.</mo><mn>3</mn><mo>,</mo><mn>2</mn><mo>.</mo><mn>0</mn></mrow></mfenced></mrow></math></span> with a significantly reduced amount of experimental data. This represents a notable advancement compared to the current state of the art, as it could allow for a more detailed analysis of aeroelastic instability in flat solar trackers with fewer experimental requirements. This article presents the developed formulation and the process followed to obtain the discovered similarity solution. Relying on the similarity solution, a simplified model of <span><math><msubsup><mrow><mi>A</mi></mrow><mrow><mn>2</mn></mrow><mrow><mo>∗</mo></mrow></msubsup></math></span> has been proposed as a function of the reduced velocity and <span><math><mrow><mi>H</mi><mo>/</mo><mi>B</mi></mrow></math></span> for <span><math><mrow><msubsup><mrow><mi>α</mi></mrow><mrow><mi>m</mi><mi>e</mi><mi>a</mi><mi>n</mi></mrow><mrow><mi>e</mi><mi>f</mi><mi>f</mi></mrow></msubsup><mo>=</mo><mn>0</mn><mo>°</mo></mrow></math></span>.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105932"},"PeriodicalIF":4.2,"publicationDate":"2024-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142660458","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Multiple hazards caused by tropical cyclones (TCs), such as heavy rains and strong winds, result in substantial property losses and casualties worldwide each year. TC wind field models, describing the development of the wind hazard, are key within early warning realizations and associated risk assessments. Different to conventional parametric, analytical or meteorological numerical models, this study aims to develop a machine-learning-based approach for modeling TC wind fields by incorporating multiple meteorological parameters. The wind field model considers linear and nonlinear modeling respectively, where the input data includes various meteorological parameters such as surface pressure gradient (SPG), geopotential (GEO), boundary layer height (BLH), and forecast surface roughness (FSR). The output data is the TC wind field data of the Regional and Mesoscale Meteorology Branch (RAMMB) extracted by image recognition method, and assimilated with the wind field from the fifth generation of the European Center for Medium-Range Weather Forecasts (ECMWF) atmospheric reanalysis dataset ERA5. In the linear model, various combinations of parameters are considered, yet always yielding unsatisfactory results. The best results in the linear model were obtained using all four parameter combinations, where the root mean square error (RMSE) was 2.60 m/s and the coefficient of determination value was 0.44. To increase performance, three nonlinear machine learning methods—Fully Connected Deep Neural Networks (FC-DNN), Convolutional Neural Networks (CNN), and Transformer—are introduced to the training process. Comparing the wind field continuity, RMSE and between the three models, it is found that the Transformer outperforms all other models, with value of 0.877 and an RMSE of 2.23. As a final step, the trained Transformer model was used to predict the evolution of wind speed of the Typhoon Lekima (1909), in what could serve as effective model validation.
{"title":"Machine-learning-based tropical cyclone wind field model incorporating multiple meteorological parameters","authors":"Miaomiao Wei , Genshen Fang , Nikolaos Nikitas , Yaojun Ge","doi":"10.1016/j.jweia.2024.105936","DOIUrl":"10.1016/j.jweia.2024.105936","url":null,"abstract":"<div><div>Multiple hazards caused by tropical cyclones (TCs), such as heavy rains and strong winds, result in substantial property losses and casualties worldwide each year. TC wind field models, describing the development of the wind hazard, are key within early warning realizations and associated risk assessments. Different to conventional parametric, analytical or meteorological numerical models, this study aims to develop a machine-learning-based approach for modeling TC wind fields by incorporating multiple meteorological parameters. The wind field model considers linear and nonlinear modeling respectively, where the input data includes various meteorological parameters such as surface pressure gradient (SPG), geopotential (GEO), boundary layer height (BLH), and forecast surface roughness (FSR). The output data is the TC wind field data of the Regional and Mesoscale Meteorology Branch (RAMMB) extracted by image recognition method, and assimilated with the wind field from the fifth generation of the European Center for Medium-Range Weather Forecasts (ECMWF) atmospheric reanalysis dataset ERA5. In the linear model, various combinations of parameters are considered, yet always yielding unsatisfactory results. The best results in the linear model were obtained using all four parameter combinations, where the root mean square error (RMSE) was 2.60 m/s and the coefficient of determination <span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> value was 0.44. To increase performance, three nonlinear machine learning methods—Fully Connected Deep Neural Networks (FC-DNN), Convolutional Neural Networks (CNN), and Transformer—are introduced to the training process. Comparing the wind field continuity, RMSE and <span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> between the three models, it is found that the Transformer outperforms all other models, with <span><math><mrow><msup><mi>R</mi><mn>2</mn></msup></mrow></math></span> value of 0.877 and an RMSE of 2.23. As a final step, the trained Transformer model was used to predict the evolution of wind speed of the Typhoon Lekima (1909), in what could serve as effective model validation.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"255 ","pages":"Article 105936"},"PeriodicalIF":4.2,"publicationDate":"2024-11-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142578652","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-29DOI: 10.1016/j.jweia.2024.105935
Binbin Yao , Zhisong Wang , Zhiyuan Fang , Zhengliang Li
Downbursts, as a strong localized wind event, have caused significant damage to engineering structures throughout the world. However, given the spatial and temporal randomness of such strong winds, on-site measurements are often difficult to obtain a sufficient amount of valid wind field information in a short period of time. To refine the resolution of the wind field, this study proposes a physics-informed neural network network-based (PINN-based) approach to reconstruct the downburst from limited observed data. The Navier-Stokes (N-S) equations are embedded into the fully connected neural network as a physical constraint to construct the PINN. The PINN model is then validated by the reconstruction of numerical downburst generated by large eddy simulations. The reconstruction of the sparse downburst wind field by PINN performs well in both interpolation and extrapolation prediction. The optimal construction of the PINN has been evaluated through parameter analysis of the influence of training data and network parameters. Finally, the optimal PINN construction is used to reconstruct the wind field of the experimental data with a relative error of 5% for the horizontal wind velocity.
{"title":"Reconstruction of downburst wind fields using physics-informed neural network","authors":"Binbin Yao , Zhisong Wang , Zhiyuan Fang , Zhengliang Li","doi":"10.1016/j.jweia.2024.105935","DOIUrl":"10.1016/j.jweia.2024.105935","url":null,"abstract":"<div><div>Downbursts, as a strong localized wind event, have caused significant damage to engineering structures throughout the world. However, given the spatial and temporal randomness of such strong winds, on-site measurements are often difficult to obtain a sufficient amount of valid wind field information in a short period of time. To refine the resolution of the wind field, this study proposes a physics-informed neural network network-based (PINN-based) approach to reconstruct the downburst from limited observed data. The Navier-Stokes (N-S) equations are embedded into the fully connected neural network as a physical constraint to construct the PINN. The PINN model is then validated by the reconstruction of numerical downburst generated by large eddy simulations. The reconstruction of the sparse downburst wind field by PINN performs well in both interpolation and extrapolation prediction. The optimal construction of the PINN has been evaluated through parameter analysis of the influence of training data and network parameters. Finally, the optimal PINN construction is used to reconstruct the wind field of the experimental data with a relative error of 5% for the horizontal wind velocity.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"254 ","pages":"Article 105935"},"PeriodicalIF":4.2,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142538414","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The univariate design method may not match the wind resistance demands of bridges in mountainous areas. Therefore, it is crucial to comprehensively consider the joint effect of multiple wind parameters for determining wind-resistant design parameters of bridges. To address challenges such as short measurement periods and difficulties in expanding the extreme value model of wind parameters, a Bootstrap resampling strategy incorporating seasonal wind speed trends was developed, verified, and applied to long-term probabilistic modeling; thus, the uncertainty of the probability model of average wind parameters was investigated. Then, taking the environmental contour of wind speed and attack angle under varying wind directions as the basis, a technical framework for wind-resistant bridges based on multi-parameter joint design is proposed. Meanwhile, the main girder's longitudinal and lateral design wind speeds are derived under the joint influence of attack angle and yaw angle. The results show that the control wind direction of longitudinal and lateral design wind speed is different. The joint design considering multiple wind parameters effectively makes up the limitations of traditional methods. It provides valuable insights for wind-resistant design and lifecycle toughness evaluation of bridges in mountainous areas.
{"title":"Study on joint design method of multiple wind parameters for long-span bridges in deep-cutting gorge areas based on field measurement","authors":"Jinxiang Zhang , Fanying Jiang , Mingjin Zhang , Haoxiang Zheng , Yongle Li , Junsong Liang","doi":"10.1016/j.jweia.2024.105930","DOIUrl":"10.1016/j.jweia.2024.105930","url":null,"abstract":"<div><div>The univariate design method may not match the wind resistance demands of bridges in mountainous areas. Therefore, it is crucial to comprehensively consider the joint effect of multiple wind parameters for determining wind-resistant design parameters of bridges. To address challenges such as short measurement periods and difficulties in expanding the extreme value model of wind parameters, a Bootstrap resampling strategy incorporating seasonal wind speed trends was developed, verified, and applied to long-term probabilistic modeling; thus, the uncertainty of the probability model of average wind parameters was investigated. Then, taking the environmental contour of wind speed and attack angle under varying wind directions as the basis, a technical framework for wind-resistant bridges based on multi-parameter joint design is proposed. Meanwhile, the main girder's longitudinal and lateral design wind speeds are derived under the joint influence of attack angle and yaw angle. The results show that the control wind direction of longitudinal and lateral design wind speed is different. The joint design considering multiple wind parameters effectively makes up the limitations of traditional methods. It provides valuable insights for wind-resistant design and lifecycle toughness evaluation of bridges in mountainous areas.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"254 ","pages":"Article 105930"},"PeriodicalIF":4.2,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142538413","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-10-29DOI: 10.1016/j.jweia.2024.105929
Lizhi Wen , Kazuyoshi Nishijima
Plate-type debris is a typical type of windborne debris, often originating from roof tiles and shingles. Numerical simulation using aerodynamic models provides a practical method to predict trajectories of windborne debris. In this paper, we first propose a revised model for the 3-degree-of-freedom (3-DOF) flight motion of square plates in winds by integrating experimental data from previous studies. Thereby, we divide the aerodynamic force and moment into a translational part and a rotational part. In addition, we propose conditions of autorotation in the revised model. The calculation of the rotational force and moment depends on whether these conditions are fulfilled. The revised model is validated by comparing the numerical results with experimental results of plate trajectories. Next, based on the revised model for the 3-DOF motion, we propose an aerodynamic model for the 6-DOF motion by incorporating the findings about the rotational force and moment, which were obtained from the authors’ previous study on the 6-DOF motion of square plates. Based on these findings, the model is developed in the way that the direction of the rotational force depends on the relative wind velocity and the angular velocity of plate, and the direction of the rotational moment depends on the translational moment. By doing so the proposed model in this paper avoids directly using a database of aerodynamics, which is large and difficult to obtain. Validation using the experimental results of plate trajectories shows that the proposed model, which has a relatively simple form, performs generally well.
{"title":"An aerodynamic model for 6-DOF flight motion of windborne debris of square plates","authors":"Lizhi Wen , Kazuyoshi Nishijima","doi":"10.1016/j.jweia.2024.105929","DOIUrl":"10.1016/j.jweia.2024.105929","url":null,"abstract":"<div><div>Plate-type debris is a typical type of windborne debris, often originating from roof tiles and shingles. Numerical simulation using aerodynamic models provides a practical method to predict trajectories of windborne debris. In this paper, we first propose a revised model for the 3-degree-of-freedom (3-DOF) flight motion of square plates in winds by integrating experimental data from previous studies. Thereby, we divide the aerodynamic force and moment into a translational part and a rotational part. In addition, we propose conditions of autorotation in the revised model. The calculation of the rotational force and moment depends on whether these conditions are fulfilled. The revised model is validated by comparing the numerical results with experimental results of plate trajectories. Next, based on the revised model for the 3-DOF motion, we propose an aerodynamic model for the 6-DOF motion by incorporating the findings about the rotational force and moment, which were obtained from the authors’ previous study on the 6-DOF motion of square plates. Based on these findings, the model is developed in the way that the direction of the rotational force depends on the relative wind velocity and the angular velocity of plate, and the direction of the rotational moment depends on the translational moment. By doing so the proposed model in this paper avoids directly using a database of aerodynamics, which is large and difficult to obtain. Validation using the experimental results of plate trajectories shows that the proposed model, which has a relatively simple form, performs generally well.</div></div>","PeriodicalId":54752,"journal":{"name":"Journal of Wind Engineering and Industrial Aerodynamics","volume":"254 ","pages":"Article 105929"},"PeriodicalIF":4.2,"publicationDate":"2024-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142538412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号