The high-speed boundary layer transition directly affects the refined design of the flight vehicle’s aerodynamic and thermal properties.However,current research in this field generally focuses on simplified geometrie...The high-speed boundary layer transition directly affects the refined design of the flight vehicle’s aerodynamic and thermal properties.However,current research in this field generally focuses on simplified geometries,such as cones or flat plates,resulting in a limited understanding of high-speed boundary layer transition in prac-tical vehicle configurations.This study aims to carry out an in-depth investigation of high-speed boundary layer transitions in a fin-cone configuration with a double-cone design.PCB sensors are used to characterize the evolution of instability waves along the streamwise direction on both the leeward and windward sides of the model.Additionally,focused laser differential interferometry(FLDI)is deployed to measure the instability characteristics along the wall-normal direction of the high-speed bound-ary layer.Furthermore,an infrared camera is employed to obtain surface temperatures.In addition to the second-mode instabilities observed on the leeward side,streamwise vortex-induced local peaks on the windward side are also observed with PCB sensors.Infrared images reveal that the thermal characteristics are different between the wind-ward and leeward sides due to transition mechanisms.The study reveals that at simi-lar Reynolds numbers,the area fully developed to turbulence on the windward side is considerably larger than that on the leeward side for the cone region,while the fin region displays different trends.These findings provide invaluable insights for the advancement of aerodynamic and thermal design in high-speed aircraft.展开更多
文摘[研究目的]解决多源数据融合过程中参与者贡献与收益的匹配问题、参与者选择缺乏灵活性以及联邦系统的动态适应性不足问题,提升多源数据融合的公平性和合理性。[研究方法]提出一种基于动态自适应联邦学习的多源数据融合框架(Federated Learning and Dynamic Improvement,FLDI),并设计预算分配机制、参与者选择机制以及参与者动态进出机制确保多源数据融合过程的安全、公平和可持续。分别在分类任务的专利、论文以及媒体数据集和预测任务的MNIST、FMNIST和CIFAR-10数据集上展开性能测试,并在不同场景中评估框架性能。[研究结果/结论]FLDI在面对复杂场景时,其准确率相较于FedAvg和FedProx提升了3%~4%;在干净数据集场景下,FLDI在分类任务的平均准确率达到67.01%,在预测任务的平均准确率达到81.56%;进行增强实验后,FLDI在分类任务的平均准确率上升了4.54%,在预测任务的平均准确率上升了3.31%;FLDI框架在分类任务和预测任务中较之FedAvg和FedProx更具性能优势。
基金supported by an Experimental Identification Technology Research Project(No.2022G02000250007).
文摘The high-speed boundary layer transition directly affects the refined design of the flight vehicle’s aerodynamic and thermal properties.However,current research in this field generally focuses on simplified geometries,such as cones or flat plates,resulting in a limited understanding of high-speed boundary layer transition in prac-tical vehicle configurations.This study aims to carry out an in-depth investigation of high-speed boundary layer transitions in a fin-cone configuration with a double-cone design.PCB sensors are used to characterize the evolution of instability waves along the streamwise direction on both the leeward and windward sides of the model.Additionally,focused laser differential interferometry(FLDI)is deployed to measure the instability characteristics along the wall-normal direction of the high-speed bound-ary layer.Furthermore,an infrared camera is employed to obtain surface temperatures.In addition to the second-mode instabilities observed on the leeward side,streamwise vortex-induced local peaks on the windward side are also observed with PCB sensors.Infrared images reveal that the thermal characteristics are different between the wind-ward and leeward sides due to transition mechanisms.The study reveals that at simi-lar Reynolds numbers,the area fully developed to turbulence on the windward side is considerably larger than that on the leeward side for the cone region,while the fin region displays different trends.These findings provide invaluable insights for the advancement of aerodynamic and thermal design in high-speed aircraft.
