目的桨叶运动参数是直升机设计到生产的重要指标,传统的视觉测量方法直接应用于室外环境下,由于受复杂光照背景影响,存在找不到桨叶区域、不能进行准确测量的问题。据此,本文提出一种融合多特征与自注意力的旋转目标检测器(fusion multi...目的桨叶运动参数是直升机设计到生产的重要指标,传统的视觉测量方法直接应用于室外环境下,由于受复杂光照背景影响,存在找不到桨叶区域、不能进行准确测量的问题。据此,本文提出一种融合多特征与自注意力的旋转目标检测器(fusion multi-feature and self-attention rotating detector,FMSA-RD)。方法首先,针对YOLOv5s(you only look once)特征提取能力不足和冗余问题,在主干网络中设计了更为有效的多特征提取和融合模块,结合不同时刻位置与尺度下的特征信息以提高网络对室外桨叶的检测精度;并去掉部分无关卷积层以简化模块结构参数。其次,融合多头自注意力机制与CSP(crossstage partial convolution)瓶颈结构,整合全局信息以抑制室外复杂光照背景干扰。最后,引入倾斜交并比(skew intersection over union,SKEWIOU)损失和角度损失,改进损失函数,进一步提升桨叶检测精度。结果本文进行了多组对比实验,分别在自制的室外直升机桨叶数据集和公共数据集DOTA-v1.0(dataset for object detection in aerial images)上进行验证,对比基线YOLOv5s目标检测网络,本文模型平均精度均值(mean average precision,mAP)分别提高6.6%和12.8%,帧速率(frames per second,FPS)分别提高21.8%和47.7%。结论本文设计的旋转目标检测模型,提升了室外复杂光照背景下桨叶的检测精度和速度。展开更多
对多旋翼电动垂直起降(electric vertical takeoff and landing,eVTOL)航空器推进系统进行了可靠性分析与分配。首先,针对多旋翼eVTOL航空器可靠性历史数据积累不足的问题,使用模糊贝叶斯网络(fuzzy Bayesian network,FBN)建立可靠性分...对多旋翼电动垂直起降(electric vertical takeoff and landing,eVTOL)航空器推进系统进行了可靠性分析与分配。首先,针对多旋翼eVTOL航空器可靠性历史数据积累不足的问题,使用模糊贝叶斯网络(fuzzy Bayesian network,FBN)建立可靠性分析模型,对其可靠性先验数据进行了补充,并进行可靠性后验推理,辅助定位推进系统关键环节。其次,基于FBN可靠性分析模型,提出一种改进电子设备可靠性咨询组(advisory group on reliability of electronic equipment,AGREE)可靠性分配方法,对不同构型多旋翼eVTOL推进系统进行可靠性分配。结果表明,FBN可靠性分析模型补充了推进系统可靠性数据,可有效识别系统薄弱环节。改进AGREE分配法的可靠性分配结果符合SC-VTOL-01中对eVTOL航空器的可靠性要求,同时该方法得到的可靠性分配结果更为合理,体现了不同构型、子系统、部件间的差异。展开更多
Accurate measurement of helicopter rotor motion parameters(flap,lead-lag,torsion,and azimuth angles)is essential for rotor blade design,helicopter dynamics modeling,and flight safety and health monitoring.However,the ...Accurate measurement of helicopter rotor motion parameters(flap,lead-lag,torsion,and azimuth angles)is essential for rotor blade design,helicopter dynamics modeling,and flight safety and health monitoring.However,the existing methods face challenges in testing equipment installation,calibration,and data transmission,resulting in limited reports on real-time in-flight measurements of blade motion parameters.This paper proposes a non-contact optoelectronic method based on two-dimensional position-sensitive detectors for in-flight measurement and a ground calibration system to obtain real-time rotor motion parameters during helicopter flight.The proposed method establishes the time evolution relationship of rotor motion parameters and verifies the performance of the in-flight measurement system regarding measurement resolution and accuracy through the construction of a blade motion posture experimental platform.The proposed method has been applied to the flight measurement of a medium-sized single-rotor helicopter,and the obtained results have been compared with theoretical analysis outcomes.Furthermore,this paper examines the characteristics of blade motion parameters during flight and discusses the challenges and potential solutions for measuring rotor motion parameters during helicopter flight using the proposed method.展开更多
To reduce the vibration of the Coaxial Helicopter Main Transmission System(CHMTS)considering both level and vertical flight conditions,a vibration evaluation and optimization model for the CHMTS was built.The vibratio...To reduce the vibration of the Coaxial Helicopter Main Transmission System(CHMTS)considering both level and vertical flight conditions,a vibration evaluation and optimization model for the CHMTS was built.The vibration simulation model of the CHMTS was set up by gear dynamics theory and loaded contact analysis.For better evaluation of the system vibration,a vibration evaluation method for the CHMTS was established by the G1 method-variation coefficient method.A hybrid Gravitational Search Algorithm-Simulated Annealing(GSA-SA)algorithm was combined to balance convergence speed and searching accuracy.The principle test was conducted to prove the accuracy of theoretical method,in which the maximum relative error is16.26%.The optional results show that the vibration of the optimized transmission system decreases significantly,in which the maximum reduction of key vibration indicators reaches more than 20%.The theoretical results have been compared to the experiment to verify the effectiveness of the vibration optimization method.The proposed method could be extended to other fields.展开更多
文摘目的桨叶运动参数是直升机设计到生产的重要指标,传统的视觉测量方法直接应用于室外环境下,由于受复杂光照背景影响,存在找不到桨叶区域、不能进行准确测量的问题。据此,本文提出一种融合多特征与自注意力的旋转目标检测器(fusion multi-feature and self-attention rotating detector,FMSA-RD)。方法首先,针对YOLOv5s(you only look once)特征提取能力不足和冗余问题,在主干网络中设计了更为有效的多特征提取和融合模块,结合不同时刻位置与尺度下的特征信息以提高网络对室外桨叶的检测精度;并去掉部分无关卷积层以简化模块结构参数。其次,融合多头自注意力机制与CSP(crossstage partial convolution)瓶颈结构,整合全局信息以抑制室外复杂光照背景干扰。最后,引入倾斜交并比(skew intersection over union,SKEWIOU)损失和角度损失,改进损失函数,进一步提升桨叶检测精度。结果本文进行了多组对比实验,分别在自制的室外直升机桨叶数据集和公共数据集DOTA-v1.0(dataset for object detection in aerial images)上进行验证,对比基线YOLOv5s目标检测网络,本文模型平均精度均值(mean average precision,mAP)分别提高6.6%和12.8%,帧速率(frames per second,FPS)分别提高21.8%和47.7%。结论本文设计的旋转目标检测模型,提升了室外复杂光照背景下桨叶的检测精度和速度。
文摘对多旋翼电动垂直起降(electric vertical takeoff and landing,eVTOL)航空器推进系统进行了可靠性分析与分配。首先,针对多旋翼eVTOL航空器可靠性历史数据积累不足的问题,使用模糊贝叶斯网络(fuzzy Bayesian network,FBN)建立可靠性分析模型,对其可靠性先验数据进行了补充,并进行可靠性后验推理,辅助定位推进系统关键环节。其次,基于FBN可靠性分析模型,提出一种改进电子设备可靠性咨询组(advisory group on reliability of electronic equipment,AGREE)可靠性分配方法,对不同构型多旋翼eVTOL推进系统进行可靠性分配。结果表明,FBN可靠性分析模型补充了推进系统可靠性数据,可有效识别系统薄弱环节。改进AGREE分配法的可靠性分配结果符合SC-VTOL-01中对eVTOL航空器的可靠性要求,同时该方法得到的可靠性分配结果更为合理,体现了不同构型、子系统、部件间的差异。
基金the funding provided by the National Helicopter Development Project of China。
文摘Accurate measurement of helicopter rotor motion parameters(flap,lead-lag,torsion,and azimuth angles)is essential for rotor blade design,helicopter dynamics modeling,and flight safety and health monitoring.However,the existing methods face challenges in testing equipment installation,calibration,and data transmission,resulting in limited reports on real-time in-flight measurements of blade motion parameters.This paper proposes a non-contact optoelectronic method based on two-dimensional position-sensitive detectors for in-flight measurement and a ground calibration system to obtain real-time rotor motion parameters during helicopter flight.The proposed method establishes the time evolution relationship of rotor motion parameters and verifies the performance of the in-flight measurement system regarding measurement resolution and accuracy through the construction of a blade motion posture experimental platform.The proposed method has been applied to the flight measurement of a medium-sized single-rotor helicopter,and the obtained results have been compared with theoretical analysis outcomes.Furthermore,this paper examines the characteristics of blade motion parameters during flight and discusses the challenges and potential solutions for measuring rotor motion parameters during helicopter flight using the proposed method.
基金funded by the National Natural Science Foundation of China(No.52105060)the Special Transmission Project,China(No.KY-1044-2023-0458)。
文摘To reduce the vibration of the Coaxial Helicopter Main Transmission System(CHMTS)considering both level and vertical flight conditions,a vibration evaluation and optimization model for the CHMTS was built.The vibration simulation model of the CHMTS was set up by gear dynamics theory and loaded contact analysis.For better evaluation of the system vibration,a vibration evaluation method for the CHMTS was established by the G1 method-variation coefficient method.A hybrid Gravitational Search Algorithm-Simulated Annealing(GSA-SA)algorithm was combined to balance convergence speed and searching accuracy.The principle test was conducted to prove the accuracy of theoretical method,in which the maximum relative error is16.26%.The optional results show that the vibration of the optimized transmission system decreases significantly,in which the maximum reduction of key vibration indicators reaches more than 20%.The theoretical results have been compared to the experiment to verify the effectiveness of the vibration optimization method.The proposed method could be extended to other fields.
