无风速计的多普勒激光雷达室内校准：基于光纤与蒙特卡洛模拟
Anemometer-free indoor calibration of Doppler lidar using fiber optics and Monte Carlo simulation
浏览论文内容
中文总结 AI 辅助
本文提出一种无风速计的多普勒激光雷达室内校准方法，通过光纤测试台测量视线速度不确定度，并结合蒙特卡洛模拟推导风速不确定度，在保证与IEC校准相当精度的同时大幅缩短校准时间。
中文摘要 AI 辅助
现有的用于风能应用的多普勒激光雷达校准和分类方法耗时较长，并且高估了激光雷达的测量不确定度。这些缺点源于参考仪器：气象桅杆上的风速计。现场校准中的输入信号是真实的风况，并包含与测风桅杆和地形相关的不确定度。本研究提出了一种新的激光雷达校准方法并对其进行了验证。新方法分为两个步骤。首先，测量激光雷达视线速度（LOS）的不确定度。其次，利用模拟风场，由这些中间不确定度推导出水平风速的不确定度。视线速度不确定度通过一个光纤测试台进行评估，该测试台作为一个合成风洞，称为“光纤多速度大气模拟”（SAFO-MV）。使用SAFO-MV，激光雷达的视线速度通过声光调制器（AOM）进行频移，并从光纤卷盘上背向散射，模拟来自均匀、低湍流风场的背向散射。参考信号通过校准AOM的射频驱动器实现国际单位制（SI）溯源。对测试台和实验室过程中的不确定度来源进行了检查，发现其影响很小。风速不确定度通过蒙特卡洛不确定度传播推导得出。风场使用PyConTurb（KSEC湍流模型的一种实现）进行模拟。模拟中的虚拟激光雷达复现了根据IEC 61400-50-2分类活动得出的激光雷达灵敏度。设备灵敏度的复现以及该方法的可追溯性构成了对测量模型的验证，使其可用于激光雷达校准。所提出的方法产生的不确定度与IEC激光雷达校准相当，同时大大减少了校准时间。这项新技术可以简化激光雷达的部署，并降低风资源评估中的不确定度。
英文摘要
Existing methods for calibrating and classifying Doppler lidars for wind energy applications are time-consuming and overestimate lidar measurement uncertainty. These shortcomings are due to the reference instruments: anemometers on meteorological masts. The input signals in field calibrations are real wind conditions, and include uncertainties associated with the met mast and the terrain. This research presents validation of a new method for lidar calibration. The new approach has two steps. First, the uncertainty of the lidar line of sight velocities (LOS) are measured. Second, those intermediate uncertainties are used to derive the uncertainty of the horizontal wind speed using a simulated wind field. LOS uncertainties are assessed using a fiber optic bench as a synthetic wind tunnel called Simulation of the Atmosphere with Fiber Optics Multi-Velocity (SAFO-MV). Using SAFO-MV, the lidar LOS is shifted using an acousto-optic modulator (AOM) and backscattered from a fiber spool, mimicking backscatter from a uniform, low turbulence wind field. The reference signals are SI-traceable via calibration of the AOM's RF-driver. Sources of uncertainty in the bench and the laboratory process are examined and found to be small. Wind speed uncertainties are derived via Monte Carlo uncertainty propagation. Wind fields are simulated using PyConTurb, an implementation of the KSEC turbulence model. A virtual lidar in the simulation replicates lidar sensitivities derived by IEC 61400-50-2 classification campaigns. Replication of device sensitivities and the traceability of the approach comprise a validation of the measurement model, enabling its use for lidar calibration. The proposed method yields uncertainties comparable to IEC lidar calibration, while greatly reducing calibration time. This new technology can streamline lidar deployment and reduce uncertainty in wind resource assessment.
发表机构
- Vaisala France SAS（Vaisala法国公司）
- DeutscheWindGuard GmbH（德国风卫有限公司）
机构由 AI 辅助整理，请以论文原文为准。
