A Dual-Mode Millimeter-Wave Sensor Network for Structural Monitoring in Wind Farms

该项目将通过研究有效和自主的涡轮机检查来增强美国风能基础设施的弹性和可持续性。传统上，风力结构监测的主要形式是使用基于接触的传感器，如加速度计和应变计。然而，这些接触式传感器的应用受到现场集成、维护和重新配置方面的挑战的限制。光探测和测距（LIDAR）系统是非接触式的，具有高精度，但它们容易受到天气条件的影响，并且通常安装在固定位置。摄像机可以安装在无人驾驶飞行器（uav）上，扫描涡轮机的表面。然而，这种方法需要停止涡轮叶片的旋转，并且只能检查表面。为了应对这些挑战，该项目将开发一个由统一决策框架驱动的双模毫米波传感器网络，以优化风电场的结构检查。由于射频信号对环境光线和天气条件具有很强的抗干扰能力，因此位于涡轮机附近的固定平台将在正常运行期间提供强大的连续监测。另一方面，基于蜂群-无人机的传感器网络的形成可以在连续监测传感器识别初始问题或在定期维护期间合成大观测孔径以进行高分辨率成像。本项目涉及毫米波传感、无人机自适应编队、蜂群飞行控制、系统级巡检进度优化统一决策框架等多学科交叉。它将为电力传输网络、石油/天然气管道和运输网络等关键基础设施的结构健康监测产生新的知识和方法。该项目为学生提供了一个宝贵的机会，培养他们在系统可靠性优化、自主机器人和微波/毫米波技术领域的兴趣。pi将开发综合研究和教育项目，以吸引代表性不足的学生和K-12学生进入工程领域，并让本科生参与研究。它还将鼓励以成功的技术开发为基础的学生创业。该项目的创新之处在于，它将独特的微型毫米波遥感能力与先进的无人机控制方法相结合，用于网络化相干探测。将系统级的长期巡检规划与基于机组级短期信息的动态预测相结合。该项目将研究：1)安装在风力涡轮机附近的固定式传感器，在涡轮机运行时提供不间断的监测，其中将集成分析和机器学习方法，从旋转涡轮机叶片产生的微多普勒特征中分析叶片畸变；2)基于无人机的传感器阵列，利用高分辨率合成孔径成像技术扫描涡轮叶片细节；3)新颖的固有飞行控制策略，使无人机群以足够的精度、能量效率和最小的抖动实现所需的编队，用于合成孔径成像；4)统一的决策框架，从系统和动态的角度优化风电机组巡检计划。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

This project will enhance the resiliency and sustainability of the Nation’s wind energy infrastructures by investigating effective and autonomous turbine inspection. The major forms of structural monitoring of wind structures have traditionally been accomplished using contact-based sensors such as accelerometers and strain gauges. However, application of these contact-based sensors is limited by challenges in onsite integration, maintenance, and reconfiguration. Light Detection and Ranging (LIDAR) systems are non-contact and have high precision, but they are susceptible to weather conditions and are often installed at fixed locations. Cameras can be mounted on unmanned aerial vehicles (UAVs) to scan the surface of turbines. However, this approach requires stopping the rotation of turbine blades and can only inspect the surface. To address these challenges, this project will develop a dual-mode millimeter-wave sensor network driven by a unified decision-making framework that optimizes structural inspection in wind farms. Since radio frequency signals are robust against ambient light and weather conditions, a stationary platform located near turbines will offer robust continuous monitoring during normal operation. On the other hand, a formation of swarm-UAV-based sensor network can synthesize a large observation aperture for high-resolution imaging when an initial problem is identified by continuous-monitoring sensors, or during scheduled maintenance. This project has an interdisciplinary nature involving millimeter-wave sensing, adaptive UAV formation, swarm flight control, and unified decision-making framework for system-level inspection schedule optimization. It will generate new knowledge and methodologies for structural health monitoring of critical infrastructures such as power transmission networks, oil/gas pipelines, and transportation networks. The project provides a valuable opportunity for students to develop their interest in the fields of system reliability optimization, autonomous robotics, and microwave/millimeter-wave technologies. The PIs will develop integrated research and education programs to attract students from underrepresented groups and K-12 students into engineering and involve undergraduate students into research. It will also encourage student entrepreneurship based on successful technology development.The project is innovative in that it integrates unique miniature millimeter-wave remote sensing capability with advanced UAV control methods for networked coherent detection. Furthermore, it unifies the long-term inspection planning at the system level and the dynamic prognosis based on the short-term information at the turbine level. This project will investigate: 1) stationary sensors mounted near wind turbines to provide uninterrupted monitoring while turbines are in operation, where analytic and machine learning methods will be integrated to analyze blade distortion from the micro-Doppler signatures generated by rotating turbine blades; 2) UAV-based sensor arrays to scan details of turbine blades with high-resolution synthetic aperture imaging; 3) novel intrinsic flight control strategies to enable swarms of UAVs to realize the desired formation with sufficient precision, energy efficiency, and minimal jitter for synthetic aperture imaging; 4) a unified decision-making framework to optimize the inspection schedule of wind turbines from a systems and dynamic perspective.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.