可解释的机器学习用于风力轮机叶片的塔雷达监测:在不断变化的操作条件下识别细粒度叶片
Sercan Alipek1, Christian Kexel2, Jochen Moll1
1Department of Mechanical Engineering, University of Siegen, Paul-Bonatz-Straße 9-11, 57076 Siegen, Germany.
Sensors (Basel, Switzerland)
|February 27, 2026
概括
雷达测量可以识别风力轮机叶片的独特结构特征,即使在不断变化的环境条件下也可以进行分类. 这种方法有助于精确监控旋翼叶片.
科学领域:
- 工程 工程师 工程师 工程师
- 人工智能的人工智能
- 信号处理 信号处理
背景情况:
- 使用塔雷达系统的风力轮机叶片的操作分类在可转移性,环境条件适应性和可解释性方面面临挑战.
- 深度神经网络越来越多地用于分析工业应用中的复杂传感器数据.
研究的目的:
- 通过塔雷达测量,评估运行风力轮机叶片的数据驱动分类方法.
- 解决可转移性,不断变化的环境和操作条件 (EOC) 以及塔雷达监控 (TRM) 中深层神经决策的解释性方面的挑战.
主要方法:
- 使用连续的塔雷达测量被压缩成一个二维的慢时间到范围表示 (雷达图).
- 采用频率调制连续波 (FMCW) 雷达系统,在33.4-36 GHz频率范围内工作.
- 应用了对类敏感的可视化技术,指导梯度加权类激活映射 (GuidedGradCAM),以识别卷积神经网络的相关雷达图特征.
主要成果:
- 单个转子叶片表现出由雷达传感器检测到的特征性结构特征,允许基于神经网络的歧视.
- 这些已识别的特征证明了弹性,并且不会因改变EOC而变得无效.
- 像素级的扭曲突出了低级信息对于精确的旋翼叶片分类的关键作用.
结论:
- 塔雷达监测 (TRM) 可以通过雷达识别的独特结构特征来有效地分类单个风力轮机叶片.
- 开发的机器学习方法在适应不同EOC方面表现有前途,提高了TRM系统的稳定性.
- 像GuidedGradCAM这样的可解释性技术对于理解和验证TRM应用中深度神经网络做出的决定至关重要.
相关概念视频
Wind Turbine Machine Models
635
In the growing field of wind energy, incorporating wind turbine models into transient stability analysis is essential. Induction and synchronous machines are the primary models used, with induction machines being prevalent due to their simplicity and reliability.
Induction machines interact through the rotating magnetic field generated by the stator and the rotor. The key parameter is slip, which is the difference between synchronous speed and rotor speed relative to synchronous speed. Slip is...
Induction machines interact through the rotating magnetic field generated by the stator and the rotor. The key parameter is slip, which is the difference between synchronous speed and rotor speed relative to synchronous speed. Slip is...
635
Turbine-Governor Control
1.0K
Turbine-governor control is crucial for maintaining power system stability by balancing turbine mechanical power output with electrical load demand. This mechanism ensures that generator frequency and rotor speed are within acceptable limits during load variations. Turbine-generator units store kinetic energy due to their rotating masses; this energy is released to meet the load requirement when the load increases. The electrical torque of turbines rises to meet the demand, whereas the...
1.0K
Design Example: Calculating Safe Diameter for Wind-Exposed Disc
369
Assessing safety in wind-exposed installations is crucial to preventing potential failures. This example explores the calculation and design adjustments needed to mount a circular disc on a building facade, where wind forces are a primary concern. A 4-meter diameter disc was initially designed as an aesthetic feature facing winds at a velocity of 25 meters per second, with an air density of 1.25 kilograms per cubic meter. Given these conditions, the drag force on the disc was determined using...
369
Application of Linearization and Approximation
121
A drone flying through complex terrain often relies on more than one sensing method to estimate small changes in altitude. Along with direct measurements, air pressure provides a useful indirect indicator of vertical movement. Atmospheric pressure decreases as altitude increases, and this relationship is commonly described using an exponential model. Although accurate, converting pressure measurements into altitude values requires calculations that are too complex to perform repeatedly during...
121


