Department of Energy and Marine Propulsion, University of A Coruña, ETSNyM, Paseo de Ronda 51, 15011 A Coruña, Spain.
Department of Mechanical Engineering, Catholic University of Ávila, C/Canteros, s/n, 05005 Avila, Spain.
Sensors (Basel). 2020 Dec 23;21(1):40. doi: 10.3390/s21010040.
There are different monitoring procedures in wind farms with two main objectives: (i) to improve energy production by the capability of the national electrical network and (ii) to reduce the stooped hours due to preventive and or corrective maintenance activities. In this sense, different sensors are employed to sample in real-time the working conditions of equipment, the electrical production and the weather conditions. Despite this, just the anemometer measurement can be related to the more important errors of interruption of power regulation and anemometer errors. Both errors are related to gusty winds and contribute to more than 33% of the cost of a wind farm. The present paper reports some mathematical relations between weather and maintenance but there are no extreme values of each variable that let us predict a near failure and its corresponding loss of working hours. To achieve this, statistical analysis identifies the relation between weather variables and errors and different models are obtained. What is more, due to the difficulty and economic implications involving the implementation of complex algorithms and techniques of artificial intelligence, it is still a challenge to optimize this process. Finally, the obtained results show a particular case study that can be extrapolated to other wind farms after different case studies to adjust the model to different weather regions, and serve as a useful tool for weather maintenance.
风电场有不同的监测程序，主要有两个目标：（i）通过国家电网的能力提高能源产量，（ii）减少因预防性和/或纠正性维护活动而导致的停机时间。在这方面，不同的传感器被用来实时采样设备的工作条件、电力生产和天气条件。尽管如此，只有风速计的测量值可以与更重要的功率调节中断误差和风速计误差相关联。这两个误差都与阵风有关，占风电场成本的 33%以上。本文报告了一些天气和维护之间的数学关系，但没有每个变量的极值，这使得我们无法预测即将发生的故障及其对应的工作时间损失。为了实现这一点，统计分析确定了天气变量和误差之间的关系，并获得了不同的模型。更重要的是，由于实施复杂的算法和人工智能技术涉及到的难度和经济影响，优化这个过程仍然是一个挑战。最后，得到的结果显示了一个特殊的案例研究，可以在不同的案例研究之后推广到其他风电场，以调整模型适应不同的天气区域，并作为天气维护的有用工具。
