针对不停机风力发电机改进的自动调焦系统  

Improved auto-focusing system for non-stop wind turbine blade

作　　者：田航 陈果 赵辉[1] 陶卫[1] 吕娜 TIAN Hang;CHEN Guo;ZHAO Hui;TAO Wei;LÜNa(School of Sensing Science and Engineering,School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学电子信息与电气工程学院感知科学与工程学院,上海200240

出　　处：《现代电子技术》2024年第21期139-148,共10页Modern Electronics Technique

摘　　要：风能是一种重要的可再生资源,因此对风力发电机组进行损伤检测具有重要意义。由于叶片图像的清晰度对损伤检测有很大影响,因此需要通过自动调焦步骤获得清晰的图像。不停机风力发电机叶片追踪图像中叶片的位置和尺寸改变,无法准确评估图像清晰度。因此,通过聚焦搜索策略无法实现自动聚焦步骤。文中提出一种基于改进的清晰度评价方法的自动调焦系统。通过基于面积补偿系数的方法补偿清晰度评价,并进行仿真实验和风力发电机模型实验。叶片目标变化对清晰度评价的影响从0.218降至0.030,叶片目标变化对清晰度的影响大大降低。在改进的清晰度评价方法的基础上,采用爬山法的搜索策略实现自动调焦步骤,同时,由于需要考虑算法程序对于不停机风力发电机的实时性,计算了自动调焦步骤的消耗时间,通过降低图像分辨率和调用NumPy库,将耗时从13.79 s减小到0.1 s以下。最后,实现了不停机风力发电机叶片追踪图像的自动调焦步骤,为提高自动调焦技术的性能提供了借鉴。Wind energy is an important renewable resource,so the damage detection of wind turbines is of great significance.Since the definition of the blade image has a great influence on damage detection,it is necessary to obtain clear images by the step of auto-focusing.The blade position and size of the blade in the blade tracking image of the non-stop wind turbine are changing,so it is difficult to evaluate the image definition accurately.As a result,the step of auto-focusing cannot be realized by focusing search strategy.In this paper,an auto-focusing system based on the improved definition evaluation method is proposed.The definition evaluation is compensated by the method based on the area compensation coefficient.The simulation experiments and wind turbine model experiments are carried out.The experimental results show that the influence of the change of blade(the object)on definition evaluation decreases from 0.218 to 0.030.It can be seen that the effect from change of blade(the object)on definition is reduced greatly.Then,based on the improved definition evaluation method,the search strategy of mountain climbing method is used to realize the step of auto-focusing.It is necessary to take into account the real-time performance of the algorithm program for the non-stop wind turbine,so the consuming time of the step of auto-focusing is calculated,and the consuming time is reduced from 13.79 s to less than 0.1 s by reducing the image resolution and calling NumPy library.Finally,the step of auto-focusing for the blade tracking image of the non-stop wind turbine is realized.To sum up,it provides a reference for improving the auto-focusing technology.

关 键 词：风力发电机 自动调焦 叶片变化 清晰度评价 调焦搜索 损伤检测 

分 类 号：TN911-34[电子电信—通信与信息系统]