WO2024021260A1 - 一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法 - Google Patents

一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法 Download PDF

Info

Publication number
WO2024021260A1
WO2024021260A1 PCT/CN2022/119571 CN2022119571W WO2024021260A1 WO 2024021260 A1 WO2024021260 A1 WO 2024021260A1 CN 2022119571 W CN2022119571 W CN 2022119571W WO 2024021260 A1 WO2024021260 A1 WO 2024021260A1
Authority
WO
WIPO (PCT)
Prior art keywords
speed camera
steam turbine
blade
flaw detection
detection system
Prior art date
Application number
PCT/CN2022/119571
Other languages
English (en)
French (fr)
Inventor
关淳
翁振宇
薛海亮
赵洪羽
祝海义
马义良
潘劭平
初世明
李宇峰
赫广迅
宋彬
梁天赋
王健
尉坤
郭魁俊
Original Assignee
哈电发电设备国家工程研究中心有限公司
哈尔滨汽轮机厂有限责任公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 哈电发电设备国家工程研究中心有限公司, 哈尔滨汽轮机厂有限责任公司 filed Critical 哈电发电设备国家工程研究中心有限公司
Publication of WO2024021260A1 publication Critical patent/WO2024021260A1/zh

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/91—Investigating the presence of flaws or contamination using penetration of dyes, e.g. fluorescent ink
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques

Definitions

  • the invention belongs to the field of steam turbine blade flaw detection technology, and specifically relates to an intelligent flaw detection system and working method for steam turbine blades based on high-speed camera technology.
  • the common method of inspecting the surface condition of blades is to use the endoscope method.
  • the endoscope is inserted into the turbine, the position of the front lens of the endoscope is manually adjusted, and the blade to be inspected is aligned for static inspection.
  • this method requires maintenance personnel to manually install and adjust the position every time.
  • high-definition photography is used, the shooting effect of the rotating blades is poor and the inspection efficiency is low.
  • the problem to be solved by this invention is to improve the efficiency and comprehensiveness of steam turbine blade inspection, and propose an intelligent flaw detection system and working method for steam turbine blades based on high-speed camera technology.
  • An intelligent flaw detection system for steam turbine blades based on high-speed camera technology including a high-speed camera device, auxiliary device, transmission device and imaging analysis system;
  • the high-speed camera device includes a high-speed camera and protective tooling; the auxiliary device includes a lighting device and a reagent injector; the transmission device includes a transmission telescopic rod and a transmission controller; the imaging analysis system includes a collection module, an imaging display module, and a transmission controller. Image recognition module;
  • a lighting device and a reagent injector are respectively installed on the left and right sides of the housing of the high-speed camera.
  • a protective tooling is installed on the outside of the high-speed camera. The protective tooling and the transmission telescopic rod are connected by bolts;
  • the transmission telescopic rod is connected to the transmission controller
  • the imaging analysis system is connected to one end of the signal line, and the other end of the signal line is connected to the high-speed camera device through a transmission telescopic rod.
  • the high-speed camera is a high-speed, high-definition camera with no less than 8 million pixels, a focus range of 2mm-12mm, a visual distance of no less than 5m, a photo frequency of no less than 500 frames/second, an aperture of F3.0, and 0-180° wide-angle shooting.
  • the protective tooling is a stainless steel sleeve, which is set on the outside of the high-speed camera for carrying and protecting the high-speed camera.
  • the outside of the protective tooling is welded to the steam turbine exhaust guide ring.
  • the lighting device rotates synchronously with the high-speed camera.
  • the reagent injector rotates synchronously with the high-speed camera.
  • the transmission telescopic rod is electrically adjusted, with a maximum extension of 1 m.
  • the transmission telescopic rod is made of stainless steel, and the transmission controller is used to control the telescopic length of the transmission telescopic rod.
  • the acquisition module is used to collect and store original video;
  • the imaging display module is used to display the collected blade images, control high-speed camera devices, auxiliary devices and transmission devices, and freely adjust the imaging display rate, call historical records, and Picture saving, picture comparison, and inspection report output;
  • the picture recognition module is used to automatically identify, analyze, and diagnose the collected pictures of blades.
  • the module has a built-in picture library of typical blade faults.
  • a working method of an intelligent flaw detection system for steam turbine blades based on high-speed camera technology relying on the above-mentioned intelligent flaw detection system for steam turbine blades based on high-speed camera technology to achieve:
  • the transmission telescopic rod is extended to a designated position through the transmission controller. Then, the high-speed camera of the high-speed camera device extends out of the protective tooling, and the angle and focus of the high-speed camera are adjusted to ensure The high-speed camera clearly captures the blade surface;
  • the lighting device and reagent injection device in the auxiliary device are installed on the shell of the high-speed camera, and the protective tooling is extended along with the high-speed camera.
  • the lighting device Before starting to shoot, the lighting device is turned on. According to the surface conditions of the blades, it is selected whether to turn on the reagent injection device.
  • the turbine disk After the vehicle starts, start shooting and start shooting and recording the blades. Adjust the shooting parameters of the optical camera according to the turning speed of the turbine, including focal length, aperture, and shooting frame rate. It is required to shoot one blade in each frame during the shooting process to complete the analysis of the blades. Numbering and identification.
  • the turning speed is 3-5 rpm
  • the shooting time is 5 minutes
  • picture records of more than 15 laps are saved, and the collected pictures are used for picture diagnosis analysis.
  • the picture recognition module has a built-in picture library of typical blade faults, including five typical faults: blade shroud wear, blade cracks, blade collisions, blade water erosion, and tie rod fractures.
  • the fault picture samples come from the steam turbines of Harbin Steam Turbine Factory Co., Ltd.
  • the blade When the number of effective comparison pictures of a blade exceeds 80% of the total number of images collected for the blade, the blade is considered to have a fault; when all collected images have been compared and analyzed, but the number of effective comparison images is less than 80% of the total number of images collected for the blade. %, it is considered that the blade has no fault problem, fault identification is performed on the blade with fault problem, and the fault diagnosis result is output.
  • An intelligent flaw detection system for steam turbine blades based on high-speed camera technology uses a high-speed camera optical probe to replace manual inspection, and avoids the problem that endoscopes cannot take high-definition shots of rotating blades. Combined with the steam turbine turning process, It can not only comprehensively and quickly obtain the surface damage status of low-pressure final blades, but also reduce the risk of manual drilling. It can also save photo records of each inspection, thereby improving the power station's operation and maintenance efficiency of flexible deep peaking units and ensuring The safety of long-term flexible operation of the unit.
  • Figure 1 is a block diagram of an intelligent flaw detection system for steam turbine blades based on high-speed camera technology according to the present invention
  • Figure 2 is a schematic diagram of the working state of an intelligent flaw detection system for steam turbine blades based on high-speed camera technology according to the present invention
  • Figure 3 is a schematic diagram of a non-working state of an intelligent flaw detection system for steam turbine blades based on high-speed camera technology according to the present invention
  • Figure 4 is a picture recognition module fault diagnosis flow chart of an intelligent flaw detection system for steam turbine blades based on high-speed camera technology according to the present invention.
  • An intelligent flaw detection system for steam turbine blades based on high-speed camera technology including a high-speed camera device 1, an auxiliary device 2, a transmission device 3 and an imaging analysis system 4;
  • the high-speed camera device 1 includes a high-speed camera 1-1 and protective equipment 1-2; the auxiliary device 2 includes a lighting device 2-1 and a reagent injector 2-2; the transmission device 3 includes a transmission telescopic rod 3-1 and transmission controller 3-2; the imaging analysis system 4 includes a collection module 4-1, an imaging display module 4-2 and a picture recognition module 4-3;
  • a lighting device 2-1 and a reagent injector 2-2 are respectively installed on the left and right sides of the housing of the high-speed camera 1-1.
  • a protective tooling 1-2 is installed on the outside of the high-speed camera 1-1.
  • the protective tooling 1-2 and the transmission telescopic rod 3-1 are connected by bolts;
  • the transmission telescopic rod 3-1 is connected to the transmission controller 3-2;
  • the imaging analysis system 4 is connected to one end of the signal line, and the other end of the signal line is connected to the high-speed camera device 1 through the transmission telescopic rod 3-1.
  • the high-speed camera 1-1 is a high-speed, high-definition camera with no less than 8 million pixels, a focus range of 2mm-12mm, a visual distance of no less than 5m, a photo frequency of no less than 500 frames/second, and an aperture of F3. 0, with 0-180° wide-angle shooting.
  • the protective tooling 1-2 is a stainless steel sleeve, which is set on the outside of the high-speed camera 1-1 to carry and protect the high-speed camera 1-1.
  • the outer side of the protective tooling 1-2 is welded to the turbine exhaust. On the guide ring.
  • the lighting device 2-1 rotates synchronously with the high-speed camera 1-1.
  • the reagent injector 2-2 rotates synchronously with the high-speed camera 1-1.
  • the transmission telescopic rod 3-1 is electrically adjusted, and the maximum elongation is 1m.
  • the transmission telescopic rod 3-1 is made of stainless steel, and the transmission controller 3-2 is used to control the transmission telescopic rod 3. -1 telescopic length.
  • the collection module 4-1 is used to collect and store the original video;
  • the imaging display module 4-2 is used to display the collected blade images, control the high-speed camera device 1, the auxiliary device 2 and the transmission device 3, and freely collect the images. Adjust the imaging display rate, call history records, save pictures, compare pictures, and output inspection reports;
  • the picture recognition module 4-3 is used to automatically identify, analyze, and diagnose the collected pictures of blades.
  • the module has a built-in picture library of typical blade failures.
  • the reagent injection device 2-2 is equipped with a flaw detection colorant for the steam turbine blades.
  • the reagent injection device can be manually controlled to spray the colorant to the designated area for coloring inspection.
  • the flaw detection colorant can be replenished regularly, and the remaining amount of colorant has a real-time display function.
  • the reagent injection device can rotate synchronously with the high-speed camera 1-1 to ensure that the injection area is consistent with the probe observation area.
  • the transmission controller 3-2 is used to control the telescopic length of the transmission telescopic rod 3-1, that is, to control the vertical position of the high-speed camera 1-1 to ensure that the entire blade can be observed.
  • the transmission telescopic rod is extended to a designated position through the transmission controller. Then, the high-speed camera of the high-speed camera device extends out of the protective tooling, and the angle and focus of the high-speed camera are adjusted to ensure The high-speed camera clearly captures the blade surface;
  • the lighting device and reagent injection device in the auxiliary device are installed on the shell of the high-speed camera, and the protective tooling is extended along with the high-speed camera.
  • the lighting device Before starting to shoot, the lighting device is turned on. According to the surface conditions of the blades, it is selected whether to turn on the reagent injection device.
  • the turbine disk After the vehicle starts, start shooting and start shooting and recording the blades. Adjust the shooting parameters of the optical camera according to the turning speed of the turbine, including focal length, aperture, and shooting frame rate. It is required to shoot one blade in each frame during the shooting process to complete the analysis of the blades. Numbering and identification.
  • the turning speed is 3-5 rpm
  • the shooting time is 5 minutes
  • picture records of more than 15 laps are saved, and the collected pictures are used for picture diagnosis analysis.
  • the picture recognition module has a built-in picture library of typical blade faults, including five typical faults: blade shroud wear, blade cracks, blade collisions, blade water erosion, and tie rod fractures.
  • the fault picture samples come from the steam turbines of Harbin Steam Turbine Factory Co., Ltd.
  • the blade When the number of effective comparison pictures of a blade exceeds 80% of the total number of images collected for the blade, the blade is considered to have a fault; when all collected images have been compared and analyzed, but the number of effective comparison images is less than 80% of the total number of images collected for the blade. %, it is considered that the blade has no fault problem, fault identification is performed on the blade with fault problem, and the fault diagnosis result is output.
  • the diagnostic results are given through fuzzy comparison of the freeze-frame pictures collected from the video and the fault pictures in the picture library. Due to the limited content of the fault picture library, it is difficult to comprehensively cover fault defects at all locations. For this reason, the current fault diagnosis algorithm has strong flexibility and is mainly related to the type of blade damage and is not sensitive to the location of blade damage. That is to say, regardless of any location, even if it is different from the actual damaged location in the picture library, as long as typical faults such as cracks, water erosion, bumps, and falling blocks occur, corresponding diagnostic evaluations will be given.
  • the intelligent flaw detection system for steam turbine blades based on high-speed camera technology completes the imaging, it needs to be protected to avoid the impact and damage of the camera system by airflow during actual operation of the steam turbine.
  • the high-speed camera, auxiliary device, and transmission telescopic rod are retracted into the protective tooling, and the protective tool is installed outside the steam turbine exhaust guide ring.
  • the intelligent flaw detection system for steam turbine blades based on high-speed camera technology stops when the power is turned off.
  • the steam turbine blade intelligent flaw detection system based on high-speed camera technology will restart when the next time the steam turbine is shut down for cranking inspection. This can effectively avoid the direct impact of airflow on the intelligent flaw detection system and transmission device of steam turbine blades based on high-speed camera technology during actual operation of the steam turbine. It can also reduce water erosion damage and improve service life.

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法，属于汽轮机叶片探伤技术领域。为提高汽轮机叶片检查的效率及全面性。包括高速摄像装置(1)、辅助装置(2)、传动装置(3)和成像分析系统(4)；高速摄像装置(1)包括高速摄像头(1-1)、防护工装(1-2)；辅助装置(2)包括照明装置(2-1)和试剂喷射器(2-2)；传动装置(3)包括传动伸缩杆(3-1)和传动控制器(3-2)；成像分析系统(4)包括采集模块(4-1)、成像显示模块(4-2)和图片识别模块(4-3)；高速摄像头(1-1)的外壳上左右两侧分别安装有照明装置(2-1)和试剂喷射器(2-2)，高速摄像头(1-1)的外部安装有防护工装(1-2)，防护工装(1-2)和传动伸缩杆(3-1)采用螺栓连接；传动伸缩杆(3-1)和传动控制器(3-2)连接；成像分析系统(4)连接信号线的一端，信号线的另一端穿过传动伸缩杆(3-1)和高速摄像装置(1)连接，检查全面且快速。

Description

一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法 技术领域
本发明属于汽轮机叶片探伤技术领域，具体涉及一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法。
 背景技术
为了促进电网对风、光等新能源的消纳，常规火电汽轮机要求具备更高的灵活性，能够长期处于深度调峰、频繁变负荷等非设计工况运行，而对于非设计工况运行，低压末级长叶片往往在颤振、鼓风、水蚀等更为恶劣的环境下运行，导致低压末级长叶片运行风险严重升高。为了安全运行，电站运维人员需要在大修期内择机检查叶片受损情况，检查频率一般为每年一次，检查方法为在机组停机时安排检修人员通过低压缸人孔钻入汽轮机，人为观察低压末级长叶片表面健康状态，此种检查方法既增加了人为操作的安全风险，也因人眼观察有限而很难掌握到每一只叶片的受损状态，导致运维风险高而效率低。
目前常见的叶片表面状态检查方法是采用内窥镜方法，将内窥镜伸入汽轮机内，手动调整内窥镜前置镜头的位置，对准需要检查的叶片，进行静态检查。但这种方式每次都需要维修人员手动安装和调整位置，虽然采用了高清摄像的方式，但对旋转叶片拍摄效果差，检查效率低。
 技术问题
本发明要解决的问题是提高汽轮机叶片检查的效率及全面性，提出一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法。
 技术解决方案
为实现上述目的，本发明通过以下技术方案实现：
一种基于高速摄像技术的汽轮机叶片智能探伤系统，包括高速摄像装置、辅助装置、传动装置和成像分析系统；
所述高速摄像装置包括高速摄像头、防护工装；所述辅助装置包括照明装置和试剂喷射器；所述传动装置包括传动伸缩杆和传动控制器；所述成像分析系统包括采集模块、成像显示模块和图片识别模块；
所述高速摄像头的外壳上左右两侧分别安装有照明装置和试剂喷射器，所述高速摄像头的外部安装有防护工装，所述防护工装和传动伸缩杆采用螺栓连接；
所述传动伸缩杆和传动控制器连接；
所述成像分析系统连接信号线的一端，信号线的另一端穿过传动伸缩杆和所述高速摄像装置连接。
进一步的，所述高速摄像头为高速、高清照相机，像素不低于800万，聚焦范围2mm-12mm，可视距离不小于5m，拍照频率不低于500帧/秒，光圈采用F3.0，具备0-180°广角拍摄。
进一步的，所述防护工装为不锈钢套筒，套装于高速摄像头的外部，用于承载和保护高速摄像头，所述防护工装的外侧焊接在汽轮机排汽导流环上。
进一步的，所述照明装置与高速摄像头同步转动。
进一步的，所述试剂喷射器与高速摄像头同步转动。
进一步的，所述传动伸缩杆通过电动调节，最大伸长量为1m，所述传动伸缩杆的材质为不锈钢，所述传动控制器用于控制传动伸缩杆伸缩长度。
进一步的，采集模块用于采集和存储原始录像；成像显示模块用于显示采集到的叶片图像、控制高速摄像装置、辅助装置和传动装置，并针对采集图像自由调整成像显示速率、调用历史记录、图片保存、图片对比、输出检查报告；图片识别模块用于自动识别、分析和诊断叶片的采集图片，模块内置叶片典型故障图片库。
一种基于高速摄像技术的汽轮机叶片智能探伤系统的工作方法，依托于所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统实现：
一种基于高速摄像技术的汽轮机叶片智能探伤系统工作时，传动伸缩杆通过传动控制器伸长至指定位置，而后，高速摄像装置的高速摄像头伸出防护工装，并调整高速摄像头角度和焦距，确保高速摄像头清楚的拍摄到叶片表面；
辅助装置中的照明装置和试剂喷射装置安装在高速摄像头的外壳上，并随高速摄像头一起伸出防护工装，开始拍摄前，照明装置开启，根据叶片表面情况选择是否开启试剂喷射装置，当汽轮机盘车开始后，启动拍摄，对叶片开始拍摄记录，根据汽轮机盘车转速调整光学摄像头的拍摄参数，包括焦距、光圈、拍摄帧率，要求拍摄过程中每一帧拍摄一只叶片，完成对叶片的编号和分辨。
进一步的，盘车转速为3-5转/分钟，拍摄时间5分钟，保存15圈以上的图片记录，得到的采集图片用于图片诊断分析。
进一步的，图片识别模块内置叶片典型故障图片库，包括叶片围带碰磨、叶片裂纹、叶片磕碰、叶片水蚀、拉筋断裂5种典型故障，故障图片样本来源于哈尔滨汽轮机厂有限责任公司的汽轮机低压长叶片典型故障图库，每种故障类型的图片超过10000张；
采用神经网络算法，建立图片对比模型，用于采集图片和故障图片的对比分析；
当采集图片与故障图片样本相似度超过90%，即认为对比为有效对比，否则认为对比无效；
当叶片有效对比图片数量超过该叶片所采集的图片总数的80%时，认为该叶片存在故障问题；当全部采集图片都完成对比分析，但有效对比图片数量小于该叶片所采集的图片总数的80%时，认为该叶片无故障问题，对存在故障问题的叶片进行故障识别，并输出故障诊断结果。
 有益效果
本发明所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，采用高速摄像的光学探头来取代人为检查，并规避了内窥镜无法对旋转叶片高清拍摄的问题，结合汽轮机盘车过程，既能全面且快速地获取低压末级叶片表面受损状态，又能降低人为钻缸风险，还能保存每次检查的照片记录，从而提高电站对灵活性深度调峰机组的运维效率，确保机组长期灵活性运行的安全性。
 附图说明
图1为本发明所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统的组成框图；
图2为本发明所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统的工作状态示意图；
图3为本发明所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统的非工作状态示意图；
图4为本发明所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统的图片识别模块故障诊断流程图。
 本发明的实施方式
为了使本发明的目的、技术方案及优点更加清楚明白，以下结合附图及具体实施方式，对本发明进行进一步详细说明。应当理解，此处所描述的具体实施方式仅用以解释本发明，并不用于限定本发明，即所描述的具体实施方式仅仅是本发明一部分实施方式，而不是全部的具体实施方式。通常在此处附图中描述和展示的本发明具体实施方式的组件可以以各种不同的配置来布置和设计，本发明还可以具有其他实施方式。
因此，以下对在附图中提供的本发明的具体实施方式的详细描述并非旨在限制要求保护的本发明的范围，而是仅仅表示本发明的选定具体实施方式。基于本发明的具体实施方式，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他具体实施方式，都属于本发明保护的范围。
为能进一步了解本发明的发明内容、特点及功效，兹例举以下具体实施方式，并配合附图1-4详细说明如下 ：
具体实施方式一：
一种基于高速摄像技术的汽轮机叶片智能探伤系统，包括高速摄像装置1、辅助装置2、传动装置3和成像分析系统4；
所述高速摄像装置1包括高速摄像头1-1、防护工装1-2；所述辅助装置2包括照明装置2-1和试剂喷射器2-2；所述传动装置3包括传动伸缩杆3-1和传动控制器3-2；所述成像分析系统4包括采集模块4-1、成像显示模块4-2和图片识别模块4-3；
所述高速摄像头1-1的外壳上左右两侧分别安装有照明装置2-1和试剂喷射器2-2，所述高速摄像头1-1的外部安装有防护工装1-2，所述防护工装1-2和传动伸缩杆3-1采用螺栓连接；
所述传动伸缩杆3-1和传动控制器3-2连接；
所述成像分析系统4连接信号线的一端，信号线的另一端穿过传动伸缩杆3-1和所述高速摄像装置1连接。
进一步的，所述高速摄像头1-1为高速、高清照相机，像素不低于800万，聚焦范围2mm-12mm，可视距离不小于5m，拍照频率不低于500帧/秒，光圈采用F3.0，具备0-180°广角拍摄。
进一步的，所述防护工装1-2为不锈钢套筒，套装于高速摄像头1-1的外部，用于承载和保护高速摄像头1-1，所述防护工装1-2的外侧焊接在汽轮机排汽导流环上。
进一步的，所述照明装置2-1与高速摄像头1-1同步转动。
进一步的，所述试剂喷射器2-2与高速摄像头1-1同步转动。
进一步的，所述传动伸缩杆3-1通过电动调节，最大伸长量为1m，所述传动伸缩杆3-1的材质为不锈钢，所述传动控制器3-2用于控制传动伸缩杆3-1伸缩长度。
进一步的，采集模块4-1用于采集和存储原始录像；成像显示模块4-2用于显示采集到的叶片图像、控制高速摄像装置1、辅助装置2和传动装置3，并针对采集图像自由调整成像显示速率、调用历史记录、图片保存、图片对比、输出检查报告；图片识别模块4-3用于自动识别、分析和诊断叶片的采集图片，模块内置叶片典型故障图片库。
进一步的，试剂喷射装置2-2装有汽轮机叶片的探伤着色剂，当需要对叶片进行局部着色探伤检查时，可手动控制试剂喷射装置将着色剂喷向指定区域，进行着色检查。探伤着色剂可定期补充，着色剂剩余量具备实时显示功能。试剂喷射装置能够与高速摄像头1-1同步转动，以确保喷射区域与探头观察区域一致。
进一步的，传动控制器3-2用于控制传动伸缩杆3-1伸缩长度，即控制高速摄像头1-1的竖直位置，确保能够观察到叶片全貌。
具体实施方式二：
一种基于高速摄像技术的汽轮机叶片智能探伤系统的工作方法，依托于具体实施方式一所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统实现：
一种基于高速摄像技术的汽轮机叶片智能探伤系统工作时，传动伸缩杆通过传动控制器伸长至指定位置，而后，高速摄像装置的高速摄像头伸出防护工装，并调整高速摄像头角度和焦距，确保高速摄像头清楚的拍摄到叶片表面；
辅助装置中的照明装置和试剂喷射装置安装在高速摄像头的外壳上，并随高速摄像头一起伸出防护工装，开始拍摄前，照明装置开启，根据叶片表面情况选择是否开启试剂喷射装置，当汽轮机盘车开始后，启动拍摄，对叶片开始拍摄记录，根据汽轮机盘车转速调整光学摄像头的拍摄参数，包括焦距、光圈、拍摄帧率，要求拍摄过程中每一帧拍摄一只叶片，完成对叶片的编号和分辨。
进一步的，盘车转速为3-5转/分钟，拍摄时间5分钟，保存15圈以上的图片记录，得到的采集图片用于图片诊断分析。
进一步的，图片识别模块内置叶片典型故障图片库，包括叶片围带碰磨、叶片裂纹、叶片磕碰、叶片水蚀、拉筋断裂5种典型故障，故障图片样本来源于哈尔滨汽轮机厂有限责任公司的汽轮机低压长叶片典型故障图库，每种故障类型的图片超过10000张；
采用神经网络算法，建立图片对比模型，用于采集图片和故障图片的对比分析；
当采集图片与故障图片样本相似度超过90%，即认为对比为有效对比，否则认为对比无效；
当叶片有效对比图片数量超过该叶片所采集的图片总数的80%时，认为该叶片存在故障问题；当全部采集图片都完成对比分析，但有效对比图片数量小于该叶片所采集的图片总数的80%时，认为该叶片无故障问题，对存在故障问题的叶片进行故障识别，并输出故障诊断结果。
进一步的，通过对采集录像的定格图片和图片库故障图片的模糊比对，给出诊断结果。由于故障图片库的内容有限，很难全面覆盖所有位置的故障缺陷，为此，当前的故障诊断算法具有较强的灵活性，主要与叶片受损类型相关，与叶片受损位置并不敏感。即无论任何位置，即便与图片库真实受损位置不同，但只要出现裂纹、水蚀、磕碰缺口、掉块等典型故障，均会给出相应的诊断评估。
进一步的，所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统完成摄像后，需要进行保护，避免汽轮机实际工作时，气流对摄像系统的冲击和破坏。非工作状态下，高速摄像头、辅助装置、传动伸缩杆缩回防护工装中，防护工装置于汽轮机排汽导流环外侧，所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统断电停止工作，待下次汽轮机停机后盘车检查时，所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统重新启机工作。如此能够有效避免汽轮机实际工作时，气流对所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统和传动装置的直接冲击，也能够减轻水蚀破坏，提高使用寿命。
需要说明的是，术语“第一”和“第二”等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。而且，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者设备不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者设备所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括所述要素的过程、方法、物品或者设备中还存在另外的相同要素。
虽然在上文中已经参考具体实施方式对本申请进行了描述，然而在不脱离本申请的范围的情况下，可以对其进行各种改进并且可以用等效物替换其中的部件。尤其是，只要不存在结构冲突，本申请所披露的具体实施方式中的各项特征均可通过任意方式相互结合 起来使用，在本说明书中未对这些组合的情况进行穷举性的描述仅仅是出于省略篇幅和节 约资源的考虑。因此，本申请并不局限于文中公开的特定具体实施方式，而是包括落入权利要求的范围内的所有技术方案。

Claims (10)

  1. 一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：包括高速摄像装置（1）、辅助装置（2）、传动装置（3）和成像分析系统（4）；
    所述高速摄像装置（1）包括高速摄像头（1-1）、防护工装（1-2）；所述辅助装置（2）包括照明装置（2-1）和试剂喷射器（2-2）；所述传动装置（3）包括传动伸缩杆（3-1）和传动控制器（3-2）；所述成像分析系统（4）包括采集模块（4-1）、成像显示模块（4-2）和图片识别模块（4-3）；
    所述高速摄像头（1-1）的外壳上左右两侧分别安装有照明装置（2-1）和试剂喷射器（2-2），所述高速摄像头（1-1）的外部安装有防护工装（1-2），所述防护工装（1-2）和传动伸缩杆（3-1）采用螺栓连接；
    所述传动伸缩杆（3-1）和传动控制器（3-2）连接；
    所述成像分析系统（4）连接信号线的一端，信号线的另一端穿过传动伸缩杆（3-1）和所述高速摄像装置（1）连接。
  2. 根据权利要求1所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：所述高速摄像头（1-1）为高速、高清照相机，像素不低于800万，聚焦范围2mm-12mm，可视距离不小于5m，拍照频率不低于500帧/秒，光圈采用F3.0，具备0-180°广角拍摄。
  3. 根据权利要求1或2所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：所述防护工装（1-2）为不锈钢套筒，套装于高速摄像头（1-1）的外部，用于承载和保护高速摄像头（1-1），所述防护工装（1-2）的外侧焊接在汽轮机排汽导流环上。
  4. 根据权利要求3所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：所述照明装置（2-1）与高速摄像头（1-1）同步转动。
  5. 根据权利要求4所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：所述试剂喷射器（2-2）与高速摄像头（1-1）同步转动。
  6. 根据权利要求5所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：所述传动伸缩杆（3-1）通过电动调节，最大伸长量为1m，所述传动伸缩杆（3-1）的材质为不锈钢，所述传动控制器（3-2）用于控制传动伸缩杆（3-1）伸缩长度。
  7. 根据权利要求6所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统，其特征在于：采集模块（4-1）用于采集和存储原始录像；成像显示模块（4-2）用于显示采集到的叶片图像、控制高速摄像装置（1）、辅助装置（2）和传动装置（3），并针对采集图像自由调整成像显示速率、调用历史记录、图片保存、图片对比、输出检查报告；图片识别模块（4-3）用于自动识别、分析和诊断叶片的采集图片，模块内置叶片典型故障图片库。
  8. 一种基于高速摄像技术的汽轮机叶片智能探伤系统的工作方法，依托于权利要求1-7之一所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统实现，其特征在于：
    一种基于高速摄像技术的汽轮机叶片智能探伤系统工作时，传动伸缩杆通过传动控制器伸长至指定位置，而后，高速摄像装置的高速摄像头伸出防护工装，并调整高速摄像头角度和焦距，确保高速摄像头清楚的拍摄到叶片表面；
    辅助装置中的照明装置和试剂喷射装置安装在高速摄像头的外壳上，并随高速摄像头一起伸出防护工装，开始拍摄前，照明装置开启，根据叶片表面情况选择是否开启试剂喷射装置，当汽轮机盘车开始后，启动拍摄，对叶片开始拍摄记录，根据汽轮机盘车转速调整光学摄像头的拍摄参数，包括焦距、光圈、拍摄帧率，要求拍摄过程中每一帧拍摄一只叶片，完成对叶片的编号和分辨。
  9. 根据权利要求8所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统的工作方法，其特征在于：盘车转速为3-5转/分钟，拍摄时间5分钟，保存15圈以上的图片记录，得到的采集图片用于图片诊断分析。
  10. 根据权利要求9所述的一种基于高速摄像技术的汽轮机叶片智能探伤系统的工作方法，其特征在于：图片识别模块内置叶片典型故障图片库，包括叶片围带碰磨、叶片裂纹、叶片磕碰、叶片水蚀、拉筋断裂5种典型故障，故障图片样本来源于哈尔滨汽轮机厂有限责任公司的汽轮机低压长叶片典型故障图库，每种故障类型的图片超过10000张；
    采用神经网络算法，建立图片对比模型，用于采集图片和故障图片的对比分析；
    当采集图片与故障图片样本相似度超过90%，即认为对比为有效对比，否则认为对比无效；
    当叶片有效对比图片数量超过该叶片所采集的图片总数的80%时，认为该叶片存在故障问题；当全部采集图片都完成对比分析，但有效对比图片数量小于该叶片所采集的图片总数的80%时，认为该叶片无故障问题，对存在故障问题的叶片进行故障识别，并输出故障诊断结果。
PCT/CN2022/119571 2022-07-29 2022-09-19 一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法 WO2024021260A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210911977.3A CN115219515A (zh) 2022-07-29 2022-07-29 一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法
CN202210911977.3 2022-07-29

Publications (1)

Publication Number Publication Date
WO2024021260A1 true WO2024021260A1 (zh) 2024-02-01

Family

ID=83616652

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/119571 WO2024021260A1 (zh) 2022-07-29 2022-09-19 一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法

Country Status (2)

Country Link
CN (1) CN115219515A (zh)
WO (1) WO2024021260A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102539438A (zh) * 2011-12-02 2012-07-04 上海电机学院 风力发电机组叶片实时状态监测与故障诊断系统及方法
CN102706885A (zh) * 2012-05-15 2012-10-03 广东电网公司电力科学研究院 一种风力发电机组叶片损伤在线检测系统
US20180100489A1 (en) * 2015-04-24 2018-04-12 Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. Method and device for determining a position of defects or damage on rotor blades of a wind turbine in an installed state
CN209976699U (zh) * 2019-03-19 2020-01-21 内蒙古工业大学 一种风力机叶片故障智能监测装置
CN112720518A (zh) * 2020-12-22 2021-04-30 台州佳沃科技有限公司 风电机组叶片检修机器人
JP2021196360A (ja) * 2020-06-09 2021-12-27 エアロファシリティー株式会社 風力発電機ブレード点検装置、風力発電機ブレード点検システム及び風力発電機ブレード点検方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102539438A (zh) * 2011-12-02 2012-07-04 上海电机学院 风力发电机组叶片实时状态监测与故障诊断系统及方法
CN102706885A (zh) * 2012-05-15 2012-10-03 广东电网公司电力科学研究院 一种风力发电机组叶片损伤在线检测系统
US20180100489A1 (en) * 2015-04-24 2018-04-12 Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. Method and device for determining a position of defects or damage on rotor blades of a wind turbine in an installed state
CN209976699U (zh) * 2019-03-19 2020-01-21 内蒙古工业大学 一种风力机叶片故障智能监测装置
JP2021196360A (ja) * 2020-06-09 2021-12-27 エアロファシリティー株式会社 風力発電機ブレード点検装置、風力発電機ブレード点検システム及び風力発電機ブレード点検方法
CN112720518A (zh) * 2020-12-22 2021-04-30 台州佳沃科技有限公司 风电机组叶片检修机器人

Also Published As

Publication number Publication date
CN115219515A (zh) 2022-10-21

Similar Documents

Publication Publication Date Title
US7271894B2 (en) Imaging system for robotically inspecting gas turbine combustion components
CN107356339B (zh) 一种无人机光伏电站检测方法
CN109494617B (zh) 一种清障巡检机器人及其清障方法
CN111043964B (zh) 基于图像分析的土压平衡盾构机滚刀磨损监测装置及方法
CN102706885A (zh) 一种风力发电机组叶片损伤在线检测系统
CN102539438A (zh) 风力发电机组叶片实时状态监测与故障诊断系统及方法
CN112686886A (zh) 一种电力巡检系统及其设备故障诊断方法
CN111708380B (zh) 风电机组外观缺陷的检测方法、平台、无人机以及系统
CN110345992A (zh) 一种基于高温红外成像的垃圾焚烧电厂积灰监测方法及装置
CN112880837A (zh) 一种设备故障分析方法
CN113034599A (zh) 一种航空发动机的孔探检测装置和方法
CN114866744A (zh) 一种风电叶片自动追踪扫描探测系统与方法
CN212635748U (zh) 配电网巡检机器人
WO2024021260A1 (zh) 一种基于高速摄像技术的汽轮机叶片智能探伤系统及工作方法
CN107356200B (zh) 基于渣块轨迹的煤粉锅炉炉内落渣测量方法和系统
CN117193143A (zh) 一种基于视觉的火电厂吹灰器运行监控系统
CN212379309U (zh) 一种锅炉受热面外观缺陷在线检测与识别装置
CN105203315A (zh) 一种设在巡检车上的风机浆叶监测装置
CN116482119A (zh) 一种航空发动机孔探一次成像检查系统及方法
CN213152220U (zh) 一种槽式光热电站全天候监控系统
Zhang et al. Wind turbine blade defect image acquisition system
CN115938668A (zh) 一种中低压线缆工作状态检测系统
JP7023498B2 (ja) 太陽光発電パネルの検査装置
CN107621664A (zh) 一种用于太阳能采集区域的云遮预判装置与方法
KR200431176Y1 (ko) 고정판 상단부위의 전열관 내경 절단장치

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22952701

Country of ref document: EP

Kind code of ref document: A1