阅读说明：本技术 一种风电机组现场快速诊断方法 (On-site rapid diagnosis method for wind turbine generator ) 是由 张士龙 卢成志 于 2019-08-30 设计创作，主要内容包括：本发明公开了一种风电机组现场快速诊断方法,风电机组现场快速诊断装置包括现场拍照模块、图像识别模块、快速诊断模块、数据传输导线和电源；现场拍照模块通过拍摄工具在风力发电场站现场进行拍照,实现风电机组网络一体化诊断平台的原始数据需求,图像识别模块进行字符识别、定位、缺陷检测、图片分类,快速诊断模块根据图像识别模块的结果数据,使用快速诊断模块中的算法快速判断叶片的故障,数据传输导线为现场拍照模块、图像识别模块、快速诊断模块之间的数据通道,电源为所有设备提供电能。本发明结构简单、操作方便、实用性强,适合风电行业使用,具有灵敏度高、计算速度快、结果直观形象等优点。(The invention discloses a wind turbine generator on-site rapid diagnosis method.A wind turbine generator on-site rapid diagnosis device comprises an on-site photographing module, an image recognition module, a rapid diagnosis module, a data transmission lead and a power supply; the on-site photographing module photographs on the site of the wind power plant station through a photographing tool to meet the original data requirement of the wind turbine network integrated diagnosis platform, the image recognition module performs character recognition, positioning, defect detection and picture classification, the rapid diagnosis module uses an algorithm in the rapid diagnosis module to rapidly judge the fault of the blade according to the result data of the image recognition module, the data transmission lead is a data channel among the on-site photographing module, the image recognition module and the rapid diagnosis module, and the power supply provides electric energy for all equipment. The invention has the advantages of simple structure, convenient operation, strong practicability, high sensitivity, high calculation speed, visual and visual result and the like, and is suitable for being used in the wind power industry.)

1. A wind turbine on-site rapid diagnosis method is characterized in that a wind turbine on-site rapid diagnosis device comprises an on-site photographing module (1), an image recognition module (2), a rapid diagnosis module (3), a data transmission wire (4) and a power supply (5); the on-site photographing module (1) is used for photographing on the site of the wind power plant station through a photographing tool, the original data requirement of the wind turbine generator network integrated diagnosis platform is met, the image recognition module (2) is used for character recognition, positioning, defect detection and picture classification, the rapid diagnosis module (3) is used for rapidly judging the faults of the blades according to the result data of the image recognition module (2) by using an algorithm in the rapid diagnosis module (3), the data transmission lead (4) is used for the on-site photographing module (1), the image recognition module (2) and a data channel between the rapid diagnosis modules (3), and the power supply (5) is used for providing electric energy for all equipment.

2. The wind turbine generator system on-site rapid diagnosis method as claimed in claim 1, wherein AiDitron in the image recognition module (2) is software developed for machine vision based on artificial intelligence deep learning, and AiDitron artificial intelligence software/intelligent camera is not required to be programmed and can be trained in a self-learning mode.

Technical Field

The invention relates to a field rapid diagnosis method for a wind turbine generator, and belongs to the field of wind power fault diagnosis.

Background

The wind power generation comprises a wind generating set, a tower frame for supporting the generating set, a storage battery charging controller, an inverter, an unloader, a grid-connected controller, a storage battery pack and the like; the wind generating set comprises a wind wheel, a gear box and a generator; the wind wheel comprises blades, a hub, a reinforcing member and the like.

The maintenance of the unit can be divided into first-level maintenance, second-level maintenance and third-level maintenance (namely first-level maintenance, second-level maintenance and third-level maintenance). One-off, namely, the found faults are quickly eliminated, and damaged parts are quickly and timely replaced or repaired, which is very important for prolonging the service life of the unit. Two guaranties, which means that the unit operates for about one year. The method is characterized in that the fan blades, the steering device and other parts are disassembled, putty is cleaned, the parts are replaced or repaired if damaged, and proper amount of calcium-based lubricating oil (butter) is added to each bearing and the movable part. The three-protection means that the unit is operated for three to five years, one-time comprehensive inspection and successive protection is carried out, and if parts are damaged, the parts are replaced or repaired. The long-term operation of the fan needs maintenance, but the maintenance can not guarantee that the fan does not break down and influence normal power generation.

The reasons for the generation of fan faults are various: 1. imperfect design, 2, natural causes, 3, improper operation and maintenance, 4, out of control, 5, lack of preventive maintenance, etc.

The fan can generate faults and a fan fault diagnosis device is needed. The fan system fault diagnosis is to judge the running state and abnormal condition of the fan system and provide basis for system fault recovery according to the judgment made by the diagnosis. The method is characterized in that a fan system is required to be detected for fault diagnosis, when a system fault occurs, the type, the fault position and the reason of the fault are diagnosed, and a solution is finally given to realize fault removal.

The performance indicators for evaluating a failure diagnosis apparatus are:

1) timeliness of fault detection: the method refers to the capability of a fault diagnosis system to detect the fault in the shortest time after the system has the fault. The shorter the time from the occurrence of the fault to the detection of the fault, the better the timeliness of the fault detection.

2) Sensitivity of early detection: refers to the detection capability of the fault diagnosis system for tiny fault signals. The smaller the fault signal that the fault diagnosis system can detect, the higher the sensitivity of its early detection.

3) False alarm rate and missing report rate of fault: the false alarm indicates that the system fails to go out but is detected by mistake to have a fault; a false positive is a system failure that is not detected. A reliable fault diagnosis system should minimize the false alarm rate and the false negative rate as much as possible.

4) Fault isolation capability: refers to the ability of the diagnostic system to distinguish between different faults. The stronger the fault separation capability is, the stronger the distinguishing capability of the diagnosis system on different faults is, and the more accurate the fault positioning is.

5) Fault identification capability: refers to the ability of the diagnostic system to identify fault magnitude and time-varying characteristics. The higher the fault identification capability is, the more accurate the diagnosis system identifies the fault, and the more beneficial the fault evaluation and maintenance are.

6) Robustness: the method is characterized in that the fault diagnosis task is correctly completed by a diagnosis system under the condition of noise, interference and the like, and meanwhile, the capabilities of low false alarm rate and low missing report rate are kept. The more robust the reliability is, the higher the reliability of the diagnostic system is.

7) Self-adaptive capacity: the fault diagnosis system has self-adaptive capacity for the changed measured object and can fully utilize new information generated by the change to improve the fault diagnosis system.

In practical application, the performance indexes need to be analyzed and judged according to actual conditions, which performances are primary and which are secondary, then the diagnosis method is analyzed, and a final diagnosis scheme is obtained after proper selection and selection.

During routine operation and maintenance of the fan, the fault of the fan is often difficult to find, and a wind turbine generator field rapid diagnosis device and a diagnosis method are urgently needed. If daily maintenance is carried out in place, high maintenance cost in the future can be avoided, and economic loss caused by shutdown can be reduced.

Disclosure of Invention

The invention aims to overcome the defects in the prior art and provides a wind turbine generator on-site rapid diagnosis method which is reasonable in design, simple in system and has operability.

The technical scheme adopted by the invention for solving the problems is as follows: a wind turbine on-site rapid diagnosis method is characterized in that a wind turbine on-site rapid diagnosis device comprises an on-site photographing module, an image recognition module, a rapid diagnosis module, a data transmission lead and a power supply; the on-site photographing module photographs on the site of the wind power plant station through a photographing tool to meet the original data requirement of the wind turbine network integrated diagnosis platform, the image recognition module performs character recognition, positioning, defect detection and picture classification, the rapid diagnosis module uses an algorithm in the rapid diagnosis module to rapidly judge the fault of the blade according to the result data of the image recognition module, the data transmission lead is a data channel among the on-site photographing module, the image recognition module and the rapid diagnosis module, and the power supply provides electric energy for all equipment.

Compared with the prior art, the invention has the following advantages and effects:

1. the wind power generation system is simple in structure, compact in system, strong in practicability, suitable for being used in the wind power industry, and has operability, and practice proves that the method is a good method.

2. The operation is convenient.

3. The method has the advantages of high sensitivity, high calculation speed, visual and visual result and the like.

4. The novel design, the test analysis is rapid, and the graph line is clear.

5. The comparability is strong.

6. And collecting and storing multiple parameters.

7. Has wide applicability.

Drawings

Fig. 1 is a schematic structural diagram of a wind turbine on-site rapid diagnosis device in an embodiment of the invention.

In the figure: the system comprises a field photographing module 1, an image recognition module 2, a rapid diagnosis module 3, a data transmission lead 4 and a power supply 5.

Detailed Description

The present invention will be described in further detail below by way of examples with reference to the accompanying drawings, which are illustrative of the present invention and are not to be construed as limiting the present invention.

Referring to fig. 1, the wind turbine generator on-site rapid diagnosis device in the embodiment includes an on-site photographing module 1, an image recognition module 2, a rapid diagnosis module 3, a data transmission wire 4, a power supply 5, and the like.

Device use parameter of wind turbine generator on-site rapid diagnosis device

Operating temperature	0-50	℃
			Device power supply	220	AC
Weather conditions for live filming	Weather without adverse light, rain and snow

The on-site photographing module 1: the module has the function of shooting on the site of the wind power generation station through a special shooting tool to meet the original data requirement of the wind turbine network integrated diagnosis platform.

The image recognition module 2: AiDitron is a piece of software developed for machine vision based on artificial intelligence deep learning.

The main functions are as follows:

1. character recognition

2. And (6) positioning.

3. And detecting the defects.

4. And (5) classifying the pictures.

The AiDitron artificial intelligence software/intelligent camera can be trained in a self-learning mode like a person without programming, has powerful functions and has the advantages that:

1. the accuracy is high: in the field of industrial detection, the accuracy is infinitely close to 100% through the learning and continuous repetition of big data.

2. The problem of difficulty is solved: one parameter is adjusted in AiDitron artificial intelligence software, and the other parameter is not changed, so that the detection requirements of all defects can be met.

3. A short time may result: on items with short time requirements, as long as there are enough pictures and identification is performed, in principle, one day can achieve the desired result.

4. The later maintenance is convenient: the traditional algorithm has problems on site, a desired result cannot be detected, and a software programming technician needs to go to a production site to debug software. The AiDitron artificial intelligence software can directly learn the undetected pictures again in the production field to achieve the desired result.

5. The operating threshold is low: people without programming base can use the software to quickly detect and identify complex defects.

The detection results achieved are as follows:

realized technical index achievement

Index name	Index parameter	Unit of	Remarks for note
				Minimum detected defect size	5	mm	Size of cracks and damage
Precision of defect location	50	cm	Positional accuracy of defects
				Color of picture	Black and white
Shooting distance	65-80	Rice and its production process
				Monitoring array resolution	12	Mp	Colour display
High definition array resolution	72	Mp	Black and white
				Shooting range	74*10	m	65m shooting distance
Shooting range	86*12	m	Shooting distance of 70m
				Shooting range	99*13	m	80m shooting distance

The rapid diagnosis module 3: and (4) rapidly judging the fault of the blade by using an algorithm in the rapid diagnosis module 3 according to the result data of the image recognition module 2.

Data transmission wire 4: the data channel among the on-site photographing module 1, the image recognition module 2, the rapid diagnosis module 3 and the like is formed.

Power supply 5: all devices are supplied with electrical energy.

Although the present invention has been described with reference to the above embodiments, it should be understood that the scope of the present invention is not limited thereto, and that various changes and modifications can be made by those skilled in the art without departing from the spirit and scope of the present invention.