Hydropower station pressing plate state inspection equipment and inspection method

Abstract

The invention discloses hydropower station pressing plate state inspection equipment which comprises a machine body, a plurality of driving devices are arranged on the machine body, mechanical arms rotating in the vertical direction are installed on the driving devices, a plurality of wings are rotationally connected to the mechanical arms, and a shooting device used for collecting and recording pressing plate information is installed on the machine body. The system can be suitable for high-precision shooting of pressing plate images at different positions such as far, near, high and low positions, and the inspection efficiency is improved to a great extent. According to the hydropower station pressing plate state inspection method disclosed by the invention, pressing plates at different positions and at different heights are shot through mobile inspection equipment, and original images of the pressing plates are obtained; the position of the pressing plate is identified by calculating the brightness value proportion of the three channels R, G and B of the pixels of the original image; the pressing plate position is combined, the pixel gradient direction of the pressing plate edge image is calculated, the pressing plate state is identified through gradient direction clustering, all-directional monitoring of the pressing plate state in different environments can be achieved, and support is provided for hydropower station safety.

Description

Technical Field

[0001] The present invention belongs to the technical field of health monitoring of hydropower station facilities and equipment, and in particular relates to hydropower station pressure plate status inspection equipment, and also relates to a hydropower station pressure plate status inspection method. Background Art

[0002] Hydropower has the advantages of flexibility and high efficiency, and is the main direction of future clean energy development. As an important relay protection device in hydropower stations, the correctness of the pressure plate state is crucial to the safe and stable operation of hydropower stations. At the same time, the monitoring of the pressure plate state is also an important part of the daily inspection of hydropower stations.

[0003] At present, the status of the pressure plate of a hydropower station is often realized by traditional manual inspection, that is, the operation and maintenance personnel record the status of the pressure plate. However, this method has the disadvantages of high labor cost and low efficiency, which are specifically reflected in the following aspects: First, the hydropower station operates 24 hours a day, and the status of the pressure plate needs to be inspected many times a day, and the inspection tasks are many; second, with the intelligent development of hydropower stations, a large number of relay protection equipment are put into operation, the number of pressure plates is increasing, and the workload of a single inspection is large; third, due to visual fatigue, mental slackness and other reasons, the wrong check and missed check, as well as the influence of different inspection personnel's experience, limit the quality of manual inspection, and in serious cases, lead to hidden dangers in the secondary circuit.

[0004] In recent years, with the improvement of the intelligence level of hydropower stations, especially the development of image processing technology, pressure plate recognition technology using image processing has been gradually applied to practice. However, there are still the following problems for the recognition of pressure plate status in hydropower stations: First, the current image-based pressure plate status recognition method mainly adopts deep learning algorithms such as neural networks. This algorithm requires the use of a large number of samples to train the model, with high training costs and weak remote migration capabilities; Second, the current image-based pressure plate status recognition method is mainly used in substations and is used in conjunction with quadruped robots. However, considering the small space of the hydropower station and the pressure plates at different positions such as high, low, far and near, it is necessary to use auxiliary means such as lens tilt shooting, continuous zoom and two-stage monitoring combined with positioning identification. The acquired images have problems such as angle deviation, lighting changes, and blur, which directly affect the recognition accuracy of the pressure plate status. Summary of the invention

[0005] The first purpose of the present invention is to provide a hydropower station pressure plate status inspection device, which solves the technical problems of low efficiency, time-consuming and labor-intensive, large workload, high intensity and low inspection efficiency of the existing manual pressure plate daily inspection of the hydropower station.

[0006] The second object of the present invention is to provide a method for inspecting the status of a pressure plate in a hydropower station.

[0007] The first technical solution adopted by the present invention is that the hydropower station pressure plate state inspection equipment includes a fuselage, a plurality of driving devices are arranged on the fuselage, a mechanical arm rotating along the vertical direction is installed on the driving device, a plurality of wings are rotatably connected to the mechanical arm, and a shooting device for collecting and recording pressure plate information is installed on the fuselage; The wing includes a blade and a circular wheel, and two ends of the blade are fixedly connected to the inner wall of the circular wheel and are tangent to the circular wheel.

[0008] The first technical solution of the present invention is also characterized in that: The blades are made of carbon fiber composite materials, and the circular wheels are made of engineering plastics or metal alloys.

[0009] The driving device comprises a motor and a reducer which are connected to each other, the reducer is connected to a transmission mechanism, and the transmission mechanism is connected to the mechanical arm.

[0010] The shooting device includes a high-definition camera, an image sensor and a data storage module; the high-definition camera is used to capture the image information of the pressure plate, the image sensor converts the optical signal into an electrical signal, and the processed data is stored in the data storage module.

[0011] The second technical solution adopted by the present invention is a method for inspecting the status of a pressure plate in a hydropower station, using the above-mentioned inspection device for inspecting the status of a pressure plate in a hydropower station, comprising the following steps: S1: The mobile inspection equipment takes photos of the pressure plates at different positions and heights to obtain the original images of the pressure plates; S2: Identify the position of the pressing plate by calculating the ratio of the brightness values ​​of the three channels R, G, and B of the original image pixels; S3: Combined with the platen position, the pixel gradient direction of the platen edge image is calculated, and the platen state is identified by clustering the gradient direction.

[0012] The second technical solution of the present invention is also characterized in that: S1 is specifically: For low-position platens without fences, the walking mode is used to capture images of the platen status; For middle and high-layer platens, switch to flight mode to capture images of the platen status; For pressure plates in narrow areas blocked by fences, the flight mode is used to capture images of the pressure plate status; Among them: when the flying mode is adopted, the wings rotate in the horizontal direction; when the walking mode is adopted, the wings rotate in the vertical direction.

[0013] S2 is specifically: Calculate the position in the original image The sum of the brightness values ​​of the three channels R, G, and B of the pixel at the position is:

[0014] Secondly, calculate the ratio of the brightness values ​​of the three channels R, G, and B to the total brightness value:

[0015] Based on the location obtained The position of the pressure plate is identified by the ratio of the brightness values ​​of the three channels R, G, and B of the pixel to the total brightness value:

[0016] In the above formula: , , The location The ratio of the brightness values ​​of the three channels R, G, and B of the pixel at the position to the total brightness value; S3 is specifically: Construct an image gradient detection operator to calculate the pixel gradient of the platen edge image, namely:

[0017] Where: and are the gradients of the platen edge image pixels in the horizontal and vertical directions, f is the image pixel matrix, represents the convolution operation of the gradient operator and the image, Hx and Hy is the gradient operator in the horizontal and vertical directions, expressed as:

[0018] The gradient amplitude of the image pixels is calculated using the gradients in the horizontal and vertical directions, and the edge contour of the platen is identified using the pixel gradient amplitude, that is:

[0019] Calculate the gradient direction of the platen contour pixel, that is:

[0020] The K-means clustering algorithm is used to cluster the pixel gradient direction of the edge of the pressure plate contour:

[0021] Where: Ex is the clustering result of the pixel gradient direction of the platen contour edge. When it is the minimum value, it is the best clustering. v is the type of clustering, n is the number of image pixels, is the mean value of the pixel gradient direction at the edge of the platen contour; Based on the platen edge pixel clustering results, considering the edge pixel gradient relationship between the closed and open states, we can get:

[0022] Finally, high-precision recognition of the platen state is achieved through clustering results of image pixel gradient directions.

[0023] Compared with the prior art, the present invention has the following beneficial effects: The hydropower station pressure plate state inspection equipment of the present invention can simultaneously adapt to the high-precision shooting of pressure plate images at different positions such as far, near, high and low, thereby greatly improving the inspection efficiency; at the level of state monitoring methods, according to the pressure plate state characteristics, different types of pressure plate position recognition based on pixel brightness ratio analysis and pressure plate state recognition methods based on pixel gradient direction clustering are proposed, which can realize all-round monitoring of pressure plate states under different environments and provide support for the safety of hydropower stations; Based on the differences between different types of pressure plates in different color channels and the differences in shapes of different pressure plates in different states, the present invention proposes a method for identifying the positions of different types of pressure plates by pixel brightness ratio analysis and a method for identifying the states of pressure plates based on pixel gradient direction clustering, which has the advantages of high monitoring accuracy and strong migration ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a schematic diagram of the flow of the method for inspecting the status of a pressure plate of a hydropower station according to the present invention; Figure 2a It is a schematic diagram of switching between walking and flying models of a composite wheeled robot in the hydropower station pressure plate status inspection equipment of the present invention; Figure 2b yes Figure 2a A top view of Figure 3 It is a schematic diagram of the process flow of the pressure plate status identification method in the hydropower station pressure plate status inspection method of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0026] Example 1 like Figure 2a and Figure 2bAs shown, the hydropower station pressure plate state inspection device disclosed in the present invention comprises a fuselage, a plurality of driving devices are arranged on the fuselage, a mechanical arm that rotates along the vertical direction is installed on the driving device, a plurality of wings are rotatably connected to the mechanical arm, and a shooting device for collecting and recording pressure plate information is installed on the fuselage; The wing includes a blade and a circular wheel. Both ends of the blade are fixedly connected to the inner wall of the circular wheel and are tangent to the circular wheel. The blade is made of carbon fiber composite material, and the circular wheel is made of engineering plastic or metal alloy.

[0027] In this embodiment, both ends of the paddle are fixedly connected to the inner wall of the circular wheel, and the paddle is tangent to the circular wheel. This structural design enables the paddle to generate auxiliary propulsion or direction adjustment force when the inspection equipment moves, while the circular wheel is used for moving and walking. The material is selected from engineering plastics or metal alloys with high wear resistance and high strength to ensure stable and reliable operation in the complex hydropower station environment.

[0028] In this embodiment, the blades and circular wheels can be switched between walking and flying modes by rotating the robotic arm according to the scenario. In the walking mode, the wings rotate in the vertical direction and are converted into wheels for walking. In the flying mode, the wings rotate in the horizontal direction and are converted into wings for flying.

[0029] Example 2 On the basis of Example 1, the driving device includes a motor and a reducer connected to each other, the reducer is connected to a transmission mechanism, and the transmission mechanism is connected to the mechanical arm. The shooting device includes a high-definition camera, an image sensor and a data storage module; the high-definition camera is used to capture the image information of the pressure plate, the image sensor converts the optical signal into an electrical signal, and the processed data is stored in the data storage module.

[0030] In this embodiment, after the motor reduces the speed and increases the torque through the speed reducer, it drives the mechanical arm to achieve precise vertical rotation through the transmission mechanism, so that the mechanical arm can flexibly adjust its position to meet the inspection needs of pressure plates in different positions. The transmission mechanism can change the angle of the wing relative to the mechanical arm through motor drive or manual adjustment according to the actual inspection situation, thereby optimizing the movement and inspection effect of the equipment. The high-definition camera is responsible for capturing the image information of the pressure plate, and the image sensor converts the optical signal into an electrical signal. The processed data is stored in the data storage module, which has the characteristics of large-capacity storage and fast reading and writing, and can meet the storage needs of long-term and large amounts of data. In actual applications, a battery is installed on the fuselage, and the battery powers the motor and the shooting device.

[0031] Example 3 like Figure 1 and Figure 3As shown, the present invention also discloses a method for inspecting the status of a pressure plate in a hydropower station, which uses the inspection device for inspecting the status of a pressure plate in a hydropower station of Embodiment 1-2, and includes the following steps: S1: The mobile inspection equipment takes photos of the pressure plates at different positions and heights to obtain the original images of the pressure plates; S2: Identify the position of the pressing plate by calculating the ratio of the brightness values ​​of the three channels R, G, and B of the original image pixels; S3: Combined with the platen position, the pixel gradient direction of the platen edge image is calculated, and the platen state is identified by clustering the gradient direction.

[0032] Example 4 Based on Example 3, S1 specifically includes: for a low-position pressure plate without a fence, using a walking mode to capture an image of the pressure plate state; For middle and high-layer platens, switch to flight mode to capture images of the platen status; For pressure plates in narrow areas blocked by fences, the flight mode is used to capture images of the pressure plate status; Among them: when the flying mode is adopted, the wings rotate in the horizontal direction; when the walking mode is adopted, the wings rotate in the vertical direction.

[0033] For pressure plates in different positions and environments, the walking mode and flying mode are used for image capture, which can quickly and specifically obtain pressure plate images. For example, the walking mode is used for low-position unfenced pressure plates, and the flying mode is used for pressure plates in middle and high layers and those blocked by fences and in narrow areas. This avoids the tedious process of multiple traversals of pressure plates in different positions during manual inspection, greatly improves inspection efficiency, and effectively solves the problem of multiple tasks and heavy workload for manual inspection.

[0034] By automatically switching modes to obtain platen images, the large amount of manpower required for manual inspections is reduced, thus reducing labor costs. At the same time, compared with the high cost of deep learning algorithms that require a large number of samples to train models, this solution achieves platen status monitoring through reasonable mode design, without relying on a high-cost training process, thus reducing monitoring costs overall.

[0035] Selecting appropriate image capture modes according to different locations and environments can avoid image acquisition problems caused by the special environment of the hydropower station to a certain extent. For example, for pressure plates in different positions, different flying or walking modes can be used to better adapt to the situation of narrow space, high and low, near and far positions, and reduce the impact of angle deviation, lighting changes, blur and other problems on the accuracy of pressure plate state recognition, thus overcoming the defects of existing image processing and recognition technology when applied to hydropower stations.

[0036] Example 5 Based on Example 3, S2 is specifically: Considering the differences in color between different types of pressure plates, the positions of different types of pressure plates are identified by calculating the ratio of the brightness values ​​of the three channels of image pixels R, G, and B. Considering that the pressure plates are divided into three types: red, yellow, and gray, and the color differences between different types of pressure plates and the surrounding environment, the distribution positions of different types of pressure plates are realized and the positions are calculated. The sum of the brightness values ​​of the three channels R, G, and B of the pixel at the position is:

[0037] Secondly, calculate the ratio of the brightness values ​​of the three channels R, G, and B to the total brightness value:

[0038] Then, the color difference of different types of platens is considered to realize the recognition of pixel positions, that is:

[0039] In the above formula: , , The location The ratio of the brightness values ​​of the three channels R, G, and B of the pixel at the position to the total brightness value; At this point, different types of pressure plate positions can be distinguished.

[0040] Example 6 On the basis of Example 3, S3 is specifically: On the basis of determining the pixel position coordinates of different pressure plates, considering the differences in spatial morphology when the pressure plate is in different states (pressure plate closed and open), the pressure plate state recognition is realized by calculating the gradient direction of the pressure plate edge image pixel and using the gradient direction clustering method. Specifically: First, we construct an image gradient detection operator and calculate the pixel gradient of the pressure plate image based on the position distribution of the pressure plate, namely:

[0041] Where: and are the gradients of image pixels in the horizontal and vertical directions, respectively. f is the image pixel matrix, represents the convolution operation of the gradient operator and the image, Hx and Hy The gradient operators in the horizontal and vertical directions used in the present invention are expressed as:

[0042] Secondly, the gradient amplitude of the image pixels is calculated using the gradients in the horizontal and vertical directions, and the edge contour of the platen is identified using the pixel gradient amplitude, that is:

[0043] Furthermore, the gradient direction of the platen contour pixels is calculated, namely:

[0044] Finally, considering the differences in the gradient directions of the edge pixels of the pressure plate under different states, the pressure plate state recognition is realized by clustering the gradient directions of the edge pixels of the pressure plate contour. Specifically, the K-means clustering algorithm is used to realize the pressure plate state recognition. The formula is as follows:

[0045] Where: Ex is the clustering result of the pixel gradient direction of the platen contour edge. When it is the minimum value, it is the best clustering. v is the type of clustering. In the present invention, v =2, that is, two states: closed and open. n is the number of image pixels, is the mean value of the pixel gradient direction at the edge of the platen contour.

[0046] Based on the platen edge pixel clustering results, considering the edge pixel gradient relationship between the closed and open states, we can get:

[0047] Finally, high-precision recognition of the platen state is achieved through clustering results of image pixel gradient directions.

[0048] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0049] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0050] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. Hydropower station pressure plate status inspection equipment, characterized in that: It includes a fuselage, on which are disposed a plurality of driving devices, on which are mounted a mechanical arm that rotates in a vertical direction, on which are rotatably connected a plurality of wings, and on which is mounted a photographing device for collecting and recording information of the pressure plate; The wing includes a blade and a circular wheel, and two ends of the blade are fixedly connected to the inner wall of the circular wheel and are tangent to the circular wheel.

2. The hydropower station pressure plate status inspection equipment according to claim 1 is characterized in that: The blades are made of carbon fiber composite materials, and the circular wheels are made of engineering plastics or metal alloys.

3. The hydropower station pressure plate status inspection equipment according to claim 1 is characterized in that: The driving device comprises a motor and a reducer which are connected to each other, the reducer is connected to a transmission mechanism, and the transmission mechanism is connected to the mechanical arm.

4. The hydropower station pressure plate status inspection equipment according to claim 1 is characterized in that: The shooting device includes a high-definition camera, an image sensor and a data storage module; the high-definition camera is used to capture the image information of the pressure plate, the image sensor converts the optical signal into an electrical signal, and the processed data is stored in the data storage module.

5. A method for inspecting the status of a pressure plate in a hydropower station, using the device for inspecting the status of a pressure plate in a hydropower station as claimed in any one of claims 1 to 4, characterized in that: The following steps are involved: S1: The mobile inspection equipment takes photos of the pressure plates at different positions and heights to obtain the original images of the pressure plates; S2: Identify the position of the pressing plate by calculating the ratio of the brightness values ​​of the three channels R, G, and B of the original image pixels; S3: Combined with the platen position, the pixel gradient direction of the platen edge image is calculated, and the platen state is identified by clustering the gradient direction.

6. The method for inspecting the status of a pressure plate of a hydropower station according to claim 5, characterized in that: The S1 is specifically: For low-position platens without fences, the walking mode is used to capture images of the platen status; For middle and high-layer platens, switch to flight mode to capture images of the platen status; For pressure plates in narrow areas blocked by fences, the flight mode is used to capture images of the pressure plate status; Among them: when the flying mode is adopted, the wings rotate in the horizontal direction; when the walking mode is adopted, the wings rotate in the vertical direction.

7. The method for inspecting the status of a pressure plate of a hydropower station according to claim 6, characterized in that: The S2 is specifically: Calculate the position in the original image The sum of the brightness values ​​of the three channels R, G, and B of the pixel at the position is: Secondly, calculate the ratio of the brightness values ​​of the three channels R, G, and B to the total brightness value: Based on the location obtained The position of the pressure plate is identified by the ratio of the brightness values ​​of the three channels R, G, and B of the pixel to the total brightness value: In the above formula: , , The location The ratio of the brightness values ​​of the three channels R, G, and B of the pixel at the position to the total brightness value.

8. The method for inspecting the status of a pressure plate of a hydropower station according to claim 7, characterized in that: The S3 is specifically: Construct an image gradient detection operator to calculate the pixel gradient of the platen edge image, namely: Where: and are the gradients of the platen edge image pixels in the horizontal and vertical directions, f is the image pixel matrix, represents the convolution operation of the gradient operator and the image, Hx and Hy is the gradient operator in the horizontal and vertical directions, expressed as: The gradient amplitude of the image pixels is calculated using the gradients in the horizontal and vertical directions, and the edge contour of the platen is identified using the pixel gradient amplitude, that is: Calculate the gradient direction of the platen contour pixel, that is: The K-means clustering algorithm is used to cluster the pixel gradient direction of the edge of the pressure plate contour: Where: Ex is the clustering result of the pixel gradient direction of the platen contour edge. When it is the minimum value, it is the best clustering. v is the type of clustering, n is the number of image pixels, is the mean value of the pixel gradient direction at the edge of the platen contour; Based on the platen edge pixel clustering results, considering the edge pixel gradient relationship between the closed and open states, we can get: Finally, high-precision recognition of the platen state is achieved through clustering results of image pixel gradient directions.

Images