Hydraulic model hub discharge capacity measuring device and method based on machine vision
Through the hydraulic model hub discharge capacity measurement device based on machine vision, high-speed cameras and image processing technology are used to solve the problem of large water surface line measurement error in hydraulic model tests, and the accurate and rapid acquisition of water level data of multiple measurement points is achieved. , improves measurement accuracy and efficiency.
Patent Information
- Application Number
- CN202411273266.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-11-08
AI Technical Summary
In hydraulic model tests, the leakage capacity measurement has the problem of large measurement errors in the water surface line and it is difficult to quickly capture multiple measurement points at the same time. Traditional methods such as opening and ruler measurement have problems such as high reading errors and high cost and large measurement errors.
Using a hydraulic model hub discharge capacity measurement device based on machine vision, including a hub drainage gate model, a water surface line capture system and a water surface line image post-processing system, the water surface line image is collected through a high-speed camera, and the threshold segmentation method and contour line extraction algorithm are used. Identify and extract water surface line data to achieve accurate and rapid acquisition of water level data from multiple measurement points.
The downstream water level adjustment efficiency and the accuracy of the water level-flow relationship curve during the drainage capacity measurement process are improved, accurate and rapid measurement is achieved, and measurement errors are reduced.
Smart Images

Figure CN118913621A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a device and a method for measuring the discharge capacity of a hydraulic model hub based on machine vision, and belongs to the technical field of fluid pressure testing based on image data processing. Background Art
[0002] The cross-sectional model test in the hydraulic model test is mainly used to demonstrate the relevant hydraulic problems under the design scheme of the water conservancy hub. The discharge capacity curve is the basic characteristic curve of the hub dispatching. Verifying the discharge capacity of the hub under typical flow conditions is the most important part of the hydraulic model test results. The measurement of discharge capacity has very strict requirements on the measurement accuracy and density of the upstream and downstream water surface lines. The error of a few millimeters in the water surface line in the model can be converted into an error of tens of centimeters in the prototype, which has a great impact on the calculation of the hub discharge capacity. The current methods for measuring the water surface line are: (1) opening a hole at the bottom of the water tank, connecting the water body in the water tank to the measuring cylinder through a water pipe, and using a measuring needle to measure the water level of the measuring cylinder; (2) using a ruler to directly measure the distance between the water surface and the bottom of the water tank. The first method usually causes the water pipe to be blocked by impurities such as mud and sand in the water tank, resulting in a large reading error. In addition, the cost of opening holes and connecting water pipes is high, and it is difficult to realize the arrangement of a large number of measuring points along the water surface line. In the second method, the ruler extending into the water tank will affect the water level at the measuring point, and the inclination of the ruler will also lead to a large measurement error.
[0003] In order to overcome the problems of large water surface line measurement errors and difficulty in simultaneously and quickly capturing multiple measuring points in the measurement of the discharge capacity of hydraulic model test hubs, it is urgent to develop a set of accurate and fast measurement devices and methods. Summary of the invention
[0004] In order to solve the above problems, the present invention discloses a device and method for measuring the discharge capacity of a hydraulic model hub based on machine vision, and the specific technical solution is as follows: A device for measuring the discharge capacity of a hydraulic model hub based on machine vision includes a hub sluice model, a water surface line capture system, and a water surface line image post-processing system.
[0005] The hub sluice gate model includes a circulating water tank system, a sluice gate pier and a needle-type water level meter. The circulating water tank system includes an upstream reservoir, a water delivery pipeline, a tempered glass water tank, a downstream tail gate, and a downstream reservoir. A water pump, a valve and an electromagnetic flowmeter are arranged on the water delivery pipeline. The water delivery pipeline connects the upstream reservoir and the downstream reservoir, and the water in the downstream reservoir is pumped back to the upstream reservoir by a water pump. The downstream tail gate is set at the downstream port of the tempered glass water tank; the sluice gate pier is installed in the tempered glass water tank, and the water level of the upstream flow changes after passing through the sluice gate pier. The needle-type water level meter is placed on the outside of the tempered glass water tank to determine the water level of a single point on the water surface line of the test area as a reference height for adjusting the height of the high-speed camera;
[0006] The water surface line capture system includes a slide rail with a scale, two brackets, a laser level, a high-speed camera and a camera calibration object. The two brackets are installed on the slide rail so as to be movable up and down. The high-speed camera is placed on an upper bracket, and the laser level is placed on a lower bracket. The camera calibration object is attached to the outer wall of the tempered glass water tank, which is lower than the water surface. The laser level is used to determine the elevation of the camera calibration object, the high-speed camera is used to shoot the water surface line within the measurement range, and the camera calibration object is used to convert pixel coordinates and physical coordinates.
[0007] The water surface line image post-processing system is loaded with a water surface line recognition method and a water surface line data extraction method. The water surface line recognition method segments the image through a threshold segmentation method to identify the water body, air and camera calibration objects in the image. The water surface line data extraction method uses a contour line extraction algorithm to extract the contours of the water body and the camera calibration object.
[0008] Furthermore, a concrete foundation is constructed at the bottom of the tempered glass water tank, a bottom plate is mounted on the concrete foundation, and a sluice gate pier is arranged at a position of the tempered glass water tank close to the upstream end.
[0009] Furthermore, the sluice gate pier includes two piers, the space between the two piers is a gate hole, and the space between the two piers and the side wall of the tempered glass water tank on the corresponding side is half a gate hole each, which is used to simulate the discharge process of the two gate holes of the sluice gate. The two gate holes refer to the middle gate hole and half gate holes on both sides.
[0010] Furthermore, the gate chamber includes gate piers, gate holes and gate bottom plates. The horizontal bottom plate of the upstream section of the gate chamber is the gate bottom plate. The downstream side of the gate bottom plate is connected to the slope, and the slope section is called the connecting section. The horizontal bottom plate connected to the downstream of the slope is the energy dissipation pool bottom plate. Pressure measuring points are arranged along the longitudinal center line of the connecting section and the energy dissipation pool bottom plate. The bottom of the bottom plate is close to the side wall of the tempered glass water tank and is supported by a brick wall. The bottom of the longitudinal center line of the bottom plate is hollow, and the data transmission lines of the pressure measuring points are arranged.
[0011] Furthermore, the bottom of the needle-type water level meter is connected to the upstream and downstream bottoms of the water tank through a water pipe to obtain the water level of a single point in the upstream and downstream test areas as a reference height for adjusting the camera height.
[0012] Furthermore, the laser level is a surface light source, and the laser horizontal line formed on the side wall of the water tank is below the water surface. The camera calibration object is a rectangular white sticker (size w*h mm), and the lower edge of the sticker is aligned with the laser horizontal line and affixed to the side wall of the water tank.
[0013] The method for measuring the discharge capacity of a hydraulic section model hub based on the hydraulic model hub discharge capacity measuring device based on machine vision comprises the following steps: Step 1: Obtain different design discharge flows Q according to the hub discharge capacity operating table 设计 The design value of the upstream water level of the hub under the condition H 设计上游 and downstream water level design value H 设计下游 ; Step 2: Geometric scale based on the model , calculate the flow scale of the model , and the design discharge flow of the model test is obtained ; Step 3: Start the upstream water pump of the circulating water tank in the hub sluice model, and control the flow rate q in the water tank through the valve and electromagnetic flowmeter 试验 =q 设计 , the prototype flow rate corresponding to the model test flow rate is ; Step 4: Adjust the height of the laser level so that the laser horizontal line coincides with the height of the model gate bottom plate. Mark the scale on the slide rail as 0 mm. Then adjust the height of the laser level so that the laser horizontal line is located near the downstream water surface line. Align the lower edge of the camera calibration object with the laser horizontal line and stick it to the side wall of the water tank to obtain the distance h from the lower edge of the calibration object to the model gate bottom plate. 标定物 ; Step 5: Read the reading of the needle-type water level meter, adjust the high-speed camera to the height of the needle, complete the rough adjustment of the high-speed camera height, and continue to fine-tune the height and focal length of the high-speed camera according to the position of the water surface line and the camera calibration object in the high-speed camera window, so that the water surface line and the camera calibration object appear clearly in the window; Step 6: After obtaining the water surface line image, convert the image into a grayscale image, and use the threshold segmentation method to binarize the grayscale image to identify the air, water and camera calibration objects in the image; Step 7: Use the bwboundaries function in Matlab to obtain the water surface contour and the calibration object contour in the binary image, and directly obtain the pixel coordinates of each point on the contour line from the image; Step 8: Take a corner point of the camera calibration object as the coordinate origin, and the actual size of the camera calibration object as a reference, convert the pixel coordinate system in the image into the physical coordinate system, convert the pixel coordinates of each point on the water surface line into physical coordinates, and obtain the water level downstream of the hub through scale conversion. ; Step 9: Change the water level by adjusting the tailgate downstream of the circulating water tank ,until ; Step 10: Repeat steps 4 to 8 to measure the water level upstream of the hub ; Step 11: Repeat steps 3 to 10 to draw a series of Q 试验 -H试验上游 Scatter, test Q 试验 5 to 8 operating conditions should be selected in order from small to large within the design and operation range of the hub. 试验 The value of should include the minimum and maximum values of the design flow, and the discharge capacity curve of the hub should be obtained and compared with Q 设计 -H 设计上游 The curves are compared. If the H 试验上游 Both lower than H 设计上游 , indicating that the hub’s discharge capacity meets the requirements.
[0014] The beneficial effects of the present invention are: The present invention collects water surface line images through a high-speed camera, captures them from the images through machine vision technology, and quantitatively describes the water surface line. Compared with traditional needle water level meters, this device can accurately, quickly and synchronously obtain water level data of multiple measuring points within a certain range of the hydraulic model, thereby improving the efficiency of downstream water level regulation and the accuracy of drawing the water level-flow relationship curve during the discharge capacity measurement process. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 is a front view of the present invention; Figure 2 yes Figure 1 Right view of; Figure 3 It is the original picture of the water surface line photographed by the high-speed camera of the present invention; Figure 4 It is the grayscale image of the water surface line photograph of the present invention; Figure 5 is a histogram of pixels of different gray levels of the present invention; Figure 6 It is the water surface line image after binarization processing of the present invention; Figure 7 It is the water surface contour and calibration object contour (camera coordinate system) read from the binary image by the present invention; Figure 8 It is the water surface contour and calibration object contour (physical coordinate system) read from the binary image by the present invention; Fig. 9 is a hub discharge capacity curve designed and tested in an embodiment of the present invention; Fig.10 It is a flow chart of water surface line capture and extraction; Fig.11 It is a schematic diagram of the downstream tailgate structure of the present invention.
[0016] List of reference numerals: 1—tempered glass water tank, 2—gate pier, 3—probe type water level gauge, 4—slide rail, 5—bracket, 6—high-speed camera, 7—laser level, 8—water surface line, 9—camera calibration object, 10—water surface contour line, 11—camera calibration object contour line. DETAILED DESCRIPTION
[0017] The present invention is further explained below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following specific embodiments are only used to illustrate the present invention and are not used to limit the scope of the present invention.
[0018] Combined with Figure 1-2 It can be seen that the hydraulic model hub discharge capacity measurement device based on machine vision includes a hub sluice model, a water surface line capture system and a water surface line image post-processing system.
[0019] The following is an introduction to the specific structure of each system: The names of specific components involved include: tempered glass water tank 1, gate pier 2, needle-type water level meter 3, slide rail 4, bracket 5, high-speed camera 6, laser level 7, water surface line 8, camera calibration object 9, water surface contour line 10, camera calibration object contour line 11.
[0020] The hub sluice gate model includes a circulating water tank system, a sluice gate pier and a needle-type water level meter. The circulating water tank system includes an upstream reservoir, a water delivery pipeline, a tempered glass water tank, a downstream tail gate, and a downstream reservoir. The water delivery pipeline is arranged with a water pump, a valve and an electromagnetic flowmeter. The water delivery pipeline connects the upstream reservoir and the downstream reservoir, and the water in the downstream reservoir is pumped back to the upstream reservoir by a water pump. The downstream tail gate is set at the downstream port of the tempered glass water tank. The sluice gate pier is installed in the tempered glass water tank. The water level changes after the upstream flow passes through the sluice gate pier. The needle-type water level meter is placed on the outside of the tempered glass water tank to determine the water level of a single point on the water surface line of the area to be measured, which serves as a reference height for adjusting the camera height. The downstream tail gate adopts a grating tail gate, that is, it is composed of multiple gratings connected on both sides vertically. Each grating can be independently controlled to rotate around the vertical center line, or all gratings can be controlled to rotate synchronously through a master switch. The angle of the rotating grating adjusts the discharge flow to control the tail gate water level. For details, please refer to the attached Fig.11 .
[0021] A concrete foundation is built at the bottom of the tempered glass water tank, a bottom plate is arranged on the concrete foundation, and a sluice gate pier is arranged at a position of the tempered glass water tank close to the upstream end.
[0022] The sluice gate piers include two piers, the space between the two piers is a gate hole, and the space between the two piers and the side wall of the tempered glass water tank on the corresponding side is half a gate hole each, which is used to simulate the discharge process of the two gate holes of the sluice gate. The two gate holes refer to the middle gate hole and the half gate holes on both sides. The upstream and downstream vertical ends of the piers are set to arc shape.
[0023] The gate chamber includes gate piers, gate holes and gate bottom plates. The horizontal bottom plate of the upstream section of the gate chamber is the gate bottom plate. The downstream side of the gate bottom plate is connected to the slope, which is called the connecting section. The horizontal bottom plate connected to the downstream of the slope is the stilling pool bottom plate. Pressure measuring points are arranged along the longitudinal center line of the connecting section and the stilling pool bottom plate. The bottom of the bottom plate is supported by a brick wall close to the side wall of the tempered glass water tank. The bottom of the longitudinal center line of the bottom plate is hollow, and the data transmission lines of the pressure measuring points are arranged.
[0024] The bottom of the needle-type water level meter is connected to the upstream and downstream bottoms of the water tank through a rubber tube to obtain the water level of a single point in the upstream and downstream test areas, which serves as a reference height for height adjustment of the high-speed camera.
[0025] The water surface line capture system includes a slide rail with scale, two brackets, a laser level, a high-speed camera and a camera calibration object. The two brackets are installed on the slide rail so that they can move up and down. The high-speed camera is placed on the upper bracket, and the laser level is placed on the lower bracket. The camera calibration object is attached to the outer wall of the tempered glass water tank, which is lower than the water surface. The laser level is used to determine the elevation of the camera calibration object, and the high-speed camera is used to shoot the water surface line within the measurement range. The camera calibration object is used to convert pixel coordinates and physical coordinates. The two brackets are set one above and one below on the slide rail with scale. The scale on the slide rail is convenient for reading the height of the bracket at any time. A C-shaped notch is provided on the side where the bracket contacts the slide rail. The C-shaped notch is stuck on the slide rail. The slide rail is a column with a scale. The C-shaped notch and the slide rail cooperate with mechanical labyrinth contact. The bracket will not detach from the slide rail, and an adjusting nut is provided on the bracket. The nut end of the adjusting nut is located on the outside. The screw end passes through the bracket at the C-shaped notch and contacts the slide rail. When adjustment is required, loosen the adjusting nut and slide up and down. When fixation is required, tighten the adjusting nut, and the screw end of the adjusting nut squeezes the slide rail and fixes it. A base is also provided at the bottom of the slide rail, and a pulley is provided below the base. The base is often selected to be circular, rectangular or square, and the pulleys are evenly arranged at the bottom corners of the base or evenly dispersed around the circumference. It is convenient to push the position of the mobile water surface line capture system when testing tempered glass sinks in different positions.
[0026] The laser level is a surface light source. The laser horizontal line formed on the side wall of the tempered glass water tank is below the water surface. The camera calibration object is a rectangular white sticker (size w*h mm). The lower edge of the sticker is aligned with the laser horizontal line and affixed to the side wall of the tempered glass water tank.
[0027] The water surface line image post-processing system is loaded with a water surface line recognition method and a water surface line data extraction method. The water surface line recognition method uses a threshold segmentation method to segment the image and identify the water body, air and camera calibration object in the image. The water surface line data extraction method uses a contour line extraction algorithm to extract the contours of the water body and the camera calibration object.
[0028] The distance between the needle-type water level meter 3 and the scaled slide rail 4 can be adjusted according to the size of the site. The distance between the camera calibration object 9 and the water surface line 8 is adjusted according to the size of the camera window.
[0029] The method for measuring the discharge capacity of a hydraulic section model hub based on the hydraulic model hub discharge capacity measuring device based on machine vision comprises the following steps: Step 1: Obtain different design discharge flow rates Q according to the hub discharge capacity operating condition table (Table 1) provided by the design department 设计 The design value of the upstream water level of the hub under the condition H 设计上游 and downstream water level design value H 设计下游 ; Step 2: Geometric scale based on the model , calculate the flow scale of the model , and the design discharge flow of the model test is obtained ; Step 3: Open the upstream water pump of the circulating water tank (i.e., the tempered glass water tank) in the hub sluice gate model, and control the flow rate q in the water tank through the valve and electromagnetic flowmeter 试验 =q 设计 , the prototype flow rate corresponding to the model test flow rate is ; Step 4: Adjust the height of the laser level so that the laser horizontal line coincides with the height of the model gate bottom plate. Mark the scale on the slide rail as 0 mm. Then adjust the height of the laser level so that the laser horizontal line is located near the downstream water surface line. Align the lower edge of the camera calibration object with the laser horizontal line and stick it to the side wall of the water tank to obtain the distance h from the lower edge of the calibration object to the model gate bottom plate. 标定物 ; Step 5: Read the reading of the needle-type water level meter, adjust the high-speed camera to the height of the needle, complete the rough adjustment of the camera height, and continue to fine-tune the camera height and focal length according to the position of the water surface line and the camera calibration object in the camera window, so that the water surface line and the camera calibration object appear clearly in the window; Step 6: After obtaining the water surface line image, use the imread function to import the image into the MATLAB software, and use the rgb2gray function to convert the color image into a grayscale image (such as Figure 4 ); Step 7: Use the image threshold segmentation method to segment the grayscale image to distinguish the water, air and camera calibration objects in the image. The threshold is determined by manual selection, and the imhist function is used to analyze the histogram of different grayscale pixels in the grayscale image (such as Figure 5 ), select the gray value between the foreground and background of the image, and generally select the gray value at the bottom of the valley between the two peaks in the histogram as the threshold.
[0030] Step 8: Define a threshold variable and assign it to the threshold selected in step 7. Compare the grayscale of each pixel in the grayscale image with the threshold. If it is greater than the threshold, the pixel is set to 1 (i.e. white). Otherwise, the pixel is set to 0 (i.e. black). Thus, a binary image is obtained (e.g. Figure 6 as shown).
[0031] Step 9: Use the bwboundaries function to detect the boundaries in the binary image, obtain the boundary list B, use for k=1:length(B) to read each boundary in the list, define boundary=B{k}, and use plot(boundary(:,2),boundary(:,1),'w','LineWidth',2) to draw the boundaries in the image, where boundary(:,2) and boundary(:,1) represent the x and y coordinates of each point on the boundary, 'w' means the boundary line is white, and 'LineWidth',2 means the line width is 2. The obtained water surface contour line 10 and camera calibration object contour line 11 are as follows: Figure 7 shown.
[0032] Step 10: Click the mouse on the contour line graph to directly read the coordinates of the left end point a, the right end point b of the lower boundary of the calibration object and any point c on the water surface line in the pixel coordinate system (x a , y a ), (x b, y b ) and (x c, y c ). The pixel coordinate system is as follows Figure 7 Shown in the red coordinate system.
[0033] Step 11: Establish a physical coordinate system with point a as the origin (such as Figure 8 The green coordinate system in the middle shows the coordinates of point c. The conversion from pixel coordinate system to physical coordinate system requires two steps: (1) translation and flipping of the coordinate system. The coordinates of point c are converted to (x c -x a , -y c +y a ); (2) Determine the proportional relationship between pixel coordinates and physical coordinates. Use the size of the camera calibration object to calculate the proportional relationship between the camera coordinate system and the physical coordinate system: (x b -x a Finally, we get the pixel coordinate system point c (x c, y c ) in the physical coordinate system is ( ); Step 12: From step 4, we know that the distance between point a and the bottom plate of the model gate is h 标定物mm, then the distance between point c on the water surface and the bottom plate of the model gate is mm, and quickly obtain the coordinates of 10 points on the water surface line downstream of the hub in this way, and then take the average as the distance from the downstream water surface to the bottom plate of the model gate (h 试验水面 ), converted to the prototype downstream water level: ; Step 13: Change the water level by adjusting the tailgate downstream of the circulating water tank H 试验下游 ,until H 试验下游 = H 设计下游 ; Step 14: Repeat steps 4 to 12 to measure the water level upstream of the hub H 试验上游 ; Step 15: Repeat steps 3 to 14 to draw a series of Q 试验 -H 试验上游 Scatter points to obtain the discharge capacity curve of the hub and compare it with Q 设计 -H 设计上游 The curves are compared ( Fig. 9 ), the results show that the discharge capacity curve obtained from the test is lower than the designed discharge capacity curve, indicating that the designed hub discharge capacity meets the requirements.
[0034] The data of step one comes from the hub discharge capacity operating table provided by the design department. Table 1 is an embodiment of the hub discharge capacity operating table. The hub's designed discharge capacity is calculated by the design institute based on the structure of the sluice gate, the cross-sectional form of the upstream and downstream river channels, and hydrological data. The data is directly given by the design institute. The present invention verifies whether its designed discharge capacity can be achieved through model tests.
[0035] Table 1 Hub discharge capacity results
[0036] The flow chart of water surface line capture and extraction is as follows Fig.10 As shown, the specific method is as follows: Adjust the height of the laser level 7 so that the laser horizontal line is below the water surface, align the lower edge of the camera calibration object 9 with the laser horizontal line, and place it on the outer wall of the tempered glass tank.
[0037] Read the reading of the needle-type water level meter 3 to obtain the water level of a single measuring point within the measuring range, adjust the camera bracket to the needle height, complete the coarse adjustment of the camera height, and continue to fine-tune the height and focal length of the high-speed camera 6 according to the position of the water surface line and the calibration object in the camera window, so that the water surface line 8 and the camera calibration object 9 appear clearly in the window.
[0038] Convert the collected photos into grayscale images, analyze the grayscale of the image, and obtain the grayscale histogram as follows: Figure 5 As shown, select Figure 5 The gray value at the bottom of the valley between the two peaks is used as the threshold for image threshold segmentation. The binary image obtained after threshold segmentation is as follows Figure 6 As shown;
[0039] The bwboundaries function in Matlab is used to extract the water surface contour and the calibration object contour in the binary image. Figure 7 shown.
[0040] In the contour line diagram, click the contour line with the mouse to obtain the pixel coordinates of each point, establish a physical coordinate system with point a as the coordinate origin, and obtain the physical coordinates of any point c on the water surface line through conversion.
[0041] In order to test the effect of the present invention, a water surface line measurement experiment was carried out. The measurement results are as follows: Figure 7 As shown in the figure, the pixel coordinates of the vertex a at the lower left of the camera calibration object are (117,512), and the pixel coordinates of the vertex at the lower right are (416,512). As the length of the calibration object is 100 mm, the ratio between the pixel coordinates and the physical coordinates is 299:100. The pixel coordinates of a point c on the water surface line in the figure are (637,146), which is converted into physical coordinates of (173.9, 122.4). As the elevation of the laser horizontal line shows, the elevation of point a is 140 mm, and the elevation of point c is 262.4 mm. Similarly, the elevation of any point on the water surface line can be obtained.
[0042] The term “connection” as used in this application may mean a direct connection between components or an indirect connection between components via other components.
[0043] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.
Claims
1. A hydraulic model hub discharge capacity measurement device based on machine vision, characterized in that: Including hub sluice model, water surface line capture system and water surface line image post-processing system, The hub sluice gate model includes a circulating water tank system, a sluice gate pier and a needle-type water level meter. The circulating water tank system includes an upstream reservoir, a water delivery pipeline, a tempered glass water tank, a downstream tail gate and a downstream reservoir. The water delivery pipeline is arranged with a water pump, a valve and an electromagnetic flowmeter. The water delivery pipeline connects the upstream reservoir and the downstream reservoir, and the water in the downstream reservoir is pumped back to the upstream reservoir by a water pump. The downstream tail gate is arranged at the downstream port of the tempered glass water tank. The sluice gate pier is installed in the tempered glass water tank. The water level of the upstream flow changes after passing through the sluice gate pier. The needle-type water level meter is placed on the side of the outer side of the tempered glass water tank to determine the water level of a single point on the water surface line of the test area as a reference height for adjusting the height of the high-speed camera. The water surface line capture system includes a slide rail with a scale, two brackets, a laser level, a high-speed camera and a camera calibration object. The two brackets are installed on one side of the slide rail so as to be movable up and down. The high-speed camera is placed on an upper bracket, and the laser level is placed on a lower bracket. The camera calibration object is attached to the outer wall of the tempered glass water tank, which is lower than the water surface. The laser level is used to determine the elevation of the camera calibration object, the high-speed camera is used to shoot the water surface line within the measurement range, and the camera calibration object is used to convert pixel coordinates and physical coordinates. The water surface line image post-processing system is loaded with a water surface line recognition method and a water surface line data extraction method. The water surface line recognition method segments the image through a threshold segmentation method to identify the water body, air and camera calibration objects in the image. The water surface line data extraction method uses a contour line extraction algorithm to extract the contours of the water body and the camera calibration object.
2. The device for measuring the discharge capacity of a hydraulic model hub based on machine vision according to claim 1 is characterized in that: A concrete foundation is constructed at the bottom of the tempered glass water tank, a bottom plate is mounted on the concrete foundation, and a sluice gate pier is arranged at a position of the tempered glass water tank close to the upstream end.
3. The device for measuring the discharge capacity of a hydraulic model hub based on machine vision according to claim 2 is characterized in that: The sluice gate pier includes two sluice gate piers, the space between the two sluice gate piers is a gate hole, and the space between the two sluice gate piers and the inner wall of the tempered glass water tank on the corresponding side is half a gate hole each, which is used to simulate the discharge process of the two gate holes of the sluice gate. The two gate holes refer to the middle gate hole and the half gate holes on both sides, and the bottom plate under the gate hole is called the gate bottom plate.
4. The device for measuring the discharge capacity of a hydraulic model hub based on machine vision according to claim 3 is characterized in that: The gate pier, gate hole and gate bottom plate together constitute a gate chamber. The gate bottom plate located in the upstream section of the gate chamber is horizontally arranged, and the downstream side of the gate bottom plate is connected to a slope, and the slope section is called a connecting section. The horizontal bottom plate connected to the downstream of the slope is an energy dissipation pool bottom plate. Pressure measuring points are arranged along the longitudinal center line of the connecting section and the energy dissipation pool bottom plate. A brick wall is used for support below the bottom plate near the side wall of the tempered glass water tank. The bottom plate is hollow below the longitudinal center line, and data transmission lines for pressure measuring points are arranged.
5. The device for measuring the discharge capacity of a hydraulic model hub based on machine vision according to claim 1 is characterized in that: The bottom of the needle-type water level meter is connected to the upstream or downstream bottom of the water tank through a water pipe to obtain the water level of a single point in the upstream or downstream area to be measured, which serves as a reference height for adjusting the camera height.
6. The device for measuring the discharge capacity of a hydraulic model hub based on machine vision according to claim 1 is characterized in that: The laser level is a surface light source, and the laser horizontal line formed on the side wall of the tempered glass water tank is below the water surface. The camera calibration object is a rectangular white sticker, and the lower edge of the sticker is aligned with the laser horizontal line and affixed to the side wall of the tempered glass water tank.
7. A method for measuring the discharge capacity of a hydraulic section model hub based on the hydraulic model hub discharge capacity measuring device based on machine vision according to any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: Obtain different design discharge flows Q according to the hub discharge capacity operating table 设计 The design value of the upstream water level of the hub under the condition H 设计上游 and downstream water level design value H 设计下游 ; Step 2: Geometric scale based on the model , calculate the flow scale of the model , and the design discharge flow of the model test is obtained ; Step 3: Open the upstream water pump of the tempered glass water tank in the hub sluice model, and control the flow rate q in the tempered glass water tank through the valve and electromagnetic flowmeter 试验 =q 设计 , the prototype flow rate corresponding to the model test flow rate is ; Step 4: Adjust the height of the laser level so that the laser horizontal line coincides with the height of the bottom plate of the hub sluice gate model. Mark the scale on the slide rail as 0 mm. Then adjust the height of the laser level so that the laser horizontal line is located near the bottom of the downstream water surface line. Align the lower edge of the camera calibration object with the laser horizontal line and stick it to the side wall of the tempered glass water tank to obtain the distance h from the lower edge of the calibration object to the bottom plate of the hub sluice gate model. 标定物 ; Step 5: Read the reading of the needle-type water level meter, adjust the high-speed camera to the height of the needle, complete the rough adjustment of the high-speed camera height, and continue to fine-tune the height and focal length of the high-speed camera according to the position of the water surface line and the camera calibration object in the high-speed camera window, so that the water surface line and the camera calibration object appear clearly in the window; Step 6: After obtaining the water surface line image, convert the image into a grayscale image, and use the threshold segmentation method to binarize the grayscale image to identify the air, water and camera calibration objects in the image; Step 7: Use the bwboundaries function in Matlab to obtain the water surface contour and the calibration object contour in the binary image, and directly obtain the pixel coordinates of each point on the contour line from the image; Step 8: Take a corner point of the camera calibration object as the coordinate origin, and the actual size of the camera calibration object as a reference. Convert the pixel coordinate system in the image to the physical coordinate system, convert the pixel coordinates of each point on the water surface line to the physical coordinates, and obtain the water level downstream of the hub through scale conversion. ; Step 9: Change the water level by adjusting the tailgate downstream of the circulating water tank ,until ; Step 10: Repeat steps 4 to 8 to measure the water level upstream of the hub ; Step 11: Repeat steps 3 to 10 to draw a series of Q 试验 -H 试验上游 Scatter, test Q 试验 5 to 8 operating conditions should be selected in order from small to large within the design and operation range of the hub. 试验 The value of includes the minimum and maximum value of the design flow, and the discharge capacity curve of the hub is obtained, and is compared with Q 设计 -H 设计上游 The curve is compared. If the leakage flow rate under each working condition obtained by the test is 试验上游 Both lower than H 设计上游 , indicating that the hub’s discharge capacity meets the requirements.
