Method and system for detecting scaling and blocking degree of water inlet filter screen of drainage pump of hydropower station

Abstract

The invention relates to the technical field of water inlet filter screen pore scaling detection, and discloses a water inlet filter screen scaling and blocking degree detection method and system for a hydropower station drainage pump, and the method comprises the steps: obtaining a hydropower station drainage pump image, and extracting a water inlet filter screen edge contour image from the hydropower station drainage pump image based on a Canny edge detection algorithm; separating a filter screen pore color component image from the water inlet filter screen edge contour image through an HSV color model; based on the HSV color Euclidean distance between the pixel points in the filter screen pore color component image, calculating the pixel area of the filter screen pore so as to detect the scaling and blocking degree of the water inlet filter screen of the hydropower station drainage pump, and further, regularly detecting the pore area of the water inlet filter screen of the hydropower station drainage pump so as to detect the scaling and blocking degree of the water inlet filter screen of the hydropower station drainage pump. And a detection result is reported to a control center, so that the blocking degree of the water inlet filter screen of the drainage pump of the hydropower station can be dynamically tracked and monitored, and operation and inspection personnel can conveniently clean the scale of the filter screen in time.

Description

Technical Field

[0001] The invention relates to the technical field of water inlet filter pore scaling detection, and in particular to a method and system for detecting the scaling blockage degree of a water inlet filter of a drainage pump of a hydropower station. Background Art

[0002] As a commonly used drainage facility, drainage pumps are widely used in hydropower stations, mines, river management and other fields. In large hydropower stations in southwest my country, due to the special terrain, the powerhouses are mostly located in the mountains on both sides, and the terrain is often low. Therefore, the station is equipped with a complete drainage system to discharge the accumulated water from mountain seepage, heavy rainfall, etc., to avoid flooding accidents. Taking the drainage of the diversion tunnel of a hydropower station as an example, the mountain seepage water is led to the collection pool through an open channel, and then the accumulated water is discharged through a drainage pump. The diversion tunnel collection pool generally has a lot of gravel, and the water quality is seriously calcified. In order to avoid the inhalation of gravel and calcified fragments into the water pump and cause damage, a filter is often added to the water inlet flow channel of the water pump. However, due to water quality reasons, the pores of the filter are very easy to scale and clog, resulting in reduced hydraulic performance of the water pump and even damage to the motor.

[0003] At present, different prevention and control measures are adopted for the problem of flow channel blockage according to the actual working environment. For example, in conventional river drainage and irrigation water delivery pump stations, trash racks are added in front of the water inlet to block domestic garbage, etc., to avoid blockage caused by foreign matter in the pump station. For high-sand-content water quality pump stations, some scholars have greatly reduced the accumulation of sediment near the impeller mouth ring by reducing the placement angle of the double-suction pump blades and opening wedge-shaped grooves on the outside of the impeller cover; some scholars have concluded that the appropriate forebay shape can effectively avoid backflow and thus reduce sediment accumulation; some scholars have studied the mechanism of the installation elevation, angle, and wedge shape of the guide plate on reducing sediment accumulation in the forebay of the pump station. For small-channel fluid facilities, taking drip irrigation equipment as an example, some scholars have experimentally analyzed the degree of sediment blockage of various drippers with different structures under the conditions of Yellow River drip irrigation, and concluded that the non-compensated cylindrical dripper with a water inlet boss structure can effectively increase the service life of the drip irrigation system; some scholars have studied the influence of the number of water inlets, the number of trapezoidal flow channel teeth, and the number of water outlets on the anti-blockage performance of the single-wing labyrinth drip irrigation belt. The results show that when the size of the trapezoidal labyrinth flow channel is the same, the fewer the number of trapezoidal flow channel teeth, the greater the flow velocity inside the flow channel, which is more conducive to the discharge of sediment particles from the flow channel. In addition, some scholars have conducted in-depth research on the structural design and self-cleaning methods of filter devices of different fluid equipment, and then summarized effective filtering and anti-blocking solutions. For example, some scholars have removed ammonium carbamate crystals from the pores of the filter by heating the compressor circulation section, heating the ammonia synthesis loop circulation, and heating with steam in the pipe; some scholars have adjusted the process to reduce the generation of coke powder, thereby alleviating the clogging of the inlet filter of the heavy fuel oil pump; and some scholars have designed a mesh-type cyclone self-cleaning pre-pump filter based on the combination of cyclone separation and negative pressure adsorption, which solves the problem of frequent clogging and difficulty in self-cleaning of the pre-pump filter.

[0004] It can be seen that the existing technologies are mostly focused on the analysis of flow patterns and their blockage mechanisms, as well as the optimization of flow channel structure design to deal with flow channel blockage. Although a very small number of scholars have constructed blockage detection models for porous small flow channel equipment from a predictive perspective, these methods are based on the ability to accurately extract blockage-related operating and environmental characteristic variables, as well as sufficient high-quality blockage positive and negative training sample data sets. The preconditions for the work are too harsh. For special working scenarios such as the diversion tunnel sump of a hydropower station, the problem of water quality calcification cannot be avoided, and the optimization space for the water pump inlet filter structure is limited. Therefore, how to detect the degree of scaling and blockage of the drainage pump inlet filter when the optimization space for the drainage pump inlet filter structure is limited, so as to facilitate dynamic tracking and monitoring of the blockage degree of the hydropower station drainage pump inlet filter, and also facilitate early notification of relevant operation and maintenance personnel to clean the filter, so as to avoid the occurrence of phenomena such as reduced hydraulic performance and motor damage of the hydropower station drainage pump, has become an urgent problem to be solved.

[0005] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of the present invention is to provide a method and system for detecting the degree of scaling and blockage of the water inlet filter of a drainage pump in a hydropower station, aiming to solve the problem that the prior art is unable to detect the degree of scaling and blockage of the water inlet filter of a drainage pump when the space for optimization of the water inlet filter structure of the drainage pump is limited, so as to dynamically track and monitor the blockage degree of the water inlet filter of the drainage pump in the hydropower station, and to notify the relevant operation and maintenance personnel in advance to clean the filter, so as to avoid the technical problems such as the reduction of hydraulic performance of the drainage pump in the hydropower station and the damage of the motor.

[0007] To achieve the above object, the present invention provides a method for detecting the degree of scaling and clogging of a water inlet filter of a drainage pump of a hydropower station, the method comprising the following steps:

[0008] Acquire a drainage pump image of a hydropower station, and extract an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm;

[0009] Separating a filter mesh pore color component image from the water inlet filter mesh edge contour image using an HSV color model;

[0010] The pixel area of ​​the filter pores is calculated based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0011] Optionally, the step of acquiring an image of a drainage pump of a hydropower station and extracting an edge contour image of a water inlet filter from the image of the drainage pump of the hydropower station based on a Canny edge detection algorithm comprises:

[0012] Obtaining a coordinate set of multiple groups of pixel points of the edge contour of the water inlet filter screen annotated based on the image of the drainage pump of the hydropower station;

[0013] The Canny edge detection algorithm is used to extract the contour edge of the water inlet filter from the image of the drainage pump of the hydropower station;

[0014] According to the coordinate set and the contour edge, the drainage pump image of the hydropower station is intercepted into a mask area including the complete structure of the water inlet filter screen to obtain an edge contour image of the water inlet filter screen.

[0015] Optionally, the step of extracting the edge of the water inlet filter from the image of the hydropower station drainage pump using the Canny edge detection algorithm comprises:

[0016] The image of the drainage pump of the hydropower station is subjected to grayscale processing and Gaussian filtering in sequence to obtain a filtered grayscale image;

[0017] Calculating the gradient intensity and gradient direction of each pixel in the filtered grayscale image using the Sobel operator, and performing non-maximum suppression processing according to the gradient intensity and gradient direction of each pixel to obtain a local maximum pixel;

[0018] The edge of the water inlet filter is identified by using the Canny edge detection algorithm based on the local maximum pixel point using a double threshold method.

[0019] Optionally, the step of separating the filter screen pore color component image from the water inlet filter screen edge contour image by using the HSV color model comprises:

[0020] Obtaining the RGB components of the water inlet filter edge contour image, and performing HSV conversion processing on the RGB components to obtain an HSV conversion result;

[0021] Determining the optimal color component of the separation filter pores based on the HSV conversion result;

[0022] Accordingly, the step of calculating the pixel area of ​​the filter pores based on the HSV color Euclidean distance between each pixel point in the filter pore color component image to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station includes:

[0023] The pixel area of ​​the filter mesh pores is calculated based on the HSV color Euclidean distance between each pixel point in the optimal color component, so as to detect the degree of scaling and clogging of the water inlet filter mesh of the drainage pump of the hydropower station.

[0024] Optionally, the step of calculating the pixel area of ​​the filter pores based on the HSV color Euclidean distance between each pixel point in the filter pore color component image to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station includes:

[0025] Calculating the HSV color Euclidean distance between each pixel in the filter pore color component image;

[0026] Clustering all pixel points based on the HSV color Euclidean distance to obtain a pixel point clustering result;

[0027] The pixel areas of the filter pores are calculated based on the pixel points corresponding to all the filter pores in the pixel point clustering result, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0028] In addition, to achieve the above purpose, the present invention also proposes a system for detecting the degree of scaling and clogging of the water inlet filter screen of a drainage pump of a hydropower station, wherein the system comprises:

[0029] A contour extraction module, used to obtain a drainage pump image of a hydropower station, and extract an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm;

[0030] An HSV separation module is used to separate a filter mesh pore color component image from the water inlet filter mesh edge contour image by using an HSV color model;

[0031] The pore calculation module is used to calculate the pixel area of ​​the filter pore based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and blockage of the water inlet filter of the drainage pump of the hydropower station.

[0032] Optionally, the contour extraction module is further used to obtain a coordinate set of multiple groups of pixel points of the edge contour of the water inlet filter screen annotated based on the image of the drainage pump of the hydropower station;

[0033] The contour extraction module is also used to extract the contour edge of the water inlet filter from the image of the hydropower station drainage pump using the Canny edge detection algorithm;

[0034] The contour extraction module is also used to intercept the image of the hydropower station drainage pump into a mask area that includes the complete structure of the water inlet filter according to the coordinate set and the contour edge, so as to obtain an edge contour image of the water inlet filter.

[0035] Optionally, the contour extraction module is further used to sequentially perform grayscale processing and Gaussian filtering on the image of the drainage pump of the hydropower station to obtain a filtered grayscale image;

[0036] The contour extraction module is further used to calculate the gradient intensity and gradient direction of each pixel in the filtered grayscale image using a Sobel operator, and perform non-maximum suppression processing according to the gradient intensity and gradient direction of each pixel to obtain a local maximum pixel;

[0037] The contour extraction module is also used to identify the edge of the water inlet filter screen using a double threshold method based on the local maximum pixel point using a Canny edge detection algorithm.

[0038] Optionally, the HSV separation module is further used to obtain the RGB components of the water inlet filter edge contour image, and perform HSV conversion processing on the RGB components to obtain an HSV conversion result;

[0039] The HSV separation module is further used to determine the optimal color component of the separation filter pores based on the HSV conversion result;

[0040] Correspondingly, the pore calculation module is also used to calculate the pixel area of ​​the filter pore based on the HSV color Euclidean distance between each pixel point in the optimal color component, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0041] Optionally, the pore calculation module is further used to calculate the HSV color Euclidean distance between each pixel point in the filter pore color component image;

[0042] The pore calculation module is further used to cluster all pixel points based on the HSV color Euclidean distance to obtain a pixel point clustering result;

[0043] The pore calculation module is also used to calculate the pixel area of ​​the filter pores based on the pixel points corresponding to all the filter pores in the pixel point clustering result, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0044] In the present invention, an image of a drainage pump of a hydropower station is obtained, and an edge contour image of an inlet filter is extracted from the image of the drainage pump of the hydropower station based on a Canny edge detection algorithm; a filter pore color component image is separated from the edge contour image of the inlet filter by an HSV color model; the pixel area of ​​the filter pore is calculated based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and blockage of the inlet filter of the drainage pump of the hydropower station; further, the pore area of ​​the inlet filter of the drainage pump of the hydropower station is regularly detected, and the detection result is reported to the control center, so as to dynamically track and monitor the blockage degree of the inlet filter of the drainage pump of the hydropower station, and also to notify relevant operation and maintenance personnel in advance to perform filter cleaning treatment, so as to avoid the occurrence of phenomena such as reduced hydraulic performance and motor damage of the drainage pump of the hydropower station. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of a first embodiment of a method for detecting the degree of fouling and clogging of a water inlet filter screen of a drainage pump of a hydropower station according to the present invention;

[0046] Figure 2 It is a diagram of the deployment of water pumps and cameras involved in the first embodiment of the method for detecting the degree of fouling and blockage of the water inlet filter screen of a drainage pump of a hydropower station according to the present invention;

[0047] Figure 3 A diagram of the extraction process of the water inlet filter involved in the first embodiment of the method for detecting the degree of fouling and clogging of the water inlet filter of a drainage pump of a hydropower station according to the present invention;

[0048] Figure 4 The pixel gradient direction map involved in the first embodiment of the method for detecting the degree of fouling and clogging of the water inlet filter of the drainage pump of a hydropower station according to the present invention;

[0049] Figure 5 It is a diagram of the edge connection process involved in the first embodiment of the method for detecting the degree of fouling and clogging of the water inlet filter screen of the drainage pump of a hydropower station according to the present invention;

[0050] Figure 6 The HSV color space model diagram involved in the first embodiment of the method for detecting the degree of fouling and clogging of the water inlet filter screen of the drainage pump of a hydropower station according to the present invention;

[0051] Figure 7 The RGB conversion HSV result diagram of the water inlet filter involved in the first embodiment of the method for detecting the scale blockage degree of the water inlet filter of the drainage pump of the hydropower station of the present invention;

[0052] Figure 8 It is a mask image of the pores of the water inlet filter involved in the first embodiment of the method for detecting the degree of scaling and clogging of the water inlet filter of the drainage pump of a hydropower station of the present invention;

[0053] Fig. 9 It is a diagram for eliminating abnormal areas of the water inlet filter involved in the first embodiment of the method for detecting the degree of fouling and clogging of the water inlet filter of the drainage pump of a hydropower station of the present invention;

[0054] Fig.10 It is a structural block diagram of the first embodiment of the system for detecting the fouling and clogging degree of the water inlet filter screen of the drainage pump of a hydropower station according to the present invention.

[0055] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0056] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0057] The embodiment of the present invention provides a method for detecting the degree of fouling and clogging of the water inlet filter screen of a drainage pump of a hydropower station, referring to Figure 1 , Figure 1 It is a flow chart of a first embodiment of a method for detecting the degree of fouling and clogging of a water inlet filter screen of a drainage pump of a hydropower station according to the present invention.

[0058] In this embodiment, the method for detecting the degree of fouling and blockage of the water inlet filter of the drainage pump of the hydropower station comprises the following steps:

[0059] Step S10: acquiring a drainage pump image of a hydropower station, and extracting an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm;

[0060] It is easy to understand that for special working scenarios such as the diversion tunnel collection pool of a hydropower station, the problem of water calcification is unavoidable, and the optimization space for the drainage pump water inlet filter structure is limited. It is necessary to dynamically track the degree of fouling and blockage of the filter pores to detect problems in advance and intervene manually in time. Therefore, this embodiment proposes a method for detecting pore scaling of the water inlet filter based on computer vision. By deploying high-definition cameras, water pump images are collected regularly, and then the Canny edge detection algorithm is used to extract the water inlet filter part in the image, and then the HSV (Hue, Saturation, Value) color model (HSV color model) is used to separate pores and non-pores (including scale and filter body), and the area ratio of pores to the filter is calculated, so as to issue a threshold alarm for the remaining pore ratio and notify relevant operation and maintenance personnel in advance to clean the filter. Reference Figure 2 , Figure 2 This is a diagram of the deployment of water pumps and cameras involved in the first embodiment of the method for detecting the degree of fouling and clogging of the water inlet filter of a drainage pump in a hydropower station according to the present invention. Figure 2 In this paper, a drainage pump for a diversion tunnel sump of a hydropower plant is taken as an example. The pump is installed vertically, and a filter is installed at the water inlet to filter gravel and calcification. Figure 2 (right) The camera is installed on the wall of the water collection tank closest to the water pump. Figure 2 (Left) shown.

[0061] In this embodiment, a coordinate set of multiple groups of pixel points of the edge contour of the water inlet filter marked based on the hydropower station drainage pump image can be first obtained; then, the Canny edge detection algorithm is used to extract the contour edge of the water inlet filter from the hydropower station drainage pump image; and then, based on the coordinate set and the contour edge, the hydropower station drainage pump image is intercepted into a mask area that includes the complete structure of the water inlet filter to obtain an edge contour image of the water inlet filter.

[0062] In a specific implementation, when extracting the water inlet filter contour, the water inlet filter contour can be labeled first. For example, the labeling tool LabelImg can be used to manually label the edge contour of the water inlet filter to obtain a set of (x, y) coordinates of multiple sets of pixel points of the edge contour. Figure 3 , Figure 3 This is a diagram of the extraction process of the water inlet filter involved in the first embodiment of the method for detecting the degree of scaling and clogging of the water inlet filter of a drainage pump of a hydropower station according to the present invention. Figure 3 In the example, the picture taken from the camera's perspective is as follows Figure 3 (Left) shown.

[0063] Then, the edge can be extracted by the Canny edge detection algorithm. After grayscale processing of the drainage pump image, the Canny operator is used to perform Gaussian filtering, gradient calculation, non-maximum suppression processing and double threshold selection on the image in turn. Since there are many irregular lines in the image, the key to the algorithm is the selection of double thresholds. This embodiment dynamically adjusts the double threshold through the trackbar of OpenCV and selects a suitable threshold combination.

[0064] Next, the mask area selection can be performed. Using the coordinate set of the edge contour of the water inlet filter extracted above, combined with all the edges extracted by the Canny edge detection algorithm, the image is intercepted into the mask area that includes the complete structure of the water inlet filter, such as Figure 3 (Top right). Finally, the complete edge contour image of the water inlet filter is obtained, as shown in Figure 3 (lower right) shown.

[0065] Among them, the Canny operator detects edges in a very robust way, can extract image edges without interfering with image features, and separate noise from the input image. It is currently one of the most practical and functional detection operators. In this embodiment, the image is first grayed, and then the grayed image is smoothed using a two-dimensional Gaussian filter template. Next, the derivative operator is used to calculate the gradient of the pixel grayscale in the x and y directions, and the amplitude and direction of the gradient are calculated. Then, non-maximum suppression is performed to find the maximum value of the pixel in the gradient direction to determine whether to provisionally define the central pixel point as an edge point. Finally, a double threshold selection is performed, and a final judgment is made on the provisional edge point obtained in the previous step to select the final edge.

[0066] When the edge of the water inlet filter is detected by the Canny operator, the image of the drainage pump of the hydropower station may be grayed and Gaussian filtered in sequence to obtain a gray image after filtering;

[0067] It is easy to understand that the use of filters can remove image noise, that is, the use of filters and images for convolution. The purpose of filtering is to eliminate noise. The noise in the image corresponds to the high-frequency part of the image, so the noise can be eliminated in the frequency domain. Usually, edge detection and filtering are in conflict, because after filtering, the image is pixel-weighted averaged according to the template, which reduces the noise but blurs the edge of the image at the same time, which increases the uncertainty of locating the edge. It is necessary to create a compromise between the two. The kernel determined by the Gaussian function can resist noise interference and will not have much impact on the sensitivity of edge detection. In this embodiment, the two-dimensional Gaussian kernel is defined as follows:

[0068]

[0069] Where K represents the Gaussian function, σ represents the standard deviation of the filter, and (x, y) represents the pixel point.

[0070] Then, the Sobel operator is used to calculate the gradient intensity and gradient direction of each pixel in the filtered grayscale image, and non-maximum suppression processing is performed according to the gradient intensity and gradient direction of each pixel to obtain a local maximum pixel;

[0071] It is easy to understand that the edge in the image can be in any direction. The gradient strength G and direction θ of each pixel can be calculated by calculating the first-order derivatives in the horizontal and vertical directions of each pixel, and then using the following formula:

[0072]

[0073] In the formula, G represents the gradient intensity of the pixel point, θ represents the gradient direction of the pixel point, and G x Represents the gradient value in the horizontal direction, G y Represents the gradient value in the vertical direction, the positive direction of x is from left to right (i.e. horizontal direction), and the positive direction of y is from top to bottom (i.e. vertical direction).

[0074] Among them, the first-order derivatives of the horizontal gradient and the vertical gradient can be obtained by filtering with operators such as Roberts, Prewitt, and Sobel. This embodiment uses the Sobel operator instead of the traditional 2*2 template to reduce the generation of isolated edge points and pseudo edge points. It uses a larger weight on the center coefficient and can provide better image smoothing. The Sobel gradient operator along the x and y directions is shown as follows:

[0075]

[0076] In the formula, the Sobel operator is two 3×3 matrices, namely the gradient operator g in the x direction x and the gradient operator g in the y direction y .

[0077] It should be noted that non-maximum suppression (NMS) can be understood as finding the local gradient maximum of pixels along the pixel gradient direction, and setting the grayscale value of the pixel corresponding to the local non-maximum to 0. Its function is to remove false edges, refine and accurately locate the image edges.

[0078] Reference Figure 4 , Figure 4 It is a pixel gradient direction map involved in the first embodiment of the method for detecting the degree of fouling and blockage of the water inlet filter of the drainage pump of a hydropower station according to the present invention. Figure 4 In the example, let A be the central pixel and there are eight pixels in its neighborhood, let them be ah (i.e. a, b, c, d, e, f, g, h). First, determine whether the grayscale gradient value of the central pixel A is the largest in its eight-value neighborhood. Figure 4 (Left) The oblique line (G1G2) is the gradient direction of point A, and its local maximum is distributed on this line, that is, in addition to point A, the local gradient maximum may also be points G1 and G2. By calculating the gradients of these three points, we can compare whether point A is the local gradient maximum point. If the grayscale gradient value of A is less than any of G1 and G2, point A can be excluded as an edge point. This is the principle of non-maximum suppression. Figure 4 (Right) shows the result of non-maximum suppression.

[0079] Next, the edge of the water inlet filter is identified by using a Canny edge detection algorithm based on the local maximum pixel point using a double threshold method.

[0080] It is easy to understand that the double threshold method sets two thresholds TH1 and TH2 (TH1>TH2) to suppress the local maximum pixel x generated in the non-maximum suppression. ‘ Filtering is performed to achieve the purpose of noise reduction and edge continuity. In this process, the pixels are filtered into three sets: the strong edge set S s , weak edge set S w And discard the edge set S d , respectively defined as follows:

[0081] {S w |TH1>x'>TH2}

[0082] {S s |x'>TH1}

[0083] {S d |TH2>x'}

[0084] In the formula, x' is the local maximum pixel, S s is a strong edge set, S w is a weak edge set, Sd To discard the edge set, TH1 and TH2 are double thresholds set in the Canny edge detection algorithm, and TH1>TH2.

[0085] Among them, when {S s} split point, then search in the 3*3 window to find whether there is {S w}, if so, connect it to {S s} collection, refer to Figure 5 , Figure 5 This is a diagram of the edge connection process involved in the first embodiment of the method for detecting the degree of scaling and clogging of the water inlet filter of a drainage pump in a hydropower station according to the present invention. Figure 5 In the above example, the final set of strong edge points is the identified image edge.

[0086] Step S20: Separating a filter mesh pore color component image from the water inlet filter mesh edge contour image using an HSV color model;

[0087] In this embodiment, the RGB components of the water inlet filter edge contour image are obtained, and the RGB components are subjected to HSV conversion processing to obtain HSV conversion results; then, the optimal color component of the separation filter pores is determined based on the HSV conversion results; then, the pixel area of ​​the filter pores is calculated based on the HSV color Euclidean distance between each pixel point in the optimal color component, so that the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station can be detected.

[0088] Among them, the HSV (Hue, Saturation, Value) color space was created by AR Smith in 1978. Figure 6 , Figure 6 This is the HSV color space model diagram involved in the first embodiment of the method for detecting the degree of fouling and blockage of the water inlet filter of the drainage pump of a hydropower station of the present invention. H represents the hue of a certain pixel, that is, the color attribute it represents when describing a pure color. It is measured by angle and has a value range of 0° to 360°. S represents the saturation of the pixel, which is proportional to its value and has a value range of 0 to 255. V represents the brightness of the pixel, which embodies the concept of colorless intensity and has a value range of 0.0 to 1.0, with 0.0 representing black and 1.0 representing white.

[0089] It is easy to understand that the HSV model is evolved from the RGB cube model, so it is possible to convert from the RGB color space to the HSV color space. According to the evolution principle, the conversion formula shown in the following formula in this implementation can be obtained:

[0090]

[0091] Where R, G, and B are the three components of the RGB model, and H, S, and V are the three components of the HSV model.

[0092] When performing HSV separation of filter pores, the HSV color component is obtained by converting RGB. Therefore, the RGB component of the water inlet filter image can be obtained first, and then converted to the HSV component according to the above formula. During the conversion process, the thresholds of hue, saturation, and brightness need to be adjusted to enhance the filter pore characteristics. Figure 7 , Figure 7 This is a result diagram of the conversion of RGB to HSV of the water inlet filter involved in the first embodiment of the method for detecting the degree of fouling and blockage of the water inlet filter of the drainage pump of a hydropower station according to the present invention. Figure 7 It can be seen that the RGB component image cannot clearly distinguish the difference between the filter pores and other areas. The HSV S component image has a better effect and the filter pores are clearly distinguished. Therefore, it can be used as the optimal color component to separate the filter pores.

[0093] Based on the HSV color space, the color regions of all filter pores can be extracted, and further, the contours of the filter pores can be obtained through the color region mask. Figure 8 , Figure 8 This is a mask diagram of the pores of the water inlet filter involved in the first embodiment of the method for detecting the degree of scaling and clogging of the water inlet filter of a drainage pump of a hydropower station according to the present invention. Figure 8 The result of obtaining the outline of the filter pores through the color area mask is shown in Figure .

[0094] Step S30: Calculate the pixel area of ​​the filter pores based on the HSV color Euclidean distance between each pixel point in the filter pore color component image to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0095] In this embodiment, the HSV color Euclidean distance between each pixel in the filter pore color component image can be calculated, and then all the pixels are clustered based on the HSV color Euclidean distance to obtain the pixel clustering result. Then, the pixel area of ​​the filter pore is calculated based on the pixel points corresponding to all the filter pores in the pixel clustering result to detect the degree of scaling and blockage of the water inlet filter of the drainage pump of the hydropower station. As can be seen from the above, the HSV S component image has a better effect and the filter pores are clearly distinguished. Therefore, the filter pore color component image is preferably the S component image, and the S component image is taken as an example below.

[0096] In the HSV color space, since there are parts of the filter with similar color components to the pores, there are abnormal areas in the extracted water inlet filter pores. Fig. 9 , Fig. 9This is a diagram for eliminating abnormal areas of the water inlet filter involved in the first embodiment of the method for detecting the degree of fouling and blockage of the water inlet filter of a drainage pump of a hydropower station according to the present invention. Fig. 9 As shown in (left), the abnormal area is screened by the area threshold, and the part exceeding the threshold is removed, and the result is as follows Fig. 9 (Right) Pore diagram of the inlet filter after removal.

[0097] Further, based on Fig. 9 (right), the HSV color Euclidean distance between each pixel in the image can be calculated, and then all pixels can be clustered into two categories. The first category is the pixels corresponding to the pores of the filter (i.e. Fig. 9 (right) middle white bright spot), and the other is the pixel points corresponding to other black areas. Then, by accumulating the number of pixel points of the first category, the pixel area of ​​the filter pores can be obtained as 29405. With this scheme, the pore area of ​​the water pump inlet filter is regularly tested, and the test results are reported to the control center, so that the operation and maintenance personnel can intervene in time to clear the blockage. This embodiment was implemented in a hydropower plant, and the results showed that this embodiment can effectively detect the degree of scaling and blockage of the drainage pump inlet filter, laying a feasible practical foundation for the safe and stable operation of the drainage system of the hydropower station.

[0098] This embodiment is based on computer vision technology, and uses the Canny edge detection algorithm to extract all effective contour edges of the water pump, and then combines manual annotation to select the contour mask of the water pump inlet filter to extract the complete filter contour image, and then uses the HSV color model to extract the S component that can separate the filter pores, and finally clusters the pixel points of the filter S component image by color Euclidean distance, and accumulates the pixel points that characterize the filter pores to obtain the pore pixel area. By regularly calculating the filter pore pixel area of ​​the collected water pump images and pushing the results to the hydropower station operation and maintenance control room, dynamic tracking and monitoring of the blockage degree of the water pump inlet filter can be achieved.

[0099] In this embodiment, an image of a drainage pump of a hydropower station is obtained, and an edge contour image of an inlet filter is extracted from the image of the drainage pump of the hydropower station based on a Canny edge detection algorithm; a filter pore color component image is separated from the edge contour image of the inlet filter by using an HSV color model; and the pixel area of ​​the filter pore is calculated based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and blockage of the inlet filter of the drainage pump of the hydropower station. Furthermore, the pore area of ​​the inlet filter of the drainage pump of the hydropower station is regularly detected, and the detection result is reported to the control center, so as to dynamically track and monitor the blockage degree of the inlet filter of the drainage pump of the hydropower station, and also to notify relevant operation and maintenance personnel in advance to perform filter cleaning treatment, so as to avoid the occurrence of phenomena such as reduced hydraulic performance and motor damage of the drainage pump of the hydropower station.

[0100] Reference Fig.10 , Fig.10 It is a structural block diagram of the first embodiment of the system for detecting the fouling and clogging degree of the water inlet filter screen of the drainage pump of a hydropower station according to the present invention.

[0101] like Fig.10 As shown, the scale blockage degree detection system of the water inlet filter screen of the drainage pump of the hydropower station proposed in the embodiment of the present invention includes:

[0102] The contour extraction module 100 is used to obtain a drainage pump image of a hydropower station, and extract an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm;

[0103] An HSV separation module 200 is used to separate a filter screen pore color component image from the water inlet filter screen edge contour image using an HSV color model;

[0104] The pore calculation module 300 is used to calculate the pixel area of ​​the filter pore based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and blockage of the water inlet filter of the drainage pump of the hydropower station.

[0105] Optionally, the contour extraction module 100 is further used to obtain a coordinate set of multiple groups of pixel points of the edge contour of the water inlet filter screen annotated based on the image of the drainage pump of the hydropower station;

[0106] The contour extraction module 100 is also used to extract the contour edge of the water inlet filter from the image of the hydropower station drainage pump using the Canny edge detection algorithm;

[0107] The contour extraction module 100 is further used to intercept the image of the drainage pump of the hydropower station into a mask area including the complete structure of the water inlet filter according to the coordinate set and the contour edge, so as to obtain an edge contour image of the water inlet filter.

[0108] Optionally, the contour extraction module 100 is further used to sequentially perform grayscale processing and Gaussian filtering on the image of the drainage pump of the hydropower station to obtain a filtered grayscale image;

[0109] The contour extraction module 100 is further used to calculate the gradient intensity and gradient direction of each pixel in the filtered grayscale image using a Sobel operator, and perform non-maximum suppression processing according to the gradient intensity and gradient direction of each pixel to obtain a local maximum pixel;

[0110] The contour extraction module 100 is also used to identify the edge of the water inlet filter screen using a double threshold method based on the local maximum pixel point using a Canny edge detection algorithm.

[0111] Optionally, the HSV separation module 200 is further used to obtain the RGB components of the water inlet filter edge contour image, and perform HSV conversion processing on the RGB components to obtain an HSV conversion result;

[0112] The HSV separation module 200 is further used to determine the optimal color component of the separation filter pores based on the HSV conversion result;

[0113] Correspondingly, the pore calculation module 300 is also used to calculate the pixel area of ​​the filter pore based on the HSV color Euclidean distance between each pixel point in the optimal color component, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0114] Optionally, the pore calculation module 300 is further used to calculate the HSV color Euclidean distance between each pixel in the filter pore color component image;

[0115] The pore calculation module 300 is further used to cluster all pixel points based on the HSV color Euclidean distance to obtain a pixel point clustering result;

[0116] The pore calculation module 300 is also used to calculate the pixel area of ​​the filter pores based on the pixel points corresponding to all the filter pores in the pixel point clustering result, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

[0117] In this embodiment, an image of a drainage pump of a hydropower station is obtained, and an edge contour image of an inlet filter is extracted from the image of the drainage pump of the hydropower station based on a Canny edge detection algorithm; a filter pore color component image is separated from the edge contour image of the inlet filter by using an HSV color model; and the pixel area of ​​the filter pore is calculated based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and blockage of the inlet filter of the drainage pump of the hydropower station. Furthermore, the pore area of ​​the inlet filter of the drainage pump of the hydropower station is regularly detected, and the detection result is reported to the control center, so as to dynamically track and monitor the blockage degree of the inlet filter of the drainage pump of the hydropower station, and also to notify relevant operation and maintenance personnel in advance to perform filter cleaning treatment, so as to avoid the occurrence of phenomena such as reduced hydraulic performance and motor damage of the drainage pump of the hydropower station.

[0118] Other embodiments or specific implementations of the system for detecting the degree of fouling and blockage of the water inlet filter of the drainage pump of a hydropower station of the present invention can refer to the above-mentioned method embodiments and will not be described in detail here.

[0119] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system 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 system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0120] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0121] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0122] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for detecting the degree of fouling and blockage of the water inlet filter of a drainage pump of a hydropower station, characterized in that: The method for detecting the degree of fouling and clogging of the water inlet filter screen of the drainage pump of the hydropower station comprises the following steps: Acquire a drainage pump image of a hydropower station, and extract an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm; Separating a filter mesh pore color component image from the water inlet filter mesh edge contour image using an HSV color model; The pixel area of ​​the filter pores is calculated based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

2. The method for detecting the degree of fouling and clogging of the water inlet filter screen of a drainage pump of a hydropower station as claimed in claim 1, characterized in that: The step of acquiring a drainage pump image of a hydropower station and extracting an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm comprises: Obtaining a coordinate set of multiple groups of pixel points of the edge contour of the water inlet filter screen annotated based on the image of the drainage pump of the hydropower station; The Canny edge detection algorithm is used to extract the contour edge of the water inlet filter from the image of the drainage pump of the hydropower station; According to the coordinate set and the contour edge, the drainage pump image of the hydropower station is intercepted into a mask area including the complete structure of the water inlet filter screen to obtain an edge contour image of the water inlet filter screen.

3. The method for detecting the degree of fouling and clogging of the water inlet filter screen of the drainage pump of a hydropower station as claimed in claim 2, characterized in that: The step of extracting the edge of the water inlet filter from the image of the hydropower station drainage pump using the Canny edge detection algorithm comprises: The image of the drainage pump of the hydropower station is subjected to grayscale processing and Gaussian filtering in sequence to obtain a filtered grayscale image; Calculating the gradient intensity and gradient direction of each pixel in the filtered grayscale image using the Sobel operator, and performing non-maximum suppression processing according to the gradient intensity and gradient direction of each pixel to obtain a local maximum pixel; The edge of the water inlet filter is identified by using the Canny edge detection algorithm based on the local maximum pixel point using a double threshold method.

4. The method for detecting the degree of fouling and clogging of the water inlet filter screen of a drainage pump of a hydropower station as claimed in claim 1, characterized in that: The step of separating the filter screen pore color component image from the water inlet filter screen edge contour image by using the HSV color model comprises: Obtaining the RGB components of the water inlet filter edge contour image, and performing HSV conversion processing on the RGB components to obtain an HSV conversion result; Determining the optimal color component of the separation filter pores based on the HSV conversion result; Accordingly, the step of calculating the pixel area of ​​the filter pores based on the HSV color Euclidean distance between each pixel point in the filter pore color component image to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station includes: The pixel area of ​​the filter mesh pores is calculated based on the HSV color Euclidean distance between each pixel point in the optimal color component, so as to detect the degree of scaling and clogging of the water inlet filter mesh of the drainage pump of the hydropower station.

5. The method for detecting the degree of fouling and clogging of the water inlet filter screen of a drainage pump of a hydropower station as claimed in claim 1, characterized in that: The step of calculating the pixel area of ​​the filter pores based on the HSV color Euclidean distance between each pixel point in the filter pore color component image to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station includes: Calculating the HSV color Euclidean distance between each pixel in the filter pore color component image; Clustering all pixel points based on the HSV color Euclidean distance to obtain a pixel point clustering result; The pixel areas of the filter pores are calculated based on the pixel points corresponding to all the filter pores in the pixel point clustering result, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

6. A system for detecting the fouling and clogging degree of the water inlet filter screen of a drainage pump of a hydropower station, characterized in that: The water inlet filter screen scaling and clogging degree detection system of the hydropower station drainage pump comprises: A contour extraction module, used to obtain a drainage pump image of a hydropower station, and extract an edge contour image of a water inlet filter from the drainage pump image of the hydropower station based on a Canny edge detection algorithm; An HSV separation module is used to separate a filter mesh pore color component image from the water inlet filter mesh edge contour image by using an HSV color model; The pore calculation module is used to calculate the pixel area of ​​the filter pore based on the HSV color Euclidean distance between each pixel point in the filter pore color component image, so as to detect the degree of scaling and blockage of the water inlet filter of the drainage pump of the hydropower station.

7. The system for detecting the fouling and clogging degree of the water inlet filter screen of the drainage pump of a hydropower station as claimed in claim 6, characterized in that: The contour extraction module is also used to obtain a coordinate set of multiple groups of pixel points of the edge contour of the water inlet filter screen annotated based on the image of the drainage pump of the hydropower station; The contour extraction module is also used to extract the contour edge of the water inlet filter from the image of the hydropower station drainage pump using the Canny edge detection algorithm; The contour extraction module is also used to intercept the image of the hydropower station drainage pump into a mask area that includes the complete structure of the water inlet filter according to the coordinate set and the contour edge, so as to obtain an edge contour image of the water inlet filter.

8. The system for detecting the fouling and clogging degree of the water inlet filter screen of the drainage pump of a hydropower station as claimed in claim 7, characterized in that: The contour extraction module is also used to sequentially perform grayscale processing and Gaussian filtering on the image of the drainage pump of the hydropower station to obtain a filtered grayscale image; The contour extraction module is further used to calculate the gradient intensity and gradient direction of each pixel in the filtered grayscale image using a Sobel operator, and perform non-maximum suppression processing according to the gradient intensity and gradient direction of each pixel to obtain a local maximum pixel; The contour extraction module is also used to identify the edge of the water inlet filter screen using a double threshold method based on the local maximum pixel point using a Canny edge detection algorithm.

9. The system for detecting the fouling and clogging degree of the water inlet filter screen of the drainage pump of a hydropower station as claimed in claim 6, characterized in that: The HSV separation module is further used to obtain the RGB components of the water inlet filter edge contour image, and perform HSV conversion processing on the RGB components to obtain an HSV conversion result; The HSV separation module is further used to determine the optimal color component of the separation filter pores based on the HSV conversion result; Correspondingly, the pore calculation module is also used to calculate the pixel area of ​​the filter pore based on the HSV color Euclidean distance between each pixel point in the optimal color component, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

10. The system for detecting the fouling and clogging degree of the water inlet filter screen of the drainage pump of a hydropower station as claimed in claim 6, characterized in that: The pore calculation module is also used to calculate the HSV color Euclidean distance between each pixel in the filter pore color component image; The pore calculation module is further used to cluster all pixel points based on the HSV color Euclidean distance to obtain a pixel point clustering result; The pore calculation module is also used to calculate the pixel area of ​​the filter pores based on the pixel points corresponding to all the filter pores in the pixel point clustering result, so as to detect the degree of scaling and clogging of the water inlet filter of the drainage pump of the hydropower station.

Images