Valve abnormal state detection method and system, electronic equipment and storage medium

Abstract

The invention relates to a valve abnormal state detection method and system, electronic equipment and a storage medium. The method comprises the following steps: acquiring a state change image of the valve; wherein the state change image is a difference image obtained by performing image subtraction operation on images acquired by an opening indicator of the valve at at least two different moments in the rotating process of the valve; identifying a change area contour in the state change image; calculating an interval variable quantity or an interval change rate through the change region contour; wherein the interval variable quantity is the variable quantity of valve rotation in a time interval between two different moments, and the interval change rate is the change rate of valve rotation in the time interval; and the abnormal state condition of the valve is evaluated through the interval variable quantity or the interval change rate. The abnormal state of the key area is evaluated according to the specific condition of the valve, the whole process of the valve opening and closing action does not need to be completely analyzed, and therefore the analyzed data size is reduced, and the requirement for rapid positioning of the overhaul area is met.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of valves, in particular to a valve abnormal state detection method and system, an electronic device and a storage medium. BACKGROUND

[0002] Valves have been widely used in our daily life, however, it is difficult to judge the health condition of the valve, so that the valve damage, rupture and other problems occur from time to time, and even cause serious economic and safety losses. The action performance of the valve is a main parameter for evaluating the performance of the valve. Therefore, the action performance of the valve is an important reference index in the daily inspection process of the valve.

[0003] To this end, the disclosure No. CN116557625A discloses a valve performance evaluation method, device, storage medium and system based on action monitoring. The method comprises: acquiring monitoring data of a valve opening and closing mechanism; matching the data segment to which the acquired monitoring data belongs; analyzing the monitoring data and obtaining the valve action state level thereof located in different data segments; and characterizing the current state of the valve according to the valve action state level. The life or health state of the valve is characterized by monitoring the parameters of the valve execution process, thereby providing data support and theoretical basis for reasonable planning of valve repair and maintenance.

[0004] However, the application is realized based on a new type of valve containing an angular velocity sensor, but in many actual operation scenarios, there are still many old type valves which do not have angular velocity sensors, and in actual inspection operation, the position of the valve is variable, and the inspector may not be able to directly contact. Moreover, the application analyzes the valve action state in different data segments, and needs to analyze all the monitoring data.

[0005] The present application aims to establish a valve abnormal state detection method, system, electronic device and storage medium to solve the above problems. SUMMARY

[0006] In order to achieve the above-mentioned purposes and other advantages according to the present application, the first object of the present application is to provide a valve abnormal state detection method, comprising:

[0007] acquiring a state change image of the valve; wherein the state change image is a difference image obtained by image subtraction operation of images collected at least two different time instants by an opening degree indicator of the valve during rotation of the valve;

[0008] identifying a change region contour in the state change image;

[0009] calculating an interval change amount or an interval change rate through the change region contour; wherein the interval change amount is a change amount of the valve rotation in a time interval between two different time instants, and the interval change rate is a change rate of the valve rotation in the time interval.

[0010] The abnormal state of the valve is evaluated according to the interval change amount or the interval change rate.

[0011] In a preferred embodiment, the interval change amount is any one of an area size, an arc length, or an edge angle size of the change region profile; and the interval change rate is a ratio of the interval change amount to an interval time length.

[0012] In a preferred embodiment, the calculating the interval change amount or the interval change rate corresponding to the state change image further comprises the steps of:

[0013] identifying an interval rotation direction corresponding to the state change image;

[0014] directionally weighting the interval change amount or the interval change rate according to the interval rotation direction.

[0015] In a preferred embodiment, the interval rotation direction corresponding to the state change image is obtained from a color of the change region in the state change image.

[0016] or from an overlap of the change region in the state change image and a change region in a time-continuous change image.

[0017] In a preferred embodiment, the evaluating the abnormal state of the valve according to the interval change amount or the interval change rate comprises the steps of:

[0018] judging whether the state change image is an abnormal image according to the interval change amount or the interval change rate;

[0019] calculating a position coordinate of an abnormal region of the valve according to a profile of the change region in the abnormal image;

[0020] outputting the abnormal state of the valve; wherein the abnormal state of the valve comprises the position coordinate of the abnormal region and one of the interval change amount or the interval change rate.

[0021] In a preferred embodiment, the judging whether the state change image is the abnormal image according to the interval change amount or the interval change rate comprises the steps of:

[0022] judging whether the interval change amount or the interval change rate is less than a first abnormal threshold value;

[0023] if the interval change amount or the interval change rate is less than the first abnormal threshold value, the state change image is the abnormal image.

[0024] In a preferred embodiment, the judging whether the state change image is the abnormal image according to the interval change amount or the interval change rate further comprises the steps of:

[0025] judging whether the interval change amount or the interval change rate is greater than a second abnormal threshold value; wherein the second abnormal threshold value is greater than the first abnormal threshold value.

[0026] If the interval change amount or the interval change rate is greater than the second abnormal threshold, the state change image is an abnormal image;

[0027] If the interval change amount or the interval change rate is greater than or equal to the first abnormal threshold and less than or equal to the second abnormal threshold, the state change image is a non-abnormal image.

[0028] In a preferred embodiment, the position coordinates of the valve abnormal region are calculated by the change region contour of the abnormal image, comprising the steps of:

[0029] determining whether the change region contour area is less than the region identification threshold;

[0030] If the change region contour area is less than the region identification threshold, the change region contour of the state change image which is time-continuous with the current state change image is identified, and the step of determining whether the change region contour area is less than the region identification threshold is returned to;

[0031] If the change region contour area is greater than or equal to the region identification threshold, the position coordinates of the abnormal region are obtained according to the position of the change region contour in the image.

[0032] In a preferred embodiment, the abnormal state of the valve further includes the abnormal degree of the valve;

[0033] The step of outputting the abnormal state of the valve further includes the steps of:

[0034] inputting the state data of the valve into the trained valve abnormal degree analysis model to obtain the motion abnormal degree of the valve; wherein the state data includes the position coordinates of the normal region, and one of the interval change amount or the interval change rate; the valve abnormal degree analysis model is obtained by iteratively training the state data labeled with the abnormal degree through a deep learning network.

[0035] In a preferred embodiment, the training of the valve abnormal degree analysis model comprises the steps of:

[0036] obtaining training data; wherein the training data is state data labeled with the abnormal degree obtained according to the historical maintenance data of the valve; the state data includes the position coordinates of the normal region, and one of the interval change amount or the interval change rate.

[0037] iteratively training the training data through a deep learning network to obtain the trained valve abnormal degree analysis model; wherein the abnormal state data is input and the motion abnormal degree is output during the iterative training process.

[0038] In a preferred embodiment, the state change image of the valve is obtained, specifically comprising the steps of:

[0039] Obtaining a detection image sequence of the valve; wherein the detection image sequence is a sequence of images of an opening indicator of the valve during rotation of the valve;

[0040] Setting an interval time for sampling of the detection image sequence;

[0041] Image sampling of the detection image sequence according to the interval time to obtain a sampling image sequence;

[0042] Subtraction operation on adjacent images in the sampling image sequence to obtain a state change image.

[0043] A second object of the present application is to provide a valve abnormal state detection system, comprising:

[0044] A contour extraction unit for identifying a change region contour in the state change image;

[0045] A change calculation unit for calculating an interval change amount or an interval change rate through the change region contour; wherein the interval change amount is a change amount of the valve rotation in a time interval between two different time points, and the interval change rate is a change rate of the valve rotation in the time interval;

[0046] A change analysis unit for evaluating the abnormal state of the valve through the interval change amount or the interval change rate.

[0047] In a preferred embodiment, the change analysis unit further comprises a trained valve abnormality degree analysis model, which is used to analyze the motion abnormality degree of the valve through the state data, wherein the state data comprises the position coordinates of the normal region, and one of the interval change amount or the interval change rate; the valve abnormality degree analysis model is obtained by iterative training of the state data labeled with the abnormality degree through a deep learning network.

[0048] In a preferred embodiment, the valve abnormal state detection system further comprises:

[0049] A model training unit for obtaining training data and iteratively training the training data through a deep learning network to obtain the trained valve abnormality degree analysis model; wherein the training data is state data labeled with the abnormality degree obtained according to historical maintenance data of the valve; the state data comprises the position coordinates of the normal region, and one of the interval change amount or the interval change rate; during the iterative training process, the abnormal state data is input and the motion abnormality degree is output.

[0050] A third object of the present application is to provide an electronic device comprising a memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the valve motion performance abnormality detection method.

[0051] A fourth object of the present application is to provide a computer storage medium having computer instructions stored thereon; wherein the computer instructions are executed by a processor to implement the valve action performance anomaly detection method.

[0052] Compared with the prior art, the present application has the following beneficial effects:

[0053] (1) The valve anomaly state detection method of the present application is obtained through the state change image of the valve, so the method is suitable for old valves that cannot be installed or are difficult to install.

[0054] (2) The present application is an analysis of the state change image, so the anomaly state of the key area can be evaluated according to the specific situation of the valve, without the need for complete analysis of the entire process of valve opening and closing action, thereby reducing the amount of data analyzed to meet the rapid positioning of the repair area.

[0055] (3) The state change image used in the present application is a difference image obtained by image subtraction operation, and through this method, the background of the image is difficult to produce redundant image noise, and the extraction of the image pattern contour is facilitated.

[0056] (4) The data of the abnormal area of the acceleration sensor is not visual enough for the operation and maintenance personnel, and is not convenient for actual maintenance work, while the present application facilitates the positioning and repair of the abnormal area of the valve by the valve operation and maintenance personnel in the actual maintenance work through the image algorithm.

[0057] (5) The present application further analyzes the motion anomaly degree of the repair area of the valve through the deep learning network, thereby providing data support and theoretical basis for reasonable planning of valve repair and maintenance. BRIEF DESCRIPTION OF DRAWINGS

[0058] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0059] Figure 1 A flowchart of a valve anomaly state detection method provided for embodiment 1;

[0060] Figure 2 A schematic diagram of the generation of the state change image in embodiment 1;

[0061] Figure 3 A flowchart of obtaining the state change image of the valve in embodiment 1;

[0062] Figure 4 A flowchart of direction weighting of the interval change amount or the interval change rate according to the interval rotation direction in embodiment 1;

[0063] Figure 5 Flowchart for the case of evaluating the abnormal state of the valve by the interval change amount or the interval change rate in Embodiment 1;

[0064] Figure 6 Flowchart for judging whether the state change image is an abnormal image according to the calculated interval change amount or interval change rate in Embodiment 1;

[0065] Figure 7 Flowchart for calculating the position coordinates of the abnormal region of the valve by the change region profile of the abnormal image in Embodiment 1;

[0066] Figure 8 Flowchart for the case of evaluating the abnormal state of the valve by the interval change amount or the interval change rate in Embodiment 1;

[0067] Figure 9a Principle diagram of a first valve action performance abnormality detection system provided in Embodiment 2;

[0068] Figure 9b Principle diagram of a second valve action performance abnormality detection system provided in Embodiment 2;

[0069] Figure 9c Principle diagram of a third valve action performance abnormality detection system provided in Embodiment 2;

[0070] Figure 10 Principle diagram of an electronic device according to Embodiment 3 of the present application;

[0071] Figure 11 Principle diagram of a computer storage medium according to Embodiment 4 of the present application. DETAILED DESCRIPTION

[0072] Hereinafter, the present application will be described more fully with reference to the accompanying drawings, in which specific embodiments of the application are shown. As those skilled in the art would realize, the described embodiments can be modified in various different ways, all without departing from the scope of the present application as defined by the appended claims. Accordingly, no limitations are implied from the description below.

[0073] In the following description, the suffixes "module", "part" or "unit" used for an element are merely intended for facilitating description of the present application, and they are not intended to limit the present application. Therefore, "module", "part" or "unit" can be mixedly used.

[0074] It is to be understood that the steps shown in the flowcharts of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowcharts, the steps shown or described can be executed in a different order from that shown in some cases.

[0075] Embodiment 1

[0076] According to an aspect of the embodiments of the present application, as shown in Figure 1 a valve abnormal state detection method is provided, comprising:

[0077] S1, obtaining a state change image of the valve; wherein the state change image is a difference image obtained by image subtraction operation of images collected at least two different time points of the valve opening degree indicator during the rotation of the valve.

[0078] The above-mentioned state change image is an image obtained by recording the motion state of the valve between two time points in the process of opening or closing, to represent the state change between the two time points.

[0079] It should be understood that the process of opening or closing of the valve is a continuous time process, and the valve usually displays the state of the valve through the opening degree indicator, and the display area of the opening degree indicator changes during the process of opening or closing of the valve, so the above-mentioned state change image refers to the image in which the valve opening degree indicator changes;

[0080] In addition, the number of state change images depends on the number of selected time intervals, and the state change image obtained in step S1 is not limited, and multiple different state change images can be obtained.

[0081] S2, identifying the change area contour in the state change image;

[0082] The state change image obtained by the above-mentioned step S1 is a difference image obtained by image subtraction operation, and the background of the image obtained by this method is difficult to produce redundant image noise, so that it is not necessary to filter the noise in the background area of the image by additional algorithm; in step S2, the state change image can extract the change area contour by contour detection (edge detection) algorithm.

[0083] The above-mentioned contour detection algorithm can be a first-order differential based edge detection, which detects the edge of the image by calculating the gradient value of the image, such as Roberts operator, Prewitt operator and Sobel operator, etc.; it can also be to detect the edge by seeking the zero-crossing point in the second derivative, such as Laplacian operator and Canny operator, etc.; the environment for realizing the contour detection algorithm is not limited to MATLAB, OpenCV, Python, etc.

[0084] S3, calculating the interval change amount or interval change rate through the change area contour; wherein the interval change amount is the change amount of the valve rotation in the time interval between two different time points, and the interval change rate is the change rate of the valve rotation in the time interval;

[0085] S4, evaluate the abnormal state of the valve by the interval change amount or the interval change rate.

[0086] In one embodiment, for step S1, Figure 2 The schematic diagram of the state change image generation, wherein, Figure 2 (a) is an image 91 in the sampling image sequence, the image 91 includes an opening indicator area 911 and a background image area 912; the valve displays the state of the valve at the current time through the opening indicator area 911, and the opening indicator area 911 displays the rotation degree of the valve at the current time by displaying a first area 9111 and a second area 9112 area, and the larger the area of the first area 9111, the greater the rotation angle of the valve opening; Figure 2 (b) and Figure 2 (c) is for different time images 91a and 91b in the image sequence; wherein the image 91a is an image at t n time, and the image 91b is an image at t n+k time; wherein k is the sampling interval time, it can be seen that the part of the opening indicator area 911 of the image 91b changes compared with the image 91a, and the background image area 912 does not change; the image 91b is subtracted from the image 91a to obtain a state change image 92 as shown in Figure 2 (d), it can be seen that a change area 921 appears on the state change image 92, and other image areas in the state change image 92 that do not change are displayed as blank areas 922 through image subtraction operation, wherein the blank area 922 includes an area 9221 in the indicator area 911 that does not change.

[0087] In one embodiment, as shown in Figure 3 , the state change image of the valve is obtained, specifically including the following steps:

[0088] S101, obtain a detection image sequence of the valve; wherein the detection image sequence is a sequence of opening indicator images of the valve during valve rotation;

[0089] Optionally, the detection image sequence is obtained by shooting a video stream; the shooting angle, distance and light of each frame in the above detection image sequence are the same;

[0090] S102, set the sampling interval time of the detection image sequence;

[0091] S103, image sampling is performed on the detection image sequence according to the interval time to obtain a sampling image sequence;

[0092] Specifically, for steps S102-103 in the above, step S101 adopts the way of shooting video stream to obtain the detection image sequence, and the interval of processing is calculated according to the total frame number of the video and the maximum processing frame number. Let N represent the total frame number of the video file, and T represent the target saved frame number. If N is less than or equal to T, the sampling interval is 1, otherwise the sampling interval Interval=N / T;

[0093] All frames are traversed and average extraction is performed. For the frames that need to be saved, the corresponding images are obtained according to the mask for saving, and the saved pictures are named according to the frame number. Let Count represent the counter of the current frame, and Saved represent the saved frame number. The theoretical saved frame number calculated according to the sampling interval is Interval×Saved, and if Count≥Interval×Saved, the current frame is saved and arranged in sequence. Finally, when the saved frame number is greater than or equal to the target frame number, and the total frame number of the video file is greater than or equal to the target frame number, the frame extraction is stopped; finally, the stored folder path and the number of stored pictures are returned to obtain the sampling image sequence.

[0094] S104, subtracting adjacent images in the sampling image sequence to obtain a state change image.

[0095] Optionally, the implementation environment of the above subtraction operation is not limited to MATLAB, OpenCV, Python, etc.

[0096] In one specific embodiment, the interval change amount in step S3 is any one of the area size, arc length or edge angle size of the change region contour; and the interval change rate is the ratio of the interval change amount to the interval length. The interval change amount is obtained by identifying the change region contour in the state change image; and the interval change amount is any one of the area size, arc length or edge angle size of the change region contour. In one exemplary embodiment, the interval change amount is obtained by identifying the change region contour in the state change image through edge identification by OpenCV. Generally, the shape of the change region is a sector, and the area size of the contour, the arc length of the curved side of the sector, and the angle size generated by the two straight sides can be calculated by OpenCV, which can represent the interval change amount of the change region through the above values;

[0097] It is considered that in some embodiments, different state change images may have different sampling interval lengths, and the interval change amount cannot be used to determine whether the state change image is an abnormal image. Therefore, the interval change rate is used for judgment. The interval change rate is the ratio of the interval change amount to the interval length. Specifically, the interval change rate is obtained by dividing the interval change amount by the interval length.

[0098] In one embodiment, the interval rotation direction corresponding to the state change image is determined according to the overlapping of the change regions of the time-continuous change images; for example, there are two time-continuous state change images, which are state change image 92a and state change image 92b. The state change image 92a is the state change image from t1 to t2, and the state change image 92b is the state change image from t2 to t3. If there is no reverse rotation from t1 to t2, but there is a reverse rotation from t2 to t3, the corresponding P1 and P2 will overlap.

[0099] S301, identifying the interval rotation direction corresponding to the state change image;

[0100] S302, directionally weighting the interval change quantity or the interval change rate according to the interval rotation direction.

[0101] In one embodiment, as shown in Figure 4 S302, directionally weighting the interval change quantity or the interval change rate according to the interval rotation direction, includes the steps of:

[0102] S3021, judging the relationship between the interval rotation direction and the overall rotation direction of the valve rotation process; if the interval rotation direction is different from the overall rotation direction, step S3022 is executed; otherwise, step S3023 is executed.

[0103] S3022, multiplying the value of the interval change quantity or the interval change rate by -1.

[0104] S3023, the value of the interval change quantity or the interval change rate remains unchanged.

[0105] In one embodiment, the interval rotation direction corresponding to the state change image in the above step S302 is determined according to the overlapping of the change regions of the time-continuous change images; for example, there are two time-continuous state change images, which are state change image 92a and state change image 92b. The state change image 92a is the state change image from t1 to t2, and the state change image 92b is the state change image from t2 to t3. If there is no reverse rotation from t1 to t2, but there is a reverse rotation from t2 to t3, the corresponding P1 and P2 will overlap.

[0106] In another case embodiment, the interval rotation direction corresponding to the state change image in the above step S3021 is determined according to the color of the change region in the state change image; for example, the state change image 92a is the state change image from t1 to t2. This image is obtained by performing image subtraction operation on the image at t2 and the image at t1. In the case of forward rotation, the color of the change region is a third color generated by subtracting a second color from a first color, and in the case of reverse rotation, the color of the change region is a fourth color generated by subtracting the first color from the second color. Therefore, the interval rotation direction corresponding to the state change image can be determined by the third color or the fourth color of the state change image.

[0107] In one embodiment, as shown in Figure 5As shown, the abnormal state of the valve is evaluated by the interval change amount or the interval change rate, including steps of:

[0108] S401, judging whether the state change image is an abnormal image according to the interval change amount or the interval change rate;

[0109] S402, calculating the position coordinates of the valve abnormal area through the change area profile of the abnormal image;

[0110] The position coordinates of the valve abnormal area can be obtained by the center of gravity of the profile calculated by OpenCV positioning the change area, and can also be obtained by the included angle between the straight side of the profile and the straight side of the sector vertex of the opening indicator.

[0111] S404, outputting the abnormal state of the valve; wherein the abnormal state includes the position coordinates of the abnormal area, and one of the interval change amount or the interval change rate.

[0112] In a specific implementation, as shown in the figure, whether the state change image is an abnormal image is judged according to the calculation of the interval change amount or the interval change rate, specifically including steps of: Figure 6

[0113] S4011, judging whether the interval change amount or the interval change rate is less than a first abnormal threshold; if the interval change amount or the interval change rate is less than the first abnormal threshold, step S4013 is executed.

[0114] S4013, identifying the current state change image as an abnormal image.

[0115] The step S4011 is used to determine that the interval change amount or the interval change rate is too small, indicating that the valve may have a motion abnormality such as sticking during opening or closing.

[0116] In a preferred scheme, if the situation that the valve rotation speed is too fast also needs to be considered, as shown in the figure, the step S4011 also includes specific steps: Figure 6

[0117] S4012, judging whether the interval change amount or the interval change rate is greater than a second abnormal threshold; if the interval change amount or the interval change rate is greater than the second abnormal threshold, step S4013 is executed, otherwise step S4014 is executed;

[0118] S4014, identifying the current state change image as a non-abnormal image.

[0119] The first abnormal threshold or the second abnormal threshold will change according to the acceleration of the normal valve during rotation according to the change of time, and in this process, the second abnormal threshold is greater than the first abnormal threshold in each time period. ​​

[0120] In one embodiment, the interval change amount is used to determine whether the state change image is an abnormal image, wherein the interval change amount is the area size of the change region contour calculated by OpenCV, the interval between the two time points of the state change image is k seconds; the total area of the opening degree indicator in the image is Ss, the total standard time length of the opening or closing of the valve under normal circumstances is Ts, then the first abnormal threshold value The second abnormal threshold value Wherein α < 1 < β; for example, the interval between the two time points of the state change image is 0.05 seconds; the total area of the opening degree indicator in the image is 20, the total standard time length of the opening or closing of the valve under normal circumstances is 1.25 seconds, α = 0.6, β = 1.4; then the first abnormal threshold value M1 = 0.48, the second abnormal threshold value M2 = 1.12.

[0121] In a specific embodiment, as shown in Figure 7 The position coordinates of the valve abnormal region are calculated by the change region contour of the abnormal image, including the steps of:

[0122] S4021, determine whether the area of the change region contour 921 is less than the region identification threshold value; if the area of the change region contour 921 is less than the region identification threshold value, execute step S4022; if the area of the change region contour 921 is greater than or equal to the region identification threshold value, execute step S4023.

[0123] S4022, identify the change region contour 921 of the state change image which is time continuous with the current state change image, and return to S4021.

[0124] S4023, obtain the region data of the abnormal performance of the valve according to the position of the change region contour in the image.

[0125] Through steps S4021-4023 in the steps of the embodiment, the region data of the abnormal performance of the valve is difficult to be obtained because there is no change or the change is too small in the valve within a unit time.

[0126] Specifically, in step S32, the state change image 92b which is time continuous with the current state change image 92a is set as the state change image 92a from t k to t k+1 , then the state change image 92b is the state change image from t k+1 to t k+2 , if t k+1 to t k+2If the state change image is still less than the region recognition threshold, the new state change image 92b is selected in the same time sequence, and in this embodiment, the new state change image 92b is selected in the same time sequence as t k+2 to t k+3 .

[0127] In addition, the state change image 92b can also be selected in the reverse time sequence, but if the state change image is still less than the region recognition threshold, the new state change image 92b needs to be selected in the same time sequence.

[0128] In a specific embodiment, the abnormal state condition also includes the abnormal degree of the valve; as shown, the output valve abnormal state condition further includes the following steps: Figure 8

[0129] S403, input the state data of the valve into the trained valve abnormal degree analysis model to obtain the motion abnormal degree of the valve; wherein the state data includes the position coordinates of the normal region, and one of the interval change amount or the interval change rate; the valve abnormal degree analysis model is obtained by iteratively training the state data labeled with the abnormal degree through a deep learning network.

[0130] The above step S4 further analyzes the motion abnormal degree of the maintenance area of the valve through a deep learning network, thereby providing data support and theoretical basis for reasonable planning of valve maintenance and maintenance. In a specific embodiment, for the valve motion performance abnormality analysis model in the above step S4, the training method before the model is used, including the following steps:

[0131] S501, obtain training data; wherein the training data is state data labeled with the abnormal degree obtained according to the valve historical maintenance data; the state data includes the position coordinates of the normal region, and one of the interval change amount or the interval change rate.

[0132] S502, iteratively train the training data through a deep learning network to obtain a trained valve abnormal degree analysis model; wherein the abnormal state data is input and the motion abnormal degree is output during the iterative training process.

[0133] The above steps S501-502 use deep learning to input the position of the motion performance abnormality and the motion of the position of the motion performance abnormality as input, and output the motion abnormal degree, so as to obtain a motion performance abnormality analysis model which can analyze the motion abnormal degree by inputting the position of the motion performance abnormality and the motion of the position of the motion performance abnormality.

[0134] Embodiment 2 ​

[0135] According to another aspect of the embodiments of the present application, there is also provided a valve abnormal state detection system 600, as shown in Figure 9a comprising:

[0136] an image acquisition unit 601 configured to acquire a state change image of the valve; wherein the state change image is a difference image obtained by image subtraction operation on images of an opening degree indicator of the valve acquired at at least two different time instants during rotation of the valve;

[0137] The state change image is an image obtained by recording the motion state of the valve between two time instants during opening or closing of the valve, to represent the state change between the two time instants. It should be understood that the process of opening or closing of the valve is a continuous time process, and the opening degree indicator of the valve usually displays the state of the valve, and the display area of the opening degree indicator changes during the process of opening or closing of the valve, so the state change image refers to the image in which the opening degree indicator of the valve changes; in addition, the number of state change images depends on the number of selected time intervals, and the state change images acquired in the image acquisition unit 601 are not limited, and multiple different state change images can be acquired.

[0138] a contour extraction unit 602 configured to identify a change region contour in the state change image;

[0139] The state change image obtained by the image acquisition unit 601 is a difference image obtained by image subtraction operation, and the background of the image obtained by this method is difficult to produce redundant image noise, so that it is not necessary to filter the noise in the background region of the image by using additional algorithms; in the contour extraction unit 602, the state change image can extract the change region contour by a contour detection (edge detection) algorithm. The contour detection algorithm can be a first-order differential based edge detection, which detects the edge of the image by calculating the gradient value of the image, such as Roberts operator, Prewitt operator and Sobel operator, etc.; it can also be to detect the edge by seeking the zero-crossing point in the second derivative, such as Laplacian operator and Canny operator, etc.; the environment in which the contour detection algorithm is implemented is not limited to MATLAB, OpenCV, Python, etc.

[0140] The change calculation unit 603 is configured to calculate an interval change amount or an interval change rate of the valve rotation through the change region contour; the interval change amount is a change amount of the valve rotation in a time interval between two different time points, and the interval change rate is a change rate of the valve rotation in the time interval; in a specific embodiment, the interval change amount in the change calculation unit 603 is any one of an area size, an arc length, or an edge angle size of the change region contour; and the interval change rate is a ratio of the interval change amount to the interval time length. The interval change amount is obtained by identifying the change region contour in the state change image; and the interval change amount is any one of the area size, the arc length, or the edge angle size of the change region contour. In an exemplary embodiment, the interval change amount is obtained by identifying the change region contour in the state change image through edge identification by OpenCV; generally, the shape of the change region is a sector, and the area size of the contour, the arc length of the curved side of the sector, and the angle size generated by the two straight sides can be calculated by OpenCV, and the above values can be used to represent the interval change amount of the change region. It is considered that, in some embodiments, different state change images may correspond to different interval lengths, and the interval change amount cannot be used to determine whether the state change image is an abnormal image; therefore, the interval change rate is used for determination; the interval change rate is a ratio of the interval change amount to the interval time length. Specifically, the interval change rate is obtained by dividing the interval change amount by the interval time length.

[0141] The change analysis unit 604 is configured to evaluate the abnormal state of the valve through the interval change amount or the interval change rate. The abnormal state includes the position coordinates of the abnormal region, and one of the interval change amount or the interval change rate.

[0142] In a specific embodiment, for the image acquisition unit 601, Figure 2 a schematic diagram generated for the state change image, wherein Figure 2 (a) is an image 91 in the sampling image sequence, the image 91 includes an opening indicator region 911 and a background image region 912; the valve displays the state of the valve at the current time through the opening indicator region 911, and the opening indicator region 911 displays the rotation degree of the valve at the current time through the display of a first region 9111 and a second region 9112, and the larger the area of the first region 9111, the greater the rotation angle of the valve opening; Figure 2 (b) and Figure 2 (c) is for different time images 91a and 91b in the image sequence; the image 91a is an image at time t n , and the image 91b is an image at time t n+kThe image at time 91b; where k is the sampling interval, it can be seen that compared with image 91a, a part of the aperture indicator region 911 has changed, while the background image region 912 has not changed; by subtracting image 91a from image 91b, we can obtain the image as shown below. Figure 2 (d) shows the state change image 92. It can be seen that the state change image 92 has a change area 921. Other image areas in the state change image 92 that have not changed are displayed as blank areas 922 by image subtraction. Among them, the blank areas 922 include the area 9221 of the indicator area 911 that has not changed.

[0143] In one specific embodiment, such as Figure 9b As shown, the change analysis unit 604 also includes a trained valve anomaly analysis model 6041. The valve anomaly analysis model 6041 is used to analyze the degree of valve movement anomaly through state data. The state data includes the position coordinates of the normal region and one of interval change amount or interval change rate. The valve anomaly analysis model is obtained by iteratively training the state data labeled with anomaly degree through a deep learning network.

[0144] In one specific embodiment, such as Figure 9c As shown, the valve abnormal state detection system further includes: a model training unit 605, used to acquire training data and iteratively train the training data through a deep learning network to obtain a trained valve abnormality analysis model; wherein, the training data is state data labeled with the degree of abnormality obtained from the valve's historical maintenance data; the state data includes the position coordinates of the normal area, and, one of the interval change amount or interval change rate; during the iterative training process, the abnormal state data is used as input and the degree of motion abnormality is used as output.

[0145] Example 3

[0146] According to another aspect of the embodiments of the present invention, such as Figure 10 As shown, an electronic device 700 is also provided, including a memory 701 and a processor 702; wherein the memory 701 is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor 702 to implement the valve abnormal state detection method described above. For a detailed description of the method, please refer to the corresponding description in the above method embodiments, which will not be repeated here.

[0147] Example 4

[0148] According to another aspect of the embodiments of the present invention, such as Figure 11As shown, a storage medium 800 is also provided, and computer instructions are stored on the storage medium 800; wherein the computer instructions are executed by a processor to implement the valve abnormal state detection method of any one of the above. For detailed description of the method, reference can be made to the corresponding description in the method embodiments above, which will not be repeated here.

[0149] The program instructions are stored in a computer readable storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or on a network, and include a plurality of computer program instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above method according to the embodiments of the present application

[0150] From the above description of the embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or by software in combination with necessary hardware. Although the embodiments of the present application have been disclosed as above, they are not limited to the application examples and embodiments listed in the specification, and can be fully applied to various fields suitable for the present application, and additional modifications can be easily implemented by those skilled in the art, and therefore the present application is not limited to specific details and the figures shown and described herein, but falls within the general concept defined by the claims and their equivalents.

[0151] The device, electronic equipment, non-volatile computer storage medium and method provided by the embodiments of the present application are corresponding, and therefore the device, electronic equipment and non-volatile computer storage medium also have similar beneficial technical effects to the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding device, electronic equipment and non-volatile computer storage medium will not be repeated here.

[0152] Those skilled in the art also know that, in addition to implementing the controller in the form of pure computer readable program code, the same function can also be implemented by logically programming the method steps to make the controller in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers and embedded microcontrollers. Therefore, such a controller can be considered as a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can even be considered as both software modules implementing the method and structures within the hardware component.

[0153] The systems, apparatuses, modules, or units disclosed in the above embodiments can be specifically implemented by a computer chip or entity, or by a product with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0154] For the sake of description, the above apparatuses are described in various units by functions for description. Of course, the functions of each unit can be implemented in one or more software and / or hardware when implementing one or more embodiments of the present specification.

[0155] Those skilled in the art will understand that the embodiments of the present specification can be provided as a method, a system, or a computer program product. Therefore, the embodiments of the present specification can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects.

[0156] The present specification is described with reference to flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present specification. It should be understood that each flow and / or block in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that performs the functions specified in one or more blocks or flows.

[0157] These computer program instructions can also be stored in a computer-readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction apparatuses that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus that performs the functions specified in one or more blocks or flows.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a process for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1the functions specified in the one or more blocks.

[0159] In one typical arrangement, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0160] The memory can include non-persistent memory, random access memory (RAM), and / or non-volatile memory, such as read only memory (ROM) or flash memory, among others. The memory is an example of computer readable media.

[0161] It is also important to note that the term "comprising" or "including" or any other variation thereof is intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0162] The specification can be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like, that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of steps and methods described in this specification is not the only sequence or method that can be implemented. Other sequences of steps can be provided and still be performed by the generic computer-executable instructions associated with computer-readable media.

[0163] Embodiments in the specification are described progressively, and the same or similar parts among the embodiments can be mutually referred to. Each embodiment focuses on the difference from other embodiments. In particular, the system embodiments are described relatively simply because they are substantially similar to the method embodiments, and the relevant parts can be referred to the description of the method embodiments.

[0164] The above only describes the embodiments of the specification and does not limit one or more embodiments of the specification. One or more embodiments of the specification can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of one or more embodiments of the specification shall be included in the scope of the claims of one or more embodiments of the specification.

Claims

1. A method for detecting abnormal state of a valve, comprising: obtaining a state change image of the valve; wherein the state change image is a difference image obtained by image subtraction operation of images of an opening indicator of the valve at at least two different time instants during rotation of the valve; identifying a change region contour in the state change image; calculating an interval change amount or an interval change rate based on the change region contour; wherein the interval change amount is a change amount of the valve rotation in a time interval between the two different time instants, and the interval change rate is a change rate of the valve rotation in the time interval; evaluating an abnormal state of the valve based on the interval change amount or the interval change rate.

2. The method of detecting abnormal valve conditions according to claim 1, characterized by The interval change amount is any one of an area size, an arc length or an edge angle size of the change region contour; and the interval change rate is a ratio of the interval change amount to the interval time length.

3. The method of detecting abnormal valve conditions according to claim 2, wherein The calculation of the interval change amount or the interval change rate corresponding to the identified state change image further comprises steps of: identifying an interval rotation direction corresponding to the state change image; directionally weighting the interval change amount or the interval change rate according to the interval rotation direction.

4. The method of detecting abnormal valve conditions according to claim 3, wherein The interval rotation direction corresponding to the state change image is obtained from a color of the change region in the state change image; or obtained from an overlap of the change region in the state change image and a change region in a change image continuous in time.

5. The method of detecting abnormal valve conditions according to claim 1, wherein The evaluation of the abnormal state of the valve based on the interval change amount or the interval change rate comprises steps of: judging whether the state change image is an abnormal image according to the interval change amount or the interval change rate; calculating a position coordinate of an abnormal region of the valve based on the change region contour of the abnormal image; outputting the abnormal state of the valve; wherein the abnormal state of the valve comprises one of the position coordinate of the abnormal region, the interval change amount and the interval change rate.

6. The method of detecting abnormal valve conditions according to claim 5, wherein The judgment of whether the state change image is an abnormal image according to the interval change amount or the interval change rate comprises steps of: judging whether the interval change amount or the interval change rate is less than a first abnormal threshold; if the interval change amount or the interval change rate is less than the first abnormal threshold, the state change image is an abnormal image.

7. The method of detecting abnormal valve conditions according to claim 6, wherein The judgment of whether the state change image is an abnormal image according to the interval change amount or the interval change rate further comprises steps of: judging whether the interval change amount or the interval change rate is greater than a second abnormal threshold; wherein the second abnormal threshold is greater than the first abnormal threshold; if the interval change amount or the interval change rate is greater than the second abnormal threshold, the state change image is an abnormal image; if the interval change amount or the interval change rate is greater than or equal to the first abnormal threshold and less than or equal to the second abnormal threshold, the state change image is a non-abnormal image.

8. The method of detecting abnormal valve conditions according to claim 5, wherein The calculation of the position coordinate of the abnormal region of the valve based on the change region contour of the abnormal image comprises steps of: judging whether an area of the change region contour is less than a region identification threshold; If the area of the change region profile is less than the region identification threshold, a change region profile of a state change image that is time-continuous with the current state change image is identified, and the step of determining whether the area of the change region profile is less than the region identification threshold is returned to. If the area of the change region profile is greater than or equal to the region identification threshold, position coordinates of an abnormal region are obtained according to the position of the change region profile in the image.

9. The method for detecting abnormal valve state according to claim 5, characterized by, The abnormal state of the valve also includes the degree of abnormality of the valve; The output of the abnormal state of the valve also includes the step of: Inputting the state data of the valve into the trained valve abnormality degree analysis model to obtain the motion abnormality degree of the valve; wherein the state data includes the position coordinates of the normal region, and one of the interval change amount or the interval change rate; the valve abnormality degree analysis model is obtained by iteratively training the state data labeled with the degree of abnormality through a deep learning network.

10. The method of detecting abnormal valve conditions according to claim 9, wherein The training of the valve abnormality degree analysis model includes the steps of: Obtaining training data; wherein the training data is state data labeled with the degree of abnormality obtained from valve historical maintenance data; the state data includes the position coordinates of the normal region, and one of the interval change amount or the interval change rate. Iteratively training the training data through a deep learning network to obtain a trained valve abnormality degree analysis model; wherein the abnormal state data is input during the iterative training process, and the motion abnormality degree is output.

11. The method for detecting abnormal valve state according to claim 1, characterized by, Obtaining a state change image of a valve, specifically including the steps of: Obtaining a detection image sequence of the valve; wherein the detection image sequence is a sequence composed of images of an opening indicator of the valve during rotation of the valve; Setting the interval time of sampling of the detection image sequence; Image sampling of the detection image sequence according to the interval time to obtain a sampled image sequence; Subtracting adjacent images in the sampled image sequence to obtain a state change image.

12. A valve abnormal state detection system characterized by comprising: It includes: An image acquisition unit for obtaining a state change image of a valve; wherein the state change image is a difference image obtained by image subtraction operation of images of an opening indicator of the valve collected at least two different times during rotation of the valve; A profile extraction unit for identifying a change region profile in the state change image; A change calculation unit for calculating an interval change amount or an interval change rate through the change region profile; wherein the interval change amount is the change amount of the valve rotation within a time interval between two different times, and the interval change rate is the change rate of the valve rotation within the time interval; A change analysis unit for evaluating the abnormal state of the valve through the interval change amount or the interval change rate.

13. The valve abnormality state detection system of claim 12, wherein The change analysis unit further comprises a trained valve abnormality degree analysis model for analyzing the motion abnormality degree of the valve through state data, wherein the state data comprises one of position coordinates of a normal region, and an interval change amount or an interval change rate; and the valve abnormality degree analysis model is obtained by iteratively training state data labeled with an abnormality degree through a deep learning network.

14. The valve abnormal state detection system according to claim 13, characterized by, The valve abnormality state detection system further comprises: A model training unit is configured to obtain training data and iteratively train the training data through a deep learning network to obtain a trained valve abnormality degree analysis model; wherein the training data is state data labeled with an abnormality degree obtained according to valve historical maintenance data; the state data comprises one of position coordinates of a normal region, and an interval change amount or an interval change rate; and in the iterative training process, the abnormality state data is used as input and the motion abnormality degree is used as output.

15. An electronic device comprising a memory and a processor; wherein, The memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method steps of any one of claims 1-11.

16. A storage medium having stored thereon computer instructions; wherein, The computer instructions, when executed by the processor, implement the method steps of any one of claims 1-11.

Images