Transformer oil level early warning method and early warning device based on image recognition

Abstract

The present invention provides a transformer oil level early warning method and early warning device based on image recognition, which belongs to the technical field of power transmission and transformation equipment; the early warning method includes the following steps: collecting image information of the oil level gauge of the transformer to obtain first image information; determining the type information of the oil level gauge based on the first image information; determining the oil level information of the transformer based on the type information; and generating transformer oil level alarm prompt information based on the oil level information. The early warning method provided by the present invention can collect the image information of the oil level gauge of the transformer, thereby determining the type of the oil level gauge based on the first image information, realizing automatic classification of the oil level gauge, determining the oil level information of the transformer; and generating transformer oil level alarm prompt information based on the oil level information, so as to realize transformer oil level early warning, effectively improve the operation and maintenance level of the substation, and reduce the failure rate of the equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission and transformation equipment, and in particular to the field of transformer safety monitoring; specifically, to a transformer oil level early warning method and early warning device based on image recognition. Background Art

[0002] Transformers are an indispensable and important equipment in the power system, playing a vital role in voltage conversion and power transmission. In order to ensure the normal operation of the transformer, regular inspection and maintenance are required, and one of the important tasks is to observe the oil level of the transformer.

[0003] The oil level of the transformer is usually obtained through a bushing oil level gauge. The page height in the bushing directly reflects the oil level in the transformer tank. However, the setting position of the transformer bushing oil level gauge is relatively high and there are various types. The reading methods of different types of oil level gauges are different. For example, Legend Electric (Shenyang) Co., Ltd. uses a pointer oil level gauge to indicate the oil level in 220kV and some 110kV oil bushings, Shanghai MWB Transformer Co., Ltd. uses a grid-type glass oil mirror to indicate the oil level, and Hefei ABB Transformer Co., Ltd. produces ridge-type single-glass oil mirrors and double-glass oil mirrors.

[0004] With the development of economy, the power grid is expanding, the number of transformers is increasing year by year, and the oil level of transformer bushing has not yet been digitally collected. Manual observation and reading of oil level is time-consuming, labor-intensive and inefficient. At present, drone and robot inspections are developing vigorously. After taking photos of transformer bushing oil level with drones and robots, corresponding algorithms can be designed to realize automatic identification and estimation of oil level, thereby reducing the operation and maintenance pressure of substation operation and maintenance personnel and improving the equipment operation and maintenance level.

[0005] However, in the existing technology, most studies focus on pointer oil level gauges in substations, ignoring the calculation of other types of transformer bushing oil level gauges. In the relevant solutions for pointer oil level gauges, the calculations are complex using machine learning, fuzzy matching, deep learning and other methods, and the characteristics of transformer pointer oil level gauges are not considered. The numerical calculation method of predictive information and image processing is not combined, and a simple and efficient pointer bushing oil level calculation cannot be achieved; and the calculation of other types of oil level gauges is almost blank at present.

[0006] Therefore, there is a technical problem in the prior art that it is impossible to realize automatic reading and early warning of oil levels for different types of oil level gauges. Summary of the invention

[0007] The main purpose of the present invention is to provide a transformer oil level warning method and warning device based on image recognition, so as to solve the technical problem in the prior art that it is impossible to realize automatic reading and warning of oil level for different types of oil level gauges.

[0008] In order to achieve the above-mentioned object, according to one aspect of the present invention, there is provided a method comprising the following steps: collecting image information of an oil level gauge of a transformer to obtain first image information; determining type information of the oil level gauge based on the first image information; determining oil level information of the transformer based on the type information; generating alarm prompt information of the transformer oil level based on the oil level information; wherein the method of determining the type information of the oil level gauge based on the first image information comprises: binarizing the first image information to obtain second image information; determining a category characteristic index of the second image information based on the second image information, wherein the category characteristic index at least comprises a Hough circle quantity statistical index D 1 , Line gradient index D 2 and histogram transfer cost index D 3 ; Determine the type information of the oil level gauge based on the category characteristic index.

[0009] In some embodiments, after determining the type information of the oil level gauge based on the category characteristic indicator, the method further includes: verifying the determined type information.

[0010] In some embodiments, checking the determined type information includes using an unsupervised algorithm DBSCAN to calculate the Hough circle quantity statistical index D 1 , Line gradient index D 2 and histogram transfer cost index D 3 Perform density-based clustering and manually identify and classify automatically identified outliers.

[0011] In some embodiments, the Hough circle quantity statistical index D 1 Obtained through:

[0012] The coordinate index of any pixel point on the second image information is (x i ,y j ), the search step is s, the Hough circle transform algorithm is used to traverse all images of the second image information, and the circle radius obtained by voting is saved as a set r k , where the maximum circle radius and the second largest circle radius obtained by Hough transform are r max 、r s.max , then the Hough circle quantity statistical index D 1 for:

[0013] .

[0014] In some embodiments, the line gradient index D 2 Obtained through:

[0015] Assume that the angle θ of the line segment about the horizontal axis of the second image information to the vertical line is θ, and the angle interval between the highest oil level and the lowest oil level of the oil level gauge pointer is [a°, b°], where -90<a<0, 0≤b<90, then the probability distribution function f(θ) that θ falls on any point in [a°, b°] is:

[0016] ;

[0017] When θ∈[-90°,90°], it is the second type of oil level meter, and the probability P(θ) of obtaining θ by random sampling is:

[0018] ;

[0019] Perform length(θ) random sampling on the calculated θ, where length(•) is the number of calculation arrays, and calculate its average probability as the line gradient index D 2 for:

[0020] .

[0021] In some embodiments, the histogram transfer cost indicator D 3 Obtained through:

[0022] Assuming that the number of bits of the grayscale value of a unit pixel of the second image information is d, the grayscale visible range of the second image information is [0,2 d -1], let the second image information be in M A The number of grayscale pixels is U A (M A ), M A The gray level becomes M after histogram equalization. B Grayscale, W is the number of horizontal pixels, H is the number of vertical pixels, WH is the size of the image to be analyzed, and the normalized histogram equalization is:

[0023] ;

[0024] From the above formula, we can know that M B ≥M A , define the product of the grayscale difference before and after histogram equalization and the number of normalized pixels as the transfer cost S c ,So:

[0025] .

[0026] The transfer cost is normalized using the neural network activation function Sigmoid function as the histogram transfer cost index D 3 ,So:

[0027] .

[0028] According to another aspect of the present invention, the present invention also provides a transformer oil level warning device based on image recognition, including: an acquisition module, used to acquire image information of the oil level gauge of the transformer to obtain first image information; a first determination module, used to determine the type information of the oil level gauge based on the first image information; a second determination module, used to determine the oil level information of the transformer based on the type information; a generation module, used to generate alarm prompt information of the transformer oil level based on the oil level information, wherein determining the type information of the oil level gauge based on the first image information includes: binarizing the first image information to obtain second image information; determining the category characteristic index of the second image information based on the second image information, the category characteristic index at least including a Hough circle quantity statistical index D1, a line gradient index D2 and a histogram transfer cost index D3; determining the type information of the oil level gauge based on the category characteristic index.

[0029] According to another aspect of the present invention, the present invention further provides a safety belt storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the transformer oil level warning method based on image recognition described in any one of the above are implemented.

[0030] According to another aspect of the present invention, the present invention also provides an electronic device, comprising at least a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program on the memory, the steps of the transformer oil level warning method based on image recognition as described in any of the above items are implemented.

[0031] According to another aspect of the present invention, the present invention further provides a computer program product, the computer program product comprising a computer program, and when the computer program is executed by a processor, the transformer oil level warning method based on image recognition as described in any of the above items is implemented.

[0032] By applying the transformer oil level early warning method and early warning device based on image recognition provided by the present invention, by collecting image information of the oil level gauge of the transformer, the type of the oil level gauge can be determined based on the first image information, and automatic classification of the oil level gauge can be achieved. By determining the oil level information of the transformer; and generating alarm prompt information of the transformer oil level based on the oil level information, the transformer oil level early warning can be achieved, which effectively improves the operation and maintenance level of the substation and reduces the failure rate of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0034] Figure 1 Schematic diagram of four types of oil level gauges involved in the present invention;

[0035] Figure 2 It is a flow chart of the transformer oil level early warning method based on image recognition provided by the present invention;

[0036] Figure 3 is a flow chart of determining the type information of the oil level gauge based on the image information provided by the present invention;

[0037] Figure 4 It is a structural block diagram of a transformer oil level warning device based on image recognition provided by the present disclosure;

[0038] Figure 5 It is a schematic diagram of the structure of the electronic device provided by the present disclosure. DETAILED DESCRIPTION

[0039] The specific embodiments of the present disclosure are described in detail below with reference to the accompanying drawings, but are not intended to limit the present disclosure.

[0040] It should be understood that various modifications may be made to the embodiments disclosed herein. Therefore, the above description should not be considered as limiting, but merely as an example of an embodiment. Other modifications within the scope and spirit of the present disclosure will occur to those skilled in the art.

[0041] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the general description of the present disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the present disclosure.

[0042] These and other characteristics of the present disclosure will become apparent from the following description of a preferred form of embodiment given as a non-limiting example, with reference to the attached drawings.

[0043] It should also be understood that, although the present disclosure has been described with reference to some specific examples, those skilled in the art will be able to realize many other equivalent forms of the present disclosure that have the features described in the claims and are therefore within the scope of protection defined thereby.

[0044] The above and other aspects, features and advantages of the present disclosure will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings.

[0045] Specific embodiments of the present disclosure are described hereinafter with reference to the accompanying drawings; however, it should be understood that the disclosed embodiments are merely examples of the present disclosure, which may be implemented in a variety of ways. Well-known and / or repeated functions and structures are not described in detail to avoid obscuring the present disclosure with unnecessary or redundant details. Therefore, the specific structural and functional details disclosed herein are not intended to be limiting, but only to serve as a basis and representative basis for the claims to teach those skilled in the art to use the present disclosure in a variety of ways with substantially any suitable detailed structure.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0047] This specification may use the phrases "in one embodiment," "in another embodiment," "in a further embodiment," or "in other embodiments," all of which may refer to one or more of the same or different embodiments according to the present disclosure.

[0048] The present disclosure is further described below in conjunction with the accompanying drawings and specific embodiments.

[0049] The amount of oil inside the transformer is one of the important factors affecting the safe operation of the transformer. The oil level of the transformer is usually obtained through a bushing oil level gauge.

[0050] There are many types of existing transformer bushing oil level gauges according to different equipment manufacturers and manufacturing processes. Figure 1 As shown, the oil level gauge can be divided into the following 4 types according to the reading method of the oil level gauge, and the specific differences are as follows:

[0051] Figure 1 The first type of oil level gauge shown in A is a pointer type oil level gauge. It is connected to the gear transmission system through a floating connecting rod inside the casing, and indicates the oil level with a straight long pointer. This type of oil level gauge generally only marks the highest and lowest oil levels, that is, there are only two scale lines.

[0052] Figure 1The second type of oil level gauge shown in B is a grid oil level gauge. It is a type of glass oil mirror. Its display mode is that there are grid ridges composed of horizontal and vertical lines inside the oil level gauge. When the oil level gauge is full of oil, the grid ridges cannot be seen by the naked eye. If there is no oil in the oil level gauge, the grid ridges can be seen. The oil level indication is based on the lowest visible grid ridge as the standard.

[0053] Figure 1 The third type of oil level gauge shown in C is a single oil mirror type oil level gauge. It is a type of glass oil mirror, which is different from the second type of oil mirror. It has only vertical frosted edges inside, and the edges have a strong reflective effect. When it is full of oil, there is no reflective effect, showing low brightness and low saturation characteristics; when there is no oil, it shows high brightness characteristics due to the reflective effect. The oil level indication is based on the lowest visible vertical edge.

[0054] Figure 1 The fourth type of oil level gauge shown in D is a double oil mirror type oil level gauge. It is consistent with the third type of oil level gauge in terms of type and observation characteristics, but has two independent oil mirrors from top to bottom, divided into an upper oil mirror and a lower oil mirror. When the upper oil mirror is full of oil, the lower oil mirror must also be full of oil, and the casing oil level is the highest at this time; when the lower oil mirror is full of oil and the upper oil mirror is oil-free, it indicates that the casing oil level is in the middle, which is a normal operating state; when the lower oil mirror is oil-free, the upper oil mirror must be oil-free, and the casing oil level is the lowest at this time.

[0055] From the above analysis, it can be seen that there are differences in the reading methods of various oil level gauges, and there is no automatic liquid level reading device that can be simultaneously applicable to different types of oil level gauges in the prior art; in order to solve the technical problem that the prior art cannot realize automatic oil level reading and early warning for different types of oil level gauges, the present invention provides a transformer oil level early warning method and early warning device based on image recognition.

[0056] Figure 2 A flowchart of the transformer oil level early warning method based on image recognition provided by the present invention is shown as follows: Figure 2 As shown, the transformer oil level early warning method based on image recognition includes the following steps:

[0057] S101, collecting image information of an oil level gauge of a transformer to obtain first image information.

[0058] In this step, image information of the oil level gauge of the transformer is collected to obtain first image information, wherein the image information at least includes an image of a liquid level instrument area of ​​the oil level gauge.

[0059] Specifically, image information of the oil level gauge is captured by a camera, wherein the image information at least includes an image of a liquid level instrument area of ​​the oil level gauge.

[0060] In order to improve the image acquisition accuracy, this step also includes enhancing and denoising the acquired image, specifically as follows: grayscale processing is performed on the acquired image of the liquid level meter area; the single-scale Retinex algorithm is used to enhance the regional image taken under insufficient lighting conditions; and the bilateral filtering algorithm is used to remove the image noise. By enhancing and denoising the acquired image, the brightness and contrast of the image can be improved, which is conducive to improving the recognition accuracy of the transformer oil level.

[0061] S102: Determine type information of the oil level gauge based on the first image information.

[0062] After completing the above step S101, in this step, the type information of the oil level gauge is determined based on the image information.

[0063] Furthermore, if Figure 3 As shown, determining the type information of the oil level gauge based on the image information includes the following steps:

[0064] S201, binarizing the image information to obtain second image information.

[0065] In this step, the image information is binarized to obtain second image information.

[0066] Specifically, assuming that the size of the oil level gauge image to be analyzed is W×H, where W is the number of horizontal pixels and H is the number of vertical pixels, the image information is binarized to obtain the second image information.

[0067] S202: Determine, based on the second image information, a category characteristic index of the image information, where the category characteristic index at least includes a Hough circle quantity statistical index, a line gradient index, and a histogram transfer cost index.

[0068] After acquiring the second image information, in this step, the category characteristic index of the image information is determined based on the acquired second image information, wherein the category characteristic index at least includes a Hough circle quantity statistical index, a line gradient index, and a histogram transfer cost index.

[0069] Specifically, let the coordinate index of any pixel point on the second image information be (xi, yj), the search step length be s, use the Hough circle transform algorithm to traverse all images of the second image information, and save the circle radius obtained by voting as a set r k Since the second image information includes the overall image information of the oil level gauge and there is no intersection circle, the above parameters are limited as follows:

[0070] s≤x i≤Ws,i=0,1,...,floor(W / s-2);

[0071] s≤y j ≤Hs,j=0,1,...,floor(H / s-2);

[0072] ;

[0073] r max =max{r k │k=1,2,...,N};

[0074] r s.max =max{r k │k=1,2,...,N, and r k ≤r max}.

[0075] Where floor(•) is the floor function, N is the number of circles obtained by Hough transform; r max 、r s.max They represent the maximum circle radius and the second largest circle radius respectively.

[0076] Considering that the oil level gauge is the largest circle in the field of view of the oil level gauge image, when the radius of the largest circle is significantly different from the radius of the second largest circle, it indicates that the algorithm has searched for circles other than the oil level gauge, and the radii of the two oil level gauges are similar. 0.5 is set as the parameter for judging the difference between the radii of the two circles, and the Hough circle quantity statistical index D can be calculated. 1 for:

[0077] .

[0078] Furthermore, the Hough line detection method is used to calculate the straight line segments in the second image information that meet the peak points of the Hough matrix. Each detected straight line segment is composed of two oil level gauge image position endpoint data, so that the angle θ of each line segment about the horizontal axis of the second image information to the vertical line can be calculated, and the value range of θ is [-90°, 90°].

[0079] When the oil level gauge is operating normally, the pointer of the first type of oil level gauge will only fall within the range between the highest oil level and the lowest oil level under normal operating conditions, and the angular range of the first type of oil level gauge is generally lower than 180°. So, let the angular range between the highest oil level and the lowest oil level of the first type of oil level gauge be [a°, b°], where -90<a<0, 0≤b<90, a and b are known inherent parameters of the first type of oil level gauge), then, the probability of the first type of oil level gauge θ falling at any point in [a°, b°] is the same, that is, it obeys a uniform distribution, and its probability distribution function f(θ) is obtained.

[0080] ;

[0081] The probability that the oil level gauge is of the first type is P (a ≤ θ ≤ b) and is f (θ). When θ ∈ [-90°, 90°], it is of the second type. The probability P (θ) of obtaining θ in a random sampling is:

[0082] ;

[0083] In order to reduce the possible errors in the calculation process, the calculated θ is randomly sampled length(θ) times, where length(•) is the number of calculation arrays, and its average probability is calculated as the line gradient index D 2 :

[0084] .

[0085] When the line gradient index D 2 The larger it is, and the closer it is to 1, it means that the oil level gauge tends to be the second type of oil level gauge.

[0086] The third type of oil level gauge is a single oil mirror type oil level gauge, which shows a local high brightness characteristic when not full of oil, and a global low brightness characteristic when full of oil, that is, it will show high values ​​at both ends of the histogram, while the histograms of the first and second type oil level gauges are relatively balanced in any state, and will not show local high value characteristics. Therefore, by comparing the histogram before and after equalization, it can be determined whether it is a third type of oil level gauge.

[0087] Specifically, assuming that the number of bits of the grayscale value of a unit pixel of the second image information is d, the grayscale visible range of the second image information is [0,2 d -1], let the second image information be in M A The number of grayscale pixels is U A (M A ), M A The gray level becomes M after histogram equalization. B Grayscale, so that the normalized histogram equalization can be obtained as:

[0088] ;

[0089] From the above formula, we can know that M B ≥M A , define the product of the grayscale difference before and after histogram equalization and the number of normalized pixels as the transfer cost S c ,So:

[0090] ;

[0091] Considering that the transfer costs of oil level gauge images of different sizes vary greatly, the transfer cost is normalized using the neural network activation function Sigmoid function as the histogram transfer cost index D 3 ,So:

[0092] .

[0093] Among them, due to the transfer cost S c ∈(0,+∞), then the histogram transfer cost index D 3 ∈(0.5,1). Histogram transfer cost index D 3 The larger the value is, the greater the difference between the second image information and the oil level gauge image after histogram equalization. Since the third type of oil level gauge presents a local high gray value characteristic, its transfer cost is greater than that of the first type of oil level gauge and the second type of oil level gauge.

[0094] S203: Determine the type information of the oil level gauge based on the type characteristic index.

[0095] After completing the above step S202, in this step, the type information of the oil level gauge is determined based on the category characteristic index.

[0096] Specifically, since the fourth type of oil level gauge is a dual oil mirror type oil level gauge, it has two oil mirrors; while the other three types of oil level gauges have only one oil mirror, therefore, by calculating the Hough circle number statistical index D 1 To determine whether the oil level gauge is a fourth type oil level gauge.

[0097] The first type of oil level gauge is a pointer type oil level gauge, which necessarily contains a line with an unknown gradient, i.e., the pointer, while the second type of oil level gauge has a line with a horizontal gradient and a vertical gradient when the oil is not full, and does not contain a line with any gradient when the oil is full. Therefore, by calculating the line gradient index D 2 To determine whether the oil level gauge is a second type oil level gauge, when the line gradient indicator D 2 The larger it is, and the closer it is to 1, it means that the oil level gauge tends to be the second type of oil level gauge.

[0098] The third type of oil level gauge is a single oil mirror type oil level gauge, which shows a local high brightness characteristic when it is not full of oil, and a global low brightness characteristic when it is full of oil, that is, it will show high values ​​at both ends of the histogram, while the histograms of the first type and the second type of oil level gauge are relatively balanced in any state, and there will be no local high value characteristics. Therefore, by comparing the histogram before and after equalization, it can be determined whether it is a third type of oil level gauge. Therefore, by calculating the histogram transfer cost index, it is determined whether it is a third type of oil level gauge.

[0099] That is to say, in the judgment process, we can first calculate the Hough circle quantity statistical index D 1 Determine whether the oil level gauge is the fourth type oil level gauge, and then calculate the line gradient index D 2 To determine whether the oil level gauge is a second type oil level gauge, then, by calculating the histogram transfer cost index D 3 Determine whether it is a third type of oil level gauge.

[0100] If the type of the oil level gauge cannot be determined through the above judgments, it is considered to be the first type of oil level gauge.

[0101] In order to further improve the accuracy of the judgment, after the type information of the oil level gauge is determined based on the category characteristic index, the type information may be checked.

[0102] Specifically, due to the uncertainty in the process of image acquisition and index calculation, some data may not meet the above index parameter characteristics, that is, noise parameters appear. Therefore, the unsupervised algorithm DBSCAN is used to calculate the Hough circle number statistical index D 1 , Line gradient index D 2 and histogram transfer cost index D 3 Density-based clustering is performed, and the automatically identified outliers (i.e., noise parameters) are manually identified and classified. By automatically identifying outliers, the accuracy of identifying the type of oil level gauge can be further improved.

[0103] S103: Determine the oil level information of the transformer based on the type information and the image information.

[0104] After completing the above step S102, in this step, the oil level information of the transformer is determined based on the type information and the image information.

[0105] Specifically, after the type of the oil level gauge is acquired, the oil level information of the oil level gauge may be calculated according to the type characteristics of the oil level gauge.

[0106] For the first type of oil level gauge, the angle of the line segment calculated by the Hough line detection about the horizontal axis of the oil level gauge image to the vertical line is θ, then the angle of the line segment about the horizontal axis of the second image information is (θ+90)°, and its angle interval is [(a+90)°, (b+90)°], that is, (a+90)° corresponds to the lowest oil level, and (b+90)° corresponds to the highest oil level. Suppose the Hough line detects g line segments, and their endpoint data is {(p 1 ,q 1 ),(p 2 ,q 2 ),...,(p g ,q g )}, calculate the longest line segment , the corresponding line segment coordinate index t, the line segment endpoint (p t ,q t ) The angle between the horizontal axis of the oil level gauge image and the vertical line is θ t Since the casing oil level gauge does not have a calibrated digital oil level, in order to facilitate the comparison of different types of oil levels, the oil level is recorded in the form of percentage. The oil level of the first type of oil level gauge is set to 0. 1 , then O 1 ∈[0,100]%, 0 represents the lowest oil level and 100 represents the highest oil level; the oil level of the first type of oil level gauge is O 1 for:

[0107] .

[0108] For the second type of oil level gauge, let the second image information be I, and the radius of the oil level gauge is , create a binary morphological dilation structure element V, whose size is , calculate the expansion of V to I for: .

[0109] In the above formula, It is the reflection relative to the origin. After binary expansion, the grid edges that were originally empty are filled with expansion elements. Create a size of 2r max ×2r max The binary oil level gauge image set Y is set to 0, and the horizontal axis in Y is at r max Treatment 1, obtains a binary oil level gauge image with a vertical line in the middle.

[0110] Calculate Y and I p The intersection of Y , we can see that the oil level gauge image I Y The length of the middle vertical line is the height of the grid edge after filling.

[0111] I Y =I p ∩Y;

[0112] Using Hough Line Detection I Y The length of the vertical line in is L Y , assuming the oil level of the second type oil level gauge is O 2 , then O 2 for:

[0113] .

[0114] The third type of oil level gauge is a single oil mirror oil level gauge, and its oil level is essentially the correlation between the high brightness part reflected by the reflected sunlight when there is no oil and the low brightness part when filled with oil. Therefore, the oil level can be calculated by calculating the histogram grayscale, distinguishing the high brightness area and the low brightness area, and calculating the oil level with the pixel values ​​of different brightness areas. From the above analysis, it can be seen that the key to the third type of oil level gauge is to calculate the segmentation position of the histogram grayscale at high brightness and low brightness.

[0115] The maximum inter-class variance method is used for segmentation. The total number of pixels U of the oil level gauge image can be calculated from the above steps as follows:

[0116] ;

[0117] Among them, let the initial threshold be C, and the gray level of the low-brightness oil level gauge image pixel be Z 0 ={1,2,...,C}, the gray level of high brightness is Z 1 ={C+1,C+2,...,2 d -1}, then the gray level is randomly sampled and falls on Z 0 The probability w 0 for;

[0118] ;

[0119] Similarly, the gray level is randomly sampled and falls on Z 1 The probability w 1 for:

[0120] ;

[0121] From the above two equations, we can see that w 1 =1-w 0 . Further, calculate the different brightness groups Z 0 , Z 1 The average gray level u 0 、u 1 :

[0122] ;

[0123] ;

[0124] Then the average grayscale u of the second image information is the value of different brightness groups Z 0 , Z 1 The average gray level u 0 、u 1 Weighted sum:

[0125] ;

[0126] Use the following formula to calculate the variance The maximum value of can be obtained by C corresponding to the maximum value:

[0127] ;

[0128] Thus, the calculation of the oil level gauge is transformed into calculating the proportion of low brightness pixels in the entire pixel space. At this time, the oil level of the third type of oil level gauge is the random sampling in C above that falls on Z 0 The probability w 0 Therefore, let the oil level of the third type oil level gauge be O 3 , then O 3 for:

[0129] .

[0130] For the fourth type of oil level gauge, the only difference between the fourth type of oil level gauge and the third type of oil level gauge is that it has two oil mirrors. Assume that the oil level of the upper oil mirror is 0. up , the oil level of the lower oil mirror is O down , where O up or down The algorithm of the oil level indicated in is the same as that of the third type oil level gauge.

[0131] In the calculation process of the fourth type of oil level gauge, the oil level of the upper oil mirror is first calculated. up , if O up =0, indicating that there is no oil in the upper oil mirror, then continue to calculate the oil level of the lower oil mirror. down Since the lower oil mirror only shows half of the actual oil level, the oil level of the fourth type oil level gauge is obtained at this time. 4 for:

[0132] O 4 =0.5×O down ×100%;

[0133] If O up ≥0, indicating that the upper oil mirror has oil, then the lower oil mirror must be full of oil, and there is no need to calculate the oil level of the lower oil mirror. down At this time, the oil level of the fourth type oil level gauge is O 4 for:

[0134] O 4 =(0.5+0.5×O up )×100%;

[0135] Summarizing the above two situations, we can get the oil level of the fourth type of oil level gauge: 4 .

[0136] .

[0137] S104: Generate transformer oil level alarm information based on the oil level information.

[0138] After completing the above step S103, in this step, an alarm prompt information of the oil level of the transformer is generated based on the oil level information.

[0139] Specifically, in order to prevent the transformer from having too little oil and affecting the safe use of the transformer, the determined oil level information of the transformer can be compared with the oil warning threshold. When it is lower than the oil warning threshold, an alarm prompt message is generated and the alarm prompt message is sent to the staff; wherein the oil warning threshold can be set according to specific settings such as the type of transformer, application environment, etc., for example, it can be set to 20%.

[0140] The alarm prompt information can be sent to the operation and maintenance personnel of the substation via the Internet, text messages, etc., or it can be sent externally via alarm prompt sounds or status display lights, etc. The specific prompt method is not further limited in this application.

[0141] By applying the transformer oil level warning method based on image recognition provided by the present invention, by collecting image information of the oil level gauge of the transformer, the type of the oil level gauge can be determined based on the first image information, and automatic classification of the oil level gauge can be achieved. By determining the oil level information of the transformer; and generating alarm prompt information of the transformer oil level based on the oil level information, the transformer oil level warning can be achieved, which effectively improves the operation and maintenance level of the substation and reduces the failure rate of equipment.

[0142] Example 2

[0143] In order to better implement the above method, the second aspect of the present disclosure further provides a transformer oil level warning device based on image recognition, which can be integrated on an electronic device.

[0144] For example, Figure 4 As shown, the early warning device 200 may include: a collection module 210, a first determination module 220, a second determination module 230 and a generation module 240, which are specifically as follows:

[0145] (1) A collection module 210, wherein the collection module 210 is used to collect image information of an oil level gauge of a transformer to obtain first image information.

[0146] (2) A first determination module 220, wherein the first determination module 220 is configured to determine type information of the oil level gauge based on the first image information.

[0147] (3) A second determination module 230, wherein the second determination module 230 is configured to determine the oil level information of the transformer based on the type information.

[0148] (4) A generating module 240, wherein the generating module 240 is used to generate an alarm prompt information of the transformer oil level based on the oil level information; wherein the first determining module 220 includes a first processing unit, a first determining unit, and a second determining unit, wherein the first processing unit is used to perform binarization processing on the image information to obtain second image information, and the first determining unit is used to determine a category characteristic index of the image information based on the second image information, wherein the category characteristic index at least includes a Hough circle quantity statistical index D 1 , Line gradient index D 2 and histogram transfer cost index D 3 The second determination unit is used to determine the type information of the oil level gauge based on the category characteristic indicator.

[0149] Furthermore, the first determination module 220 includes a verification unit, and the verification unit is used to verify the determined type information.

[0150] Furthermore, the determined type information is verified by using the unsupervised algorithm DBSCAN to calculate the Hough circle quantity statistical index D 1 , Line gradient index D 2 and histogram transfer cost index D 3 Perform density-based clustering and manually identify and classify automatically identified outliers.

[0151] By applying the transformer oil level warning device based on image recognition provided by the present invention, by collecting image information of the oil level gauge of the transformer, the type of the oil level gauge can be determined based on the first image information, and automatic classification of the oil level gauge can be achieved. By determining the oil level information of the transformer; and generating alarm prompt information of the transformer oil level based on the oil level information, the transformer oil level warning can be achieved, which effectively improves the operation and maintenance level of the substation and reduces the failure rate of equipment.

[0152] Example 3

[0153] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0154] To this end, a third embodiment of the present disclosure provides a storage medium, which is a computer-readable medium and stores a computer program. When the computer program is executed by a processor, the method provided by the embodiment of the present disclosure is implemented, including the following steps S11 to S14:

[0155] S11, collecting image information of the oil level gauge of the transformer to obtain first image information;

[0156] S12, determining type information of the oil level gauge based on the first image information;

[0157] S13, determining the oil level information of the transformer based on the type information;

[0158] S14, generating transformer oil level alarm information based on the oil level information;

[0159] Wherein, determining the type information of the oil level gauge based on the first image information includes:

[0160] Performing binarization processing on the image information to obtain second image information;

[0161] Based on the second image information, determining a category characteristic index of the image information, wherein the category characteristic index at least includes a Hough circle quantity statistical index D1, a line gradient index D2, and a histogram transfer cost index D3;

[0162] The type information of the oil level gauge is determined based on the type characteristic index.

[0163] Furthermore, when the computer program is executed by a processor, other methods provided by any of the above embodiments of the present disclosure are implemented.

[0164] According to the transformer oil level warning method based on image recognition provided by the present invention, by collecting image information of the oil level gauge of the transformer, the type of the oil level gauge can be determined based on the first image information, and automatic classification of the oil level gauge can be achieved. By determining the oil level information of the transformer; and generating alarm prompt information of the transformer oil level based on the oil level information, the transformer oil level warning can be achieved, which effectively improves the operation and maintenance level of the substation and reduces the failure rate of equipment.

[0165] Example 4

[0166] A fourth embodiment of the present disclosure provides an electronic device, such as Figure 5 As shown, the electronic device at least includes a memory 310 and a processor 320. The memory 310 stores a computer program. The processor 320 implements the method provided by any embodiment of the present disclosure when executing the computer program on the memory 310. Exemplarily, the method executed by the electronic device computer program is as follows:

[0167] S21, collecting image information of the oil level gauge of the transformer to obtain first image information;

[0168] S22, determining type information of the oil level gauge based on the first image information;

[0169] S23, determining the oil level information of the transformer based on the type information;

[0170] S24, generating transformer oil level alarm information based on the oil level information;

[0171] Wherein, determining the type information of the oil level gauge based on the first image information includes:

[0172] Performing binarization processing on the image information to obtain second image information;

[0173] Based on the second image information, determining a category characteristic index of the image information, wherein the category characteristic index at least includes a Hough circle quantity statistical index D1, a line gradient index D2, and a histogram transfer cost index D3;

[0174] The type information of the oil level gauge is determined based on the type characteristic index.

[0175] In specific implementation, the acquisition module 210, the first determination module 220, the second determination module 230 and the generation module 240 are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions.

[0176] According to the transformer oil level warning method based on image recognition provided by the present invention, by collecting image information of the oil level gauge of the transformer, the type of the oil level gauge can be determined based on the first image information, and automatic classification of the oil level gauge can be achieved. By determining the oil level information of the transformer; and generating alarm prompt information of the transformer oil level based on the oil level information, the transformer oil level warning can be achieved, which effectively improves the operation and maintenance level of the substation and reduces the failure rate of equipment.

[0177] The storage medium may be included in the electronic device, or may exist independently without being installed in the electronic device.

[0178] The storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains at least two Internet Protocol addresses; sends a node evaluation request including at least two Internet Protocol addresses to a node evaluation device, wherein the node evaluation device selects an Internet Protocol address from the at least two Internet Protocol addresses and returns it; receives the Internet Protocol address returned by the node evaluation device; wherein the obtained Internet Protocol address indicates an edge node in a content distribution network.

[0179] Alternatively, the storage medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device: receives a node evaluation request including at least two Internet Protocol addresses; selects an Internet Protocol address from the at least two Internet Protocol addresses; and returns the selected Internet Protocol address; wherein the received Internet Protocol address indicates an edge node in a content distribution network.

[0180] Computer program code for performing operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the passenger computer, partially on the passenger computer, as a separate software package, partially on the passenger computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the passenger computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0181] It should be noted that the storage medium disclosed above may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. Computer readable signal media may also be any storage medium other than computer readable storage media that can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), or any suitable combination of the above.

[0182] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0183] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0184] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs).

[0185] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0186] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0187] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0188] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are only example forms of implementing the claims.

[0189] Multiple embodiments of the present disclosure are described in detail above, but the present disclosure is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications to the embodiments based on the concepts of the present disclosure, and these variations and modifications should all fall within the scope of protection required by the present disclosure.

Claims

1. A transformer oil level early warning method based on image recognition, characterized in that: The steps include: Collecting image information of an oil level gauge of the transformer to obtain first image information; determining type information of the oil level gauge based on the first image information; determining oil level information of the transformer based on the type information; Generate transformer oil level alarm information based on the oil level information; Wherein, determining the type information of the oil level gauge based on the first image information includes: Performing binarization processing on the first image information to obtain second image information; Based on the second image information, determining a category characteristic index of the second image information, wherein the category characteristic index at least includes a Hough circle quantity statistical index D1, a line gradient index D2, and a histogram transfer cost index D3; determining the type information of the oil level gauge based on the type characteristic index; The Hough circle quantity statistical index D1 is obtained by the following method: The coordinate index of any pixel point on the second image information is (x i ,y j ), the search step is s, the Hough circle transform algorithm is used to traverse all images of the second image information, and the circle radius obtained by voting is saved as a set r k , where the maximum circle radius and the second largest circle radius obtained by Hough transform are r max 、r s.max , then the Hough circle quantity statistical index D1 is: Among them, the histogram transfer cost index D3 is obtained by the following method: Assuming that the number of bits of the grayscale value of a unit pixel of the second image information is d, the grayscale visible range of the second image information is [0,2 d -1], let the second image information be in M A The number of grayscale pixels is U A (M A ), M A The gray level becomes M after histogram equalization. B Grayscale, W is the number of horizontal pixels, H is the number of vertical pixels, WH is the size of the image to be analyzed, and the normalized histogram equalization is: From the above formula, we can know that M B ≥M A , define the product of the grayscale difference before and after histogram equalization and the number of normalized pixels as the transfer cost S c ,So: The transfer cost is normalized using the neural network activation function Sigmoid function as the histogram transfer cost index D3:

2. The transformer oil level early warning method based on image recognition according to claim 1 is characterized in that: After determining the type information of the oil level gauge based on the category characteristic index, the method further includes: verifying the determined type information.

3. The transformer oil level early warning method based on image recognition according to claim 2 is characterized in that: Verification of the identified species information includes: The unsupervised algorithm DBSCAN is used to perform density-based clustering on the Hough circle quantity statistical index D1, the line gradient index D2 and the histogram transfer cost index D3, and the automatically identified outliers are manually identified and classified.

4. The transformer oil level early warning method based on image recognition according to claim 1 is characterized in that: The line gradient index D2 is obtained by: Assume that the angle θ of the line segment about the horizontal axis of the second image information to the vertical line is θ, and the angle interval between the highest oil level and the lowest oil level of the oil level gauge pointer is [a°, b°], where -90<a<0, 0≤b<90, then the probability distribution function f(θ) that θ falls on any point in [a°, b°] is: When θ∈[-90°,90°], it is the second type of oil level meter, and the probability P(θ) of obtaining θ by random sampling is: The calculated θ is randomly sampled length(θ) times, where length(·) is the number of calculation arrays, and its average probability is calculated as the line gradient index D2:

5. A transformer oil level warning device based on image recognition, characterized in that: include: A collection module, used to collect image information of the oil level gauge of the transformer to obtain first image information; A first determining module, configured to determine type information of the oil level gauge based on the first image information; A second determination module, configured to determine the oil level information of the transformer based on the type information; A generating module, used for generating transformer oil level alarm prompt information based on the oil level information; Wherein, determining the type information of the oil level gauge based on the first image information includes: Performing binarization processing on the first image information to obtain second image information; Based on the second image information, determining a category characteristic index of the second image information, wherein the category characteristic index at least includes a Hough circle quantity statistical index D1, a line gradient index D2, and a histogram transfer cost index D3; determining the type information of the oil level gauge based on the type characteristic index; The Hough circle quantity statistical index D1 is obtained by the following method: The coordinate index of any pixel point on the second image information is (x i ,y j ), the search step is s, the Hough circle transform algorithm is used to traverse all images of the second image information, and the circle radius obtained by voting is saved as a set r k , where the maximum circle radius and the second largest circle radius obtained by Hough transform are r max 、r s.max , then the Hough circle quantity statistical index D1 is: Among them, the histogram transfer cost index D3 is obtained by the following method: Assuming that the number of bits of the grayscale value of a unit pixel of the second image information is d, the grayscale visible range of the second image information is [0,2 d -1], let the second image information be in M A The number of grayscale pixels is U A (M A ), M A The gray level becomes M after histogram equalization. B Grayscale, W is the number of horizontal pixels, H is the number of vertical pixels, WH is the size of the image to be analyzed, and the normalized histogram equalization is: From the above formula, we can know that M B ≥M A , define the product of the grayscale difference before and after histogram equalization and the number of normalized pixels as the transfer cost S c ,So: The transfer cost is normalized using the neural network activation function Sigmoid function as the histogram transfer cost index D3:

6. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the transformer oil level early warning method based on image recognition described in any one of claims 1 to 4 are implemented.

7. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the transformer oil level early warning method based on image recognition according to any one of claims 1 to 4 is implemented.

8. An electronic device comprising at least a memory and a processor, wherein a computer program is stored in the memory, wherein: The processor implements the steps of the transformer oil level warning method based on image recognition as described in any one of claims 1 to 4 when executing the computer program on the memory.

Images