CN117197741B - Switch cabinet operation abnormity monitoring method based on artificial intelligence - Google Patents
Switch cabinet operation abnormity monitoring method based on artificial intelligence Download PDFInfo
- Publication number
- CN117197741B CN117197741B CN202311171646.1A CN202311171646A CN117197741B CN 117197741 B CN117197741 B CN 117197741B CN 202311171646 A CN202311171646 A CN 202311171646A CN 117197741 B CN117197741 B CN 117197741B
- Authority
- CN
- China
- Prior art keywords
- arc
- edge
- area
- discrimination
- switch cabinet
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012544 monitoring process Methods 0.000 title claims abstract description 34
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 23
- 230000002159 abnormal effect Effects 0.000 claims abstract description 33
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000010891 electric arc Methods 0.000 description 32
- 238000001514 detection method Methods 0.000 description 6
- 230000005856 abnormality Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000012935 Averaging Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008033 biological extinction Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the field of monitoring of abnormal switch cabinet, in particular to a method for monitoring abnormal switch cabinet operation based on artificial intelligence, which comprises the steps of collecting images in a switch cabinet, dividing the images into discrimination areas and calculating the area definition of each discrimination area; the arc trend direction of each discrimination area is obtained, and the increment count value of the discrimination area is obtained according to the area definition change relation of the discrimination area passing by the arc trend direction of the discrimination area; obtaining clear increment regularity of each discrimination area according to the increment count value; extracting each discrimination area conforming to the increment rule, thereby obtaining the arc certainty factor and identifying the arc area; constructing an edge bifurcation index of the arc area, and taking an edge with the bifurcation index being greater than a bifurcation threshold value as an arc edge; and acquiring the orientation deflection angle of the arc edge according to the curvature of the edge point, constructing an arc deviation index, and when the arc deviation index is higher than a deviation threshold value, operating the switch cabinet abnormally. Therefore, abnormal operation monitoring of the switch cabinet is realized, and higher monitoring precision is achieved.
Description
Technical Field
The application relates to the field of abnormal monitoring of switch cabinets, in particular to an abnormal operation monitoring method of a switch cabinet based on artificial intelligence.
Background
With the development of urban power technology, the power supply demand of the power is increased, the switch cabinet is used as power transmission and distribution power equipment, the power can be obtained from a power supply trunk line through a self annular power distribution network, even if a certain trunk line fails, the power can still be obtained from other trunk lines to continue to supply power, the power supply reliability of the switch cabinet is ensured by utilizing a mode of obtaining the power in multiple paths, and even if the occurrence time of the switch cabinet is not long, the advantages of low cost and the like due to the power supply reliability, small volume and the like are widely applied. However, since the main working processes of the switch cabinet are high-voltage power receiving, voltage reduction of the transformer and final low-voltage power distribution, high-voltage electric arcs may be generated in the switch cabinet in the process, if the electric arcs are not processed timely, serious damage may be caused to the switch cabinet, high-voltage leakage and other conditions may result, and therefore an arc chamber and an arc guiding device may be arranged in the switch cabinet to conduct arc extinction. However, the state of the arc in the switch cabinet does not necessarily move to the arc chamber to extinguish the arc according to the direction of the arc guiding device, if the guiding device has a problem or other abnormal conditions, the arc does not enter the arc chamber, and finally the high-voltage arc damages the inside of the switch cabinet, so that the safety of the switch cabinet is affected.
The traditional detection means for the inside of the switch cabinet are usually used for detecting the operation parameters of the switch cabinet equipment such as current and voltage, and the like, and corresponding sensors such as temperature, humidity and the like are arranged in the switch cabinet, so that the collected data are detected by using a data abnormality detection algorithm, and finally, the abnormal condition in the switch cabinet is judged according to the abnormal condition of the data. However, the monitoring method cannot truly reflect the actual situation in the switch cabinet, is inaccurate in monitoring the generation and trend of the electric arc, and cannot judge whether the electric arc is generated in the switch cabinet or not, and whether the electric arc enters an arc chamber to extinguish after the electric arc is generated or not.
In summary, the invention provides an artificial intelligence-based method for monitoring abnormal operation of a switch cabinet, which comprises the steps of obtaining an internal image of the switch cabinet, carrying out regional discussion on the image to construct regional definition, and carrying out clear incremental regularity of the constructed region based on the brightness influence of the electric arc on the internal image of the switch cabinet, further obtaining the certainty factor of the electric arc to judge whether the electric arc is generated in the switch cabinet, constructing a pilot arc deviation index to confirm whether the electric arc has pilot abnormality or not, and further monitoring the abnormal state of the switch cabinet.
Disclosure of Invention
In order to solve the technical problems, the invention provides an artificial intelligence-based switch cabinet operation abnormity monitoring method to solve the existing problems.
The invention discloses an artificial intelligence-based switch cabinet operation abnormity monitoring method, which adopts the following technical scheme:
one embodiment of the invention provides an artificial intelligence-based switch cabinet operation abnormality monitoring method, which comprises the following steps:
collecting an internal image of the switch cabinet and preprocessing; taking pixel points with gray values larger than a segmentation threshold value in the internal image of the switch cabinet as clear points, and carrying out regional division on the internal image of the switch cabinet to obtain each discrimination region;
obtaining the regional definition of each discrimination region according to the number of the clear points in each discrimination region; the discrimination area with the maximum gray value mean value is marked as a suspected arc area, the arc trend direction of each discrimination area is obtained, and the increment count value of the discrimination area is obtained according to the area definition change relation of the discrimination area through which the arc trend direction of the discrimination area passes; obtaining clear increment regularity of each discrimination area according to the increment count value; acquiring a discrimination area conforming to the increment rule according to the clear increment rule; obtaining the arc certainty factor according to the number of the judging areas conforming to the increment rule and the clear increment rule; when the arc certainty factor is greater than the certainty factor threshold, the suspected arc area is marked as an arc area;
regarding each edge point of the arc area, taking the edge point which only contains one edge point in the neighborhood as an edge breakpoint, obtaining a bifurcation index of the edge according to the number of the edge breakpoints and the number of the edge pixel points, and taking the edge with the bifurcation index larger than a bifurcation threshold value as an arc edge; obtaining the orientation deflection angle of the arc edge according to the curvature of each edge point of the arc edge; and obtaining an arc deviation index according to the orientation deviation angles of all the arc edges, wherein the switch cabinet operates abnormally when the arc deviation index is higher than a deviation threshold value.
Further, the obtaining the regional definition of each discrimination region according to the number of the clear points in each discrimination region includes:
the ratio of the number of the clear points in each discrimination area to the total number of the pixel points in each discrimination area is used as the area definition of each discrimination area.
Further, the acquiring the arc trend direction of each discrimination area includes: and taking a connecting line of the center point of the judging area and the center point of the suspected arc area as the arc trend direction of the judging area.
Further, the incremental count value of the discrimination area is obtained according to the area definition change relation of the discrimination area through which the electric arc trend direction of the discrimination area passes, and the expression is as follows:
in the method, in the process of the invention,to determine the incremental count value of the j-th determination region in the arc trend direction of region a,to discriminate the region definition of the j+1th discrimination region in the arc tending direction of the region a,the region definition of the j-th discrimination region in the arc trend direction of the discrimination region a.
Further, the obtaining the clear increment regularity of the discrimination area according to the increment count value includes: and calculating the sum of the increment count values of all the judging areas passing through in the arc trend direction of the judging areas, and taking the calculation result of the exponential function with the sum as the base of the exponential natural constant e as the clear increment regularity of the judging areas.
Further, the obtaining the discrimination area according with the increment rule according to the clear increment rule includes: and taking the discrimination area with the clear increment regularity being larger than the regularity threshold as the discrimination area conforming to the increment rule.
Further, the obtaining the arc certainty factor according to the number of the discrimination areas conforming to the increment rule and the clear increment rule includes:
calculating the ratio of the number of the judging areas conforming to the increment rule to the total number of the judging areas, obtaining the ratio of the clear increment rule sum value of all the judging areas conforming to the increment rule to the clear increment rule sum value of all the judging areas, and taking the product of the two ratios as the arc certainty.
Further, the obtaining the bifurcation index of the edge according to the number of edge break points and the number of edge pixel points includes:
and calculating the distance average value from all edge break points on the edge to the midpoint of the edge for each edge of the arc area, obtaining the product of the calculation result of the exponential function with the opposite number of the distance average value being the exponent natural constant e as the base number and the total number of the edge break points on the edge, and taking the ratio of the product to the total number of the pixel points on the edge as the bifurcation index of the edge.
Further, the obtaining the orientation deflection angle of the arc edge according to the curvature of each edge point of the arc edge includes:
and for each arc edge, acquiring the minimum circumscribed rectangle of the arc edge, connecting the central point of the minimum circumscribed rectangle with the edge point with the maximum curvature on the arc edge, and taking the included angle between the connecting line and the horizontal direction as the orientation deflection angle of the arc edge.
Further, the obtaining the arc deviation index according to the orientation deviation angles of all the arc edges includes:
and acquiring the mean value and standard deviation of all arc edge orientation deflection angles, and taking the product of the standard deviation and the reciprocal of the mean value as an arc deviation index.
The invention has at least the following beneficial effects:
the invention considers the abnormal state detection of the internal arc of the traditional switch cabinet, and most of the abnormal state detection is realized by arranging a sensor and detecting the running state of the switch cabinet so as to judge whether the arc is generated or not from data. According to the invention, the situation that the arc exists in the internal image of the switch cabinet is discussed in a refined manner by constructing the arc certainty factor, the problem of false identification in the arc identification process is solved, the detection precision of the arc phenomenon is improved, the deviation index of the arc in the switch cabinet is finally constructed, whether the arc is normally guided or not is judged by analyzing the consistency of the deviation direction of the arc, the abnormal situation of the switch cabinet is further judged, the judgment precision of the actual situation of the arc in the switch cabinet is improved, and the abnormal monitoring of the switch cabinet operation which is more accurate and takes the actual situation into consideration is realized.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of an artificial intelligence based method for monitoring abnormal operation of a switch cabinet;
fig. 2 is a schematic view of the edge lines of the arc zone.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the specific implementation, structure, characteristics and effects of the method for monitoring abnormal operation of the switch cabinet based on artificial intelligence according to the invention by combining the attached drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the method for monitoring abnormal operation of the switch cabinet based on artificial intelligence.
The embodiment of the invention provides an artificial intelligence-based switch cabinet operation abnormity monitoring method.
Specifically, the following method for monitoring abnormal operation of a switch cabinet based on artificial intelligence is provided, referring to fig. 1, the method comprises the following steps:
and S001, shooting an image of the interior of the running switch cabinet, acquiring RGB images of the interior of the switch cabinet, and preprocessing the images.
The embodiment aims at monitoring the operation condition of the switch cabinet according to the analysis of the image data of the switch cabinet. Before the switch cabinet is put into use, a CMOS high-definition camera is arranged in the switch cabinet, when the switch cabinet operates, the interior of the switch cabinet is subjected to image shooting, an operating high-definition RGB image in the switch cabinet is obtained, the RGB image is converted into a gray image by using a gray value averaging method, then the image is denoised by using a wavelet denoising method, and finally the image is sharpened by using a USM sharpening algorithm, so that the preprocessing of the image is completed. It should be noted that, the gray value averaging method, wavelet denoising and USM sharpening algorithm are all known techniques, and are not within the protection scope of the present embodiment, which is not described herein.
The method can be used for acquiring the internal image of the switch cabinet, and preprocessing the internal image to improve the image quality, and the internal image is used as basic data for monitoring the operation condition of the switch cabinet.
Step S002: the method comprises the steps of carrying out regional discussion on images to construct regional definition, constructing regional clear increasing regularity according to brightness distribution of images in a switch cabinet, further constructing arc certainty analysis switch cabinet to judge whether an arc is generated or not, further analyzing the condition of the generated arc, and constructing a pilot arc deviation index to detect pilot abnormality.
In the operation process of the switch cabinet, an arc is usually generated in the switch cabinet due to insulation faults, poor current-carrying loop, invasion of foreign matters, misoperation, abnormal operation of the switch cabinet and the like. When an arc is generated in the switch cabinet, the arc guiding device guides the arc to the arc chamber to extinguish the arc, and finally, the potential safety hazard of the arc in the switch cabinet is eliminated. However, the arc is generated in different modes and different generation positions, and the condition that the arc guiding device fails to guide successfully may exist, so that the arc damages other equipment in the switch cabinet. The electric arc formed in the switch cabinet is usually extremely high in temperature, the energy is concentrated in the middle part of the electric arc, strong light is emitted, the brightness is diffused from the middle part of the concentrated energy to the tip, finally the brightness is radiated into the whole switch cabinet, then the electric arc is guided to an electric arc chamber by a guiding device, the electric arc chamber segments the electric arc, the energy of the electric arc after forming small segments is gradually exhausted, the electric arc disappears in the air, and the switch cabinet is restored to the original state.
Based on the above-mentioned arc formation process in the switchgear, it is first necessary to determine whether an arc exists or not to detect the arc in the switchgear. Because the switch cabinet itself design characteristics, the switch cabinet luminance under the normal condition is lower, only local slight bright in several pilot lamp department, other regions are dim region, so only can detect near the pilot lamp a little edge when carrying out the edge detection to the image, and when appearing the electric arc, because electric arc itself is equivalent to a strong light source, can illuminate the switch cabinet inside, but because light source intensity is limited to the environment in the switch cabinet is comparatively complicated, so the effect that the electric arc lighted is limited, the illuminating effect in the switch cabinet at last is: the brighter the areas near the arc, the more sharp and more edges of the illuminated device in the image, the darker the areas further from the arc, and finally only small edges of the device in these areas can be identified in the image, with the device being more blurred. Aiming at the characteristics, the embodiment firstly carries out edge detection on the preprocessed image by using a Canny operator to acquire the edge in the image. And then a threshold segmentation algorithm of the maximum information entropy is utilized to calculate a segmentation threshold of the image, and when the gray value of a certain pixel point in the image is larger than the segmentation threshold, the pixel point is marked as a clear point. Then, the image is artificially partitioned, and the image is uniformly divided into 128 square areas, and each area is used as a discrimination area. It should be noted that, the number of square areas is not limited in this embodiment, and an implementer may divide the image into areas by himself.
For each discrimination area, counting the number of clear points and the total number of pixel points in the discrimination area, and constructing the area definition of the discrimination area based on the number of clear points and the total number of pixel points, wherein the area definition (AEL) expression of the discrimination area is as follows:
in the method, in the process of the invention,the region definition for the a-th region,in order to determine the number of distinct points in region a,to determine the total number of pixels in the area a. When the value of AEL is larger, the more clear edges in the judging area are indicated, and the overall definition of the area is larger; when the value of AEL is smaller, the smaller the sharp edge in the discrimination area is, the smaller the area overall sharpness is.
Due to the strong brightness characteristics of the arc, the region closer to the arc in the switch cabinet has higher definition, the region farther from the arc has lower definition, and the region in the image is gradually and progressively changed along the trend direction of the arc. And carrying out gray value average calculation on each discrimination area, obtaining the discrimination area with the maximum gray value average, marking the discrimination area as a suspected arc area, and connecting the central points of other discrimination areas with the central point of the suspected arc area to obtain the arc trend direction of each discrimination area. For a certain discrimination area, the area definition of all discrimination areas through which the arc trend direction passes is obtained, and the increment regularity (L) of the discrimination area is constructed, wherein the expression is specifically as follows:
in the method, in the process of the invention,to determine the clear increasing regularity of the arc trend direction in region a,to determine the number of determination regions through which the arc trend direction passes in region a,for the increment count value of the j-th discrimination area in the arc trend direction of the discrimination area a, if the difference between the area definition of the j+1th discrimination area and the area definition of the j-th discrimination area is larger than 0, the increment rule of the area definition is met, the record is 1, and if the difference between the area definition is smaller than or equal to 0, the increment rule is not met, the record is 0;to discriminate the region definition of the j+1th discrimination region in the arc tending direction of the region a,the region definition of the j-th discrimination region in the arc trend direction of the discrimination region a. When the clear increment regularity of the discrimination area is larger than the regularity threshold, the regional definition in the electric arc trend direction of the discrimination area is proved to accord with the increment rule and is taken as the discrimination area according with the increment rule; when the clear increment regularity of the discrimination area is smaller than or equal to the regularity threshold, the definition of the discrimination area in the electric arc trend direction is not in accordance with the increment rule. It should be noted that the regularity threshold implementer may set itself, and in this embodiment, the regularity threshold is set to 0.15.
For each discrimination area with corresponding clear increment regularity, repeating the method of the embodiment to obtain the clear increment regularity of each discrimination area, for a suspected arc area, the more discrimination areas conforming to the increment regularity, the larger the increment regularity is, which indicates that the suspected arc area is more likely to have arcs, counting the total number of discrimination areas conforming to the increment regularity as B, and constructing an arc certainty factor (KL), wherein the expression is:
wherein KL is arc certainty, and B isThe number of the judging areas in the image accords with the increment rule, M is the total number of the judging areas of the image,for a clear incremental regularity of the q-th decision region conforming to the incremental rule,the regularity is increased for the definition of the a-th decision area. When the value of the arc certainty factor KL is larger, the more the judging area proportion conforming to the increment rule in the image is, and the more the judging area proportion is regular, the more the arc is likely to exist in the suspected arc area in the image; when the value of the arc certainty factor KL is smaller, the judgment area conforming to the regular increment in the image is smaller, the clear increment regularity is low, and the suspected arc area in the image may not have an arc. While the theoretical maximum value of the electric arc certainty factor KL is 1, and the value of the KL is normalized by using a maximum and minimum normalization method, so that the value range of the KL is in the range of [0,1]]When the value of KL is larger than or equal to the certainty threshold, indicating that an arc exists in the suspected arc area, and marking the suspected arc area as an arc area; when the value of KL is smaller than the certainty threshold, the suspected arc area is not provided with an arc. It should be noted that, the setting implementation of the certainty threshold can be selected by the user, and the setting is set to 0.65 in this embodiment, which is not limited to this embodiment.
So far, the arc certainty KL in the switch cabinet is obtained, when an arc exists, the trend of the arc needs to be judged, and whether the arc enters an arc chamber to be extinguished under the guidance of the guiding device is judged.
In general, a switchgear is provided with an arc guiding device at a position where an arc may be generated, the guiding device is mainly configured as a guiding area composed of a plurality of metal bars, and when the arc is generated, both ends of the arc move along a direction set by the guiding device, and finally enter an arc chamber to be extinguished. However, because of the many reasons of arc generation, the position of the arc formation is not completely fixed, so that the guiding device is not successfully guided when the arc is generated, and the arc is finally not entered into the arc chamber to be extinguished, so that other equipment in the switch cabinet is damaged.
When the arc guiding device successfully guides the arc, the arc is concentrated at two ends, the middle arc bends towards the arc chamber, and when the arc guiding device does not successfully guide the arc, the two ends of the arc are free to discharge, so that a tree-shaped arc is generated. In view of the above, in the case of an arc, it is possible to divide the edges in the arc zone further in order to distinguish between true arc edges, considering that not only arc edges but also device edges in the switchgear are present in the zone. Specifically, since the arc itself generates many tiny forked arcs, and the edges of the device are more regular edges, the edge information in the arc area is further analyzed. The edge line information of the arc region is shown in fig. 2.
Firstly, an edge breakpoint of the arc edge is found, as shown in fig. 2, and when there is a break in the arc edge, it should be noted that, in this embodiment, the edge breakpoint is a pixel point on the edge, and in this embodiment, when there is only one edge point in the eight adjacent areas of the edge point, the corresponding edge point is referred to as an edge breakpoint. Since the arc edge itself carries many bifurcated arcs, there are random edge breaks in the middle and around the arc edge, and there are few breaks in the edge of the device, only the head end will have edge breaks. Aiming at an arc edge, searching edge break points on the arc edge, marking, finally counting the number of the edge break points on each edge to be recorded as Z, calculating the average distance between the edge break points and the edge midpoint to be recorded as C, and constructing a bifurcation index (ER) of the edge, wherein the expression is as follows:
where ER is the edge bifurcation index,for the normalized coefficient, Z is the total number of edge break points on the edge,and L is the total number of pixels on the edge, and is the average value of the distances between all edge break points and the edge midpoint. When the value of ER is larger, the more edge break points of the edge are indicated, the shorter the average distance is, the more obvious the bifurcation is, and the more likely the arc edge is; when the value of ER is smaller, the fewer break points that indicate the edge, the farther apart the bifurcation is less pronounced and more likely to be a device edge. Therefore, in this embodiment, the bifurcation threshold will be set, and the practitioner can set the bifurcation threshold value by himself, in this embodiment, the bifurcation threshold value is 0.6, and when the bifurcation index is higher than the bifurcation threshold value, the edge is judged to be the arc edge; when the bifurcation index is below the bifurcation threshold, then the edge is the device edge, not the edge where the arc is located.
And distinguishing an arc edge in the arc area through the steps, and further judging whether the arc is normally guided to the arc chamber according to the characteristics of the arc edge. Specifically, when the arc is properly directed, the ends of the arc edge converge in one direction, respectively, and the curved convex portion is directed toward the arc chamber path. While the unsuccessfully conducted arc is of a variety of shapes and is more diffuse.
And performing curve fitting on each arc edge according to the characteristics, obtaining a quadric equation of the arc edge, and performing curvature calculation on each edge point on each arc edge by using a curvature solving formula. The curve fitting and curvature solving formulas are known techniques, and the invention is not repeated. For each arc edge, acquiring the minimum circumscribed rectangle of each arc edge, connecting the central point of the minimum circumscribed rectangle with the edge point with the maximum curvature on the arc edge, acquiring the included angle between the connecting line and the horizontal direction, and recording as the orientation deflection angle of the arc edge. After a series of orientation angles of all arc edges in the arc region are obtained, constructing a pilot arc deviation index (SKL) by using a variation coefficient, wherein the expression is specifically as follows:
wherein SKL isThe pilot arc is deviated from an index of refraction,is the first one in the arc regionThe angle of orientation of the individual arc edges,the average value of all arc edge orientation deflection angles in the arc region is H, and the total number of arcs in the arc region.
When the deviation index of the pilot arc is larger, the more the electric arcs in the electric arc area are not pilot, the more the electric arcs are in disorder, the more the shapes are various and dispersed, and the equipment in the switch cabinet is easily damaged; when the deviation index of the pilot arc is smaller, the arc orientation in the arc area is consistent, the arc is correctly guided to the arc chamber to be extinguished, and the situation in the switch cabinet is normal. Normalizing the value of the SKL to enable the value range of the SKL to be between 0 and 1, when the value of the SKL is higher than the deviation threshold value, the electric arc in the electric arc area is easy to damage equipment in the switch cabinet, and when the value of the SKL is lower than the deviation threshold value, the electric arc in the electric arc area is guided correctly and cannot damage the equipment in the switch cabinet. The value of the deviation threshold can be set by the operator according to the actual situation, and in this embodiment, the threshold is 0.3.
By the method, the operation condition of the switch cabinet can be monitored, and the abnormal operation condition can be accurately detected, so that early warning prompt can be timely carried out, and the switch cabinet is prevented from operating under the abnormal condition.
Aiming at the problems of low accuracy, no intuitiveness, resource waste and the like of arc abnormal state detection in the traditional switch cabinet, the embodiment of the invention constructs the arc certainty in the switch cabinet, carries out refined discussion on the existence condition of the arc in the internal image of the switch cabinet, avoids the condition of arc identification error and improves the arc identification accuracy;
further, the embodiment of the invention constructs the deviation index of the electric arc in the switch cabinet, judges whether the electric arc is normally guided or not by analyzing the consistency of the deviation direction of the electric arc, further judges the abnormal condition of the switch cabinet, improves the judgment precision of the actual condition of the electric arc in the switch cabinet, and realizes the abnormal monitoring of the operation of the switch cabinet which is more accurate and takes the actual condition into consideration.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and the same or similar parts of each embodiment are referred to each other, and each embodiment mainly describes differences from other embodiments.
The above embodiments are only for illustrating the technical solution of the present application, and are not limiting; the technical solutions described in the foregoing embodiments are modified or some of the technical features are replaced equivalently, so that the essence of the corresponding technical solutions does not deviate from the scope of the technical solutions of the embodiments of the present application, and all the technical solutions are included in the protection scope of the present application.
Claims (10)
1. The method for monitoring the abnormal operation of the switch cabinet based on the artificial intelligence is characterized by comprising the following steps of:
collecting an internal image of the switch cabinet and preprocessing; taking pixel points with gray values larger than a segmentation threshold value in the internal image of the switch cabinet as clear points, and carrying out regional division on the internal image of the switch cabinet to obtain each discrimination region;
obtaining the regional definition of each discrimination region according to the number of the clear points in each discrimination region; the discrimination area with the maximum gray value mean value is marked as a suspected arc area, the arc trend direction of each discrimination area is obtained, and the increment count value of the discrimination area is obtained according to the area definition change relation of the discrimination area through which the arc trend direction of the discrimination area passes; obtaining clear increment regularity of each discrimination area according to the increment count value; acquiring a discrimination area conforming to the increment rule according to the clear increment rule; obtaining the arc certainty factor according to the number of the judging areas conforming to the increment rule and the clear increment rule; when the arc certainty factor is greater than the certainty factor threshold, the suspected arc area is marked as an arc area;
regarding each edge point of the arc area, taking the edge point which only contains one edge point in the neighborhood as an edge breakpoint, obtaining a bifurcation index of the edge according to the number of the edge breakpoints and the number of the edge pixel points, and taking the edge with the bifurcation index larger than a bifurcation threshold value as an arc edge; obtaining the orientation deflection angle of the arc edge according to the curvature of each edge point of the arc edge; and obtaining an arc deviation index according to the orientation deviation angles of all the arc edges, wherein the switch cabinet operates abnormally when the arc deviation index is higher than a deviation threshold value.
2. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the obtaining the regional definition of each discrimination region according to the number of the clear points in each discrimination region comprises:
the ratio of the number of the clear points in each discrimination area to the total number of the pixel points in each discrimination area is used as the area definition of each discrimination area.
3. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the step of obtaining the arc trend direction of each discrimination area comprises: and taking a connecting line of the center point of the judging area and the center point of the suspected arc area as the arc trend direction of the judging area.
4. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the incremental count value of the discrimination area is obtained according to the area definition change relation of the discrimination area through which the direction of trend of the arc of the discrimination area passes, and the expression is:
in the method, in the process of the invention,for the increment count value of the j-th discrimination area in the arc trend direction of discrimination area a, +.>To discriminate the zone definition of the j+1th discrimination zone in the arc trend direction of zone a,/>The region definition of the j-th discrimination region in the arc trend direction of the discrimination region a.
5. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the step of obtaining clear incremental regularity of a discrimination area according to the incremental count value comprises: and calculating the sum of the increment count values of all the judging areas passing through in the arc trend direction of the judging areas, and taking the calculation result of the exponential function with the sum as the base of the exponential natural constant e as the clear increment regularity of the judging areas.
6. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the step of obtaining the discrimination area conforming to the increment rule according to the clear increment rule comprises the steps of: and taking the discrimination area with the clear increment regularity being larger than the regularity threshold as the discrimination area conforming to the increment rule.
7. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the obtaining the arc certainty factor according to the number of discrimination areas conforming to the increment rule and the clear increment rule comprises:
calculating the ratio of the number of the judging areas conforming to the increment rule to the total number of the judging areas, obtaining the ratio of the clear increment rule sum value of all the judging areas conforming to the increment rule to the clear increment rule sum value of all the judging areas, and taking the product of the two ratios as the arc certainty.
8. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the step of obtaining the bifurcation index of the edge according to the number of edge break points and the number of edge pixels comprises the steps of:
and calculating the distance average value from all edge break points on the edge to the midpoint of the edge for each edge of the arc area, obtaining the product of the calculation result of the exponential function with the opposite number of the distance average value being the exponent natural constant e as the base number and the total number of the edge break points on the edge, and taking the ratio of the product to the total number of the pixel points on the edge as the bifurcation index of the edge.
9. The method for monitoring abnormal operation of a switchgear based on artificial intelligence according to claim 1, wherein the step of obtaining an orientation deviation angle of an arc edge according to curvature of each edge point of the arc edge comprises the steps of:
and for each arc edge, acquiring the minimum circumscribed rectangle of the arc edge, connecting the central point of the minimum circumscribed rectangle with the edge point with the maximum curvature on the arc edge, and taking the included angle between the connecting line and the horizontal direction as the orientation deflection angle of the arc edge.
10. The artificial intelligence based method of monitoring for cabinet operation anomalies according to claim 1, wherein the deriving an arc deviation index from the angle of orientation of all arc edges comprises:
and acquiring the mean value and standard deviation of all arc edge orientation deflection angles, and taking the product of the standard deviation and the reciprocal of the mean value as an arc deviation index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311171646.1A CN117197741B (en) | 2023-09-12 | 2023-09-12 | Switch cabinet operation abnormity monitoring method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311171646.1A CN117197741B (en) | 2023-09-12 | 2023-09-12 | Switch cabinet operation abnormity monitoring method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117197741A CN117197741A (en) | 2023-12-08 |
| CN117197741B true CN117197741B (en) | 2024-02-13 |
Family
ID=89003104
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311171646.1A Active CN117197741B (en) | 2023-09-12 | 2023-09-12 | Switch cabinet operation abnormity monitoring method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117197741B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115633259A (en) * | 2022-11-15 | 2023-01-20 | 深圳市泰迅数码有限公司 | Automatic regulation and control method and system for intelligent camera based on artificial intelligence |
| WO2023134789A1 (en) * | 2022-10-25 | 2023-07-20 | 苏州德斯米尔智能科技有限公司 | Automatic inspection method for belt-type conveying device |
-
2023
- 2023-09-12 CN CN202311171646.1A patent/CN117197741B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023134789A1 (en) * | 2022-10-25 | 2023-07-20 | 苏州德斯米尔智能科技有限公司 | Automatic inspection method for belt-type conveying device |
| CN115633259A (en) * | 2022-11-15 | 2023-01-20 | 深圳市泰迅数码有限公司 | Automatic regulation and control method and system for intelligent camera based on artificial intelligence |
Non-Patent Citations (2)
| Title |
|---|
| 电弧闪络危害计算方法研究;方景辉;丁一岷;孙一凡;高希骏;魏玲;;电气技术;20200815(第08期);全文 * |
| 面向驾驶员状态监测的人眼凝视方向判别;周泓;刘林冲;徐海儿;林勇;;电路与系统学报;20120815(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117197741A (en) | 2023-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111242123B (en) | Power equipment fault diagnosis method based on infrared image | |
| CN108307146B (en) | System and method for detecting potential safety hazard of high-voltage transmission line | |
| US20170323435A1 (en) | Defect quantification method, defect quantification device, and defect evaluation value display device | |
| CN116309537B (en) | Defect detection method for oil stain on surface of tab die | |
| US20180340962A1 (en) | Apparatus and method for diagnosing electric power equipment using thermal imaging camera | |
| CN114549522A (en) | Textile quality detection method based on target detection | |
| CN110119680B (en) | Automatic error checking system of regulator cubicle wiring based on image recognition | |
| CN102730535B (en) | Testing device and testing method for strip component for transportation mechanism of steel wire | |
| CN104897367A (en) | Indicating light optical state automatic test method and system | |
| CN114235825B (en) | Steel wire rope quality detection method based on computer vision | |
| CN116337879B (en) | Rapid detection method for abrasion defect of cable insulation skin | |
| CN110443278B (en) | Method, device and equipment for detecting thickness abnormality of grid line of solar cell | |
| CN115018850A (en) | Method for detecting burrs of punched hole of precise electronic part based on image processing | |
| CN115689533B (en) | Data management method of intelligent operation and maintenance system for waste gas monitoring and processing | |
| CN116309191A (en) | Intelligent gas inspection display method based on image enhancement | |
| CN112819844A (en) | Image edge detection method and device | |
| CN117197741B (en) | Switch cabinet operation abnormity monitoring method based on artificial intelligence | |
| CN106340034A (en) | Transformer oil leak detection method | |
| CN109492544B (en) | Method for classifying animal fibers through enhanced optical microscope | |
| CN114187348A (en) | Method for measuring elongation at break of cable sheath based on computer vision | |
| CN116229438B (en) | Spinning quality visual identification system | |
| CN111839444A (en) | Enteroscope lens static detection method based on image recognition matching | |
| CN107239743B (en) | automatic detection method for scale mark reading based on projection method | |
| JP5297886B2 (en) | Cell image analyzer | |
| CN117173188A (en) | Glass scar identification method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |