CN118314520A - Composite monitoring method and system for box-type substation - Google Patents

Abstract

The invention relates to the technical field of substation monitoring, in particular to a compound monitoring method and system for a box-type substation. The method comprises the following steps: s10: installing video monitoring equipment in the compartment, and recording a luminous area in the image; s20: shooting an image to form an image to be detected, and acquiring the temperature and the humidity in the station to determine a judgment threshold value group; s30: carrying out graying treatment and binarization treatment on the image to be detected to form a treated image; s40: judging the number of black pixel points in a light-emitting area in a processed image and adjusting a brooken value; s50: dividing the processed image into a plurality of suspicious regions; s60: according to the maximum area S of the suspicious region and the number n of the suspicious region, a fire value and a spark value are adjusted; s70: judging an alarm state according to the brooken value, the fire value and the spark value, and determining to stop or continue to operate; s80: after the alarm is released and the machine is restarted, the values of brooken, fire and spark are zeroed. The invention can quickly respond to various early warning conditions and remarkably improve the early warning effect.

Description

Composite monitoring method and system for box-type substation

Technical Field

The invention relates to the technical field of substation monitoring, in particular to a compound monitoring method and system for a box-type substation.

Background

The box-type transformer substation is a novel transformer which organically combines the functions of transformer step-down, low-voltage distribution and the like, and the power receiving, transforming and distributing devices are generally integrally arranged in a metal cabinet, so that the box-type transformer substation has the characteristics of small volume, compact structure, simplicity and convenience in installation and rapidness in power supply, gradually replaces a civil transformer substation, and is widely applied to mines, factory enterprises, oil-gas fields and residential communities of wind power stations.

In the use process of the box-type transformer substation, the internal conditions of the box-type transformer substation are required to be monitored, and timely early warning is carried out on the emergency situations such as damage of electrical components and fire disasters. However, the existing related art generally monitors each early warning requirement through a single sensor, such as a temperature detector or a smoke detector for the open fire in a station, and the conventional monitoring mode generally has obvious hysteresis, such as a difference between the open fire position and the detector position, so that the open fire is sometimes required to burn for a while, and the temperature rise or smoke can be detected by the detector. Therefore, a need exists for an early warning method or system that is more rapid and efficient.

Disclosure of Invention

The invention provides a compound monitoring method and a compound monitoring system for a box-type transformer substation, which can effectively solve the problems of hysteresis and low early warning efficiency of the existing early warning means.

In order to achieve the above object, the present invention provides a composite monitoring method for a box-type substation, comprising the steps of:

Comprising the following steps: a compound monitoring method for a box-type substation comprises the following steps:

S10: installing video monitoring equipment in each compartment in the station, then acquiring an image in each compartment, and manually marking out a luminous area in the image; presetting a damage value brooken, a fire value fire and a spark value spark to be 0;

S20: shooting images by each video monitoring device to form an image to be detected, and acquiring the temperature and humidity in the station; determining area, quantity, damage, fire and spark thresholds and cycle values from temperature and humidity;

s30: carrying out graying treatment and binarization treatment on the image to be detected to form a treated image;

S40: judging the color of pixel points in a light-emitting area in the processed image, if the number of black pixel points is larger than a set threshold value, adding 1 to the brookfen value, otherwise, maintaining the brookfen value;

s50: marking white pixel points which are not in a light-emitting area on a processed image as suspicious pixel points, combining the connected suspicious pixel points into a suspicious area, and further dividing a plurality of suspicious areas;

S60: calculating the maximum area S of the suspicious region, and recording the number of the suspicious region as n;

If S is larger than the area threshold, adding 1 to the fire value, otherwise, maintaining the fire value;

if n is greater than the quantity threshold, adding 1 to the spark value, otherwise, maintaining the spark value;

s70: judging an alarm state:

if brooken is greater than the damage threshold, a damage alarm is sent out;

if fire is greater than the fire threshold, giving out fire alarm;

if spark is larger than the spark threshold value, a spark alarm is sent out;

if the alarm contains any one of damage and fire, stopping the machine, otherwise returning to S20;

When the number of times of returning to S20 reaches a cyclic value and the values of brooken, fire and spark are unchanged in the period, the values of brooken, fire and spark are zeroed;

S80: after the alarm is released and the machine is restarted, the values of brooken, fire and spark are zeroed.

Further, in step S10, at least two video monitoring devices are installed in each compartment in the station, and the plurality of video monitoring devices in the same compartment are all adjusted to different angles for shooting, so that the light-emitting areas in the viewing angles of each video monitoring device are not overlapped.

Further, in step S10, the illumination in the compartment is removed, the image obtained in this state is used to mark the light emitting point, and the light emitting area corresponding to the light emitting point is recorded; and then keeping the illumination in each compartment sufficient, using the image acquired in the state for checking the luminous points, screening out the luminous points which can correspond to the indicator lamps of the electrical elements, and taking the luminous areas corresponding to the luminous points as final luminous areas.

Further, when the luminous points are verified, the luminous sources of the luminous points which do not belong to the electric element indicator lamp are confirmed, and whether the unlabeled luminous points need to be reserved or not is confirmed according to the luminous sources.

Further, in step S30, when the image to be inspected is subjected to the gradation processing, the weights of the respective color channels of the image are set, and the gradation image is finally calculated by integration.

Further, in step S30, when binarizing the image to be detected, the image is divided into a plurality of areas, and the threshold value of the binarization is adjusted according to the characteristics of each area.

Further, in step S50, after the suspicious region is divided, the suspicious region information and the image to be inspected are packaged and sent to an external storage system.

Further, in step S70, after the alarm state is judged, the image to be detected and the processed image used in the calculation are deleted.

Further, in step S10, a threshold value comparison table is also established to make the temperature and the humidity with different values correspond to the area threshold value, the quantity threshold value, the damage threshold value, the fire threshold value, the spark threshold value and the cycle value with different values.

The invention also provides a compound monitoring system for the box-type substation, which is applied to the compound monitoring method for the box-type substation, and comprises the following steps:

the video monitoring devices are respectively arranged in each compartment in the station and are used for shooting images;

the temperature sensing detector is used for detecting the temperature in the cabinet;

The humidity detector is used for detecting the humidity in the cabinet;

the alarm module is used for sending out an alarm;

and the control module is used for recording and analyzing the image acquired by the video monitoring equipment and controlling the alarm system to work.

By the technical scheme of the invention, the following technical effects can be realized:

1. The invention can acquire images in the transformer substation in real time through a set of designed computer vision algorithm technology, so that the working state, open fire and spark existence of electrical elements in the transformer substation can be monitored in real time, and an alarm can be sent out timely when an abnormal condition occurs, thereby effectively improving the safety of the box-type transformer substation;

2. The invention processes and analyzes the image by using the computer vision algorithm technology, can realize automatic detection, lighten the manual burden and improve the efficiency;

3. the algorithm technology can adjust the judgment threshold value in the system according to the environmental conditions of temperature and humidity, thereby controlling the early warning sensitivity of the algorithm and enabling the algorithm to flexibly configure and process different environmental conditions.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings may be obtained according to the drawings without inventive effort to those skilled in the art.

FIG. 1 is a schematic diagram of a video monitoring device arrangement for a composite monitoring method for a box-type substation in the present invention;

Reference numerals: 1. a compartment; 2. video monitoring equipment.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.

In the description of the present invention, it should be noted that the directions or positional relationships indicated as being "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are directions or positional relationships based on the drawings are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements to be referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.

In the description of the present invention, it should be noted that, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; the connection may be direct or indirect via an intermediate medium, or may be internal communication between two elements. The specific meaning of the above terms in the present invention will be understood in specific cases by those of ordinary skill in the art.

The compound monitoring method for the box-type substation comprises a debugging step of S10, a cyclic monitoring step of S20-S70 and a restarting step of S80, which are combined to form a complete computer vision algorithm technology, so that data such as images, temperature and humidity can be obtained in real time, and abnormal conditions can be found in time to give an alarm by calculating and judging the luminous conditions on the images. The specific content and principle of each step are as follows:

S10: referring to fig. 1, a video monitoring apparatus 2 is installed at each compartment 1 in a station, and the video monitoring apparatus 2 uses an industrial video monitoring apparatus with its own coordinate system, and an image taken by the video monitoring apparatus has an origin of coordinates, and then each pixel point can be corresponding to a fixed coordinate value. Then, acquiring an image in each compartment 1, photographing a luminous part of an electrical element in the image, manually marking a luminous area in the image according to the luminous part in the image, namely marking coordinates of pixel points of the luminous part of the corresponding electrical element in the image, and recording the coordinates in a system; the initial values of the damage value brooken, the fire value fire, and the spark value spark are then set to 0. Thus, the preparation work of the system in the earlier stage is completed.

S20: each video monitoring device shoots images to form an image to be detected, and the temperature and the humidity in the station are obtained through a temperature sensing detector and a humidity detector; different environmental conditions can cause different influences on abnormality such as damage probability, fire probability and the like of electrical elements in a transformer substation, so that the system needs to be capable of adjusting sensitivity of the system according to different environmental conditions.

S30: carrying out graying treatment and binarization treatment on the image to be detected, wherein the graying treatment is to convert the image to be detected into a black-and-white image, so that each pixel point of the image has only a gray value; the binarization process is to adjust the gray scale of the pixel point smaller than the threshold value to 0 (i.e. black) according to the size of the set threshold value and adjust the gray scale of the pixel point larger than the threshold value to 255 (i.e. white), so as to form a processed image with only black and white at the final pixel point.

S40: judging the color of pixel points in a luminous area in a processed image, if the number of black pixel points is larger than a set threshold value, indicating that the video monitoring equipment does not shoot the bright light of an electric element indicator lamp at the moment, wherein the possibility that the electric element is damaged at the moment or the video monitoring equipment data is interfered is high, so that alarm judgment cannot be directly carried out at the moment.

Since the box-type substation is closed during operation, the interior of the substation is dim, and if other light sources are present in addition to the light source of the indicator light of the electrical component, these light sources are open flames or sparks. The positions of the light sources in the image can be rapidly screened out through the graying and binarizing treatment of the image to be detected, whether the light sources except the indicator lamp of the electrical element appear can be determined through comparing the coordinates of the pixel points corresponding to the light sources with the preset luminous areas, and then the system analysis and judgment are carried out to determine whether the light sources are sparks or open fires. The specific judging process is as follows:

S50: and marking white pixel points which are not in the light-emitting area on the processed image as suspicious pixel points, combining the connected suspicious pixel points into a suspicious area, and further dividing a plurality of suspicious areas.

S60: the suspicious region can be generally regarded as a light source, and when the area of the suspicious region exceeds a limit, there is a great possibility of a group of open fires, and according to the principle, the method calculates the maximum area S of the suspicious region; if S is larger than the area threshold value, adding 1 to the fire value to indicate that the possibility of open fire is increased, otherwise, maintaining the fire value;

When the number of suspicious regions is relatively large, namely, a plurality of unconnected light sources are newly added in the image, the light source distribution in the form is very likely to be caused by sparks, and according to the principle, the number of the suspicious regions is calculated and recorded as n by the method; if n is greater than the number threshold, then the spark value is increased by 1, indicating an increased likelihood of spark occurrence, otherwise the spark value is maintained.

S70: judging an alarm state:

if brooken is larger than the damage threshold, the suspicious value of the damage of the electrical element is accumulated to a certain extent in the previous step, and the damage of the electrical element is truly generated with the maximum probability, and then a damage alarm is sent out;

If fire is larger than the fire threshold, the possible value of open fire is accumulated to a certain extent in the previous step, and the fire alarm is sent out if the maximum probability is that the open fire is actually present;

If spark is greater than the spark threshold, indicating that the possible value of the spark has accumulated to a certain extent in the previous step, the maximum probability is that the spark is actually generated, and the spark is not dissipated in a period of time, namely, a continuous spark is generated in the station, and then a spark alarm is sent out;

if the alarm contains any one of damage and fire, the machine is stopped immediately to give an alarm, and if the alarm does not appear, the machine returns to S20 to start a new judging flow continuously.

When the number of times of returning to S20 reaches a circulation value, namely, the number of times of circulating S20-S70 to the circulation value, if brooken, fire and spark are unchanged in the period, indicating that the abnormality occurring in the period is misjudgment, and zeroing the values of brooken, fire and spark; if a change in the value of any of brooken, fire and spark occurs, the number of returns to S20 is again recorded starting from 0. The value of the cyclic value may vary during control, but such variation does not affect the count of the number of returns to S20.

S80: after the alarm is released and the machine is restarted, the values of brooken, fire and spark are zeroed.

By the method, images in the transformer substation can be acquired in real time, so that the working state, open fire and spark of electrical elements in the transformer substation can be monitored in real time, an alarm can be timely sent out when abnormal conditions occur, the safety of the box-type transformer substation is effectively improved, the whole monitoring process can be automatically completed through a computer arranged in the computer, automatic detection is realized, the labor burden is effectively reduced, and the efficiency is improved.

Because the visual angle of the video monitoring equipment is limited, the set light-emitting area can be possibly shielded by the shielding object to cause misjudgment, in the method, in the step S10, at least two video monitoring equipment are preferably arranged in each compartment in the station, a plurality of video monitoring equipment in the same compartment are all adjusted to different angles to shoot, when the light-emitting area is set, the light-emitting areas of the two video monitoring equipment are mutually corresponding, and whether the shielding condition exists is distinguished by comparing the conditions of the corresponding light-emitting areas, so that the method can perform more accurate early warning calculation judgment. The two video monitoring devices in the same compartment should be adjusted so that the light-emitting areas in the viewing angles of each video monitoring device do not overlap, and mutual interference between the light-emitting areas and the suspicious areas is avoided to influence judgment.

In order to ensure the accuracy of setting the light-emitting area, the method preferably removes the illumination in the compartment in step S10, only some bright light spot areas can be seen on the image shot by the video monitoring device at this time, the light spot areas can be irradiated by the indication lamp or other interference sources (such as reflection of a mirror surface, light entering through a wall gap, etc.), the image obtained in this state is used for marking the light-emitting point, the pixel coordinate point corresponding to the center of each light spot area is marked, and the size of the light spot area (i.e. the coordinates of all the pixels in the area) is recorded at the same time; and then keeping the illumination in each compartment sufficient (taking the condition that the electric elements corresponding to the indicator lamps can be seen clearly on the image shot by the video monitoring equipment) and using the image acquired in the state for checking the luminous points, screening the luminous points corresponding to the indicator lamps of the electric elements, and taking the luminous areas corresponding to the luminous points as final luminous areas. When the luminous points are verified, the luminous sources which do not belong to the luminous points of the indicator lamp of the electrical element are confirmed, and the light spot areas corresponding to the luminous sources can be selected to be calculated and eliminated (namely, the areas are set so that the pixel points in the areas are ignored in subsequent calculation, and the luminous points can be used for metal wall areas which cannot fire or spark at all) or eliminated (such as rubberized fabric is adhered on the surface of a reflector).

In order to extract the information in the image shot by the video monitoring device more accurately, in the method, preferably, in step S30, when the image to be detected is subjected to gray processing, weights of all color channels of the image are set, and finally, a gray image is comprehensively calculated, namely: the image shot by the video monitoring equipment is usually an RGB three-channel image, when the RGB three-channel image is converted into a gray value single-channel image through a graying method, the gray value and the R, G, B value of the RGB three-channel image have a certain conversion relation, on the basis of the traditional conversion, different weights are added to R, G, B by the method, for example, when the indicator lamps of the electrical elements are red LED lamps, the weight of R can be increased to enable the calculated gray value to be more distinguished from the background color, and therefore the extraction precision of information in the image is improved. In the method, preferably, in step S30, when the image to be detected is binarized, the image is divided into a plurality of areas, the local binarization is performed, and the threshold value of the binarization is adjusted according to the characteristics of each area, so that the local brightness information (especially, the area with larger background brightness) can be more highlighted.

In the method, in step S50, after the suspicious region is divided, the suspicious region information and the image to be detected are packaged and sent to an external storage system, and the external storage system can enable a hard disk to be locally arranged or be a control center to be remotely arranged.

During image processing calculation, the image to be detected, the corresponding gray level image and the binary image which are directly shot by the video monitoring equipment are required to be stored locally. Because the detection cycle interval of the system is usually only tens to hundreds of milliseconds, if the images are stored, a large number of pictures generated during different detection cycles are reserved locally, and if the pictures need to be called in calculation, the time for traversing to retrieve the corresponding pictures is prolonged along with the use of the system, so that the operation of the system is seriously affected. Therefore, in the method, preferably in step S70, after the alarm state is judged, the image to be detected and the processed image (i.e. the gray level image and the binary image) used in the calculation are deleted, and since the key information is already transmitted and recorded outwards in step S50, the locally reserved image is useless, i.e. can be deleted. Therefore, only one set of to-be-detected image and processing image can be ensured locally in each judgment and calculation processing, so that the system can quickly call the image, and the running speed of the method is ensured.

Preferably, in step S10, a threshold value comparison table is also established, in which the area threshold value, the number threshold value, the damage threshold value, the fire threshold value, and the spark threshold value and the cycle value are used as output values, and the temperature and the humidity of different values are associated with the area threshold value, the number threshold value, the damage threshold value, the fire threshold value, and the spark threshold value and the cycle value of different values. This arrangement allows for a strong scalability by only updating the look-up table if new thresholds need to be added or existing thresholds need to be adjusted, without modifying the entire code structure. The comparison table can be reused in different systems, and only needs to be configured correspondingly according to specific conditions, so that the comparison table has good reusability.

The invention also provides a compound monitoring system for the box-type substation, which is applied to the compound monitoring method for the box-type substation, and comprises the following steps:

the video monitoring devices are respectively arranged in each compartment in the station and are used for shooting images;

the temperature sensing detector is used for detecting the temperature in the cabinet;

The humidity detector is used for detecting the humidity in the cabinet;

the alarm module is used for sending out an alarm, and comprises a siren, an alarm bell, an audible and visual alarm and the like;

the control module takes computer technology as a core, has the functions of image input, processing, storage, display, control, remote transmission, network and the like, meets the requirements of industrial application, and is used for recording and analyzing images acquired by the video monitoring equipment and controlling the alarm system to work.

The manual alarm system can be installed in the station generally, the alarm is pressed down, the self-locking function is realized after the alarm is pressed down, the manual reset is needed, and the glass sheet can be reused.

The foregoing has outlined and described the basic principles, features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (10)

1. A composite monitoring method for a box-type substation, comprising the steps of:

S10: installing video monitoring equipment in each compartment in the station, then acquiring an image in each compartment, and manually marking out a luminous area in the image; presetting a damage value brooken, a fire value fire and a spark value spark to be 0;

S20: shooting images by each video monitoring device to form an image to be detected, and acquiring the temperature and humidity in the station; determining area, quantity, damage, fire and spark thresholds and cycle values from temperature and humidity;

s30: carrying out graying treatment and binarization treatment on the image to be detected to form a treated image;

S40: judging the color of pixel points in a light-emitting area in the processed image, if the number of black pixel points is larger than a set threshold value, adding 1 to the brookfen value, otherwise, maintaining the brookfen value;

s50: marking white pixel points which are not in a light-emitting area on a processed image as suspicious pixel points, combining the connected suspicious pixel points into a suspicious area, and further dividing a plurality of suspicious areas;

S60: calculating the maximum area S of the suspicious region, and recording the number of the suspicious region as n;

If S is larger than the area threshold, adding 1 to the fire value, otherwise, maintaining the fire value;

if n is greater than the quantity threshold, adding 1 to the spark value, otherwise, maintaining the spark value;

s70: judging an alarm state:

if brooken is greater than the damage threshold, a damage alarm is sent out;

if fire is greater than the fire threshold, giving out fire alarm;

if spark is larger than the spark threshold value, a spark alarm is sent out;

if the alarm contains any one of damage and fire, stopping the machine, otherwise returning to S20;

When the number of times of returning to S20 reaches a cyclic value and the values of brooken, fire and spark are unchanged in the period, the values of brooken, fire and spark are zeroed;

S80: after the alarm is released and the machine is restarted, the values of brooken, fire and spark are zeroed.

2. The composite monitoring method for a box-type substation according to claim 1, wherein in step S10, at least two video monitoring devices are installed in each compartment in the substation, and a plurality of video monitoring devices in the same compartment are all adjusted to different angles for shooting, so that the light emitting areas in the view angles of each video monitoring device do not overlap.

3. The composite monitoring method for a box-type substation according to claim 1, wherein in step S10, the illumination in the compartment is removed first, the image obtained in this state is used to mark the light-emitting point, and the light-emitting area corresponding to the light-emitting point is recorded;

And then keeping the illumination in each compartment sufficient, using the image acquired in the state for checking the luminous points, screening out the luminous points which can correspond to the indicator lamps of the electrical elements, and taking the luminous areas corresponding to the luminous points as final luminous areas.

4. A compound monitoring method for a box-type substation according to claim 3, wherein when verifying the luminous points, the luminous sources of the luminous points not belonging to the electric component indicator lamp are confirmed, and whether the unlabeled luminous points need to be reserved is confirmed according to the luminous sources.

5. The composite monitoring method for a box-type substation according to claim 1, wherein in step S30, when the image to be inspected is subjected to the graying process, weights of the color channels of the image are set, and finally the gray image is calculated in a combined manner.

6. The composite monitoring method for a box-type substation according to claim 1, wherein in step S30, when the image to be inspected is binarized, the image is divided into a plurality of regional memorial local binarizations, and the threshold value of the binarization is adjusted according to the characteristics of each region.

7. The composite monitoring method for a box-type substation according to claim 1, wherein in step S50, after the suspicious region is divided, suspicious region information and the image to be inspected are packaged and sent to an external storage system.

8. The composite monitoring method for a box-type substation according to claim 7, wherein in step S70, after the alarm state is judged, the image to be inspected and the processed image used for the calculation are deleted.

9. The composite monitoring method for a box-type substation according to claim 1, wherein in step S10, a threshold value comparison table is also established so that different values of temperature and humidity correspond to different values of area threshold value, number threshold value, damage threshold value, fire threshold value and spark threshold value and cycle value.

10. A composite monitoring system for a box-type substation, characterized in that it is applied to the composite monitoring method for a box-type substation according to any one of claims 1 to 9, comprising:

the video monitoring devices are respectively arranged in each compartment in the station and are used for shooting images;

the temperature sensing detector is used for detecting the temperature in the cabinet;

The humidity detector is used for detecting the humidity in the cabinet;

the alarm module is used for sending out an alarm;

and the control module is used for recording and analyzing the image acquired by the video monitoring equipment and controlling the alarm system to work.