Power cabinet alarm system based on image analysis

Abstract

The present invention discloses an electric cabinet alarm system based on image analysis, and relates to the technical field of electric cabinet monitoring; it comprises a collection and deployment module, a regional analysis module, a comprehensive evaluation module and an alarm triggering module; the present invention divides the monitoring area of ​​the electric cabinet and deploys corresponding sensors to collect images and temperature data of the corresponding divided areas, so as to quantify the abnormality degree of each divided area, and combines image analysis and temperature monitoring to improve the accuracy of the alarm, thereby solving the problem that the area of ​​the electric cabinet cannot be divided in the prior art, and when analyzing the images of each divided area, the monitoring is one-sided in combination with the temperature conditions of each divided area.

Description

Technical Field

[0001] The present invention relates to the technical field of power cabinet monitoring, and in particular to a power cabinet alarm system based on image analysis. Background Art

[0002] The power cabinet is a special cabinet structure used to accommodate and protect power equipment in the power system. It is widely used in substations, distribution rooms, industrial plants, commercial buildings and other occasions. In the daily use of the power cabinet, it is necessary to monitor the use of the power cabinet so as to trigger an alarm when an abnormal situation occurs in the power cabinet.

[0003] In the prior art, monitoring, analysis and alarming of power cabinets are realized through image analysis, but there are still the following shortcomings:

[0004] The monitoring accuracy and pertinence are not high. The power cabinet cannot be divided into areas. When analyzing the images of each divided area, combined with the temperature conditions of each divided area, the monitoring is one-sided.

[0005] Furthermore, when the power cabinet triggers an alarm, most of the time, an alarm is used to prompt, and targeted processing cannot be performed according to the cause of the power cabinet triggering the alarm. The alarm processing efficiency is low, which affects the use of the power cabinet.

[0006] Therefore, a power cabinet alarm system based on image analysis is introduced. Summary of the invention

[0007] In view of this, the present invention provides a power cabinet alarm system based on image analysis to solve the problems raised by the above background technology.

[0008] The purpose of the present invention can be achieved through the following technical solutions: A power cabinet alarm system based on image analysis, comprising:

[0009] Collection and deployment module: divide the monitoring area of ​​the power cabinet, mark the divided areas into cabinet area, line area and ventilation area, and represent the corresponding divided areas with numbers i, where i=1, 2 or 3, corresponding to the cabinet area, line area and ventilation area respectively, and deploy sensors in the corresponding divided areas. Based on the preset monitoring time interval, when the preset monitoring time interval is reached, the status parameter data packets of each divided area within the set monitoring time period are collected and sent to the regional analysis module;

[0010] The state parameters of the cabinet area include the cabinet internal temperature and the cabinet internal image;

[0011] The status parameters of the line area include cable temperature and cable insulation layer image;

[0012] The status parameters of the ventilation area include vent temperature and vent image;

[0013] Regional analysis module: receives the status parameter data packets of each divided area within the set monitoring time period, analyzes them respectively, and obtains the abnormal evaluation index of each divided area within the set monitoring time period. ; Preset abnormal assessment index for each divided area The threshold index of the corresponding divided area will be They are compared with the corresponding threshold indexes respectively. If the abnormal evaluation index of a certain divided area If the abnormal evaluation index of each divided area is greater than the corresponding threshold index, an alarm signal of the corresponding divided area is generated and sent to the alarm trigger module; the early warning signal includes the cabinet alarm signal, the line alarm signal and the ventilation alarm signal; if the abnormal evaluation index of each divided area is greater than the corresponding threshold index, an alarm signal of the corresponding divided area is generated and sent to the alarm trigger module; the early warning signal includes the cabinet alarm signal, the line alarm signal and the ventilation alarm signal; If all of them are less than the corresponding threshold index, the overall analysis signal is triggered to the comprehensive evaluation module;

[0014] Comprehensive evaluation module: When the overall analysis signal is triggered, the abnormal evaluation index of each divided area is extracted , according to each divided area, the abnormal evaluation index is obtained , construct an evaluation model of the power cabinet within the set monitoring time period, and send the constructed evaluation model to the alarm trigger module;

[0015] Alarm trigger module: receives the signal generated by the power cabinet within the set monitoring time period and executes the corresponding steps; is also used to receive the evaluation model of the power cabinet within the set monitoring time period and execute the corresponding steps.

[0016] In some embodiments, the abnormal evaluation index of each divided area within the set monitoring time period is obtained. , specifically:

[0017] Obtain the temperature values ​​of each deployment point in the cabinet area within the set monitoring time period, mark the cabinet area deployment points as k, k=1, 2 or p, where p is the total number of cabinet area deployment points; take the average of the temperature values ​​of each deployment point at each time point within the set monitoring time period, and use the average as the temperature estimate of each deployment point within the set monitoring time period;

[0018] Preset the normal reference interval of the temperature estimation of each deployment point in the cabinet area; compare the temperature estimation of each deployment point in the cabinet area with the corresponding normal reference interval. If the temperature estimation of a deployment point is greater than the corresponding normal reference interval, mark the deployment point as an abnormal temperature point;

[0019] The number of abnormal temperature points in the cabinet area within the set monitoring time period is counted, recorded as Yj; the maximum temperature value in the normal reference interval corresponding to each abnormal temperature point is further extracted, recorded as the maximum reference value, and the difference between the temperature estimate corresponding to each abnormal temperature point and the maximum temperature value is calculated, recorded as the temperature difference value;

[0020] Preset an additional coefficient for a temperature difference value corresponding to each deployment point; multiply the temperature difference value of each abnormal temperature point by the corresponding additional coefficient, and then sum them up to obtain the abnormal temperature value of the cabinet area within the set monitoring time period, which is recorded as Yt;

[0021] The cabinet internal image within the set monitoring time period is obtained. After preprocessing the internal image, each component area inside the cabinet is separated from the background, and the damage features of each component area are identified; the number of pixels of each damage feature of the corresponding component area is counted, and the actual size is converted based on the resolution of the image to obtain the damage area of ​​the corresponding component area;

[0022] A damage area weight coefficient is preset for each component area, and the damage area of ​​each component area is multiplied by the corresponding weight coefficient. Then the sum is calculated to obtain the abnormal damage value of the cabinet area within the set monitoring time period, which is recorded as Yu.

[0023] In some embodiments, the abnormal evaluation index of each divided area within the set monitoring time period is obtained. , further comprising:

[0024] Obtain the temperature value of each sub-line cable in the line area within the set monitoring time period, take the average value of the temperature value of each sub-line cable at each time point within the set monitoring time period as the cable average temperature value of each sub-line cable, recorded as Dex, where x represents the number of the sub-line cable, x=1,2 or h, h is the total number of sub-line cables, and use the standard deviation formula to calculate the temperature value of each sub-line cable at each time point within the set monitoring time period to obtain the cable temperature change value of each sub-line cable, recorded as Dnx; obtain the insulation layer image of each sub-line cable in the line area, identify the abnormal features of the insulation layer image corresponding to each sub-line cable, and separate the abnormal features of each insulation layer image from the background; the abnormal features include shedding, fracture and corrosion; count the number of pixels of the shedding feature, fracture feature and corrosion feature in each insulation layer image, and convert the actual size based on the resolution of the image to obtain the shedding area, fracture area and corrosion area of ​​each insulation layer image; and mark them as Dax, Dbx and Dcx respectively;

[0025] The temperature value of each vent in the ventilation area is obtained within the set monitoring time period; the average value of the temperature of each vent at each time point within the set monitoring time period is taken as the ventilation average temperature value of each vent, recorded as Mqf, where f represents the number of each vent, f=1, 2 or g, where g is the total number of vents; the vent images of each vent in the ventilation area are obtained, and the dust area of ​​each vent image is identified, and the dust area of ​​each vent image is separated from the background by image segmentation technology, and the total number of pixels of the dust area of ​​each vent is counted, and the dust adhesion area of ​​each vent image is obtained after conversion to the actual size based on the resolution of the image, which is recorded as Msf.

[0026] In some embodiments, the abnormal evaluation index of each divided area within the set monitoring time period is obtained. , further comprising:

[0027] According to the formula , perform weighted calculation to obtain the abnormal evaluation index of each divided area within the set monitoring time period ,in , as well as They respectively represent the maximum allowable proportion of the number of abnormal temperature points in the cabinet area to the total number p, the maximum allowable value of the abnormal temperature value, and the maximum allowable value of the abnormal damage value; , as well as is the preset influence weight factor;

[0028] , , , as well as They respectively represent the maximum allowable value of the average temperature of the cable in the line area, the maximum allowable value of the cable temperature change, the maximum allowable value of the shedding area, the maximum allowable value of the fracture area and the maximum allowable value of the corrosion area; , , , as well as is the preset influence weight factor;

[0029] and They represent the normal reference value of the average ventilation temperature in the ventilation area and the reference allowable value of the dust adhesion area respectively; and is the preset influence weight factor.

[0030] In some embodiments, an evaluation model of a power cabinet within a set monitoring time period is constructed, specifically:

[0031] Extract the abnormal evaluation index of the cabinet area, line area and ventilation area within the set monitoring time period , as well as ;

[0032] The abnormal evaluation index of the power cabinet within the set monitoring time period , as well as The three-dimensional rectangular modules are converted into length values, width values ​​and height values ​​respectively, thereby constructing a three-dimensional rectangular module, and the constructed three-dimensional rectangular module is used as an evaluation model of the power cabinet within the set monitoring time period;

[0033] Preset abnormal assessment index separately , as well as The range of values ​​of each group of indices, and the abnormal evaluation index , as well as Each group of index value ranges corresponds to a length value, a width value, and a height value, and the abnormal evaluation index of the power cabinet within the monitoring time period is set. , as well as Match them with the corresponding index value ranges respectively to obtain the length value, width value and height value of the evaluation model of the power cabinet within the set monitoring time period.

[0034] In some embodiments, a signal generated by a power cabinet within a set monitoring time period is received, and corresponding steps are performed, specifically:

[0035] M1: If a ventilation alarm signal is generated, the ventilation area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the ventilation alarm signal is triggered. After the trigger is completed, a circle is drawn with the power cabinet as the center and the set distance as the radius. The maintenance personnel who are in the selectable state within the circle are screened, and the position feedback signal is sent to their mobile terminals. After each maintenance personnel confirms, the distance between each maintenance personnel and the power cabinet within the circle is obtained, and the maintenance personnel with the shortest distance is selected as the person handling the ventilation alarm signal, and the location of the power cabinet and the abnormal evaluation index of the ventilation area are sent. to the person's mobile terminal;

[0036] M2: If a cabinet alarm signal is generated, the cabinet area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the cabinet alarm signal is triggered. After the triggering is completed, the distance of each maintenance personnel is obtained based on step M1, which is recorded as Lc; further obtain the work log of each maintenance personnel, and obtain the working hours and historical selection times of each maintenance personnel from the work log of each maintenance personnel, and mark the working hours of each maintenance personnel as Lb;

[0037] The time taken for each selection is obtained from the historical selection times of each maintenance personnel, and the average time taken by each group of maintenance personnel is taken to obtain the average solution time of each maintenance personnel, which is recorded as Lm;

[0038] According to the formula Perform weighted calculation to obtain the preferred value V of each maintenance personnel within the current circle; , as well as is the preset influence weight factor;

[0039] The maintenance personnel with the largest preferred value V are selected as the personnel who handle the cabinet alarm signal, and sent to the location of the power cabinet and the abnormal evaluation index of the cabinet area. to the person's mobile terminal;

[0040] M3: If a line alarm signal is generated, the line area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the line alarm signal is triggered. After the triggering is completed, the maintenance personnel with the largest preferred value V is selected as the person who handles the line alarm signal in the same way as step M2, and the abnormal evaluation index of the line area is sent to the location of the power cabinet and the line area. to the person’s mobile terminal.

[0041] In some embodiments, an evaluation model of a power cabinet within a set monitoring time period is received, and corresponding steps are performed, specifically:

[0042] Extract the preset model of the power cabinet, and use the evaluation model of the power cabinet within the set monitoring period to dig a groove on the preset model, thereby constructing an early warning model of the power cabinet within the set monitoring period;

[0043] Mark the center points of the bottom surface of the warning model and the bottom surface of the trough respectively, and calculate the vertical distance between the two sets of center points, which is recorded as the height distance;

[0044] Calculate the distance between the slot width line and the preset model width line, which is recorded as the width distance;

[0045] Calculate the distance between the slot length line and the preset model length line, which is recorded as the length distance;

[0046] Mark the height distance, width distance, and length distance as Zq1, Zq2, and Zq3 respectively, and substitute them into the formula , perform weighted calculation to obtain the early warning evaluation index of the power cabinet within the set monitoring time period ; , as well as is the preset influence weight factor;

[0047] Setting early warning assessment index If the threshold index of the power cabinet is set within the monitoring period, If the index is greater than the set threshold, the warning signal is triggered and the warning model and warning evaluation index within the set monitoring period are set. The data are integrated into a data packet and sent to the mobile terminal of the maintenance personnel with the shortest distance.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention divides the power cabinet into monitoring areas and deploys corresponding sensors to collect images and temperature data of the corresponding divided areas, so as to quantify the abnormality of each divided area. By combining image analysis and temperature monitoring, the accuracy of alarm is improved, and the problem that the prior art cannot divide the area of ​​the power cabinet and when analyzing the image of each divided area, combined with the temperature conditions of each divided area, the monitoring is one-sided.

[0050] The present invention can accurately identify and locate the specific area where the abnormality occurs based on the specific signal type that triggers the alarm when the alarm is triggered, and execute corresponding steps according to the specific signal type that triggers the alarm, thereby improving the efficiency of alarm processing, and solving the problem that most of the existing technologies use alarms to provide prompts, but cannot perform targeted processing according to the cause that causes the power cabinet to trigger the alarm, resulting in low alarm processing efficiency, which affects the use of the power cabinet. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Further details, features and advantages of the present application are disclosed in the following description of exemplary embodiments in conjunction with the accompanying drawings, in which:

[0052] Figure 1 It is a principle block diagram of the present invention;

[0053] Figure 2 It is a schematic diagram of the early warning model in the present invention. DETAILED DESCRIPTION

[0054] Several embodiments of the present application will be described in more detail below with reference to the accompanying drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and to fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0055] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0056] See also Figure 1-Figure 2 As shown, a power cabinet alarm system based on image analysis includes a collection and deployment module, a regional analysis module, a comprehensive evaluation module and an alarm triggering module;

[0057] The collection and deployment module is used to divide the monitoring area of ​​the power cabinet, mark the divided areas into cabinet area, line area and ventilation area, and represent the corresponding divided areas with numbers i, where i=1, 2 or 3, corresponding to the cabinet area, line area and ventilation area respectively, and deploy sensors in the corresponding divided areas. Based on the preset monitoring time interval, when the preset monitoring time interval is reached, the state parameter data packets of each divided area within the set monitoring time period are collected and sent to the regional analysis module; the deployed sensors include temperature sensors and high-definition cameras;

[0058] The state parameters of the cabinet area include the cabinet internal temperature and the cabinet internal image;

[0059] The status parameters of the line area include cable temperature and cable insulation layer image;

[0060] The status parameters of the ventilation area include vent temperature and vent image;

[0061] The regional analysis module is used to receive the status parameter data packets of each divided area within the set monitoring time period, analyze the status parameters of each divided area within the set monitoring time period after parsing them respectively, and obtain the abnormal evaluation index of each divided area within the set monitoring time period. ; Preset abnormal assessment index for each divided area The threshold index of the corresponding divided area will be They are compared with the corresponding threshold indexes respectively. If the abnormal evaluation index of a certain divided area If the abnormal evaluation index of each divided area is greater than the corresponding threshold index, an alarm signal of the corresponding divided area is generated and sent to the alarm trigger module; the early warning signal includes the cabinet alarm signal, the line alarm signal and the ventilation alarm signal; if the abnormal evaluation index of each divided area is greater than the corresponding threshold index, an alarm signal of the corresponding divided area is generated and sent to the alarm trigger module; the early warning signal includes the cabinet alarm signal, the line alarm signal and the ventilation alarm signal; If all of them are less than the corresponding threshold index, the overall analysis signal is triggered to the comprehensive evaluation module;

[0062] Get the abnormal evaluation index of each divided area within the set monitoring time period , specifically:

[0063] S1: Obtain the temperature value of each deployment point in the cabinet area within the set monitoring time period, mark the deployment point in the cabinet area as k, k=1, 2 or p, where p is the total number of deployment points in the cabinet area; deployment points include but are not limited to circuit breakers, switches and terminal blocks, which are specifically set and deployed by technical personnel; take the average of the temperature values ​​of each deployment point at each time point in the set monitoring time period, and use the average as the temperature estimate of each deployment point in the set monitoring time period;

[0064] Preset the normal reference interval of the temperature estimation of each deployment point in the cabinet area; compare the temperature estimation of each deployment point in the cabinet area with the corresponding normal reference interval. If the temperature estimation of a deployment point is greater than the corresponding normal reference interval, mark the deployment point as an abnormal temperature point;

[0065] The number of abnormal temperature points in the cabinet area within the set monitoring time period is counted, recorded as Yj; the maximum temperature value in the normal reference interval corresponding to each abnormal temperature point is further extracted, recorded as the maximum reference value, and the difference between the temperature estimate corresponding to each abnormal temperature point and the maximum temperature value is calculated, recorded as the temperature difference value;

[0066] Each deployment point is preset to correspond to an additional coefficient of the temperature difference value; the additional coefficient value range is 1.108-1.385, and the specific value is set by the technician based on the specific location of the deployment point;

[0067] Multiply the temperature difference of each abnormal temperature point by the corresponding additional coefficient, and then sum them up to get the abnormal temperature value of the cabinet area within the set monitoring time period, recorded as Yt;

[0068] Obtain the internal images of the cabinet area within the set monitoring time period, and pre-process the internal images; the pre-processing includes but is not limited to denoising, contrast enhancement and edge detection; use image segmentation technology to separate the component areas inside the cabinet from the background; the component areas include but are not limited to circuit breakers, switches and wiring terminals, which are specifically set by technical personnel; identify the damage features of each component area; damage features such as cracks, notches and corrosion; count the number of pixels of each damage feature of the corresponding component area, convert the actual size based on the resolution of the image, and obtain the damage area of ​​the corresponding component area;

[0069] Each component area is preset to correspond to a specific gravity coefficient of the damaged area; the specific gravity coefficient value range is 1.213-1.586, and the specific value is set by the technicians based on the specific location of the component area;

[0070] Multiply the damaged area of ​​each component area by the corresponding specific gravity coefficient, and then sum them up to get the abnormal damage value of the cabinet area within the set monitoring time period, which is recorded as Yu;

[0071] S2: Obtain the temperature value of each sub-line cable in the line area within the set monitoring time period, take the average value of the temperature value of each sub-line cable at each time point within the set monitoring time period as the cable average temperature value of each sub-line cable, recorded as Dex, where x represents the number of the sub-line cable, x=1,2 or h, h is the total number of sub-line cables, use the standard deviation formula to calculate the temperature value of each sub-line cable at each time point within the set monitoring time period, and obtain the cable temperature change value of each sub-line cable, recorded as Dnx;

[0072] Obtain the insulation layer images of each sub-line cable in the line area, identify the abnormal features of the insulation layer images corresponding to each sub-line cable, and use image segmentation technology to separate the abnormal features of each insulation layer image from the background; the abnormal features include shedding, fracture and corrosion; count the number of pixels of shedding features, fracture features and corrosion features in each insulation layer image, and convert the actual size based on the resolution of the image to obtain the shedding area, fracture area and corrosion area of ​​each insulation layer image; and mark them as Dax, Dbx and Dcx respectively;

[0073] S3: Obtain the temperature value of each vent in the ventilation area within the set monitoring time period; take the average value of the temperature value of each vent at each time point within the set monitoring time period as the ventilation average temperature value of each vent, recorded as Mqf, where f represents the number of each vent, f=1, 2 or g, where g is the total number of vents;

[0074] The vent images of each vent in the ventilation area are obtained, and the dust area of ​​each vent image is identified. The dust area of ​​each vent image is separated from the background by using image segmentation technology, and the total number of pixels of the dust area of ​​each vent is counted. After the actual size is converted based on the resolution of the image, the dust attachment area of ​​each vent image is obtained, which is recorded as Msf;

[0075] S4: According to the formula , perform weighted calculation to obtain the abnormal evaluation index of each divided area within the set monitoring time period ,in , and It represents the maximum allowable proportion of the number of abnormal temperature points in the total number p in the cabinet area, the maximum allowable value of the abnormal temperature value, and the maximum allowable value of the abnormal damage value; , as well as is the preset influence weight factor;

[0076] , , , as well as They respectively represent the maximum allowable value of the average temperature of the cable in the line area, the maximum allowable value of the cable temperature change, the maximum allowable value of the shedding area, the maximum allowable value of the fracture area and the maximum allowable value of the corrosion area; , , , as well as is the preset influence weight factor;

[0077] and They represent the normal reference value of the average ventilation temperature in the ventilation area and the reference allowable value of the dust adhesion area respectively; and is the preset influence weight factor;

[0078] It should be noted that the reference values ​​and allowable values ​​corresponding to the above parameters are set by technical personnel and can be adjusted later according to actual conditions.

[0079] The comprehensive evaluation module is used to extract the abnormal evaluation index of each divided area when the overall analysis signal is triggered. , according to each divided area, the abnormal evaluation index is obtained , construct an evaluation model of the power cabinet within the set monitoring time period, and send the constructed evaluation model to the alarm trigger module;

[0080] Construct an evaluation model for the power cabinet within the set monitoring time period, specifically:

[0081] Extract the abnormal evaluation index of the cabinet area, line area and ventilation area within the set monitoring time period , as well as ;

[0082] The abnormal evaluation index of the power cabinet within the set monitoring time period , as well as The three-dimensional rectangular modules are converted into length values, width values ​​and height values ​​respectively, thereby constructing a three-dimensional rectangular module, and the constructed three-dimensional rectangular module is used as an evaluation model of the power cabinet within the set monitoring time period;

[0083] Preset abnormal assessment index separately , as well as The range of values ​​of each group of indices, and the abnormal evaluation index , as well as Each group of index value ranges corresponds to a length value, a width value, and a height value, and the abnormal evaluation index of the power cabinet within the monitoring time period is set. , as well as Match them with the corresponding index value ranges of each group respectively to obtain the length value, width value and height value of the evaluation model of the power cabinet within the set monitoring time period;

[0084] It should be noted that the above-mentioned index value ranges are set by technical personnel, and the subsequent index value ranges can be adjusted according to actual conditions. The larger the index of each group, the larger the length value, width value and height value obtained by matching;

[0085] Assume that during a monitoring process within a set monitoring time period, the following abnormal evaluation index is obtained:

[0086] =2.35; =9.86; =4.97;

[0087] According to the preset index value range, (2-2.5) = 7.5cm, (9.5-10) = 5.5cm, (4.5-5) = 4cm, then the power cabinet evaluation module we constructed is a three-dimensional rectangular model with a length of 7.5cm, a width of 5.5cm, and a height of 4cm.

[0088] The alarm trigger module is used to receive the signal generated by the power cabinet within the set monitoring time period and execute the corresponding steps;

[0089] M1: If a ventilation alarm signal is generated, the ventilation area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the ventilation alarm signal is triggered. After the triggering is completed, a circle is drawn with the power cabinet as the center and the set distance as the radius, and the maintenance personnel who are in the selectable state within the circle are screened, and a position feedback signal is sent to their mobile terminals. After each maintenance personnel confirms, the distance between each maintenance personnel and the power cabinet within the circle is obtained, and the maintenance personnel with the shortest distance is selected as the person who handles the ventilation alarm signal, and the location of the power cabinet and the abnormal evaluation index of the ventilation area are sent to the person's mobile terminal;

[0090] M2: If a cabinet alarm signal is generated, the cabinet area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the cabinet alarm signal is triggered. After the triggering is completed, the distance of each maintenance personnel is obtained based on step M1, which is recorded as Lc; further obtain the work log of each maintenance personnel, and obtain the working hours and historical selection times of each maintenance personnel from the work log of each maintenance personnel, and mark the working hours of each maintenance personnel as Lb;

[0091] The time taken for each selection is obtained from the historical selection times of each maintenance personnel, and the average time taken by each group of maintenance personnel is taken to obtain the average solution time of each maintenance personnel, which is recorded as Lm;

[0092] According to the formula Perform weighted calculation to obtain the preferred value V of each maintenance personnel within the current circle; , as well as is the preset influence weight factor;

[0093] The maintenance personnel with the largest preferred value V are selected as the personnel who handle the cabinet alarm signal, and sent to the location of the power cabinet and the abnormal evaluation index of the cabinet area. to the person's mobile terminal;

[0094] M3: If a line alarm signal is generated, the line area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the line alarm signal is triggered. After the triggering is completed, the maintenance personnel with the largest preferred value V is selected as the person who handles the line alarm signal in the same way as step M2, and the abnormal evaluation index of the line area is sent to the location of the power cabinet and the line area. to the person's mobile terminal;

[0095] It should be noted that different alarm signals correspond to different alarm parameters, which specifically include but are not limited to the sound rhythm type, sound frequency difference and sound duration, which are specifically set by the technicians;

[0096] It is also used to receive the evaluation model of the power cabinet within the set monitoring time period and execute corresponding steps; specifically:

[0097] Extract the preset model of the power cabinet; set it by the technician, and the preset model is larger than the maximum size of the evaluation model; use the evaluation model of the power cabinet in the set monitoring time period to dig a groove on the preset model, thereby constructing an early warning model of the power cabinet in the set monitoring time period;

[0098] Mark the center points of the bottom surface of the warning model and the bottom surface of the trough respectively, and calculate the vertical distance between the two sets of center points, which is recorded as the height distance;

[0099] Calculate the distance between the slot width line and the preset model width line, which is recorded as the width distance;

[0100] Calculate the distance between the slot length line and the preset model length line, which is recorded as the length distance;

[0101] Mark the height distance, width distance, and length distance as Zq1, Zq2, and Zq3 respectively, and substitute them into the formula , perform weighted calculation to obtain the early warning evaluation index of the power cabinet within the set monitoring time period ; , as well as is the preset influence weight factor;

[0102] Setting early warning assessment index If the threshold index of the power cabinet is set within the monitoring period, If the index is greater than the set threshold, the warning signal is triggered and the warning model and warning evaluation index within the set monitoring period are set. Integrate into data packets and send them to the mobile terminal of the maintenance personnel with the shortest distance;

[0103] The above formulas are obtained by collecting a large amount of data and performing software simulation, and a formula close to the actual value is selected. The influencing weight factor and specific coefficient value in the formula are set by technical personnel in this field according to actual conditions, and can be adjusted and modified later.

[0104] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A power cabinet alarm system based on image analysis, characterized in that: include: Collection and deployment module: divide the monitoring area of ​​the power cabinet, mark the divided areas into cabinet area, line area and ventilation area, and represent the corresponding divided areas with numbers i, where i=1, 2 or 3, corresponding to the cabinet area, line area and ventilation area respectively, and deploy sensors in the corresponding divided areas. Based on the preset monitoring time interval, when the preset monitoring time interval is reached, the status parameter data packets of each divided area within the set monitoring time period are collected and sent to the regional analysis module; The state parameters of the cabinet area include the cabinet internal temperature and the cabinet internal image; The status parameters of the line area include cable temperature and cable insulation layer image; The status parameters of the ventilation area include vent temperature and vent image; Regional analysis module: receives the status parameter data packets of each divided area within the set monitoring time period, analyzes them respectively, and obtains the abnormal evaluation index of each divided area within the set monitoring time period. ; Preset abnormal assessment index for each divided area The threshold index of the corresponding divided area will be They are compared with the corresponding threshold indexes respectively. If the abnormal evaluation index of a certain divided area If the abnormal evaluation index of each divided area is greater than the corresponding threshold index, an alarm signal of the corresponding divided area is generated and sent to the alarm trigger module; the early warning signal includes the cabinet alarm signal, the line alarm signal and the ventilation alarm signal; if the abnormal evaluation index of each divided area is greater than the corresponding threshold index, an alarm signal of the corresponding divided area is generated and sent to the alarm trigger module; the early warning signal includes the cabinet alarm signal, the line alarm signal and the ventilation alarm signal; If all of them are less than the corresponding threshold index, the overall analysis signal is triggered to the comprehensive evaluation module; Get the abnormal evaluation index of each divided area within the set monitoring time period , the specific steps are: Obtain the temperature values ​​of each deployment point in the cabinet area within the set monitoring time period, mark the cabinet area deployment points as k, k=1, 2 or p, where p is the total number of cabinet area deployment points; take the average of the temperature values ​​of each deployment point at each time point within the set monitoring time period, and use the average as the temperature estimate of each deployment point within the set monitoring time period; Preset the normal reference range of temperature estimation for each deployment point in the cabinet area; Compare the temperature estimates of each deployment point in the cabinet area with the corresponding normal reference interval. If the temperature estimate of a deployment point is greater than the corresponding normal reference interval, mark the deployment point as an abnormal temperature point. The number of abnormal temperature points in the cabinet area within the set monitoring time period is counted, recorded as Yj; the maximum temperature value in the normal reference interval corresponding to each abnormal temperature point is further extracted, recorded as the maximum reference value, and the difference between the temperature estimate corresponding to each abnormal temperature point and the maximum temperature value is calculated, recorded as the temperature difference value; Preset an additional coefficient for a temperature difference value corresponding to each deployment point; multiply the temperature difference value of each abnormal temperature point by the corresponding additional coefficient, and then sum them up to obtain the abnormal temperature value of the cabinet area within the set monitoring time period, which is recorded as Yt; Obtain the cabinet interior image within the set monitoring time period, pre-process the interior image, separate each component area inside the cabinet from the background, and identify the damage characteristics of each component area; Count the number of pixels of each damage feature in the corresponding component area, convert the actual size based on the resolution of the image, and obtain the damage area of ​​the corresponding component area; A damage area ratio coefficient is preset for each component area, and the damage area of ​​each component area is multiplied by the corresponding ratio coefficient, and then the sum is obtained to obtain the abnormal damage value of the cabinet area within the set monitoring time period, which is recorded as Yu; Obtain the temperature value of each sub-line cable in the line area within the set monitoring time period, take the average value of the temperature value of each sub-line cable at each time point within the set monitoring time period as the cable average temperature value of each sub-line cable, recorded as Dex, where x represents the number of the sub-line cable, x=1,2 or h, h is the total number of sub-line cables, and use the standard deviation formula to calculate the temperature value of each sub-line cable at each time point within the set monitoring time period to obtain the cable temperature change value of each sub-line cable, recorded as Dnx; obtain the insulation layer image of each sub-line cable in the line area, identify the abnormal features of the insulation layer image corresponding to each sub-line cable, and separate the abnormal features of each insulation layer image from the background; the abnormal features include shedding, fracture and corrosion; count the number of pixels of the shedding feature, fracture feature and corrosion feature in each insulation layer image, and convert the actual size based on the resolution of the image to obtain the shedding area, fracture area and corrosion area of ​​each insulation layer image; and mark them as Dax, Dbx and Dcx respectively; Obtain the temperature value of each vent in the ventilation area within the set monitoring time period; take the average value of the temperature value of each vent at each time point within the set monitoring time period as the ventilation average temperature value of each vent, recorded as Mqf, where f represents the number of each vent, f=1,2 or g, where g is the total number of vents; obtain the vent image of each vent in the ventilation area, and identify the dust area of ​​each vent image, use image segmentation technology to separate the dust area of ​​each vent image from the background, and count the total number of pixels of the dust area of ​​each vent, and after converting the actual size based on the resolution of the image, obtain the dust attachment area of ​​each vent image, recorded as Msf; According to the formula , perform weighted calculation to obtain the abnormal evaluation index of each divided area within the set monitoring time period ,in , as well as They respectively represent the maximum allowable proportion of the number of abnormal temperature points in the cabinet area to the total number p, the maximum allowable value of the abnormal temperature value, and the maximum allowable value of the abnormal damage value; , as well as is the preset influence weight factor; , , , as well as They respectively represent the maximum allowable value of the average temperature of the cable in the line area, the maximum allowable value of the cable temperature change, the maximum allowable value of the shedding area, the maximum allowable value of the fracture area and the maximum allowable value of the corrosion area; , , , as well as is the preset influence weight factor; and They represent the normal reference value of the average ventilation temperature in the ventilation area and the reference allowable value of the dust adhesion area respectively; and is the preset influence weight factor; Comprehensive evaluation module: When the overall analysis signal is triggered, the abnormal evaluation index of each divided area is extracted , according to each divided area, the abnormal evaluation index is obtained , construct an evaluation model of the power cabinet within the set monitoring time period, and send the constructed evaluation model to the alarm trigger module; Alarm trigger module: receives the signal generated by the power cabinet within the set monitoring time period and executes the corresponding steps; is also used to receive the evaluation model of the power cabinet within the set monitoring time period and execute the corresponding steps.

2. The power cabinet alarm system based on image analysis according to claim 1 is characterized in that: Construct an evaluation model for the power cabinet within the set monitoring time period, specifically: Extract the abnormal evaluation index of the cabinet area, line area and ventilation area within the set monitoring time period , as well as ; The abnormal evaluation index of the power cabinet within the set monitoring time period , as well as The three-dimensional rectangular modules are converted into length values, width values ​​and height values ​​respectively, thereby constructing a three-dimensional rectangular module, and the constructed three-dimensional rectangular module is used as an evaluation model of the power cabinet within the set monitoring time period; Preset abnormal assessment index separately , as well as The range of values ​​of each group of indices, and the abnormal evaluation index , as well as Each group of index value ranges corresponds to a length value, a width value, and a height value, and the abnormal evaluation index of the power cabinet within the monitoring time period is set. , as well as Match them with the corresponding index value ranges respectively to obtain the length value, width value and height value of the evaluation model of the power cabinet within the set monitoring time period.

3. The power cabinet alarm system based on image analysis according to claim 1 is characterized in that: Receive the signal generated by the power cabinet within the set monitoring time period and execute the corresponding steps, specifically: M1: If a ventilation alarm signal is generated, the ventilation area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the ventilation alarm signal is triggered. After the trigger is completed, a circle is drawn with the power cabinet as the center and the set distance as the radius. The maintenance personnel who are in the selectable state within the circle are screened, and the position feedback signal is sent to their mobile terminals. After each maintenance personnel confirms, the distance between each maintenance personnel and the power cabinet within the circle is obtained. The maintenance personnel with the shortest distance is selected as the person handling the ventilation alarm signal, and the location of the power cabinet and the abnormal evaluation index of the ventilation area are sent. to the person's mobile terminal; M2: If a cabinet alarm signal is generated, the cabinet area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the cabinet alarm signal is triggered. After the triggering is completed, the distance of each maintenance personnel is obtained based on step M1, which is recorded as Lc; further obtain the work log of each maintenance personnel, and obtain the working hours and historical selection times of each maintenance personnel from the work log of each maintenance personnel, and mark the working hours of each maintenance personnel as Lb; The time taken for each selection is obtained from the historical selection times of each maintenance personnel, and the average time taken by each group of maintenance personnel is taken to obtain the average solution time of each maintenance personnel, which is recorded as Lm; According to the formula Perform weighted calculation to obtain the preferred value V of each maintenance personnel within the current circle; , as well as is the preset influence weight factor; The maintenance personnel with the largest preferred value V are selected as the personnel who handle the cabinet alarm signal, and sent to the location of the power cabinet and the abnormal evaluation index of the cabinet area. to the person's mobile terminal; M3: If a line alarm signal is generated, the line area of ​​the power cabinet within the set monitoring time period is marked as an abnormal area, and the alarm corresponding to the line alarm signal is triggered. After the triggering is completed, the maintenance personnel with the largest preferred value V is selected as the person who handles the line alarm signal in the same way as step M2, and the abnormal evaluation index of the line area is sent to the location of the power cabinet and the line area. to the person’s mobile terminal.

4. The power cabinet alarm system based on image analysis according to claim 1 is characterized in that: Receive the evaluation model of the power cabinet within the set monitoring time period and execute corresponding steps, specifically: Extract the preset model of the power cabinet, and use the evaluation model of the power cabinet within the set monitoring period to dig a groove on the preset model, thereby constructing an early warning model of the power cabinet within the set monitoring period; Mark the center points of the bottom surface of the warning model and the bottom surface of the trough respectively, and calculate the vertical distance between the two sets of center points, which is recorded as the height distance; Calculate the distance between the slot width line and the preset model width line, which is recorded as the width distance; Calculate the distance between the slot length line and the preset model length line, which is recorded as the length distance; Mark the height distance, width distance, and length distance as Zq1, Zq2, and Zq3 respectively, and substitute them into the formula , perform weighted calculation to obtain the early warning evaluation index of the power cabinet within the set monitoring time period ; , as well as is the preset influence weight factor; Setting early warning assessment index If the threshold index of the power cabinet is set within the monitoring period, If the index is greater than the set threshold, the warning signal is triggered and the warning model and warning evaluation index within the set monitoring period are set. The data are integrated into a data packet and sent to the mobile terminal of the maintenance personnel with the shortest distance.

Images