Auxiliary detection tracking registration method and system for power distribution cabinet

By real-time monitoring of power load values ​​and deep learning models to identify abnormal areas, the problem of identifying high loads and abnormal areas in traditional distribution cabinet detection is solved, the intelligent and response speed of distribution cabinet management is improved, and the stability of power supply is ensured. Sex and equipment safety.

CN118822112BActive Publication Date: 2024-12-03ZHEJIANG JINDUN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411304703.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-19
Publication Date
2024-12-03
Estimated Expiration
2044-09-19

AI Technical Summary

Technical Problem

Traditional distribution cabinet detection, tracking and registration methods are difficult to accurately identify high-load time periods and abnormal areas, resulting in the inability to detect potential safety hazards in time, slow response speed, and unable to effectively avoid equipment losses.

Method used

By comparing the power load value in real time with preset standards, deep learning models and image processing technology are used to identify abnormal areas, and tracking registration procedures and fault warning mechanisms are triggered, detection components are dynamically adjusted to improve response speed and identification accuracy.

Benefits of technology

Accurate evaluation and visual display of the load status of the distribution cabinet, improve the response sensitivity and monitoring efficiency of the detection components, reduce manual inspection costs, improve the intelligence and safety of distribution room management, and ensure power supply stability and safe operation of the equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118822112B_ABST
    Figure CN118822112B_ABST
Patent Text Reader

Abstract

The present invention discloses an auxiliary detection tracking and registration method and system for a distribution cabinet, and relates to the technical field of distribution cabinet detection. The method comprises the following steps: deploying a sensor component in any selected distribution room, obtaining a real-time electrical parameter set corresponding to the distribution room and each distribution cabinet in the distribution room, building a comprehensive calculation model based on the electrical parameter set to generate a current power load value of the distribution room, and using a chart drawing tool to draw a power load curve; comparing the real-time power load value with a preset standard load; and outputting a comparison result for determining whether the distribution room is in a high-load state; the technical key points are: accurately determining the load state of the distribution room by comparing the power load value with the preset standard load in real time, and using a deep learning model and image processing technology to improve the accuracy of identifying abnormal areas of the distribution cabinet, and accelerating the response speed of abnormal processing by triggering a tracking registration program and a fault warning mechanism.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution cabinet detection, and in particular to an auxiliary detection tracking and registration method and system for a power distribution cabinet. Background Art

[0002] Regarding the detection, tracking and registration of distribution cabinets, this mainly involves the monitoring, tracking and registration management of distribution cabinets. The following is a detailed explanation:

[0003] 1. Monitoring of power distribution cabinet

[0004] Distribution cabinet monitoring is a vital part of the power system. It can monitor, manage and maintain the power distribution system in real time to ensure the reliability and safety of power supply. By introducing an intelligent distribution cabinet monitoring system, key parameters such as power load, voltage, current and equipment status can be monitored in real time.

[0005] 2. Tracking of distribution cabinet

[0006] Tracking of distribution cabinets mainly refers to continuous attention and recording of the operating status, maintenance records, and fault history of distribution cabinets. By tracking the operation of distribution cabinets, potential problems can be discovered in a timely manner and preventive measures can be taken. At the same time, tracking maintenance records can help maintenance personnel understand the historical status of distribution cabinets and provide reference for subsequent maintenance and repair work. In intelligent distribution cabinet management systems, tracking functions are usually combined with data analysis, early warning systems, and other functions to achieve comprehensive monitoring and management of distribution cabinets.

[0007] 3. Registration and management of distribution cabinets

[0008] The registration management of distribution cabinets refers to the registration, numbering, classification and file management of distribution cabinets; this helps enterprises understand the distribution, usage status and maintenance history of distribution cabinets; in the registration management process, it is usually necessary to record the basic information of the distribution cabinet (such as model, specification, manufacturer, etc.), installation location, user unit, maintenance records and other key information; this information is of great significance for subsequent inspection, maintenance and troubleshooting.

[0009] When detecting and tracking multiple distribution cabinets in a distribution room, it is usually necessary to first consider the load conditions of the corresponding distribution room, and perform high-frequency monitoring of each distribution cabinet in the distribution room during high-load or high-load time periods. However, how to determine the high-load time period? The traditional method is generally to define the time interval based on historical data, and use the time interval as the high-load time period. The accuracy of the load level assessment for a specific distribution room needs to be improved; at the same time, traditional distribution room management often relies on manual inspections and simple temperature monitoring methods, which cannot accurately identify abnormal areas of the distribution cabinet, resulting in potential safety hazards that cannot be discovered and handled in a timely manner; after discovering the abnormality of the distribution cabinet, the traditional processing method often involves multiple links and departments, with a slow response speed, and it is impossible to take timely measures to avoid or reduce the losses caused by equipment abnormalities. Summary of the invention

[0010] 1. Technical issues to be resolved

[0011] In view of the shortcomings of the prior art, the present invention provides an auxiliary detection, tracking and registration method and system for a distribution cabinet, which accurately determines the load status of the distribution room by comparing the power load value with the preset standard load in real time, and uses deep learning models and image processing technology to improve the accuracy of abnormal area identification in the distribution cabinet, and accelerates the response speed of abnormal processing by triggering the tracking registration program and fault warning mechanism, thereby solving the problems raised in the background technology.

[0012] (II) Technical solution

[0013] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0014] An auxiliary detection tracking and registration method for a power distribution cabinet comprises the following steps:

[0015] Deploy sensor components in any selected power distribution room to obtain the real-time electrical parameter set corresponding to the power distribution room and each power distribution cabinet in the power distribution room, build a comprehensive calculation model based on the electrical parameter set to generate the current power load value of the power distribution room, and use a chart drawing tool to draw a power load curve;

[0016] Compare the real-time power load value with the preset standard load;

[0017] Output the comparison result to determine whether the power distribution room is in a high load state;

[0018] The comparison result of the receiving judgment is triggered, and when it is judged to be a high load state, the adjustment mechanism is triggered to adjust the detection components configured in the power distribution room; when it is judged to be a low load state, no response action is taken;

[0019] Preprocess the thermal data collected by the detection component, obtain the historical data of each distribution cabinet in the corresponding distribution room, use the historical data to train the deep learning model, input the preprocessed thermal data into the deep learning model, output the identified potential area, and compare the temperature value on the potential area with the historical temperature during the same period;

[0020] If the temperature difference between the two exceeds the temperature difference threshold, the potential area is determined to be an abnormal area;

[0021] If the temperature difference between the two does not exceed the temperature threshold, the potential area is judged to be normal;

[0022] If the potential area is determined to be an abnormal area, the detection component is used to select the image of the abnormal area and trigger the tracking and registration program. Through image recognition technology, the abnormal position is locked and the abnormal event is registered in the system's built-in database.

[0023] While locking the distribution cabinets in abnormal areas, the fault warning mechanism is triggered simultaneously;

[0024] The corresponding time period T when the distribution room is in a high load state is obtained regularly, and the number of distribution cabinets with warning notification signals in the time period T is counted, and adjustment instructions are issued to the repaired distribution cabinets to execute the change strategy.

[0025] Furthermore, the electrical parameter set includes but is not limited to current, voltage, temperature and humidity; before building a comprehensive calculation model based on the electrical parameter set, data cleaning is performed on each data in the electrical parameter set.

[0026] Furthermore, the process of building a comprehensive calculation model to calculate the current power load value of the distribution room is as follows:

[0027]

[0028]

[0029]

[0030] In formula (1), Indicates the power load value of the distribution room. Indicates the basic power of the corresponding distribution room. represents the correction factor, represents the simultaneous coefficient; in formula (2), Ir and Vr represent the output current and output voltage of the distribution room respectively, represents the power factor, that is, the ratio of active power to apparent power. The if given later is the constraint condition. su represents the output circuit. uc and pc represent the single-phase circuit and the three-phase circuit respectively. In formula (3), A1 and A2 are weight coefficients, and the value range of A1 and A2 is 0~1. ce and rh represent the temperature and humidity of the distribution room respectively. G represents the adjustment factor, 0<G<1.

[0031] Furthermore, the calculation process of the simultaneous coefficient is as follows:

[0032] Collect the rated power and normal load of each distribution cabinet; add the normal load of each distribution cabinet to calculate the total normal load; count the frequency of all distribution cabinets reaching the maximum load at the same time within the past preset time period M, and use it as the estimated coefficient; add the total normal load of all distribution cabinets and multiply it by the estimated coefficient to obtain the total simultaneous load of the distribution room; add the rated power of all distribution cabinets to obtain the total rated power of the distribution room; divide the total simultaneous load of the distribution room by the total rated power to estimate the simultaneous coefficient.

[0033] Furthermore, the comparison results for determining whether the power distribution room is in a high load state are as follows:

[0034] When the real-time power load value exceeds the standard load, it means that the corresponding distribution room is in a high-load state;

[0035] When the real-time power load value does not exceed the standard load, it means that the corresponding distribution room is in a low load state.

[0036] Furthermore, the contents that trigger the adjustment mechanism are as follows:

[0037] Collect the current power load value and standard load, combine with the reference speed of the detection component, build an analysis and calculation model, generate the speed adjustment estimate value corresponding to the detection component in the distribution room, and adjust the deceleration component according to the speed adjustment estimate value. The formula used to calculate the speed adjustment estimate value is as follows:

[0038]

[0039]

[0040] In the formula, Indicates the estimated speed adjustment value. represents the reference speed, L represents the adjustment coefficient, and the value range of L is 0~1, ΔP represents the difference between the current power load value and the standard load, and Pb represents the standard load.

[0041] Furthermore, the detection assembly includes a track, a drive motor, and a combined high-definition camera and infrared thermal imager;

[0042] Among them, the combined high-definition camera and infrared thermal imager are movably configured on a circular track, and the driving motor is used to drive the combined high-definition camera and infrared thermal imager to move on the circular track. The high-definition camera and infrared thermal imager are used to monitor each distribution cabinet and are used to capture real-time images and thermal maps of each distribution cabinet respectively.

[0043] Furthermore, the synchronous triggering of the fault warning mechanism is: issuing a warning notification signal and executing a power failure repair strategy;

[0044] Among them, the forms of early warning notification signals include but are not limited to sound and light alarms and SMS notifications;

[0045] The power outage recovery strategy implemented is:

[0046] Cut off the power supply to the locked distribution cabinet and notify the maintenance personnel to go to the corresponding location for maintenance operations according to the abnormal event.

[0047] Furthermore, the process of executing the change strategy is as follows:

[0048] According to the established maintenance frequency, combined with the time period T and the number of distribution cabinets with early warning notification signals, a risk assessment-based model is built to generate a maintenance frequency adjustment value that matches the corresponding distribution cabinet, and maintenance adjustments are made according to the maintenance frequency adjustment value;

[0049] When building a risk assessment-based model, the formula is as follows:

[0050]

[0051] In the formula, Indicates the maintenance frequency adjustment value. It represents the established maintenance frequency, C represents the adjustment coefficient, which is used to control the increase of the maintenance frequency, and 1>C>0. Mr represents the number of distribution cabinets with early warning notification signals within the time period T, and T represents the time period.

[0052] An auxiliary detection tracking and registration system for a power distribution cabinet, the system comprising:

[0053] The load detection module deploys sensor components in any selected power distribution room to obtain the real-time electrical parameter set corresponding to the power distribution room and each power distribution cabinet in the power distribution room, builds a comprehensive calculation model based on the electrical parameter set to generate the current power load value of the power distribution room, and uses a chart drawing tool to draw a power load curve;

[0054] A comparison and determination module compares the real-time power load value with the preset standard load;

[0055] Output the comparison result to determine whether the power distribution room is in a high load state;

[0056] The execution adjustment module receives the comparison result of the determination. When it is determined to be a high load state, the adjustment mechanism is triggered to adjust the detection components configured in the power distribution room; when it is determined to be a low load state, no response action is taken;

[0057] The tracking and registration module pre-processes the thermal data collected by the detection component, obtains the historical data of each distribution cabinet in the corresponding distribution room, uses the historical data to train the deep learning model, inputs the pre-processed thermal data into the deep learning model, outputs the identified potential area, and compares the temperature value of the potential area with the historical temperature of the same period;

[0058] If the temperature difference between the two exceeds the temperature difference threshold, the potential area is determined to be an abnormal area;

[0059] If the temperature difference between the two does not exceed the temperature threshold, the potential area is judged to be normal;

[0060] If the potential area is determined to be an abnormal area, the detection component is used to select the image of the abnormal area and trigger the tracking and registration program. Through image recognition technology, the abnormal position is locked and the abnormal event is registered in the system's built-in database.

[0061] The early warning trigger module locks the distribution cabinets in abnormal areas and simultaneously triggers the fault early warning mechanism;

[0062] The maintenance module is changed to regularly obtain the corresponding time period T when there is a high load state in the distribution room, and count the number of distribution cabinets with early warning notification signals in the time period T, issue adjustment instructions to the repaired distribution cabinets, and execute the change strategy.

[0063] (III) Beneficial effects

[0064] The present invention provides an auxiliary detection tracking registration method and system for a power distribution cabinet, which has the following beneficial effects:

[0065] Real-time monitoring of the electrical parameter sets of the power distribution room and distribution cabinets, building a comprehensive calculation model to generate power load values, and drawing power load curves, to achieve accurate evaluation and visual display of the power load in the power distribution room; at the same time, by optimizing the estimation method of the simultaneity coefficient and improving the accuracy of electrical parameter data processing, the accuracy and reliability of power load evaluation are further improved, which helps to ensure the stability of power supply and the safe operation of equipment, and provide stronger support for the management and maintenance of the power distribution room;

[0066] Compare the power load value with the preset standard load in real time, accurately determine the load status of the distribution room, and dynamically adjust the speed of the detection component under high load conditions, thereby improving the response sensitivity and monitoring efficiency of the detection component; at the same time, by optimizing the configuration and operation mode of the detection component, the cost and risk of manual inspections are reduced, and the automation and intelligence level of distribution room management is improved;

[0067] By using deep learning models and image processing technology, the accuracy of identifying abnormal areas in the distribution cabinet is improved, and the response speed of abnormal processing is accelerated by triggering the tracking registration program and the fault warning mechanism. At the same time, the warning notification methods are enriched, the intelligent level of distribution room management is improved, and the management efficiency and accuracy are improved, providing a stronger guarantee for the safe operation of the distribution room.

[0068] The correlation analysis between the high load status of the distribution room and the early warning notification signal, as well as the scientific and reasonable adjustment of the maintenance frequency, solves the problems existing in the traditional distribution room management. This helps to more accurately assess the overall risk status of the distribution room and ensure that maintenance resources are reasonably allocated. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a flowchart of the overall steps of an auxiliary detection tracking and registration method for a power distribution cabinet in the present invention;

[0070] Figure 2 It is a cross-sectional view of the internal structure of the corresponding power distribution room in the present invention;

[0071] Figure 3 A curve diagram showing the change of power load value over time in the present invention;

[0072] Figure 4 It is a modular schematic diagram of an auxiliary detection, tracking and registration system for a power distribution cabinet in the present invention. DETAILED DESCRIPTION

[0073] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Embodiment 1:

[0074] See also Figures 1 to 3 ,This embodiment provides an auxiliary detection tracking registration method for a power distribution cabinet, which is aimed at a power distribution room in an industrial park, where several power distribution cabinets are laid;

[0075] For the auxiliary detection tracking registration method for distribution cabinets, the specific steps are as follows:

[0076] S1. Deploy sensor components in any selected power distribution room to obtain the real-time electrical parameter set corresponding to the power distribution room and each power distribution cabinet in the power distribution room, build a comprehensive calculation model based on the electrical parameter set to generate the current power load value of the power distribution room, and use a chart drawing tool to draw a power load curve;

[0077] like Figure 2 As shown, the selected distribution room layout can be intuitively seen, and corresponding numbers are set for different distribution cabinets to achieve the purpose of distinction and facilitate subsequent identification and locking; the sensor components include but are not limited to: current sensors, voltage sensors, and temperature and humidity sensors. The deployment location of the sensor components is determined according to actual needs. For example: for a single distribution cabinet, the deployed sensor components are used to monitor the electrical parameters of the corresponding distribution cabinet. For the entire distribution room, the deployed sensor components are used to monitor the electrical parameters of the corresponding entire distribution room;

[0078] The electrical parameter set includes, but is not limited to, current, voltage, temperature, and humidity; it may also include the rated power of the power distribution cabinet;

[0079] Among them, before building a comprehensive calculation model based on the electrical parameter set, each data in the electrical parameter set is cleaned;

[0080] The process of building a comprehensive calculation model to calculate the current power load value of the distribution room is as follows:

[0081]

[0082]

[0083]

[0084] In formula (1), Indicates the power load value of the distribution room. Indicates the basic power of the corresponding distribution room. represents the correction factor, represents the simultaneous coefficient; in formula (2), Ir and Vr represent the output current and output voltage of the distribution room respectively, represents the power factor, that is, the ratio of active power to apparent power. The if given later is a constraint condition. su represents the output circuit. uc and pc represent single-phase circuit and three-phase circuit respectively. Usually, these two types are used for the circuit design of the power distribution room. Other types of circuits will not appear in this scheme. Even if there are, they can be calculated according to the actual situation. In formula (3), A1 and A2 are weight coefficients, and the value range of A1 and A2 is 0~1. ce and rh represent the temperature and humidity of the power distribution room respectively. G represents the adjustment factor, 0<G<1, which is set according to actual needs and is usually 0.1.

[0085] Simultaneous coefficient : It is used to consider the different simultaneity of the loads of multiple distribution cabinets in the distribution room. It is usually estimated based on historical data and actual operation conditions. The coefficient value ranges from 0 to 1. The specific value reflects the actual simultaneous operation of the equipment loads.

[0086] The calculation process of the coefficient is as follows:

[0087] Collect the rated power and normal load of each distribution cabinet; add the normal load of each distribution cabinet to calculate the total normal load; count the frequency of all distribution cabinets reaching the maximum load at the same time within the preset time period M in the past, and use it as the estimation coefficient; add the total normal load of all distribution cabinets and multiply it by the estimation coefficient to obtain the total simultaneous load of the distribution room; add the rated power of all distribution cabinets to obtain the total rated power of the distribution room; divide the total simultaneous load of the distribution room by the total rated power to estimate the simultaneous coefficient; it should be noted that each distribution cabinet has a known rated power and normal load (i.e., the average load of the equipment during normal operation);

[0088] Chart drawing tools include Excel or Python's matplotlib library;

[0089] The power load curve uses time t as the horizontal axis, and the power load value of the distribution room The vertical axis is used to intuitively show the change of the power load of the corresponding distribution room over time.

[0090] By adopting the above technical solution, the electrical parameter sets of the distribution room and distribution cabinets are monitored in real time, a comprehensive calculation model is built to generate power load values, and a power load curve is drawn, thereby achieving accurate evaluation and visual display of the power load in the distribution room; at the same time, by optimizing the estimation method of the simultaneity coefficient and improving the accuracy of electrical parameter data processing, the accuracy and reliability of power load evaluation are further improved, which helps to ensure the stability of power supply and the safe operation of equipment, and provide stronger support for the management and maintenance of the distribution room.

[0091] S2, comparing the real-time power load value with the preset standard load;

[0092] Output the comparison result to determine whether the power distribution room is in a high load state;

[0093] Among them, in the power load management of the distribution room, the preset standard load is a key reference benchmark for evaluating the current power load status, that is, defining whether the current distribution room is in a high load or high load state, based on historical data analysis, the historical data here refers to the power load data of the distribution room in the past period of time, such as: one month or one quarter, for statistical analysis; identify the power load range under normal operating conditions, screen the maximum value of the peak value, and set the standard load of the corresponding distribution room according to 80% of the maximum value;

[0094] When the real-time power load value exceeds the standard load, it means that the corresponding distribution room is in a high-load state;

[0095] When the real-time power load value does not exceed the standard load, it means that the corresponding distribution room is in a low load state.

[0096] S3, receiving the comparison result of the determination, when it is determined to be a high load state, triggering the adjustment mechanism to adjust the detection components configured in the power distribution room; when it is determined to be a low load state, no response action is taken;

[0097] The contents that trigger the adjustment mechanism are as follows:

[0098] Collect the current power load value and standard load, combine with the reference speed of the detection component, build an analysis and calculation model, generate the speed adjustment estimate value corresponding to the detection component in the distribution room, and adjust the deceleration component according to the speed adjustment estimate value. The formula used to calculate the speed adjustment estimate value is as follows:

[0099]

[0100]

[0101] In the formula, Indicates the estimated speed adjustment value. represents the reference speed, L represents the adjustment coefficient, which is used to control the sensitivity of the speed adjustment, and the value range of L is 0~1, ΔP represents the difference between the current power load value and the standard load, and Pb represents the standard load;

[0102] Logic description:

[0103] For the basic speed, the speed is an initial setting value. The difference between the current power load value and the standard load indicates the degree of deviation between the current power load and the standard load. L, as an adjustment coefficient, is an empirical value or a preset parameter. The size of L should be adjusted according to the actual situation to ensure that the speed change is sensitive but not too drastic. When the difference increases, the speed adjustment estimate will also increase, which means that the detection component will speed up the movement speed in order to capture and analyze the status of the distribution room more frequently.

[0104] Example:

[0105] Assumption: Pb = 100 kW; =1 m / s; L=0.05 m / s, currently measured 120kW;

[0106] Then: ΔP=120 kW−100 kW=20 kW;

[0107] =1 m / s+0.05 m / s*20 kW=2 m / s;

[0108] Therefore, under high load conditions, the speed of the detection component will be adjusted to 2 m / s to respond more quickly to changes in the power distribution room. The design logic of this formula is intuitive, and the speed adjustment is directly controlled by the difference ΔP and the adjustment coefficient L. This design facilitates automatic control and can flexibly adjust the operating speed of the detection component according to the actual load conditions.

[0109] like Figure 2 As shown, the detection components include a track, a drive motor, and a combined high-definition camera and an infrared thermal imager;

[0110] In this solution, the track is a circular track, and the combined high-definition camera and infrared thermal imager are movably configured on the circular track. The drive motor is used to drive the combined high-definition camera and infrared thermal imager to make them rotate continuously and uniformly clockwise in the direction indicated by the arrow in the figure, and continuously monitor each distribution cabinet through the high-definition camera and infrared thermal imager. It is only necessary to ensure that the detection range of the high-definition camera and infrared thermal imager completely covers each distribution cabinet; the high-definition camera and infrared thermal imager are respectively used to capture real-time images and thermal maps of the appearance of each distribution cabinet.

[0111] By adopting the above technical solution, the power load value is compared with the preset standard load in real time, the load status of the distribution room is accurately determined, and the rotation speed of the detection component is dynamically adjusted under high load conditions, thereby improving the response sensitivity and monitoring efficiency of the detection component; at the same time, by optimizing the configuration and operation mode of the detection component, the cost and risk of manual inspections are reduced, and the level of automation and intelligence in the management of the distribution room is improved.

[0112] S4. Preprocess the thermal data collected by the infrared thermal imager, obtain the historical data of each distribution cabinet in the corresponding distribution room, use the historical data to train the deep learning model, input the preprocessed thermal data into the deep learning model, output the identified potential area, and compare the temperature value on the potential area with the historical temperature during the same period;

[0113] If the temperature difference between the two exceeds the temperature difference threshold, the potential area is determined to be an abnormal area;

[0114] If the temperature difference between the two does not exceed the temperature threshold, the potential area is judged to be normal;

[0115] If the potential area is determined to be an abnormal area, a high-definition camera is used to select the abnormal area and trigger the tracking and registration program. The abnormal position is locked through image recognition technology and the abnormal event is registered in the system's built-in database, including time, location and abnormal situation.

[0116] The preprocessing process of the thermal data collected by the infrared thermal imager is as follows: the thermal data corresponding to each distribution cabinet collected is cleaned and converted into a thermal map to visualize the temperature distribution; the thermal map uses different colors to represent different temperature ranges, which is convenient for intuitive analysis;

[0117] The purpose of preprocessing is to make the data match the data format and features used in training;

[0118] The deep learning model is trained using the historical temperature data of the corresponding power distribution cabinet under normal conditions. The deep learning model learns the temperature distribution of the corresponding power distribution cabinet under normal conditions. After the current thermal map is input into the trained model, real-time analysis is performed to compare the temperature value on the current thermal map with the historical temperature during the same period. If the temperature value of a certain area is different from the historical temperature during the same period, it is marked as a potential area and output;

[0119] Deep learning model training phase:

[0120] Collect historical temperature data of the corresponding power distribution cabinet under normal conditions and perform data cleaning;

[0121] Choose an appropriate deep learning model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN), depending on the characteristics of the data and the complexity of the problem; for thermal map analysis, CNN is more suitable because it can effectively process image data; use the preprocessed data to train the model so that it learns the temperature distribution characteristics under normal conditions; the model should be able to identify the difference between normal temperature distribution and abnormal temperature distribution;

[0122] Use cross-validation, accuracy, recall and other indicators to evaluate the performance of the model; adjust model parameters or try different model architectures based on the evaluation results to improve the accuracy and robustness of the model;

[0123] Setting of temperature difference threshold:

[0124] Set it according to the actual situation. When using the power distribution cabinet, considering the influence of factors such as environment and load, slightly higher or lower temperature is acceptable, but there is a safety value for temperature fluctuation, which corresponds to the temperature difference threshold. Usually, the temperature difference threshold is set to 10℃. If the temperature difference between the two exceeds 10℃, it is considered that there is an abnormality in the area, and the potential area is determined to be an abnormal area.

[0125] Image recognition technology uses image processing tools (such as OpenCV) to locate abnormal locations;

[0126] It should be noted that abnormal events include time, location, and abnormal circumstances;

[0127] Wherein, time represents the time point when the corresponding abnormal position appears; position represents the specific positioning coordinates of the corresponding abnormal position. For example, refer to Figure 2 It can be seen that each distribution cabinet is equipped with a number, which makes it easy to quickly find the corresponding distribution cabinet with an abnormal position; the abnormal situation indicates the image information and thermal map information of the abnormal position, which is convenient for subsequent analysis.

[0128] S5. While locking the power distribution cabinet in the abnormal area, the fault warning mechanism is triggered synchronously, a warning notification signal is issued, and a power failure repair strategy is executed;

[0129] The forms of early warning notification signals include but are not limited to: sound and light alarms, SMS notifications, and email reminders;

[0130] The power outage recovery strategy implemented is:

[0131] The locked distribution cabinet is powered off and maintenance personnel are notified to go to the corresponding location for maintenance operations according to the abnormal event, so as to facilitate timely maintenance of the abnormal distribution cabinet and avoid or reduce the losses caused by equipment abnormality.

[0132] By adopting the above technical solutions and using deep learning models and image processing technology, the accuracy of abnormal area identification in the distribution cabinet is improved, and the response speed of abnormal processing is accelerated by triggering the tracking registration program and fault warning mechanism; at the same time, the early warning notification methods are enriched, the intelligence level of distribution room management is improved, and the management efficiency and accuracy are improved, providing stronger guarantees for the safe operation of the distribution room.

[0133] S6. Regularly obtain the corresponding time period T when there is a high load state in the distribution room, and count the number of distribution cabinets with warning notification signals in the time period T, and issue adjustment instructions to the repaired distribution cabinets. After receiving the adjustment instructions, the system executes the change strategy;

[0134] Among them, the regular time frequency is based on demand, usually half a month or a month;

[0135] The process of executing a change strategy is as follows:

[0136] According to the established maintenance frequency, combined with the time period T and the number of distribution cabinets with early warning notification signals, a risk assessment-based model is built to generate a maintenance frequency adjustment value that matches the corresponding distribution cabinet, and perform maintenance adjustments based on the maintenance frequency adjustment value; the above model will consider the early warning signal as an indicator of increased risk, and change the maintenance frequency accordingly to deal with potential risks;

[0137] When building a risk assessment-based model, the formula is as follows:

[0138] ;

[0139] In the formula, Indicates the maintenance frequency adjustment value. It indicates the established maintenance frequency, the unit is (times / unit time), C indicates the adjustment coefficient, which is used to control the increase of maintenance frequency, and 1>C>0, Mr indicates the number of distribution cabinets with early warning notification signals in the time period T, and T indicates the time period (unit time);

[0140] Logic description:

[0141] The established maintenance frequency is the original data, which serves as the basis for adjustment. Mr / T represents the average number of distribution cabinets with warnings per unit time in the time period T. The higher this ratio is, the greater the risk is, so the maintenance frequency needs to be increased. C is used to control the increase in the maintenance frequency to ensure that the maintenance frequency can be appropriately increased when the warning signal increases. The specific increase is set at an interval of 0.1 according to actual needs. The 1+C*Mr / T part in the formula represents the increase in the maintenance frequency based on the number of warning signals and time factors. When M or Mr / T increases, this value will increase, resulting in an increase in the maintenance frequency adjustment value. For the time period T, please refer to Figure 3 It can be seen that;

[0142] For example:

[0143] =0.1 times / week (i.e., maintenance once every 10 weeks); T=1 week; Mr=5 distribution cabinets that have warning notification signals within the time period T; C=0.2 (adjustment coefficient);

[0144] but: =0.1*(1+0.2*15*1)=0.1*(1+1)=0.2;

[0145] Therefore, the adjusted maintenance frequency is 0.2 times / week, that is, once every 5 weeks. This frequency is higher than the established maintenance frequency (once every 10 weeks) to address the potential risks reflected by the early warning signals.

[0146] By adopting the above technical solution, the correlation analysis between the high load status of the distribution room and the early warning notification signal is realized, and the maintenance frequency is adjusted scientifically and reasonably, which solves the problems existing in the traditional distribution room management. This helps to more accurately assess the overall risk status of the distribution room and ensure that maintenance resources are reasonably allocated. Embodiment 2:

[0147] See also Figure 4 Based on Example 1, this embodiment further provides an auxiliary detection tracking and registration system for a power distribution cabinet, the system comprising:

[0148] The load detection module deploys sensor components in any selected power distribution room to obtain the real-time electrical parameter set corresponding to the power distribution room and each power distribution cabinet in the power distribution room, builds a comprehensive calculation model based on the electrical parameter set to generate the current power load value of the power distribution room, and uses a chart drawing tool to draw a power load curve;

[0149] A comparison and determination module compares the real-time power load value with the preset standard load;

[0150] Output the comparison result to determine whether the power distribution room is in a high load state;

[0151] The execution adjustment module receives the comparison result of the determination. When it is determined to be a high load state, the adjustment mechanism is triggered to adjust the detection components configured in the power distribution room; when it is determined to be a low load state, no response action is taken;

[0152] The tracking and registration module pre-processes the thermal data collected by the detection component, obtains the historical data of each distribution cabinet in the corresponding distribution room, uses the historical data to train the deep learning model, inputs the pre-processed thermal data into the deep learning model, outputs the identified potential area, and compares the temperature value of the potential area with the historical temperature of the same period;

[0153] If the temperature difference between the two exceeds the temperature difference threshold, the potential area is determined to be an abnormal area;

[0154] If the temperature difference between the two does not exceed the temperature threshold, the potential area is judged to be normal;

[0155] If the potential area is determined to be an abnormal area, the detection component is used to select the image of the abnormal area and trigger the tracking and registration program. Through image recognition technology, the abnormal position is locked and the abnormal event is registered in the system's built-in database.

[0156] The early warning trigger module locks the distribution cabinets in abnormal areas and simultaneously triggers the fault early warning mechanism;

[0157] The maintenance module is changed to regularly obtain the corresponding time period T when there is a high load state in the distribution room, and count the number of distribution cabinets with early warning notification signals in the time period T, issue adjustment instructions to the repaired distribution cabinets, and execute the change strategy.

[0158] In the application, the several formulas involved are all calculated by taking their numerical values ​​after removing the dimensions, and the formulas are established by collecting a large amount of data and performing software simulation to obtain a formula for the most recent actual situation. Some coefficients or weights in the formulas are set by technical personnel in this field according to actual conditions, so they will not be elaborated here.

[0159] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0160] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0161] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An auxiliary detection tracking and registration method for a power distribution cabinet, characterized in that: The steps include: Deploy sensor components in any selected power distribution room to obtain the real-time electrical parameter set corresponding to the power distribution room and each power distribution cabinet in the power distribution room, build a comprehensive calculation model based on the electrical parameter set to generate the current power load value of the power distribution room, and use a chart drawing tool to draw a power load curve; The electrical parameter set includes current, voltage, temperature and humidity. Before building a comprehensive calculation model based on the electrical parameter set, the data in the electrical parameter set is cleaned. The process of building a comprehensive calculation model to calculate the current power load value of the distribution room is as follows: ; ; ; In formula (1), Indicates the power load value of the distribution room. Indicates the basic power of the corresponding distribution room. represents the correction factor, represents the simultaneous coefficient; in formula (2), Ir and Vr represent the output current and output voltage of the distribution room respectively, represents the power factor, that is, the ratio of active power to apparent power. The if given later is the constraint condition. su represents the output circuit. uc and pc represent the single-phase circuit and the three-phase circuit respectively. In formula (3), A1 and A2 are weight coefficients, and the value range of A1 and A2 is 0~1. ce and rh represent the temperature and humidity of the distribution room respectively. G represents the adjustment factor, 0<G<1. Compare the real-time power load value with the preset standard load; Output the comparison result to determine whether the power distribution room is in a high load state; Receiving the comparison result of the determination, when it is determined that the load is high, triggering the adjustment mechanism to adjust the detection components configured in the power distribution room; When it is judged to be in low load state, no response action is taken; Preprocess the thermal data collected by the detection component, obtain the historical data of each distribution cabinet in the corresponding distribution room, use the historical data to train the deep learning model, input the preprocessed thermal data into the deep learning model, output the identified potential area, and compare the temperature value on the potential area with the historical temperature during the same period; If the temperature difference between the two exceeds the temperature difference threshold, the potential area is determined to be an abnormal area; If the temperature difference between the two does not exceed the temperature threshold, the potential area is judged to be normal; If the potential area is determined to be an abnormal area, the detection component is used to select the image of the abnormal area and trigger the tracking registration program. The abnormal position is locked through image recognition technology, and the abnormal position, the time when the abnormal position appears, and the image information and thermal map information of the abnormal position constitute an abnormal event, and the abnormal event is registered in the system's built-in database; While locking the distribution cabinet in the abnormal area, the fault warning mechanism is triggered simultaneously.

2. The auxiliary detection tracking and registration method for a power distribution cabinet according to claim 1 is characterized in that: The calculation process of the coefficient is as follows: Collect the rated power and normal load of each distribution cabinet; add the normal load of each distribution cabinet to calculate the total normal load; count the frequency of all distribution cabinets reaching the maximum load at the same time within the past preset time period M, and use it as the estimated coefficient; add the total normal load of all distribution cabinets and multiply it by the estimated coefficient to obtain the total simultaneous load of the distribution room; add the rated power of all distribution cabinets to obtain the total rated power of the distribution room; divide the total simultaneous load of the distribution room by the total rated power to estimate the simultaneous coefficient.

3. The auxiliary detection tracking and registration method for a power distribution cabinet according to claim 2 is characterized in that: The comparison results for determining whether the power distribution room is in a high load state are as follows: When the real-time power load value exceeds the standard load, it means that the corresponding distribution room is in a high-load state; When the real-time power load value does not exceed the standard load, it means that the corresponding distribution room is in a low load state.

4. The auxiliary detection tracking and registration method for a power distribution cabinet according to claim 3 is characterized in that: The contents that trigger the adjustment mechanism are as follows: Collect the current power load value and standard load, combine with the reference speed of the detection component, build an analysis and calculation model, generate the speed adjustment estimate value corresponding to the detection component in the distribution room, and adjust the deceleration component according to the speed adjustment estimate value. The formula used to calculate the speed adjustment estimate value is as follows: ; ; In the formula, Indicates the estimated speed adjustment value. represents the reference speed, L represents the adjustment coefficient, and the value range of L is 0~1, ΔP represents the difference between the current power load value and the standard load, and Pb represents the standard load.

5. The auxiliary detection tracking and registration method for a power distribution cabinet according to claim 4 is characterized in that: The detection components include tracks, drive motors, and a combined high-definition camera and infrared thermal imager; Among them, the combined high-definition camera and infrared thermal imager are movably configured on a circular track, and the drive motor is used to drive the combined high-definition camera and infrared thermal imager to move on the track. The high-definition camera and infrared thermal imager are used to monitor each distribution cabinet and are used to capture real-time images and thermal maps of each distribution cabinet respectively.

6. The auxiliary detection tracking and registration method for a power distribution cabinet according to claim 5, characterized in that: The synchronous triggering fault warning mechanism is: issuing a warning notification signal and executing a power failure repair strategy; Among them, the forms of early warning notification signals include sound and light alarms and SMS notifications; The power outage recovery strategy implemented is: Cut off the power supply to the locked distribution cabinet and notify the maintenance personnel to go to the corresponding location for maintenance operations according to the abnormal event.

7. The auxiliary detection tracking and registration method for a power distribution cabinet according to claim 6 is characterized in that: The steps also include: regularly obtaining the corresponding time period T when there is a high load state in the distribution room, and counting the number of distribution cabinets with warning notification signals in the time period T, issuing adjustment instructions to the repaired distribution cabinets, and executing the change strategy; The process of executing the change strategy is as follows: According to the established maintenance frequency, combined with the time period T and the number of distribution cabinets with early warning notification signals, a risk assessment-based model is built to generate a maintenance frequency adjustment value that matches the corresponding distribution cabinet, and maintenance adjustments are made according to the maintenance frequency adjustment value; When building a risk assessment-based model, the formula is as follows: ; In the formula, Indicates the maintenance frequency adjustment value. It represents the established maintenance frequency, C represents the adjustment coefficient, which is used to control the increase of the maintenance frequency, and 1>C>0. Mr represents the number of distribution cabinets with early warning notification signals within the time period T, and T represents the time period.

8. An auxiliary detection tracking and registration system for a power distribution cabinet, applied to the tracking and registration method according to any one of claims 1 to 7, characterized in that: The system includes: The load detection module deploys sensor components in any selected distribution room to obtain the real-time electrical parameter set corresponding to the distribution room and each distribution cabinet in the distribution room, builds a comprehensive calculation model based on the electrical parameter set to generate the current power load value of the distribution room, and uses a chart drawing tool to draw a power load curve; the electrical parameter set includes current, voltage, temperature and humidity; before building the comprehensive calculation model based on the electrical parameter set, the data in the electrical parameter set is cleaned; the process of building a comprehensive calculation model to calculate the current power load value of the distribution room is as follows: ; ; ; In formula (1), Indicates the power load value of the distribution room. Indicates the basic power of the corresponding distribution room. represents the correction factor, represents the simultaneous coefficient; in formula (2), Ir and Vr represent the output current and output voltage of the distribution room respectively, represents the power factor, that is, the ratio of active power to apparent power. The if given later is the constraint condition. su represents the output circuit. uc and pc represent the single-phase circuit and the three-phase circuit respectively. In formula (3), A1 and A2 are weight coefficients, and the value range of A1 and A2 is 0~1. ce and rh represent the temperature and humidity of the distribution room respectively. G represents the adjustment factor, 0<G<1. A comparison and determination module compares the real-time power load value with the preset standard load; Output the comparison result to determine whether the power distribution room is in a high load state; The execution adjustment module receives the comparison result of the determination. When it is determined to be a high load state, the adjustment mechanism is triggered to adjust the detection components configured in the power distribution room; when it is determined to be a low load state, no response action is taken; The tracking and registration module pre-processes the thermal data collected by the detection component, obtains the historical data of each distribution cabinet in the corresponding distribution room, uses the historical data to train the deep learning model, inputs the pre-processed thermal data into the deep learning model, outputs the identified potential area, and compares the temperature value of the potential area with the historical temperature of the same period; If the temperature difference between the two exceeds the temperature difference threshold, the potential area is determined to be an abnormal area; If the temperature difference between the two does not exceed the temperature threshold, the potential area is judged to be normal; If the potential area is determined to be an abnormal area, the detection component is used to select the image of the abnormal area and trigger the tracking registration program. The abnormal position is locked through image recognition technology, and the abnormal position, the time when the abnormal position appears, and the image information and thermal map information of the abnormal position constitute an abnormal event, and the abnormal event is registered in the system's built-in database; The early warning trigger module locks the distribution cabinets in abnormal areas and simultaneously triggers the fault early warning mechanism; The maintenance module is changed to regularly obtain the corresponding time period T when there is a high load state in the distribution room, and count the number of distribution cabinets with early warning notification signals in the time period T, issue adjustment instructions to the repaired distribution cabinets, and execute the change strategy.