A power cabinet monitoring system and method based on AI model

Through the power cabinet monitoring system based on AI model, the environmental and electrical data of the power cabinet are collected and analyzed in real time, the component aging value is calculated and early warning signals are generated, which solves the problem of difficult to monitor component aging in the existing technology, and reduces equipment failure and safety risk.

CN118677113BActive Publication Date: 2024-11-08JINAN THERMAL CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411163114.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-23
Publication Date
2024-11-08
Estimated Expiration
2044-08-23

AI Technical Summary

Technical Problem

Existing power cabinet monitoring technology is difficult to effectively monitor component aging, especially in outdoor environments that are not constant temperature and humidity, resulting in an increased risk of equipment shutdown or safety accidents.

Method used

Adopt the power cabinet monitoring system based on AI model, through the environmental data acquisition module, information storage module, real-life monitoring module and monitoring data analysis module, the environmental and electrical data of the power cabinet are collected and analyzed in real time, and the component aging value and environmental comparison value are calculated. Generate an early warning signal to prevent equipment failures due to component aging.

Benefits of technology

Timely monitoring of the aging of power cabinet components is achieved, equipment failure and safety risks caused by component aging are reduced, and the stable operation of power cabinets is ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118677113B_ABST
    Figure CN118677113B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of power supply cabinet monitoring technology, and in particular to a power supply cabinet monitoring system and method based on an AI model. The system analyzes electrical data of the power supply cabinet, calculates an electrical anomaly coefficient according to parameters such as equipment design life, number of starts and closes, and operation quality assessment, collects internal environmental data information and external data environment, obtains component aging values ​​according to environmental aging coefficients and electrical anomaly coefficients, generates early warning signals for target components based on environmental information of the internal environment and environmental information of the external environment, and according to component aging values, actual temperature comparison values, and actual humidity comparison values. The system combines the internal environment, external environment, power supply data, personnel behavior, etc. of the power supply cabinet to generate early warning signals in a timely manner, reduce safety risks during the operation of the power supply cabinet, and at the same time, prevents the power supply cabinet from accelerating component aging due to excessive differences in internal and external environments by judging and analyzing various types of power consumption behaviors of the power supply cabinet.
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 supply cabinet monitoring, and in particular to a power supply cabinet monitoring system and method based on an AI model. Background Art

[0002] A power cabinet is an electrical device used to distribute and control electrical energy. It is usually installed in a closed or semi-enclosed metal cabinet to provide power to various sub-circuits or loads.

[0003] Prior art CN116760197A discloses a monitoring method, device and distribution cabinet of an intelligent power distribution cabinet, which simulates according to an existing wiring scheme to obtain an existing power distribution model, simulates according to a preset wiring scheme to obtain a preset power distribution model, and performs data analysis on the preset power distribution model to determine whether the preset wiring scheme is reasonable;

[0004] At present, many power cabinets are usually placed outdoors. Since the temperature and humidity in the outdoor environment are usually determined by the weather, the environment in which the power cabinet is located is not constant temperature and humidity. However, the aging degree of each component in the power cabinet varies in different environments. If the aging degree of the components in the power cabinet is not monitored in time, the power cabinet equipment will be shut down due to component aging, and serious safety accidents may occur. Summary of the invention

[0005] The purpose of the present invention is to solve the problems in the background technology and to propose a power cabinet monitoring system and method based on an AI model.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A power cabinet monitoring system based on an AI model, including an environmental data acquisition module, an information storage module, a real-scene monitoring module, a monitoring data analysis module, and a monitoring early warning module;

[0008] The environmental data acquisition module is used to collect environmental information of the internal environment and external environment of the target power cabinet respectively, and transmit it to the monitoring data analysis module. The environmental information includes temperature and humidity;

[0009] An information storage module is used to store basic information of the target power cabinet and transmit the stored basic information to the monitoring data analysis module. The basic information includes components and installation and maintenance requirements.

[0010] The real-time monitoring module is used to collect the electrical data in the target power cabinet in real time, obtain the real-time data, and transmit it to the monitoring data analysis module;

[0011] A monitoring data analysis module is used to select a target component in a target power cabinet and calculate the component aging value and the environmental comparison value of the target component. The environmental comparison value includes an actual temperature comparison value and an actual humidity comparison value. The monitoring data analysis module includes an environmental processing unit and an aging processing unit. The monitoring data processing unit transmits the component aging value and the environmental comparison value to the monitoring early warning module.

[0012] An aging processing unit is used to select component operation data of a target component in real-time data, and first set a normal operation interval in the component operation data, and then compare the component operation data with the normal operation interval, and mark the electrical abnormality data according to the comparison result, identify the number of electrical abnormality data within the cycle time, and mark it as the number of cycle abnormalities, and then calculate the number of cycle abnormalities to obtain an electrical abnormality coefficient, and then calculate the temperature aging coefficient and the humidity aging coefficient based on the environmental information of the internal environment, respectively, combine the temperature aging coefficient with the humidity aging coefficient, and calculate the environmental aging coefficient, perform weighted operation on the electrical abnormality coefficient and the environmental aging coefficient, and mark the calculation result as the component aging value;

[0013] The environment processing unit is used to calculate the real-time collected temperatures of the external environment and the internal environment with the midpoint value of the temperature interval and the midpoint distance value according to the environment information of the internal environment of the target power cabinet and the environment information of the external environment to obtain an actual temperature comparison value. Similarly, the real-time collected humidity of the external environment and the internal environment is calculated with the midpoint value of the humidity interval and the midpoint distance value to obtain an actual humidity comparison value;

[0014] The monitoring and early warning module is used to judge the environmental aging coefficient, actual temperature comparison value and actual humidity comparison value of the target component respectively, and generate an early warning signal according to the judgment results.

[0015] As a further solution of the present invention, a method for obtaining an electrical anomaly coefficient includes:

[0016] Select any electrical component in the target power cabinet as the target component, use the initial use time of the target component as the reference time, and set the cycle time, which is the threshold value;

[0017] The electrical parameters of the target component within the cycle time are marked as component operation data, and the component operation data is compared with the normal operation range. If the component operation data is not within the normal operation range, the component operation data is marked as electrical abnormal data, otherwise, it is marked as normal data;

[0018] The electrical abnormality data of the target component within the cycle time is accumulated, and the accumulated number is marked as the cycle abnormality number ZF. Based on the formula The electrical anomaly coefficient XD of the target component is obtained, and a1 and a2 are both constant values.

[0019] As a further solution of the present invention, a method for obtaining the normal operation range is as follows:

[0020] The component operation data collected within the cycle time is marked as YXi in chronological order, i=1, 2, ..., I, indicating that data is collected for the target component I times within the cycle time;

[0021] Take n data as a group, that is, i takes the value of [1, n]. In a group of data, the number of data overlaps of the component operation data is counted, and the counted data is arranged in descending order and marked as data repetition values ​​fc, c=1, 2, ..., c1, c1 represents non-repetitive data in the component operation data YXi. Specifically, non-repetitive data refers to removing multiple identical data and retaining only one data;

[0022] Based on the inequality , select the operation data. If the inequality Fx holds, output the corresponding component operation data and mark the output data as the operation data. If the inequality Fx does not hold, set c2=c2+1, and re-enter the obtained c2 value into the inequality Fx, and repeat the above method for operation. Here, c2 starts from 1, and c2∈c1, b1 is the probability threshold;

[0023] The maximum and minimum values ​​in the operating data are identified, and the minimum value is used as the left interval endpoint, the maximum value is used as the right interval endpoint, and the interval obtained by combining the left interval endpoint and the right interval endpoint is used as the regular operating interval, which is a closed interval.

[0024] As a further solution of the present invention, a method for obtaining a component aging value includes:

[0025] First, the temperature data in the internal environmental information of the target power cabinet is processed, and the collected temperature data is marked as temperature input data Wj, where j=1, 2, ..., J, and J represents the total number of data collections within the cycle time;

[0026] According to the optimum temperature operation range of the target element, the left end point and the right end point of the optimum temperature operation range are marked as WYl and WYr respectively;

[0027] Based on the formula Qz=(WYl+WYr)÷2 and the formula QL=(WYr-WYl)÷2, we can get the interval midpoint value Qz and the midpoint distance value QL respectively;

[0028] According to the formula The temperature aging coefficient WL of the target component is obtained, where a3 and a4 are constant coefficients;

[0029] Repeat the above steps to obtain the humidity aging coefficient SL, which is calculated by the formula , the environmental aging coefficient HX of the target component is obtained, a5 is a fixed value, b1 and b2 are proportional coefficients respectively;

[0030] The electrical anomaly coefficient XD and the environmental aging coefficient HX are weightedly added to obtain the component aging value YL, YL=XD×b3+HX×b4, where b3 and b4 are proportional coefficients respectively.

[0031] As a further solution of the present invention, a method for obtaining an environmental contrast value includes:

[0032] According to the external environment information and internal environment information of the target power cabinet, the formula and The actual temperature contrast value TE and the actual humidity contrast value Rh are obtained, Wo represents the temperature of the external environment, Wt represents the temperature of the internal environment, Ro represents the humidity of the external environment, Rt represents the humidity of the internal environment, Rz is the midpoint value of the optimal humidity operation interval, and RL is the midpoint distance value of the optimal humidity operation interval.

[0033] As a further solution of the present invention, Wo and Wt, Ro and Rt are data collected at the same time, and if any denominator or numerator is equal to 0 during the calculation of TE and Rh, 1 will be used to replace the 0 in the denominator or numerator.

[0034] As a further embodiment of the present invention, if The actual temperature comparison value TE is directly output as -1. Then continue the calculation using the above TE calculation formula. Similarly, if the actual wetness ratio When the actual humidity contrast value Rh is directly output as -1, if Then continue the calculation using the above Rh calculation formula.

[0035] As a further solution of the present invention, the warning signal includes a normal operation signal, an abnormal operation signal and a pre-warning signal. The specific method for generating the warning signal is as follows:

[0036] If the component aging value YL is less than the aging threshold Ly, and the actual temperature comparison value TE and the actual humidity comparison value Rh in the environmental comparison value are both greater than or equal to 0, a normal operation signal is generated;

[0037] If the component aging value YL is less than the aging threshold Ly, and any one of the actual temperature comparison value TE and the actual humidity comparison value Rh in the environmental comparison value is less than 0, a pre-warning signal is generated;

[0038] If the component aging value YL≥aging threshold LY, an abnormal operation signal is generated.

[0039] As a further solution of the present invention, the environment data module includes an internal environment monitoring unit and an external environment monitoring unit, the internal environment monitoring unit is used to detect and collect the environmental information inside the target power cabinet in real time, and the external environment monitoring unit is used to detect and collect the environmental information of the external environment of the target power cabinet in real time;

[0040] The information storage module includes a preliminary storage unit and a real-time storage unit. The preliminary storage unit is used to store basic information in the target power cabinet, and the real-time storage unit is used to store real-time data of the target power cabinet. The real-time data includes electrical data, environmental information and analysis data collected in real time. The analysis data refers to the analysis results after analysis by the monitoring data analysis module.

[0041] A power cabinet monitoring method based on an AI model, the monitoring method specifically comprising the following steps:

[0042] Step 1: Select a target component in the target power cabinet, obtain the component operation data of the target component within the cycle time, compare the component operation data with the normal operation range, and mark the electrical abnormality data according to the comparison result, then obtain the number of electrical abnormality data within the cycle time, obtain the cycle abnormality number, and then calculate the cycle abnormality number to obtain the electrical abnormality coefficient of the target component;

[0043] Step 2: Collect the environmental information of the internal environment and the external environment of the target power cabinet, first calculate the temperature aging coefficient and the humidity aging coefficient based on the environmental information of the internal environment, and calculate the environmental aging coefficient of the target component by the temperature aging coefficient and the humidity aging coefficient, perform weighted accumulation of the electrical abnormality coefficient and the environmental aging coefficient, and mark the accumulated result as the component aging value;

[0044] Step 3: Calculate the actual temperature comparison value and the actual humidity comparison value based on the environmental information of the internal environment and the external environment of the target power cabinet;

[0045] Step 4: respectively judging the component aging value, actual temperature comparison value and actual humidity comparison value of the target component, and generating a warning signal according to the judgment results;

[0046] Step 5: Based on the identity information bound to the power cabinet, the management personnel are identified, and according to the identification result, the management personnel are granted the power cabinet operation authority.

[0047] Compared with the prior art, the advantages of the present invention are:

[0048] The present invention analyzes the electrical data of the power supply cabinet, obtains the normal operating range of the electrical data in the continuously collected electrical data, compares the normal operating range with the component operation data, obtains the electrical abnormality data, calculates the electrical abnormality coefficient according to the number of electrical abnormality data, and then calculates the environmental aging coefficient according to the environmental information of the internal environment and the environmental information of the external environment, and then obtains the component aging value by the environmental aging coefficient and the electrical abnormality coefficient, and at the same time, calculates the actual temperature comparison value and the actual humidity comparison value based on the environmental information of the internal environment and the environmental information of the external environment, and then generates a warning signal for the target component according to the component aging value and the actual temperature comparison value and the actual humidity comparison value. The present invention combines the internal environment and the external environment of the power supply cabinet, generates a warning signal in time, prevents the power supply cabinet from accelerating the aging of components in the power supply cabinet due to the large difference between the internal and external environments, and reduces the safety risks in the operation process of the power supply cabinet. At the same time, the power supply cabinet is bound to the identity information of the construction personnel, and the operation safety of the power supply cabinet is further guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0050] Figure 2 It is a schematic diagram of the process structure of the present invention. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0052] Reference Figure 1 and Figure 2 , a power cabinet monitoring system based on an AI model, including an environmental data acquisition module, an information storage module, a real-scene monitoring module, a monitoring data analysis module, and a monitoring early warning module;

[0053] The environmental data acquisition module is used to collect environmental information of the power supply cabinet, wherein the environmental data acquisition module includes an internal environment monitoring unit and an external environment monitoring unit, the internal environment monitoring unit is used to perform real-time detection and collection of environmental information inside the target power supply cabinet, and the external environment monitoring unit is used to perform real-time detection and collection of environmental information of the external environment of the target power supply cabinet, wherein the environmental information includes temperature and humidity, and the target power supply cabinet refers to the power supply cabinet monitored by the present system;

[0054] As an embodiment of the present invention, in this embodiment, when collecting the external environment of the target power cabinet, the external environment range is set at 5 meters. This distance can directly cover the surrounding environment of the power cabinet, thereby performing comprehensive environmental monitoring and evaluation of the power cabinet;

[0055] Further, in this embodiment, the measurement range of the temperature acquisition device of the target power cabinet is -40°C to +80°C, the measurement accuracy is ±0.5°C, and the resolution is 0.1°C. The measurement range of the humidity acquisition device is 10%~90%RH, the measurement accuracy is ±2%RH, and the resolution is 0.1%RH.

[0056] It should be further explained that, in the present embodiment, the collected data is processed in the intelligent AI model, which further improves the monitoring efficiency of the target power cabinet;

[0057] Afterwards, the environmental data acquisition module transmits the real-time collected environmental information to the monitoring data analysis module;

[0058] The information storage module includes a pre-storage unit and a real-time storage unit, and the information storage module is bidirectionally connected to the monitoring data analysis module;

[0059] The early storage unit is used to store basic information in the target power cabinet, including components of the target power cabinet, its internal structure distribution diagram and installation and maintenance requirements. The installation and maintenance requirements specifically refer to the normal operating temperature of the equipment, the insulation performance of the components and the maintenance time of the equipment. The internal structure distribution diagram in this embodiment refers to the 3D model diagram of the target power cabinet, and the specific 3D model diagram includes the distribution positions of various components inside the target power cabinet;

[0060] The real-time storage unit is used to store the real-time data of the target power cabinet, wherein the real-time data includes the electrical data, environmental information and analysis data collected in real time, and the analysis data refers to the analysis results after being analyzed by the monitoring data analysis module;

[0061] The real-time monitoring module is used to collect electrical data in the target power cabinet in real time. The specific electrical data includes electrical parameters such as voltage, capacitance and ground resistance. The real-time monitoring module then transmits the real-time collected electrical data to the monitoring data analysis module.

[0062] The monitoring data analysis module is used to combine and analyze the received environmental information and the real-time collected electrical data with the basic information and real-time data stored in the information storage module, and obtain the component aging value and the environmental comparison value according to the analysis results. Specifically, the monitoring data analysis module includes an environmental processing unit and an aging processing unit;

[0063] The aging processing unit is used to calculate the component information inside the target power cabinet according to the received electrical information and in combination with the basic information and real-time data of the target power cabinet, wherein the component information includes the component aging value. Specifically, the method for calculating the component information includes:

[0064] First, select an electrical component in the target power cabinet as the target component, and use the initial use time of the target component as the reference time, and set the cycle time at the same time. The cycle time is the threshold. In this embodiment, the cycle time is set to 30 days.

[0065] Identify the electrical data transmitted by the real-time monitoring module within the cycle time, obtain the electrical parameters of the target component, and mark them as component operation data, compare the component operation data with the normal operation range, and if the component operation data is not within the normal operation range, mark the component operation data as electrical abnormality data; conversely, if the component operation data is within the normal operation range, mark the component operation data as normal data. The normal operation range is determined by technicians in this field according to the actual operation status of the target component;

[0066] The electrical abnormality data of the target component within the cycle time is accumulated to obtain the cycle abnormality number ZF, and then based on the formula The electrical anomaly coefficient XD of the target element is obtained, where a1 and a2 are both constant values. In this embodiment, a2 is 1, and the value of a1 is obtained by a person skilled in the art after evaluating the quality of the power cabinet, wherein the quality evaluation includes the construction quality evaluation value of the welding point of the power cabinet equipment based on the flaw detection photos of the welding operation in the power cabinet equipment;

[0067] In another embodiment of the present invention, a specific method for obtaining the normal operation range is:

[0068] The component operation data collected within the cycle time is marked as YXi in chronological order, i=1, 2, ..., I, indicating that data is collected for the target component I times within the cycle time;

[0069] Take n data as a group, that is, i takes the value of [1, n]. In a group of data, the number of data overlaps of the component operation data is counted, and the counted data is arranged in descending order and marked as data repetition value fc, c=1, 2, ..., c1, c1 represents non-repetitive data in the component operation data YXi. Specifically, non-repetitive data refers to removing multiple identical data and retaining only one data. For example, there is a group of data {3, 5, 7, 3, 3, 7, 4, 5}. After removing the identical data in this group of data, the non-repetitive data obtained is {3, 5, 7, 4};

[0070] Then based on the inequality , select the operation data. If the inequality Fx holds, the corresponding component operation data is output and the output data is marked as the operation data. If the inequality Fx does not hold, let c2=c2+1, and re-input the obtained c2 value into the inequality Fx, and repeat the above method for operation. Here, c2 starts from 1, and c2∈c1, b1 is the probability threshold, and b1 is 0.82 in this embodiment;

[0071] Identify the maximum and minimum values ​​in the operating data, and use the minimum value as the left interval endpoint, the maximum value as the right interval endpoint, and the interval obtained by combining the left interval endpoint and the right interval endpoint as the normal operating interval. It should be further explained that the normal operating interval is a closed interval;

[0072] This operation step makes the voltage range of the target component concentrated within an interval, thereby obtaining the interval of stable operation of the target component within a period of time. If the actual voltage value exceeds this interval, it means that the electrical data of the target component has undergone electrical fluctuations within this period of time, and this fluctuation has an instantaneous impact on the target component, thereby affecting the normal aging process of the target component.

[0073] Then, according to the environmental information inside the target power cabinet transmitted in real time by the internal environment monitoring unit, taking the temperature data in the environmental information inside the target power cabinet as an example, the collected temperature data is marked as temperature input data Wj, where j=1, 2, ..., J, and J represents the total number of data collections within the cycle time;

[0074] According to the optimum temperature operation range of the target component, the left end point and the right end point of the optimum temperature operation range are marked as WYl and WYr respectively. The optimum temperature operation range is specifically determined by the factory information of the target component.

[0075] Based on the formula Qz=(WYl+WYr)÷2 and the formula QL=(WYr-WYl)÷2, we can get the interval midpoint value Qz and the midpoint distance value QL respectively;

[0076] Then based on the formula The temperature aging coefficient WL of the target component is obtained, wherein a3 and a4 are constant coefficients, and their specific values ​​are obtained by technicians in this field after calculation based on big data;

[0077] Then repeat the above steps to obtain the humidity aging coefficient SL, and based on the formula , the environmental aging coefficient HX of the target element is obtained, a5 is a fixed value, b1 and b2 are proportional coefficients, in this embodiment, b1 and b2 are 0.942 and 0.915 respectively, and a4 is 2.53;

[0078] The electrical anomaly coefficient XD and the environmental aging coefficient HX are weighted and accumulated to obtain the component aging value YL, that is, YL=XD×b3+HX×b4, where b3 and b4 are proportional coefficients respectively, and the specific values ​​of b3 and b4 are obtained by technicians in this field after big data calculation;

[0079] The environment processing unit is used to calculate the environment information of the external environment and the internal environment of the target power cabinet by formula and The actual temperature comparison value TE and the actual humidity comparison value Rh are obtained, Wo represents the temperature of the external environment, Wt represents the temperature of the internal environment, Ro represents the humidity of the external environment, Rt represents the humidity of the internal environment, Rz is the midpoint value of the optimum humidity range, and RL is the midpoint distance value of the optimum humidity range. It should be further explained that Wo and Wt, Ro and Rt are data collected at the same time, and if any denominator or numerator is equal to 0 during the calculation of TE and Rh, 1 will be used to replace the 0 in the denominator or numerator. For example, if the real-time temperature difference TE is calculated, the numerator calculation result is 5 and the denominator calculation result is 0, and the denominator is replaced by 1 instead of 0, then the calculation result of the temperature difference TE is 5;

[0080] In another embodiment of the present invention, if When the actual temperature comparison value TE is directly output as -1, that is, when When TE=-1, if Then continue the calculation using the above TE calculation formula. Similarly, if the actual wetness ratio When the actual humidity contrast value Rh is directly output as -1, if Then continue the calculation using the above Rh calculation formula;

[0081] Afterwards, the monitoring data analysis module transmits the component aging value and the environmental comparison value to the monitoring and early warning module, wherein the environmental comparison value includes the actual temperature comparison value and the actual humidity comparison value;

[0082] The monitoring and early warning module generates an early warning signal for the target power cabinet according to the component aging value and the environmental comparison value in the target power cabinet, wherein the early warning signal includes a normal operation signal, an abnormal operation signal and a pre-warning signal. The specific generation method of the early warning signal includes:

[0083] If the component aging value YL is less than the aging threshold Ly, and the actual temperature comparison value TE and the actual humidity comparison value Rh in the environmental comparison value are both greater than or equal to 0, a normal operation signal is generated;

[0084] If the component aging value YL is less than the aging threshold Ly, and any one of the actual temperature comparison value TE and the actual humidity comparison value Rh in the environmental comparison value is less than 0, a pre-warning signal is generated;

[0085] If the component aging value YL ≥ aging threshold LY, an abnormal operation signal is generated;

[0086] The aging threshold LY is determined by those skilled in the art based on actual target component information;

[0087] Afterwards, the monitoring and warning module transmits the generated warning signal to the corresponding management personnel through the terminal device, and the management personnel perform maintenance on the target power cabinet;

[0088] In another embodiment of the present invention, it also includes an identification module, which is used to identify the management personnel. When the management personnel performs maintenance operations on the target power cabinet according to the early warning signal, the operation identification module identifies the management personnel based on the bound identity information. If the identified management personnel does not match the bound identity information, the identification module prohibits the management personnel from operating the target power cabinet. On the contrary, if the identified management personnel matches the bound identity information, the management personnel is allowed to operate the target power cabinet.

[0089] A power cabinet monitoring method based on an AI model, the method specifically comprising the following steps:

[0090] Step 1: Based on the basic information and real-time data of the target power cabinet, select the target component in the target power cabinet, extract the real-time data of the target component within the cycle time, obtain the component operation data, compare the component operation data with the normal operation range, mark the electrical abnormality data, obtain the number of electrical abnormality data within the cycle time, obtain the cycle abnormality number, and then calculate the cycle abnormality number to obtain the electrical abnormality coefficient of the target component;

[0091] Step 2: Collect the environmental information of the internal environment and the external environment of the target power cabinet, first calculate the temperature aging coefficient and the humidity aging coefficient based on the environmental information of the internal environment, and calculate the environmental aging coefficient of the target component by the temperature aging coefficient and the humidity aging coefficient, perform weighted accumulation of the electrical abnormality coefficient and the environmental aging coefficient, and mark the accumulated result as the component aging value;

[0092] Step 3: Calculate the actual temperature comparison value and the actual humidity comparison value based on the environmental information of the internal environment and the external environment of the target power cabinet;

[0093] Step 4: respectively judging the component aging value, actual temperature comparison value and actual humidity comparison value of the target component, and generating a warning signal according to the judgment results;

[0094] Step 5: Based on the identity information bound to the power cabinet, the management personnel are identified, and according to the identification result, the management personnel are granted the power cabinet operation authority.

[0095] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A power cabinet monitoring system based on AI model, characterized in that: It includes environmental data collection module, information storage module, real-scene monitoring module, monitoring data analysis module, and monitoring early warning module; The environmental data acquisition module is used to collect environmental information of the internal environment and external environment of the target power cabinet respectively, and transmit it to the monitoring data analysis module. The environmental information includes temperature and humidity; An information storage module is used to store basic information of the target power cabinet and transmit the stored basic information to the monitoring data analysis module. The basic information includes components and installation and maintenance requirements. The real-time monitoring module is used to collect the electrical data in the target power cabinet in real time, obtain the real-time data, and transmit it to the monitoring data analysis module; A monitoring data analysis module is used to select a target component in a target power cabinet and calculate the component aging value and the environmental comparison value of the target component. The environmental comparison value includes an actual temperature comparison value and an actual humidity comparison value. The monitoring data analysis module includes an environmental processing unit and an aging processing unit. The monitoring data processing unit transmits the component aging value and the environmental comparison value to the monitoring early warning module. An aging processing unit is used to select component operation data of a target component in real-time data, and first set a normal operation interval in the component operation data, and then compare the component operation data with the normal operation interval, and mark the electrical abnormality data according to the comparison result, identify the number of electrical abnormality data within the cycle time, and mark it as the number of cycle abnormalities, and then calculate the number of cycle abnormalities to obtain an electrical abnormality coefficient, and then calculate the temperature aging coefficient and the humidity aging coefficient based on the environmental information of the internal environment, respectively, combine the temperature aging coefficient with the humidity aging coefficient, and calculate the environmental aging coefficient, perform weighted operation on the electrical abnormality coefficient and the environmental aging coefficient, and mark the calculation result as the component aging value; Specifically, the calculation method of the component aging value includes: Select any electrical component in the target power cabinet as the target component, use the initial use time of the target component as the reference time, and set the cycle time, which is the threshold value; The electrical parameters of the target component within the cycle time are marked as component operation data, and the component operation data is compared with the normal operation range. If the component operation data is not within the normal operation range, the component operation data is marked as electrical abnormal data, otherwise, it is marked as normal data; The electrical abnormality data of the target component within the cycle time is accumulated, and the accumulated number is marked as the cycle abnormality number ZF. Based on the formula The electrical anomaly coefficient XD of the target element is obtained, and a1 and a2 are both constant values; The temperature data in the internal environmental information of the target power cabinet is processed, and the collected temperature data is marked as temperature input data Wj, where j=1, 2, ..., J, and J represents the total number of data collections within the cycle time; The temperature data in the internal environmental information of the target power cabinet is processed, and the collected temperature data is marked as temperature input data Wj, where j=1, 2, ..., J, and J represents the total number of data collections within the cycle time; According to the optimum temperature operation range of the target element, the left end point and the right end point of the optimum temperature operation range are marked as WYl and WYr respectively; Based on the formula and formula Get the interval midpoint value Qz and midpoint distance value QL respectively; According to the formula The temperature aging coefficient WL of the target component is obtained, where a3 and a4 are constant coefficients; Repeat the above steps to obtain the humidity aging coefficient SL, which is calculated by the formula , the environmental aging coefficient HX of the target component is obtained, a5 is a fixed value, b1 and b2 are proportional coefficients respectively; The electrical anomaly coefficient XD and the environmental aging coefficient HX are weighted and accumulated to obtain the component aging value YL. , b3 and b4 are proportional coefficients respectively; The environment processing unit is used to calculate the real-time collected temperatures of the external environment and the internal environment with the midpoint value of the temperature interval and the midpoint distance value according to the environment information of the internal environment of the target power cabinet and the environment information of the external environment to obtain an actual temperature comparison value. Similarly, the real-time collected humidity of the external environment and the internal environment is calculated with the midpoint value of the humidity interval and the midpoint distance value to obtain an actual humidity comparison value; The monitoring and early warning module is used to judge the component aging value, actual temperature comparison value and actual humidity comparison value of the target component respectively, and generate an early warning signal according to the judgment results.

2. According to the AI ​​model-based power cabinet monitoring system of claim 1, it is characterized in that: The method for obtaining the normal operating range is as follows: The component operation data collected within the cycle time is marked as YXi in chronological order, i=1, 2, ..., I, indicating that data is collected for the target component I times within the cycle time; Take n data as a group, that is, i takes the value of [1, n]. In a group of data, the number of data overlaps of the component operation data is counted, and the counted data is arranged in descending order and marked as data repetition values ​​fc, c=1, 2, ..., c1, c1 represents non-repetitive data in the component operation data YXi. Specifically, non-repetitive data refers to removing multiple identical data and retaining only one data; Based on the inequality , select the operation data. If the inequality Fx holds, output the corresponding component operation data and mark the output data as the operation data. If the inequality Fx does not hold, set c2=c2+1, and re-enter the obtained c2 value into the inequality Fx, and repeat the above method for operation. Here, c2 starts from 1, and c2∈c1, b1 is the probability threshold; The maximum and minimum values ​​in the operating data are identified, and the minimum value is used as the left interval endpoint, the maximum value is used as the right interval endpoint, and the interval obtained by combining the left interval endpoint and the right interval endpoint is used as the regular operating interval, which is a closed interval.

3. The power cabinet monitoring system based on the AI ​​model according to claim 1 is characterized in that: Methods for obtaining environmental comparison values ​​include: According to the external environment information and internal environment information of the target power cabinet, the formula and The actual temperature contrast value TE and the actual humidity contrast value Rh are obtained, Wo represents the temperature of the external environment, Wt represents the temperature of the internal environment, Ro represents the humidity of the external environment, Rt represents the humidity of the internal environment, Rz is the midpoint value of the optimal humidity operation interval, and RL is the midpoint distance value of the optimal humidity operation interval.

4. The power cabinet monitoring system based on the AI ​​model according to claim 3 is characterized in that: Wo and Wt, Ro and Rt are data collected at the same time, and if any denominator or numerator is equal to 0 during the calculation of TE and Rh, 1 will be used to replace the 0 in the denominator or numerator.

5. The power cabinet monitoring system based on the AI ​​model according to claim 3 is characterized in that: like When the actual temperature comparison value TE is directly output as -1, if Then continue the calculation using the above TE calculation formula. Similarly, if the actual wetness ratio is When the actual humidity contrast value Rh is directly output as -1, if Then continue the calculation using the above Rh calculation formula.

6. The power cabinet monitoring system based on the AI ​​model according to claim 1 is characterized in that: The early warning signals include normal operation signals, abnormal operation signals and pre-warning signals. The specific generation method of early warning signals is as follows: If the component aging value YL is less than the aging threshold Ly, and the actual temperature comparison value TE and the actual humidity comparison value Rh in the environmental comparison value are both greater than or equal to 0, a normal operation signal is generated; If the component aging value YL is less than the aging threshold Ly, and any one of the actual temperature comparison value TE and the actual humidity comparison value Rh in the environmental comparison value is less than 0, a pre-warning signal is generated; If the component aging value YL≥aging threshold LY, an abnormal operation signal is generated.

7. The power cabinet monitoring system based on the AI ​​model according to claim 1 is characterized in that: The environmental data module includes an internal environment monitoring unit and an external environment monitoring unit. The internal environment monitoring unit is used to detect and collect the environmental information inside the target power cabinet in real time, and the external environment monitoring unit is used to detect and collect the environmental information of the external environment of the target power cabinet in real time; The information storage module includes a preliminary storage unit and a real-time storage unit. The preliminary storage unit is used to store basic information in the target power cabinet, and the real-time storage unit is used to store real-time data of the target power cabinet. The real-time data includes electrical data, environmental information and analysis data collected in real time. The analysis data refers to the analysis results after being analyzed by the monitoring data analysis module; It also includes an identification module, which is used to identify the management personnel. When the management personnel performs maintenance operations on the target power cabinet according to the early warning signal, the operation identification module identifies the management personnel based on the bound identity information. If the identified management personnel does not match the bound identity information, the identification module prohibits the management personnel from operating the target power cabinet. On the contrary, if the identified management personnel matches the bound identity information, the management personnel is allowed to operate the target power cabinet.

8. A power cabinet monitoring method based on an AI model, the power cabinet monitoring method being applied to a power cabinet monitoring system based on an AI model as described in any one of claims 1 to 7, characterized in that: The monitoring method specifically comprises the following steps: Step 1: Select a target component in the target power cabinet, obtain the component operation data of the target component within the cycle time, compare the component operation data with the normal operation range, and mark the electrical abnormality data according to the comparison result, then obtain the number of electrical abnormality data within the cycle time, obtain the cycle abnormality number, and then calculate the cycle abnormality number to obtain the electrical abnormality coefficient of the target component; Step 2: Collect the environmental information of the internal environment and the external environment of the target power cabinet, first calculate the temperature aging coefficient and the humidity aging coefficient based on the environmental information of the internal environment, and calculate the environmental aging coefficient of the target component by the temperature aging coefficient and the humidity aging coefficient, perform weighted accumulation of the electrical abnormality coefficient and the environmental aging coefficient, and mark the accumulated result as the component aging value; Step 3: Calculate the actual temperature comparison value and the actual humidity comparison value based on the environmental information of the internal environment and the external environment of the target power cabinet; Step 4: respectively judging the component aging value, actual temperature comparison value and actual humidity comparison value of the target component, and generating a warning signal according to the judgment results; Step 5: Based on the identity information bound to the power cabinet, the management personnel are identified, and according to the identification result, the management personnel are granted the power cabinet operation authority.