A 5G communication power supply health management method and device

Abstract

The present invention relates to the field of communication power supply health management technology, specifically a 5G communication power supply health management method and equipment, the method includes real-time acquisition of 5G communication power supply collection data; preprocessing the collection data to obtain preprocessed data; using a health management model to process the preprocessed data to obtain a control strategy for extending the life of the 5G communication power supply; adjusting the control mode of each device and computer room equipment of the 5G communication power supply according to the control strategy. Through the above method, the present invention solves the problem that the 5G communication power supply health management is not in place, resulting in the 5G communication power supply system being unable to operate safely and stably, and can manage the 5G communication power supply in a healthy manner and extend the service life of the 5G communication power supply.

Description

Technical Field

[0001] The present invention relates to the technical field of communication power supply health management, and in particular to a 5G communication power supply health management method and device. Background Art

[0002] Prognostics and Health Management (PHM) technology can evaluate and predict the reliability of products or systems in actual application environments through failure physics analysis. Internationally, PHM technology has been developed for more than 30 years. With the continuous improvement of PHM technology, its application areas have gradually expanded from complex military systems and equipment such as aerospace vehicles, aircraft, and nuclear reactors to many industrial fields. At present, it has been widely used in electronics, ships, automobiles, and engineering structure safety. A typical PHM system mainly includes: sensors, data processing, condition monitoring, fault diagnosis, life prediction, and information display.

[0003] Although the technology of 5G communication power supply is becoming more and more mature with the development of communication technology, 5G communication power supply lacks health management measures. For example, although air conditioning equipment is installed in the machine room for cooling, the operating temperature inside the 5G communication power supply is still too high for a long time, which will cause the performance of the 5G communication power supply to degrade or the line to be damaged. The health management of the 5G communication power supply is not in place, resulting in the inability of the 5G communication power supply system to operate safely and stably.

[0004] Therefore, the present invention provides a 5G communication power supply health management method and device to solve the above problems. Summary of the invention

[0005] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a 5G communication power supply health management method and equipment to solve the problem that the health management of the above-mentioned 5G communication power supply is not in place, resulting in the problem that the 5G communication power supply system cannot operate safely and stably.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] In a first aspect, a 5G communication power supply health management method includes:

[0008] Acquire the collected data of 5G communication power supply in real time, including voltage, current, operating temperature of each device, room temperature, power, and continuous operation time of multiple battery packs; pre-process the collected data to obtain pre-processed data.

[0009] The health management model is used to process the preprocessed data to obtain a control strategy for extending the life of the 5G communication power supply; the health management model includes a data extraction module, a fusion analysis module, and a strategy output module; the data extraction module extracts abnormal data and associated data related to the abnormal data in the preprocessed data based on preset management standards; the fusion analysis module uses the constructed multi-layer analysis model to perform multi-level fusion analysis on the abnormal data and associated data, evaluates the current state of the 5G communication power supply and obtains the analysis results; the strategy output module uses the strategy generation model to process the analysis results to output the control strategy for health management.

[0010] Adjust the control method of each device of 5G communication power supply and equipment in the computer room according to the control strategy.

[0011] Preferably, the data extraction module comprises extracting first-level associated data directly related to the abnormal data in the preprocessed data.

[0012] Preferably, the multi-layer analysis model includes three analysis layers. The first analysis layer is used to obtain data of a preset time period corresponding to the abnormal data, and analyze the data of the preset time period based on a preset management standard to determine the abnormality degree k. The calculation formula of the abnormality degree k is:

[0013]

[0014] Among them, within the preset time period, a max is the maximum value of abnormal data, a min is the minimum value of abnormal data, a 0max is the preset upper limit value corresponding to the abnormal data, a 0min is the preset lower limit value corresponding to the abnormal data, t max is the cumulative time duration that the abnormal data is greater than the preset upper limit value corresponding to the abnormal data, t min is the cumulative duration of abnormal data less than the preset lower limit corresponding to the abnormal data, a0 is the average of the preset upper limit and the preset lower limit, and t0 is the total duration of the preset time period. When the abnormality is less than or equal to the first set value, the operation of the third analysis layer is executed; when the abnormality is greater than the first set value, the operation of the second analysis layer is executed. The second analysis layer includes obtaining the change data extracted from the preprocessed data by the data extraction module. The change data includes power, battery pack switching times, and load quantity, and performing cause analysis on the abnormal data, abnormality, first-level associated data, and change data, determining the abnormal value r and the cause of the abnormality, and executing the operation of the third analysis layer. The calculation formula of the abnormal value r is:

[0015]

[0016] h i =f(a,p i ),

[0017] Among them, within the preset time period, f(a,p i ) is the mapping function of the correlation between abnormal data and the i-th first-level associated data, h i is the relevance, the total number of first-level associated data is n, pi max j, p im inj are the mean of the maximum and minimum values ​​of the i-th first-level correlation data, respectively, and p 0imax 、p 0imin are respectively the preset upper limit and the preset lower limit of the i-th first-level associated data, p 0i is the average of the preset upper limit and the preset lower limit of the i-th first-level associated data, the total number of change data is m, j represents the j-th change data among the m change data, is the actual data corresponding to the change data at time t1, The third analysis layer is used to perform fusion analysis on two or more of the abnormal data, abnormal degree, abnormal value, and abnormal cause, evaluate the current state of the 5G communication power supply and the abnormal level, and output the analysis results.

[0018] Preferably, the third analysis layer further comprises using Gaussian distribution to calculate the concentration of abnormal data and first-level correlation data, and taking the position of the extreme value of the concentration as the abnormal position.

[0019] Preferably, the third analysis layer also includes: if there are only abnormal data and abnormality degree, the abnormality level is level one, the 5G communication power supply is in a healthy state, and the analysis result is normal; if there are abnormal data, abnormality degree, abnormal value and abnormal cause, when the abnormality degree and abnormal value are in the first preset interval, the abnormality level is determined to be level two, the 5G communication power supply is in a sub-healthy state, and the abnormality level, sub-healthy state and abnormal cause are used as analysis results; when the abnormality degree and abnormal value are in the second preset interval, the abnormality level is determined to be level three, the 5G communication power supply is in an abnormal state, and the abnormality level, abnormal state and abnormal cause are used as analysis results.

[0020] Preferably, the control strategy includes: when the 5G communication power supply is in a charging state, when the temperature of the battery pack is greater than a first set temperature, switching from a high-power charging mode to a low-power charging mode and adjusting the internal cooling device to a low wind speed state, and when the temperature is lower than a second set temperature, switching from a low-power charging mode to a high-power charging mode; when the temperature of the battery pack is greater than a third set temperature or the continuous charging time is greater than a set value, switching the charged battery pack or putting the battery pack in a waiting state, starting the computer room cooling device, and adjusting the internal cooling device to a high wind speed state; wherein, the third set temperature is greater than the first set temperature, and the first set temperature is greater than the second set temperature.

[0021] When the 5G communication power supply is in a discharging state, the discharging state of the 5G communication power supply is adjusted according to the load data. When the 5G communication power supply starts to discharge, it is discharged according to the maximum voltage set for the load, and after the load is in a stable working state, it is discharged according to the minimum working voltage of the load. When the continuous power supply time of the battery pack exceeds the first set time, the battery packs are replaced in turn for power supply.

[0022] When the 5G communication power supply is in an abnormal state, the corresponding control strategy is executed according to the abnormal information in the analysis result, and the alarm information and manual processing strategy are sent to the monitoring personnel.

[0023] Preferably, a health management method for a 5G communication power supply also includes: regularly evaluating the health status of the 5G communication power supply based on historical data, specifically including: obtaining historical preprocessing data of the previous cycle; processing the historical preprocessing data based on preset management standards, and calculating the abnormal frequency of each historical preprocessing data; evaluating the degree of influence of each historical preprocessing data on the 5G communication power supply by experts, and using each degree of influence as the weight of the corresponding abnormal frequency; performing weighted averaging on each abnormal frequency to obtain an abnormal index; and obtaining a health index of the 5G communication power supply based on the abnormal index.

[0024] Preferably, convolutional neural networks, historical data and preset management standards are used to construct the three analysis layers of the multi-layer analysis model.

[0025] Preferably, the preprocessing includes denoising voltage data, power data and current data in the collected data.

[0026] In a second aspect, the present invention also provides an electronic device, comprising a memory, a processor, and a machine executable program stored in the memory and run on the processor, characterized in that when the processor executes the program, the health management method of the 5G communication power supply described in any one of the above items is implemented.

[0027] The beneficial effects of the present invention are:

[0028] 1. The present invention first pre-processes the collected data acquired in real time to obtain pre-processed data that is easy to analyze, and then uses a health management model to analyze and process the pre-processed data to obtain a control strategy for extending the service life of the 5G communication power supply. Among them, the health management model uses a data extraction module to extract abnormal data that does not meet the preset management standards in the pre-processed data, and then extracts the associated data related to the abnormal data in the pre-processed data according to the abnormal data, so as to accurately extract useful data; then, the health management model uses a multi-layer analysis model constructed in the fusion analysis module to perform multi-level fusion analysis on the abnormal data and the associated data to evaluate the state of the 5G communication power supply and obtain the analysis results. Finally, the health management model uses a strategy generation model to process the analysis results, outputs the control strategy of the health management 5G communication power supply, and adjusts the control mode of each device and the room equipment of the 5G communication power supply through the control strategy to complete this health management. Through the above method, the present invention solves the problem that the health management of the 5G communication power supply is not in place, which makes the 5G communication power supply system unable to operate safely and stably, and can manage the 5G communication power supply in a healthy way and extend the service life of the 5G communication power supply.

[0029] 2. In the multi-layer analysis model of the present invention, the first analysis layer is used to perform data analysis on the data of the preset time period corresponding to the acquired abnormal data and the preset management standard, and the abnormality is calculated, and then the operation of the second analysis layer is judged according to the relationship between the abnormality and the first set value. When the operation of the second analysis layer is executed, the change data in the preprocessed data is obtained, the abnormality, abnormal data, first-level associated data and change data are analyzed, the abnormal value is calculated and the abnormal cause is determined, and then the operation of the third analysis layer is executed. The third analysis layer performs a fusion analysis based on two or more of the received abnormal data, abnormality, abnormal value, and abnormal cause, evaluates the current state and abnormal level of the 5G communication power supply, and outputs the analysis results. The multi-layer analysis model used in the present invention can perform a fine analysis on the abnormal data, the associated data related to the abnormal data, and the preset management standard to obtain the analysis results, and then the strategy output module can output the control strategy of health management in a targeted manner according to the analysis results, so as to perform health management on the 5G communication power supply and extend the service life of the 5G communication power supply.

[0030] 3. The present invention processes historical preprocessing data based on preset management standards, calculates the abnormal frequency of each historical preprocessing data; evaluates the impact of each historical preprocessing data on the 5G communication power supply through experts, and uses each impact degree as the weight of the corresponding abnormal frequency; performs weighted average on each abnormal frequency to obtain an abnormal index; and obtains the health index of the 5G communication power supply according to the abnormal index. The present invention can analyze the historical preprocessing data stored in a set period in the above manner, and use the health index obtained by analysis to evaluate the health status of the 5G communication power supply. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A schematic flow chart of a 5G communication power health management method of the present invention;

[0032] Figure 2 A schematic flow chart of data processing for the multi-layer analysis model of the present invention;

[0033] Figure 3 It is a schematic structural diagram of the electronic device of the present invention. DETAILED DESCRIPTION

[0034] The following will refer to the attached Figure 1 To Attachment Figure 3 The embodiments of the present invention are described in detail. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0035] The 5G communication power supply system is a key infrastructure to ensure the stable operation of communication equipment. The main power supply is the input end of the communication power supply system, usually the mains or a backup generator. The main power supply is responsible for providing basic power for the entire system. Rectifier / Converter The AC power provided by the main power supply is first converted into DC power by a rectifier, and then the voltage is adjusted to the standard voltage required by the communication equipment by the converter. The battery pack serves as a backup power supply, and the battery pack provides continuous power supply when the main power supply is interrupted. The batteries are usually lead-acid batteries or lithium batteries, which are kept charged by chargers so that they can be put into use immediately when needed. The power distribution unit is responsible for distributing power to various communication equipment. The power distribution unit usually includes circuit breakers, fuses, and power distribution boards to ensure the stable and safe distribution of the 5G communication power supply system. The monitoring system is used to monitor the operating status of the power supply system in real time, including parameters such as voltage, current, and temperature. The monitoring system can warn of potential problems and ensure the efficient operation of the system. Cooling system Since the power supply equipment generates heat during operation, cooling systems such as fans or air conditioning equipment are used to dissipate heat. The grounding system ensures the safety of the 5G communication power supply system and prevents electrical failures and electromagnetic interference. These components work together to ensure that the communications power system can provide a stable and reliable power supply, safeguarding the continuity and stability of the communications network.

[0036] A 5G communication power supply health management method, as shown in the attached Figure 1 As shown, the following steps are included:

[0037] Step S11: Acquire the collected data of the 5G communication power supply in real time, including voltage, current, operating temperature of each device, room temperature, power, and continuous operating time of multiple battery packs.

[0038] Step S12: pre-processing the collected data to obtain pre-processed data.

[0039] Step S13: Use the health management model to process the preprocessed data to obtain a control strategy for extending the life of the 5G communication power supply.

[0040] Step S14: Adjust the control mode of each device of the 5G communication power supply and the equipment in the computer room according to the control strategy.

[0041] The health management model includes a data extraction module, a fusion analysis module, and a strategy output module. The data extraction module extracts abnormal data and associated data related to abnormal data from preprocessed data based on preset management standards; the fusion analysis module uses a constructed multi-layer analysis model to perform multi-level fusion analysis on abnormal data and associated data, evaluate the current state of the 5G communication power supply, and obtain analysis results. The strategy output module uses a strategy generation model to process the analysis results to output the control strategy for health management.

[0042] In this embodiment, the preset management standard stores preset standard data related to the preprocessing data, including the upper and lower limit standard values ​​of each voltage, each current, and each power, the average between the upper and lower limit standard values, the mapping relationship table of abnormal data to associated data, various calculation formulas, the corresponding relationship between analysis results and control strategies, etc., which are stored in different databases respectively. The collected data also includes real-time power, the number of battery pack switching times, the number of loads connected to the 5G communication power supply, etc.

[0043] The present invention first pre-processes the collected data acquired in real time to obtain pre-processed data that is easy to analyze, and then uses a health management model to analyze and process the pre-processed data to obtain a control strategy for extending the service life of the 5G communication power supply. Among them, the health management model uses a data extraction module to extract abnormal data that does not meet the preset management standards in the pre-processed data, and then extracts the associated data related to the abnormal data in the pre-processed data according to the abnormal data, so as to accurately extract useful data; then, the health management model uses a multi-layer analysis model constructed in the fusion analysis module to perform multi-level fusion analysis on the abnormal data and the associated data to evaluate the state of the 5G communication power supply and obtain the analysis results. Finally, the health management model uses a strategy generation model to process the analysis results, outputs the control strategy of the health management 5G communication power supply, and adjusts the control mode of each device and the room equipment of the 5G communication power supply through the control strategy to complete this health management. Through the above method, the present invention solves the problem that the health management of the 5G communication power supply is not in place, which makes the 5G communication power supply system unable to operate safely and stably, and can manage the 5G communication power supply in a healthy way and extend the service life of the 5G communication power supply.

[0044] In one embodiment of the present invention, the data extraction module includes extracting first-level associated data directly related to the abnormal data in the pre-processed data.

[0045] In this embodiment, experts analyze historical preprocessing data to determine the first-level associated data directly related to the abnormal data, and establish a mapping relationship table between the abnormal data and its corresponding first-level associated data, and store it in the database for use when searching. For example, when charging a 5G communication power supply, if the charging temperature is too high, the first-level associated data is the charging voltage, charging power, charging current, and charging rate. When the first-level associated data does not need to be analyzed, the data extraction module does not perform data extraction operations to reduce the data processing process.

[0046] In one embodiment of the present invention, referring to the attached Figure 2 The multi-layer analysis model includes three analysis layers. The first analysis layer is used to obtain data of a preset time period corresponding to the abnormal data, and analyze the data of the preset time period based on the preset management standard to determine the abnormality degree k. The calculation formula of the abnormality degree k is:

[0047]

[0048] Among them, within the preset time period, a max is the maximum value of abnormal data, a min is the minimum value of abnormal data, a 0max is the preset upper limit value corresponding to the abnormal data, a 0m i n is the preset lower limit value corresponding to the abnormal data, t max is the cumulative time duration that the abnormal data is greater than the preset upper limit value corresponding to the abnormal data, t min is the cumulative duration of abnormal data less than the preset lower limit corresponding to the abnormal data, a0 is the average of the preset upper limit and the preset lower limit, and t0 is the total duration of the preset time period; when the abnormality is less than or equal to the first set value, the operation of the third analysis layer is executed; when the abnormality is greater than the first set value, the operation of the second analysis layer is executed; the second analysis layer includes obtaining the change data extracted from the preprocessed data by the data extraction module, the change data includes power, battery pack switching times, and load quantity, performing cause analysis on the abnormal data, abnormality, first-level associated data, and change data, determining the abnormal value r and the abnormal cause, and then executing the operation of the third analysis layer, and the calculation formula of the abnormal value r is:

[0049]

[0050] h i =f(a,p i ),

[0051] Among them, within the preset time period, f(a,p i ) is the mapping function of the correlation between abnormal data and the i-th first-level associated data, h iis the relevance, the total number of first-level associated data is n, p imaxj 、p iminj are the mean of the maximum and minimum values ​​of the i-th first-level correlation data, respectively, p 0imax 、p 0imin are respectively the preset upper limit and the preset lower limit of the i-th first-level associated data, p 0i is the average of the preset upper limit and the preset lower limit of the i-th first-level associated data, the total number of change data is m, j represents the j-th change data among the m change data, is the actual data corresponding to the change data at time t1, The third analysis layer is used to perform fusion analysis on two or more of the abnormal data, abnormal degree, abnormal value, and abnormal cause, evaluate the current state of the 5G communication power supply and the abnormal level, and output the analysis results.

[0052] In this embodiment, the data of the preset time period corresponding to the abnormal data means that when the abnormal data is extracted, the data of the preset time period that has been obtained before the abnormal data is obtained to determine whether the abnormal data has just appeared or has appeared multiple times, and the data of the preset time period is subsequently processed and analyzed. The second analysis layer determines the cause of the abnormality by analyzing the abnormal data, abnormal values ​​and the first-level associated data. For example, when the 5G communication power supply is in a charging state and the internal temperature of the 5G communication power supply is too high, it is determined whether the charging voltage, charging power and charging current are abnormal. If there is an abnormality, the abnormal data is the abnormal cause of the excessive temperature.

[0053] The change data includes but is not limited to the real-time power, the number of battery pack switching times, and the number of loads connected to the 5G communication power supply. By calculating the power change frequency, load number change frequency, and battery pack switching frequency within a preset time period, it is determined whether the 5G communication power supply is abnormal.

[0054] The multi-layer analysis model adopted in the present invention can perform fine processing and analysis on abnormal data, associated data related to the abnormal data, and preset management standards to obtain analysis results. Then the strategy output module can output the health management control strategy in a targeted manner according to the analysis results, thereby performing health management on the 5G communication power supply and extending the service life of the 5G communication power supply.

[0055] In one embodiment of the present invention, the third analysis layer further includes using Gaussian distribution to calculate the concentration between the abnormal data and the first-level associated data, and taking the position of the extreme value of the concentration as the abnormal position.

[0056] In this embodiment, the Gaussian distribution method is used to count the distribution of abnormal data and first-level associated data, and the concentration is the distribution. Since the abnormal data may be directly or indirectly related, it is easy to include many normal positions when determining the abnormal position based on the abnormal data, and it is more accurate to use the position of the extreme value in the distribution as the abnormal position.

[0057] In one embodiment of the present invention, the third analysis layer also includes: if there are only abnormal data and abnormality, the abnormality level is level one, the 5G communication power supply is in a healthy state, and the analysis result is normal; if there are abnormal data, abnormality, abnormal value and abnormal cause, when the abnormality and abnormal value are in the first preset interval, the abnormality level is determined to be level two, the 5G communication power supply is in a sub-healthy state, and the abnormality level, sub-healthy state and abnormal cause are used as analysis results; when the abnormality and abnormal value are in the second preset interval, the abnormality level is determined to be level three, the 5G communication power supply is in an abnormal state, and the abnormality level, abnormal state and abnormal cause are used as analysis results. Among them, the first preset interval corresponding to the abnormality or abnormal value does not overlap with the second preset interval, and the maximum value of the first preset interval is less than the minimum value of the second preset interval.

[0058] In this embodiment, when the data extraction module does not extract abnormal data, the abnormal level is zero, the 5G communication power supply is in a healthy state, and no subsequent operations are performed, but a new round of acquisition and collection data is obtained. The first preset interval in this embodiment includes two preset standard intervals, which are the preset standard interval corresponding to the abnormality when the abnormality level is level 2, and the preset standard interval corresponding to the abnormal value. The second preset interval includes two preset standard intervals, which are the preset standard interval corresponding to the abnormality when the abnormality level is level 3, and the preset standard interval corresponding to the abnormal value. For example, the first preset interval of the abnormality is [20,45], the second preset interval is [45,70], the first preset interval of the abnormal value is [30,50], and the second preset interval is [50,90]. The present invention clarifies the state and abnormality level of the 5G communication power supply through the analysis method of the third analysis layer, and sets corresponding control strategies for different abnormality levels to improve the health management of the 5G communication power supply.

[0059] In one embodiment of the present invention, by default, when the 5G communication power supply is in a sub-healthy state and in a charging state or a discharging state, the control strategy includes: when the 5G communication power supply is in a charging state, when the temperature of the battery pack is greater than the first set temperature, the high-power charging mode is switched to the low-power charging mode and the internal cooling device is adjusted to a low wind speed state, and when the temperature is less than the second set temperature, the low-power charging mode is switched to the high-power charging mode. When the temperature of the battery pack is greater than the third set temperature or the continuous charging time is greater than the set value, the charged battery pack is switched or the battery pack is in a waiting state, the cooling device of the machine room is started, and the internal cooling device is adjusted to a high wind speed state; wherein the third set temperature is greater than the first set temperature, and the first set temperature is greater than the second set temperature. The cooling device of the machine room includes but is not limited to air conditioners, fans, vents, etc. The internal cooling device of the 5G communication power supply includes fans of different sizes and different positions.

[0060] When the 5G communication power supply is in a discharging state, the discharge state of the 5G communication power supply is adjusted according to the load data. When the 5G communication power supply starts to discharge, it is discharged according to the maximum voltage set for the load, and after the load is in a stable working state, it is discharged according to the minimum working voltage of the load. When the continuous power supply time of the battery pack exceeds the first set time, the battery packs are replaced in turn for power supply.

[0061] When the 5G communication power supply is in an abnormal state, the control strategy of the 5G communication power supply is to execute the corresponding control strategy according to the abnormal information in the analysis results, and send alarm information and manual processing strategies to the monitoring personnel.

[0062] Preferably, the third set temperature is 74° C., the first set temperature is 60° C., the second set temperature is 48° C., and the first set temperature, the second set temperature and the third set temperature can be adjusted according to actual conditions.

[0063] In this embodiment, the abnormal information includes the abnormal level, abnormal cause, abnormal result and abnormal part. When the 5G communication power supply is in an abnormal state, first send an alarm message containing abnormal information to the monitoring personnel, and execute the control strategy generated by the strategy generation model after the countdown ends. The control strategy is not limited to restarting after shutting down and cooling for a period of time, and analyzing the health status of the 5G communication power supply again. If the analysis result is still abnormal, let the strategy generation model generate a manual processing strategy according to the abnormal information, send the manual processing strategy to the monitoring personnel, and let the monitoring personnel determine whether to turn off the 5G communication power supply; if the 5G communication power supply is turned off, the 5G communication power supply can be prevented from being in an abnormal state for a long time, and the life of the 5G communication power supply can be avoided from being excessively wasted; if the 5G communication power supply is kept in a working state, power outages can be avoided. The present invention can use the above-mentioned control strategy to reduce the operating temperature of the 5G communication power supply, extend the service life of the 5G communication power supply, and the above-mentioned control strategy can save power and reduce energy consumption.

[0064] In one embodiment of the present invention, a health management method for a 5G communication power supply also includes: regularly evaluating the health status of the 5G communication power supply based on historical data, specifically including: obtaining historical preprocessing data of the previous cycle; processing the historical preprocessing data based on preset management standards, and calculating the abnormal frequency of each historical preprocessing data; evaluating the degree of influence of each historical preprocessing data on the 5G communication power supply by experts, and recording the degree of influence of each preprocessing data on the 5G communication power supply in the preset management standard for easy search and use. Use each degree of influence as the weight of the corresponding abnormal frequency; perform weighted averaging on each abnormal frequency to obtain an abnormal index; obtain the health index of the 5G communication power supply based on the abnormal index, and use the health index to evaluate the health status of the 5G communication power supply. Through the above method, the present invention can understand the overall health status of the 5G communication power supply. Preferably, the periodicity and cycle are both one month.

[0065] In one embodiment of the present invention, convolutional neural networks, historical data, and preset management standards are used to construct three analysis layers of a multi-layer analysis model.

[0066] In this embodiment, the present invention uses the convolutional layer of a convolutional neural network to build a framework of a multi-layer analysis model, uses historical data as input and the analysis results in preset management standards as output, and trains the framework of the multi-layer analysis model, so that the successfully trained multi-layer analysis model can automatically analyze the input data and output the analysis results.

[0067] In one embodiment of the present invention, preprocessing includes denoising voltage data, power data, and current data in the collected data, removing obviously abnormal outliers, etc., and processing the collected data into preprocessed data that is easy to analyze.

[0068] In one embodiment of the present invention, as shown in the attached Figure 3 As shown, the present invention also provides an electronic device, including a memory, a processor, and a machine executable program stored in the memory and run on the processor, and when the processor executes the machine executable program, the health management method of the 5G communication power supply in any of the above embodiments is implemented.

[0069] In one embodiment of the present invention, the 5G communication power backup circuit, the number of batteries used during switching is less than or equal to half of the total number of batteries, and when the current battery usage duration exceeds the set value, the battery is switched to allow the battery to rest, and the switch can be performed at any time and tested. The number of power supply batteries is adjusted according to the change trend of the load.

[0070] In one embodiment of the present invention, when the main power supply is switched to the backup power supply, first, the backup power supply is connected to the circuit to put the backup power supply in a standby state. Secondly, the power supply battery pack of the main power supply is switched to the battery pack of the backup power supply, wherein the battery pack of the backup power supply has the same number and specification as the battery pack of the main power supply.

[0071] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A 5G communication power supply health management method, characterized in that: include: Acquire the collected data of 5G communication power supply in real time, including voltage, current, operating temperature of each device, room temperature, power, and continuous operation time of multiple battery packs; Preprocessing the collected data to obtain preprocessed data; The health management model is used to process the pre-processed data to obtain a control strategy for extending the life of the 5G communication power supply; the health management model includes a data extraction module, a fusion analysis module, and a strategy output module; the data extraction module extracts abnormal data and associated data related to the abnormal data in the pre-processed data based on the preset management standard; the fusion analysis module uses the constructed multi-layer analysis model to perform multi-level fusion analysis on the abnormal data and the associated data, evaluates the current state of the 5G communication power supply and obtains the analysis results; the strategy output module uses the strategy generation model to process the analysis results to output the control strategy of health management; Wherein, the data extraction module includes extracting first-level associated data directly related to the abnormal data in the preprocessed data; The multi-layer analysis model includes three analysis layers. The first analysis layer is used to obtain data of a preset time period corresponding to the abnormal data, and analyze the data of the preset time period based on a preset management standard to determine the abnormality degree k. The calculation formula of the abnormality degree k is: Among them, within the preset time period, a max is the maximum value of abnormal data, a min is the minimum value of abnormal data, a 0max is the preset upper limit value corresponding to the abnormal data, a 0min is the preset lower limit value corresponding to the abnormal data, t max is the cumulative time duration that the abnormal data is greater than the preset upper limit value corresponding to the abnormal data, t min is the cumulative duration of abnormal data less than the preset lower limit corresponding to the abnormal data, a0 is the average of the preset upper limit and the preset lower limit, and t0 is the total duration of the preset time period; when the abnormality is less than or equal to the first set value, the operation of the third analysis layer is executed; when the abnormality is greater than the first set value, the operation of the second analysis layer is executed; the second analysis layer includes obtaining the change data in the preprocessing data, the change data includes the power, the number of battery pack switching, and the number of loads, performing cause analysis on the abnormal data, the abnormality, the first-level associated data, and the change data, determining the abnormal value r and the abnormal cause, and executing the operation of the third analysis layer, and the calculation formula of the abnormal value r is: h i =f(a,p i ), Among them, within the preset time period, f(a,p i ) is the mapping function of the correlation between abnormal data and the i-th first-level associated data, h i is the relevance, the total number of first-level associated data is n, p imaxj 、p iminj are the mean of the maximum and minimum values ​​of the i-th first-level correlation data, respectively, p 0imax 、p 0imin are respectively the preset upper limit value and the preset lower limit value of the i-th first-level associated data, p 0i is the average of the preset upper limit and the preset lower limit of the i-th first-level associated data, the total number of change data is m, j represents the j-th change data among the m change data, is the actual data corresponding to the change data at time t1, The actual data of the change data at time t2; the third analysis layer is used to perform fusion analysis on two or more of the abnormal data, abnormal degree, abnormal value, and abnormal cause, evaluate the current state and abnormal level of the 5G communication power supply, and output the analysis results; Adjust the control method of each device of 5G communication power supply and equipment in the computer room according to the control strategy.

2. The health management method according to claim 1, characterized in that: The third analysis layer also includes using Gaussian distribution to calculate the concentration of abnormal data and first-level associated data, and taking the position of the extreme value of the concentration as the abnormal position.

3. The health management method according to claim 2, characterized in that: The third analysis layer also includes: if there are only abnormal data and abnormality degree, the abnormality level is level one, the 5G communication power supply is in a healthy state, and the analysis result is normal; if there are abnormal data, abnormality degree, abnormal value and abnormal cause, when the abnormality degree and abnormal value are in the first preset interval, the abnormality level is determined to be level two, the 5G communication power supply is in a sub-healthy state, and the abnormality level, sub-healthy state and abnormal cause are used as analysis results; when the abnormality degree and abnormal value are in the second preset interval, the abnormality level is determined to be level three, the 5G communication power supply is in an abnormal state, and the abnormality level, abnormal state and abnormal cause are used as analysis results.

4. The health management method according to claim 1, characterized in that: The control strategy includes: when the 5G communication power supply is in a charging state, when the temperature of the battery pack is greater than the first set temperature, switching from a high-power charging mode to a low-power charging mode and adjusting the internal cooling device to a low wind speed state, and when the temperature is less than a second set temperature, switching from a low-power charging mode to a high-power charging mode; when the temperature of the battery pack is greater than a third set temperature or the continuous charging time is greater than a set value, switching the charged battery pack or putting the battery pack in a waiting state, starting the cooling device of the machine room, and adjusting the internal cooling device to a high wind speed state; wherein the third set temperature is greater than the first set temperature, and the first set temperature is greater than the second set temperature; When the 5G communication power supply is in a discharging state, the discharging state of the 5G communication power supply is adjusted according to the load data. When the 5G communication power supply starts to discharge, it is discharged according to the maximum voltage set by the load, and after the load is in a stable working state, it is discharged according to the minimum working voltage of the load; when the continuous power supply time of the battery pack exceeds the first set time, the battery packs are replaced in turn for power supply; When the 5G communication power supply is in an abnormal state, the corresponding control strategy is executed according to the abnormal information in the analysis result, and the alarm information and manual processing strategy are sent to the monitoring personnel.

5. The health management method according to claim 1, characterized in that: Also includes: Regularly evaluate the health status of the 5G communication power supply based on historical data, specifically including: obtaining historical preprocessing data of the previous cycle; processing the historical preprocessing data based on preset management standards, and calculating the abnormal frequency of each historical preprocessing data; evaluating the impact of each historical preprocessing data on the 5G communication power supply through experts, and taking each impact degree as the weight of the corresponding abnormal frequency; taking a weighted average of each abnormal frequency to obtain an abnormal index; and obtaining the health index of the 5G communication power supply based on the abnormal index.

6. The health management method according to claim 1, characterized in that: Convolutional neural networks, historical data, and preset management standards are used to construct the three analysis layers of the multi-layer analysis model.

7. The health management method according to claim 1, characterized in that: The preprocessing includes denoising the voltage data, power data and current data in the collected data.

8. An electronic device comprising a memory, a processor, and a machine executable program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the health management method of the 5G communication power supply as described in any one of claims 1-7 is implemented.

Images