Health degree monitoring method and system for power and environment monitoring communication storage battery

Abstract

The invention discloses a health degree monitoring method and system for a dynamic environment monitoring communication storage battery. The method comprises the following steps: acquiring a measuring point selected by a user; selecting a model according to a user; inquiring a built-in model; obtaining a model confirmation result of the user, and if the confirmation result is confirmation, obtaining an early warning rule parameter configured by the user; pre-warning rule parameters configured by the user are stored; performing model training based on the pre-warning rule parameters which are stored and configured; after model training is completed, the dynamic threshold value takes effect to be used for health degree index monitoring of the dynamic environment monitoring communication storage battery. The invention aims to calculate the alarm boundary of the upper and lower thresholds for each monitored object single index, discover the abnormal condition of the environment by monitoring whether the parameters exceed the dynamic alarm threshold boundary, and provide the stability and intelligent dynamic early warning function for a dynamic environment monitoring system.

Description

Technical Field

[0001] The present invention belongs to the field of power communication technology, and in particular to a health monitoring method and system for a dynamic environment monitoring communication battery. Background Art

[0002] Dynamic environment monitoring is also known as power environment monitoring. It mainly refers to the real-time monitoring and management of the power systems and environmental conditions of data centers, computer rooms, communication base stations and other places. It generally includes the following aspects: Power system monitoring: including power supply, UPS (uninterruptible power supply), generator, distribution system, etc., to ensure the stability and reliability of power supply. Environmental system monitoring: including monitoring of environmental parameters such as temperature and humidity, smoke, and water leakage to ensure that the equipment operates under suitable environmental conditions. Security system monitoring: including access control, video surveillance, fire protection systems, etc. to ensure the physical safety of the venue. Network system monitoring: including operating status monitoring of IT facilities such as servers, network equipment, and storage devices. Therefore, realizing the health monitoring of dynamic environment monitoring communication batteries plays an important role in ensuring the reliability of dynamic environment monitoring. The existing dynamic environment monitoring communication battery health monitoring method is a single variable dynamic threshold warning method. The warning threshold setting for each monitoring object in the communication room is only set to a fixed value. There are large differences in the environmental conditions of rooms of different sizes, different locations, and different time periods. It is relatively rough to use a unified standard fixed threshold for all rooms. The inaccuracy of the threshold further affects the accuracy of the generated warning, which does not meet the strict management requirements of the current room. In order to solve the above technical problems, the current mainstream method is to use the STL (local weighted regression smooth seasonal-trend) decomposition algorithm to eliminate trends and periodicity, and then use multiple unsupervised anomaly detection algorithms for anomaly detection. The results of multiple algorithms are bagged to obtain the results as the set threshold. This solution needs to be specially tuned for weekly, monthly or quarterly cycles in combination with business indicators. If the cycle can be well identified and eliminated, it may be possible to achieve better results, but the cycle cannot be consistent in different sites and different data. The automatic cycle identification solution uses a single variable with poor results. Summary of the invention

[0003] Technical problem to be solved by the present invention: In view of the above-mentioned problems in the prior art, a method and system for monitoring the health of a dynamic environment monitoring communication battery are provided. The present invention aims to calculate the alarm boundaries of the upper and lower thresholds for each single indicator of the monitored object, and to detect abnormal conditions in the environment by monitoring whether the parameters exceed the dynamic alarm threshold boundaries, thereby providing an intelligent dynamic early warning function for the stability of the dynamic environment monitoring system.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: A method for monitoring the health of a battery for dynamic environment monitoring communication comprises the following steps: S1, obtaining the measurement points selected by the user for computer room environment monitoring and the algorithm model keywords selected or input by the user; S2, querying the algorithm model library according to the algorithm model keyword to obtain a list of matching algorithm models in the algorithm model library, wherein the algorithm model library has multiple algorithm models built in, and the algorithm models are used to perform prediction on the current data within a specified window size in the historical data of the measuring point to obtain a historical prediction value of the battery health; S3, displays the list of matching algorithm models in the algorithm model library; S4, obtaining the algorithm model confirmed by the user. If the user confirms the selection successfully, jump to step S5, otherwise exit; S5, obtaining and saving the warning rule parameters configured by the user; S6, performing algorithm model training in combination with the saved configured warning rule parameters to determine the dynamic threshold; S7, performing a threshold judgment on the current of the measuring point in combination with the determined dynamic threshold to determine whether the dynamic ring monitoring communication battery is abnormal.

[0005] Optionally, the algorithm model built into the algorithm model library in step S2 is a time series algorithm model, which is used to perform prediction on the current time series data within a specified window size in the historical data of the measurement point to obtain a historical prediction value of the battery health.

[0006] Optionally, the built-in time series algorithm models in the algorithm model library include part or all of the autoregressive model, the autoregressive and extreme value distribution algorithm model, the exponential smoothing model, and the LSTM model.

[0007] Optionally, the user-configured warning rule parameters acquired in step S5 include an upper boundary threshold and a lower boundary threshold.

[0008] Optionally, performing model training based on the saved configured warning rule parameters in step S6 includes: S6.1, obtain historical data of the measurement point selected by the user; S6.2, sliding based on the specified window size, obtaining the historical measurement value of the battery health in each window in the historical data, and using the algorithm model confirmed and selected by the user to perform prediction on the time series data in each window to obtain the historical prediction value of the battery health; S6.3, calculate the difference between the historical measurement value and the historical prediction value of the battery health in each window as the abnormal score score. If the abnormal score score of a window is negative and falls outside the lower boundary threshold, the lower boundary of the window is determined to be abnormal. If the abnormal score score of a window is positive and falls outside the upper boundary threshold, the upper boundary of the window is determined to be abnormal. Mark the abnormal value found; S6.4, the peak over-threshold algorithm POT is used in combination with the abnormal marks of each window to determine the dynamic threshold.

[0009] Optionally, in step S6.4, the function expression for determining the dynamic threshold using the peak value over-threshold algorithm in combination with the abnormal marks of each window is: , , in, To exceed the internal threshold The tail distribution function of the excess value of is the residual obtained based on historical data, which refers to the abnormal score score. When the residual data exceeds the internal threshold The difference between the residual data and the selected threshold exceeds The probability of is the residual set, Indicates the internal threshold Approximation constant The tail shape characteristics, is the shape parameter used to control the decay speed of the tail distribution, is a scaling function used to control the extent of tail distribution expansion, is the dynamic threshold, is the internal threshold, is the estimated value of the scale parameter of the generalized Pareto distribution GPD, is the estimated value of the shape parameter of the generalized Pareto distribution GPD, To exceed the dynamic threshold The target probability value is Used to set the tightness of abnormality judgment; is the total number of observations, which represents the number of all samples in the data set; is the number of observations exceeding the threshold t.

[0010] Optionally, in step S7, when a threshold judgment is performed on the current of the measuring point in combination with the determined dynamic threshold to determine whether the dynamic ring monitoring communication battery is abnormal, if the current of the measuring point exceeds the determined dynamic threshold, the dynamic ring monitoring communication battery is judged to be abnormal, otherwise the dynamic ring monitoring communication battery is judged to be normal.

[0011] In addition, the present invention also provides a health monitoring system for a dynamic environment monitoring communication battery, comprising a microprocessor and a memory connected to each other, wherein the microprocessor is programmed or configured to execute the health monitoring method for a dynamic environment monitoring communication battery.

[0012] In addition, the present invention also provides a computer-readable storage medium, in which a computer program or instruction is stored. The computer program or instruction is programmed or configured to execute the health monitoring method for dynamic environment monitoring communication battery through a processor.

[0013] In addition, the present invention also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the health monitoring method for dynamic environment monitoring communication batteries through a processor.

[0014] Compared with the prior art, the present invention mainly has the following advantages: the present invention includes obtaining the measurement point selected by the user; according to the model selected by the user; querying the built-in model; obtaining the user's model confirmation result, if the confirmation result is confirmation, obtaining the warning rule parameters configured by the user; saving the warning rule parameters configured by the user; model training based on the saved warning rule parameters; after completing the model training, the dynamic threshold is effective for the health indicator monitoring of the dynamic environment monitoring communication battery. The present invention aims to calculate the alarm boundaries of the upper and lower thresholds for each single indicator of the monitored object, and discover the abnormal situation of the environment by monitoring whether the parameter exceeds the dynamic alarm threshold boundary, so as to provide stability and intelligent dynamic warning function for the dynamic environment monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the basic flow of the method of the embodiment of the present invention.

[0016] Figure 2 Schematic diagram of the basic process of model training in an embodiment of the present invention.

[0017] Figure 3 Schematic diagram of a fitting curve of the cumulative distribution function of the extreme value distribution in an embodiment of the present invention.

[0018] Figure 4 Schematic diagram of an example of the distribution of historical measured values, historical predicted values, and abnormal points of current in an embodiment of the present invention, wherein the black curve is the historical measured value, the blue curve is the historical predicted value, and the red dots are abnormal points.

[0019] Figure 5 : is a schematic diagram of an example of anomaly score score and dynamic threshold curve in an embodiment of the present invention, wherein the purple solid line is the anomaly score score and the dotted line is the dynamic threshold.

[0020] Figure 6 This is a schematic diagram of the first segment of data for the MK trend test in an embodiment of the present invention.

[0021] Figure 7 This is a schematic diagram of the second data segment of the MK trend test in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] like Figure 1 As shown, the health monitoring method for dynamic environment monitoring communication battery in this embodiment includes the following steps: S1, obtaining the measurement points selected by the user for computer room environment monitoring and the algorithm model keywords selected or input by the user; S2, querying the algorithm model library according to the algorithm model keyword to obtain a list of matching algorithm models in the algorithm model library, wherein the algorithm model library has multiple algorithm models built in, and the algorithm models are used to perform prediction on the current data within a specified window size in the historical data of the measuring point to obtain a historical prediction value of the battery health; S3, displays the list of matching algorithm models in the algorithm model library; S4, obtaining the algorithm model confirmed by the user. If the user confirms the selection successfully, jump to step S5, otherwise exit; S5, obtaining and saving the warning rule parameters configured by the user; S6, performing algorithm model training in combination with the saved configured warning rule parameters to determine the dynamic threshold; S7, performing a threshold judgment on the current of the measuring point in combination with the determined dynamic threshold to determine whether the dynamic ring monitoring communication battery is abnormal.

[0024] In step S1 of this embodiment, when the user selects the measuring points for monitoring the computer room environment, corresponding measuring points may be set according to different monitored objects.

[0025] The algorithm model built into the algorithm model library in step S2 of this embodiment is a time series algorithm model, which is used to perform prediction on the current time series data within a specified window size in the historical data of the measurement point to obtain the historical prediction value of the battery health.

[0026] The built-in time series algorithm models in the algorithm model library of this embodiment include some or all of the autoregressive model, the autoregressive and extreme value distribution algorithm model, the exponential smoothing model, and the LSTM model. For example, as an optional implementation, in this embodiment, for periodic monitoring parameters, after using the autoregressive model to predict the time series, the absolute value of the residual is used as the anomaly score (other methods of calculating the error such as square error can be used as a substitute for the anomaly score), and the SPOT algorithm based on the extreme value theorem is used for modeling to achieve dynamic threshold anomaly detection.

[0027] The user-configured warning rule parameters acquired in step S5 of this embodiment include an upper boundary threshold and a lower boundary threshold.

[0028] In this embodiment, an autoregressive model is used to learn past historical data (mainly focusing on cycles and trends). After eliminating cycles and trends through residuals, the SPOT algorithm is used to implement dynamic thresholds. This method eliminates the problem of continuous false alarms caused by sliding windows, and at the same time, learning through autoregressive models can better automatically eliminate cycles and trends. And because the algorithm is based on autoregressive prediction, the predicted value of the next time step can be predicted in advance. Figure 2 As shown, in step S6 of this embodiment, model training based on the saved configured warning rule parameters includes: S6.1, obtain historical data of the measurement point selected by the user; S6.2, sliding based on the specified window size, obtaining the historical measurement value of the battery health in each window in the historical data, and using the algorithm model confirmed and selected by the user to perform prediction on the time series data in each window to obtain the historical prediction value of the battery health; S6.3, calculate the difference between the historical measurement value and the historical prediction value of the battery health in each window as the abnormal score score. If the abnormal score score of a window is a negative value and falls outside the lower boundary threshold, the lower boundary of the window is determined to be abnormal. If the abnormal score score of a window is a positive value and falls outside the upper boundary threshold, the upper boundary of the window is determined to be abnormal. The abnormal value found is marked as abnormal. For example, as an optional implementation, (3) the current measurement point value minus the predicted value is used to obtain the current score. When the score is a negative value and is outside the 98% (configurable) distribution quantile of the extreme value distribution, it is a lower boundary abnormality. When the score is a positive value and is outside the 98% (configurable) distribution quantile of the extreme value distribution, it is an upper boundary abnormality. S6.4, the peak over-threshold algorithm POT is used in combination with the abnormal marks of each window to determine the dynamic threshold.

[0029] In step S6.4 of this embodiment, the function expression for determining the dynamic threshold using the peak value over-threshold algorithm in combination with the abnormal marks of each window is: , , in, To exceed the internal threshold The tail distribution function of the excess value of is the residual obtained based on historical data, which refers to the abnormal score score. When the residual data exceeds the internal threshold The difference between the residual data and the selected threshold exceeds The probability of is the residual set, Indicates the internal threshold Approximation constant The tail shape characteristics, is the shape parameter used to control the decay speed of the tail distribution, is a scaling function used to control the extent of tail distribution expansion, is the dynamic threshold, is the internal threshold, is the estimated value of the scale parameter of the generalized Pareto distribution GPD, is the estimated value of the shape parameter of the generalized Pareto distribution GPD, To exceed the dynamic threshold The target probability value is Used to set the tightness of abnormality judgment; is the total number of observations, which represents the number of all samples in the data set; is the number of observations exceeding the threshold t.

[0030] Extreme Value Distribution (EVD) is a statistical model used to describe the distribution of the maximum or minimum values ​​of a random variable. The extreme value distribution (EVD) of the difference between the historical measured values ​​and the historical predicted values ​​of the battery health in each window can be expressed as: , in, Indicates that the parameter is The cumulative distribution function of the extreme value distribution of is the residual obtained based on historical data. The residual refers to the abnormal score. is the parameter of the extreme value distribution, represents a real number, To define the domain condition for the function, make sure the expression inside the parentheses is positive. Figure 3 Schematic diagram of the fitting curve of the cumulative distribution function of the extreme value distribution in the embodiment of the present invention, wherein the parameter of the extreme value distribution is The values ​​are -0.8, 0 and 1 respectively, combined with Figure 3 , represents the tail shape feature, then the tail shape feature , parameters of the extreme value distribution The relationship is as follows: (1) The tail shape has a heavy tail phenomenon: , , the extreme value distribution at this time is: Frechet distribution. (2) The tail shape feature has an e exponential tail: , , the extreme value distribution at this time is: γ distribution. (3) The tail shape feature is bounded: , , the extreme value distribution at this time is: uniform distribution. By fitting the extreme value distribution (EVD) to the tail of the unknown distribution input, the probability of these low-probability events can be found. However, it is very difficult to directly estimate the parameters in the distribution, so the Peaks-Over-Threshold (POT) method is used, and the threshold follows the generalized Pareto distribution (GPT): , in, To exceed the internal threshold The tail distribution function of the excess value of is the residual obtained based on historical data, which refers to the abnormal score score. When the residual data exceeds the internal threshold The difference between the residual data and the selected threshold exceeds The probability of is the residual set, Indicates the internal threshold Approximation constant The tail shape characteristics, is the shape parameter used to control the decay speed of the tail distribution, is a scaling function used to control the extent of the tail distribution expansion. At this time, the quantile can be calculated using the following formula: , in, is the dynamic threshold (quantile), is the internal threshold, is the estimated value of the scale parameter of the generalized Pareto distribution GPD, is the estimated value of the shape parameter of the generalized Pareto distribution GPD, To exceed the dynamic threshold The target probability value (such as 95%), Used to set the tightness of abnormality judgment; is the total number of observations, which represents the number of all samples in the data set; is the number of observations exceeding the threshold t, i.e., the number of peaks. Finally, the result obtained in this embodiment is as follows Figure 4 and Figure 5 As shown, Figure 4 Schematic diagram of an example of the distribution of historical measured values, historical predicted values, and abnormal points of current in an embodiment of the present invention, wherein the black curve is the historical measured value, the blue curve is the historical predicted value, and the red dots are abnormal points. Figure 5 Schematic diagram of anomaly score score and dynamic threshold curve example in an embodiment of the present invention, where the purple solid line is the anomaly score score and the dotted line is the dynamic threshold. In step S7 of this embodiment, when the current of the measuring point is judged by the threshold value in combination with the determined dynamic threshold value to determine whether the dynamic ring monitoring communication battery is abnormal, if the current of the measuring point exceeds the determined dynamic threshold value, the dynamic ring monitoring communication battery is judged to be abnormal, otherwise the dynamic ring monitoring communication battery is judged to be normal.

[0031] In order to verify the health monitoring method of the communication battery for dynamic environment monitoring in this embodiment, the Mann-Kendall (MK) trend test is used in this embodiment. The Mann-Kendall (MK) trend test is a non-parametric statistical method for detecting trends in time series data. It is widely used in fields such as meteorology, hydrology, and environmental science to analyze trend changes in long-term data sets. The basic principle of the MK trend test is to determine whether there is a monotonic trend (i.e., a continuous upward or downward trend) in the time series data. The advantage of this method is that it does not require the data to conform to a specific distribution (such as a normal distribution) and is robust to missing data and irregular time intervals. The specific steps include: (1) Calculate the MK statistic within the data window, including the S statistic and the variance of the S statistic. (2) Calculate the Z statistic. (3) Set a threshold and calculate the MK statistic that exceeds the threshold. Among them, calculating the S statistic includes: For a given time series data Calculate for each pair of data points and The Difference , and its calculation function expression is: , in, is the sign function; then the S statistic is calculated according to the following formula: , in, is the S statistic, is the number of data points. When the amount of data is large (usually (n>10)), the S statistic approximately follows a normal distribution, and its variance for: , in, is the size of the group with the same value in the data set. The function expression for calculating the Z statistic is: , in, is the Z statistic, is the S statistic, is the variance of the S statistic. Figure 6 and Figure 7 They are schematic diagrams of two segments of data for MK trend test in an embodiment of the present invention, wherein the red ones are identified abnormal points.

[0032] In summary, the health monitoring method for dynamic environment monitoring communication batteries in this embodiment has the following advantages: (1) Automatically identify the warning threshold cycle and improve the efficiency of alarm threshold configuration. This embodiment uses time series analysis to automatically identify the alarm threshold cycle, uses an autoregressive model for time series prediction, and an adaptive learning algorithm to develop an adaptive algorithm that can learn and adjust the threshold in real time. The threshold is dynamically adjusted according to the latest data to adapt to the ever-changing environment, and the cycle of the warning threshold is automatically identified, without the need for personnel to periodically maintain the data. By using intelligent algorithm tools, multi-level alarm strategies, and combining business scenarios for configuration, the accuracy and efficiency of alarms can be effectively improved. (2) This embodiment can realize multi-dimensional recognition of abnormalities, improve monitoring coverage, and enhance business availability; the monitoring parameters are divided into two modes of periodicity and monotonicity according to data characteristics for dynamic threshold monitoring. For abnormal situations that occur, the periodicity, overall trend, and fluctuation size of the monitoring indicators are monitored in a timely manner, and abnormal behaviors are monitored from multiple dimensions to ensure the stable operation of the system.

[0033] In addition, this embodiment also provides a health monitoring system for a dynamic environment monitoring communication battery, comprising a microprocessor and a memory connected to each other, wherein the microprocessor is programmed or configured to execute the health monitoring method for a dynamic environment monitoring communication battery.

[0034] In addition, this embodiment also provides a computer-readable storage medium, in which a computer program or instruction is stored. The computer program or instruction is programmed or configured to execute the health monitoring method for dynamic environment monitoring communication battery through a processor.

[0035] In addition, this embodiment also provides a computer program product, including a computer program or instructions, which are programmed or configured to execute the health monitoring method for dynamic environment monitoring communication batteries through a processor.

[0036] Those skilled in the art should understand that the technical solutions provided by the embodiments of the present application may be in the form of methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0037] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for monitoring the health of a battery for dynamic environment monitoring communication, characterized in that: The steps include: S1, obtaining the measurement points selected by the user for computer room environment monitoring and the algorithm model keywords selected or input by the user; S2, querying the algorithm model library according to the algorithm model keyword to obtain a list of matching algorithm models in the algorithm model library, wherein the algorithm model library has multiple algorithm models built in, and the algorithm models are used to perform prediction on the current data within a specified window size in the historical data of the measuring point to obtain a historical prediction value of the battery health; S3, displays the list of matching algorithm models in the algorithm model library; S4, obtaining the algorithm model confirmed by the user. If the user confirms the selection successfully, jump to step S5, otherwise exit; S5, obtaining and saving the warning rule parameters configured by the user; S6, performing algorithm model training in combination with the saved configured warning rule parameters to determine the dynamic threshold; S7, performing a threshold judgment on the current of the measuring point in combination with the determined dynamic threshold to determine whether the dynamic ring monitoring communication battery is abnormal.

2. The method for monitoring the health of a battery for dynamic environment monitoring communication according to claim 1, characterized in that: The algorithm model built into the algorithm model library in step S2 is a time series algorithm model, which is used to perform prediction on the current time series data within a specified window size in the historical data of the measuring point to obtain the historical prediction value of the battery health.

3. The method for monitoring the health of a battery for dynamic environment monitoring communication according to claim 2, characterized in that: The built-in time series algorithm models in the algorithm model library include part or all of the autoregressive model, the autoregressive and extreme value distribution algorithm model, the exponential smoothing model, and the LSTM model.

4. The method for monitoring the health of a battery for dynamic environment monitoring communication according to claim 1, characterized in that: The user-configured warning rule parameters acquired in step S5 include an upper boundary threshold and a lower boundary threshold.

5. The method for monitoring the health of a battery for dynamic environment monitoring communication according to claim 4, characterized in that: The model training based on the saved configured warning rule parameters in step S6 includes: S6.1, obtain historical data of the measurement point selected by the user; S6.2, sliding based on the specified window size, obtaining the historical measurement value of the battery health in each window in the historical data, and using the algorithm model confirmed and selected by the user to perform prediction on the time series data in each window to obtain the historical prediction value of the battery health; S6.3, calculate the difference between the historical measurement value and the historical prediction value of the battery health in each window as the abnormal score score. If the abnormal score score of a window is negative and falls outside the lower boundary threshold, the lower boundary of the window is determined to be abnormal. If the abnormal score score of a window is positive and falls outside the upper boundary threshold, the upper boundary of the window is determined to be abnormal. Mark the abnormal value found; S6.4, the peak over-threshold algorithm POT is used in combination with the abnormal marks of each window to determine the dynamic threshold.

6. The method for monitoring the health of a battery for dynamic environment monitoring communication according to claim 5, characterized in that: In step S6.4, the function expression for determining the dynamic threshold using the peak over-threshold algorithm in combination with the abnormal marks of each window is: ， ， in, To exceed the internal threshold The tail distribution function of the excess value of is the residual obtained based on historical data, which refers to the abnormal score score. When the residual data exceeds the internal threshold The difference between the residual data and the selected threshold exceeds The probability of is the residual set, Indicates the internal threshold Approximation constant The tail shape characteristics, is the shape parameter used to control the decay speed of the tail distribution, is a scaling function used to control the extent of tail distribution expansion, is the dynamic threshold, is the internal threshold, is the estimated value of the scale parameter of the generalized Pareto distribution GPD, is the estimated value of the shape parameter of the generalized Pareto distribution GPD, To exceed the dynamic threshold The target probability value is Used to set the tightness of abnormality judgment; is the total number of observations, which represents the number of all samples in the data set; is the number of observations exceeding the threshold t.

7. The method for monitoring the health of a battery for dynamic environment monitoring communication according to claim 1, characterized in that: In step S7, when a threshold judgment is performed on the current of the measuring point in combination with the determined dynamic threshold to determine whether the dynamic ring monitoring communication battery is abnormal, if the current of the measuring point exceeds the determined dynamic threshold, the dynamic ring monitoring communication battery is judged to be abnormal, otherwise the dynamic ring monitoring communication battery is judged to be normal.

8. A health monitoring system for dynamic environment monitoring communication batteries, comprising a microprocessor and a memory connected to each other, characterized in that: The microprocessor is programmed or configured to execute the health monitoring method for dynamic environment monitoring communication battery as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program or instruction stored therein, characterized in that: The computer program or instruction is programmed or configured to execute the health monitoring method for dynamic environment monitoring communication battery described in any one of claims 1 to 7 through a processor.

10. A computer program product comprising a computer program or instructions, characterized in that The computer program or instruction is programmed or configured to execute the health monitoring method for dynamic environment monitoring communication battery described in any one of claims 1 to 7 through a processor.

Images