UPS battery health degree prediction method and system based on multi-source data
By constructing a correspondence group of multi-source data parameters, performing feature analysis and data increment, and optimizing the parameter library, the problem of insufficient accuracy in traditional battery health prediction is solved, and more accurate battery health prediction is achieved.
Patent Information
- Application Number
- CN202510806007.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-17
- Publication Date
- 2025-11-18
- Estimated Expiration
- 2045-06-17
AI Technical Summary
Traditional battery health prediction methods rely on a single indicator, which makes it difficult to accurately reflect battery degradation mechanisms under dynamic load and environmental change scenarios, resulting in insufficient prediction accuracy.
By constructing a battery health comparison experiment, establishing a corresponding relationship group of multi-source data parameters (such as charging power and discharge temperature curves), performing feature analysis and classification, generating a mapping parameter range, performing data increment and verification, optimizing the parameter library, eliminating distorted data, and finally using the updated parameter library to predict battery health.
It improves the accuracy and robustness of battery health prediction, providing reliable maintenance support for UPS systems.
Smart Images
Figure CN120490832B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of battery health prediction technology, and in particular to a method and system for predicting UPS battery health based on multi-source data. Background Technology
[0002] With the widespread deployment of uninterruptible power supply (UPS) systems in critical infrastructure such as communication networks, industrial automation, and emergency power supplies, the health status of batteries, as a core component of UPS systems, directly affects the reliability and lifespan of the system. Traditional battery health prediction methods often rely on single indicators (such as internal resistance changes) or simple statistical models, which are insufficient to comprehensively reflect the degradation mechanisms of batteries under complex operating conditions, resulting in insufficient prediction accuracy, especially under dynamic loads and environmental changes. In recent years, advancements in multi-source data processing and intelligent algorithms have opened up new avenues for battery health prediction. By integrating multi-dimensional parameters such as charging power curves, discharging power curves, charging temperature change curves, and discharging temperature change curves, it is possible to more accurately characterize the operating characteristics and degradation patterns of batteries. Summary of the Invention
[0003] The purpose of this invention is to provide a health prediction method and system that can accurately detect the health status of batteries in a UPS system.
[0004] This invention discloses a UPS battery health prediction method based on multi-source data, including:
[0005] Step S100: Construct a battery health comparison experiment to determine several health correspondence groups that are related to battery health and battery application state parameters;
[0006] Step S200: Based on battery health, classify the health correspondence groups to obtain several health correspondence sets. Perform feature analysis on the battery application state parameters in each health correspondence group in the health correspondence set. Further classify the health correspondence groups with equivalent feature performance to obtain several health correspondence subsets.
[0007] Step S300: Determine the mapping parameter range for each type of battery application state parameter in each health status correspondence subset, and combine the mapping parameter ranges of different types of battery application state parameters into a mapping parameter range group.
[0008] Step S400: Based on the several mapping parameter range groups corresponding to each health status correspondence set, perform data increment, construct several first battery application status parameter groups that conform to the mapping parameter range groups, and construct several new battery health status comparison experiments to obtain several verification health status correspondence groups.
[0009] Step S500: The first battery application state parameter group is combined to construct a battery application state parameter library, and the battery application state parameter library is verified and tested using several verification battery application state parameter groups. The comprehensive matching parameters between the retrieved matching battery application state parameter group and the verification battery application state parameter group are analyzed. Based on the comprehensive matching parameters, the mapping parameter range in the mapping parameter range group corresponding to the matching battery application state parameter group is narrowed, and the corresponding first battery application state parameter group is removed, thereby updating the battery application state parameter library.
[0010] Step S600: Analyze the real-time battery application status parameter group using the updated battery application status parameter library to determine the real-time battery health.
[0011] In the embodiments disclosed in this invention, the types of battery application state parameters include: charging power curve, discharging power curve, charging temperature change curve, and discharging temperature change curve.
[0012] In the embodiments disclosed in this invention, the method for performing feature analysis on battery application state parameters in each health correspondence group of the health correspondence set includes:
[0013] Step S201: Align the charging power curve and the charging temperature change curve according to the time correspondence principle, and align the discharging power curve and the discharging temperature change curve.
[0014] Step S202: Determine the trend of each curve, and cut off the curve segment whose upward trend lasts for more than a preset value and whose average upward amplitude is greater than or equal to the preset value, and record it as an upward curve segment. Cut off the curve segment whose downward trend lasts for more than a preset value and whose average downward amplitude is greater than or equal to the preset value, and record it as a downward curve segment. Cut off the remaining curve segments and record them as stable curve segments.
[0015] Step S203: Determine the time length and average ordinate value of the rising curve segment, falling curve segment, and stationary curve segment, and sort them according to their size to obtain the curve segment parameter group.
[0016] Step S204: Based on the consistency between the parameter groups of the curve segments, the health correspondence groups are classified to obtain several subsets of health correspondences.
[0017] In the embodiments disclosed in this invention, the method for determining the mapping parameter range of each type of battery application state parameter in each health status correspondence subset includes:
[0018] Step S301: Overlap and compare curves of the same type in the health correspondence subset to determine the curve coverage area between the curves, and identify the curve coverage area as the mapping parameter range.
[0019] In the embodiments disclosed in this invention, the method for incremental data processing based on several mapping parameter range groups corresponding to each health level correspondence set includes:
[0020] Step S401: Evenly set several reference time points on the horizontal axis of the time covered by the curve, determine the vertical mapping line of the curve covered by each reference time point, and determine the center point of the vertical mapping line, which is recorded as the vertical center point.
[0021] Step S402: Connect the longitudinal center points to obtain the reference curve, and analyze the boundary distance characteristics of the area covered by the curve relative to the reference curve. The boundary distance characteristics are the relative distances of the boundary nodes of different areas relative to the reference curve. Sort the relative distances from large to small, select a few relative distances with the highest order, and calculate the average value between them, which is recorded as the average relative distance.
[0022] Step S403: Based on the preset distance interval to which the average relative distance belongs, determine the allowable fluctuation parameter of each curve. Based on the allowable fluctuation parameter of each curve, perform several random reconstructions on the curves corresponding to the mapping parameter range group to obtain the reconstructed curves. Combine the reconstructed curves of different types to obtain the first battery application state parameter group.
[0023] In the embodiments disclosed in this invention, the method for randomly reconstructing the curves corresponding to the mapping parameter range group several times based on the allowable fluctuation parameter of each curve includes:
[0024] Step S4031: Determine the length of the longitudinal mapping line corresponding to different time nodes in the curve coverage area, and determine the reconstruction difference at different time nodes in combination with the allowable fluctuation parameter.
[0025] Step S4032: Based on the reconstruction difference, construct reconstruction curve nodes above and below the curve nodes of the corresponding time nodes, and connect the reconstruction curve nodes to obtain the reconstruction curve.
[0026] In the embodiments disclosed in this invention, the method for determining the amount of reconstruction difference includes:
[0027] Step S40311: Based on the preset fluctuation parameter range to which the allowed fluctuation parameter belongs, determine the corresponding length ratio coefficient, calculate the product of the length ratio coefficient and the length of the longitudinal mapping line to obtain the maximum reconstruction difference, and randomly generate several reconstruction difference values within the range of values less than or equal to the maximum reconstruction difference.
[0028] In the embodiments disclosed in this invention, the method for analyzing the comprehensive matching parameters of the retrieved matching battery application state parameter set and the verification battery application state parameter set includes:
[0029] Step S501: Compare the corresponding curves in the matching battery application status parameter group and the verification battery application status parameter group. If the coverage area between each corresponding curve is less than or equal to a preset value, it is determined that the parameters between the matching battery application status parameter group and the verification battery application status parameter group are matched, and the health difference between the two is calculated.
[0030] Step S502: Determine the comprehensive matching parameters based on the number of verification comparisons and the corresponding health difference for each comparison.
[0031] ;
[0032] Where W is the overall matching parameter, K is the matching sub-parameter adjustment coefficient, b is the matching sub-parameter adjustment constant, and n is the total number of verification comparisons. Let be the measure of health difference during the i-th verification comparison. This is a function for judging differences in health status. based on The corresponding health difference interval is given, and the corresponding matching parameter is output.
[0033] In the embodiments disclosed in this invention, the method for narrowing the range of mapping parameters in the mapping parameter range group corresponding to the matching battery application state parameter group based on comprehensive matching parameters includes:
[0034] Step S503: Based on the preset matching parameter interval to which the comprehensive matching parameter belongs, determine the narrowing ratio of the mapping parameter range, and based on the narrowing ratio of the mapping parameter range, push the upper and lower boundaries of the mapping parameter range inward to narrow it.
[0035] In the embodiments disclosed in this invention, a UPS battery health prediction system based on multi-source data is also disclosed, comprising:
[0036] The first module is used to construct a battery health comparison experiment and determine several health correspondence groups that are related to battery health and battery application state parameters.
[0037] The second module is used to classify health correspondence groups based on battery health, obtain several health correspondence sets, perform feature analysis on the battery application state parameters in each health correspondence group in the health correspondence set, and further classify health correspondence groups with equivalent feature performance to obtain several health correspondence subsets.
[0038] The third module is used to determine the mapping parameter range of each type of battery application state parameter in each health status correspondence subset, and to combine the mapping parameter ranges of different types of battery application state parameters into a mapping parameter range group.
[0039] The fourth module is used to perform data increment based on several mapping parameter range groups corresponding to each health status correspondence set, construct several first battery application status parameter groups that conform to the mapping parameter range groups, and construct several new battery health status comparison experiments to obtain several verification health status correspondence groups.
[0040] The fifth module is used to construct a battery application state parameter library by combining the first battery application state parameter group, and to verify and test the battery application state parameter library using several verification battery application state parameter groups. It analyzes the comprehensive matching parameters between the retrieved matching battery application state parameter group and the verification battery application state parameter group, and based on the comprehensive matching parameters, narrows the mapping parameter range in the mapping parameter range group corresponding to the matching battery application state parameter group, removes the corresponding first battery application state parameter group, and updates the battery application state parameter library.
[0041] The sixth module is used to analyze the real-time battery application status parameter group using the updated battery application status parameter library to determine the real-time battery health.
[0042] This invention proposes a method and system for predicting UPS battery health based on multi-source data, belonging to the field of battery health prediction technology. It establishes a correspondence between battery health and multi-source parameters (such as charging power and discharge temperature curves) through a battery health comparison experiment. Based on health classification and feature analysis, a subset of health correspondences is formed, and the mapping parameter ranges for each parameter are determined and combined into range groups. Data increment is performed using these range groups to generate a first set of battery application status parameters, and a parameter library is constructed through verification experiments. The parameter library is optimized based on comprehensive matching parameters, narrowing the mapping parameter ranges and eliminating distorted data. Finally, the updated parameter library is used to analyze real-time parameters and predict battery health. This invention improves prediction accuracy and robustness through multi-source data integration, feature extraction, and dynamic optimization, providing reliable support for UPS system maintenance.
[0043] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description
[0044] Figure 1 This is a flowchart illustrating the steps of the UPS battery health prediction method based on multi-source data disclosed in this embodiment of the invention. Detailed Implementation
[0045] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0046] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. It should be understood that the preferred embodiments described herein are only for illustration and explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Those skilled in the art can make some non-essential improvements and adjustments based on the following content of the present invention. In the present invention, unless otherwise expressly specified and limited, the technical terms used in the present invention should have the ordinary meaning understood by those skilled in the art. Example
[0047] The purpose of this invention is to provide a health prediction method and system that can accurately detect the health status of batteries in a UPS system.
[0048] This invention discloses a UPS battery health prediction method based on multi-source data. (See reference...) Figure 1 ,include:
[0049] Step S100: Construct a battery health comparison experiment to determine several health correspondence groups that are related to battery health and battery application state parameters.
[0050] The principle of step S100 is to establish the correlation between battery health and multi-source application state parameters through experimental methods, providing basic data for subsequent modeling. Specifically, this involves designing a series of battery test experiments with controlled variables, collecting battery operating data under different conditions, such as charging power curves, discharging power curves, charging temperature change curves, and discharging temperature change curves, while simultaneously measuring battery health indicators (such as remaining capacity or internal resistance). By analyzing this data, the correspondence between battery health and application state parameters under each set of experimental conditions is determined, forming a health correspondence set. For example, assuming that in one set of experiments, the battery operates under a specific charging power curve (power increases slowly over time), and its health is characterized by 90% remaining capacity, then this charging power curve and 90% health constitute a correspondence set. Through multiple sets of experiments, a sufficient number of correspondence sets are accumulated to provide data support for subsequent classification and feature analysis. The essence of this process is to quantify the complex battery operating states into an analyzable correlated dataset, laying the foundation for subsequent pattern recognition and prediction.
[0051] Step S200: Based on battery health, classify the health correspondence groups to obtain several health correspondence sets. Perform feature analysis on the battery application state parameters in each health correspondence group in the health correspondence set. Further classify health correspondence groups with equivalent feature performance to obtain several health correspondence subsets.
[0052] The principle of step S200 is to cluster and perform feature analysis on the health status correspondence groups to group the data and uncover hidden patterns, thereby improving the efficiency and accuracy of subsequent modeling. First, based on battery health status (such as remaining capacity range), the correspondence groups are categorized into several health status correspondence sets. For example, correspondence groups with health status between 80% and 90% are grouped together to form one health status correspondence set. Then, feature analysis is performed on the battery application state parameters (such as charging power curves) within each health status correspondence set to extract key features, such as the upward trend, downward trend, duration, and amplitude of the stable segment of the curve. Based on these features, correspondence groups with similar feature performance are further categorized into health status correspondence subsets. For example, assuming that the charging power curves of two correspondence groups both exhibit a "short-term rapid rise followed by a long period of stability," and their health statuses are similar, they are grouped into the same subset. The essence of this process is to simplify the complex dataset into a more representative subset through dimensionality reduction and pattern recognition, facilitating the determination of subsequent parameter ranges and data increments.
[0053] Step S300: Determine the mapping parameter range for each type of battery application state parameter in each health status correspondence subset, and combine the mapping parameter ranges of different types of battery application state parameters into a mapping parameter range group.
[0054] The principle of step S300 is to analyze the overlap of parameter curves in the subset, determine the variation range of each parameter at different time points (i.e., the mapping parameter range), and combine them into multi-dimensional range groups to provide constraints for data generation. Specifically, the method involves comparing the overlap of similar curves (such as charging power curves) in the subset, identifying the overlapping areas of all curves at each time point, and defining these as mapping parameter ranges. For example, if the power values of all charging power curves in the subset are within [120W, 140W] at the 5th minute, then the mapping parameter range for that time point is [120W, 140W]. Subsequently, the mapping parameter ranges of different types of parameters (such as charging power and discharge temperature) are combined into mapping parameter range groups, such as {charging power at the 5th minute: [120W, 140W], discharge temperature at the 5th minute: [25°C, 28°C]}. The essence of this process is to quantify the allowable variation range of parameters based on the overlap characteristics of the curves, ensuring that the subsequently generated data conforms to the operating rules of the subset, and providing a reliable basis for data increment and verification.
[0055] Step S400: Based on the several mapping parameter range groups corresponding to each health status correspondence set, perform data increment, construct several first battery application status parameter groups that conform to the mapping parameter range groups, and construct several new battery health status comparison experiments to obtain several verification health status correspondence groups.
[0056] The principle of step S400 is to generate virtual data by mapping parameter ranges, expand the dataset, and verify its effectiveness to enhance the model's adaptability. Based on the mapped parameter range set, curves that conform to the range constraints are generated using random reconstruction. For example, power values are randomly generated within the power range [120W, 140W] at the 5th minute to construct the first battery application state parameter set. These generated curves need to be optimized based on the curve overlap characteristics (such as boundary distance and allowable fluctuation parameters) to ensure close resemblance to real operating data. Subsequently, through simulation or experimentation, new health comparison experiments are conducted using these parameter sets to determine their corresponding health levels, forming a verification health correspondence set. For example, after generating a set of charging power curves, if experiments verify that its health level is 87%, a verification correspondence set is formed. This process ensures the representativeness of the data by constraining data generation through the mapped parameter range (based on curve overlap), and at the same time confirms the reliability of the generated data through verification experiments, providing support for the construction of the parameter library.
[0057] Step S500: The first battery application state parameter group is combined to construct a battery application state parameter library, and the battery application state parameter library is verified and tested using several verification battery application state parameter groups. The comprehensive matching parameters between the retrieved matching battery application state parameter group and the verification battery application state parameter group are analyzed. Based on the comprehensive matching parameters, the mapping parameter range in the mapping parameter range group corresponding to the matching battery application state parameter group is narrowed, and the corresponding first battery application state parameter group is removed, thereby updating the battery application state parameter library.
[0058] The principle of step S500 is to optimize the parameter library through verification testing, evaluate data quality based on comprehensive consistency parameters, and narrow the range of mapped parameters to improve prediction accuracy. First, the first battery application state parameter group is integrated into the parameter library. Then, tests are conducted using the verification parameter group, and the comprehensive consistency parameters are calculated through curve overlap comparison and health difference measurement to evaluate the consistency between the matching parameter group and the verification parameter group. For example, if the matching charging power curve and the verification curve overlap at each time point, and the health difference is less than 1%, the consistency parameter is high. If the comprehensive consistency parameter is low, it indicates that the matching parameter group is distorted, and its corresponding mapped parameter range needs to be narrowed (e.g., the power range at the 5th minute is narrowed from [120W, 140W] to [125W, 135W]), parameter groups exceeding the new range are removed, and the parameter library is updated. This process, by triggering range narrowing through low consistency parameters, optimizes curve overlap characteristics, ensures that the parameter library data is more realistic and reliable, and improves the accuracy of the prediction model.
[0059] Step S600: Analyze the real-time battery application status parameter group using the updated battery application status parameter library to determine the real-time battery health.
[0060] In the embodiments disclosed in this invention, the types of battery application state parameters include: charging power curve, discharging power curve, charging temperature change curve, and discharging temperature change curve.
[0061] In the embodiments disclosed in this invention, the method for performing feature analysis on battery application state parameters in each health correspondence group of the health correspondence set includes:
[0062] Step S201: Align the charging power curve and the charging temperature change curve according to the time correspondence principle, and align the discharging power curve and the discharging temperature change curve.
[0063] Step S202: Determine the trend of each curve, extract the curve segment whose upward trend lasts longer than a preset value and whose average upward amplitude is greater than or equal to the preset value, and record it as an upward curve segment; extract the curve segment whose downward trend lasts longer than a preset value and whose average downward amplitude is greater than or equal to the preset value, and record it as a downward curve segment; extract the remaining curve segments and record them as stable curve segments.
[0064] Step S203: Determine the time length and average ordinate value of the rising curve segment, falling curve segment, and stationary curve segment, and sort them according to their size to obtain the curve segment parameter group.
[0065] Step S204: Based on the consistency between the parameter groups of the curve segments, the health correspondence groups are classified to obtain several subsets of health correspondences.
[0066] In the embodiments disclosed in this invention, the method for determining the mapping parameter range of each type of battery application state parameter in each health status correspondence subset includes:
[0067] Step S301: Overlap and compare curves of the same type in the health correspondence subset to determine the curve coverage area between the curves, and identify the curve coverage area as the mapping parameter range.
[0068] In the embodiments disclosed in this invention, the method for incremental data processing based on several mapping parameter range groups corresponding to each health level correspondence set includes:
[0069] Step S401: Evenly set several reference time points on the horizontal axis of the time covered by the curve, determine the vertical mapping line of the curve covered by each reference time point, and determine the center point of the vertical mapping line, which is recorded as the vertical center point.
[0070] Step S402: Connect the longitudinal center points to obtain the reference curve, and analyze the boundary distance characteristics of the area covered by the curve relative to the reference curve. The boundary distance characteristics are the relative distances of the boundary nodes of different areas relative to the reference curve. Sort the relative distances from large to small, select a few relative distances with the highest order, and calculate the average value between them, which is recorded as the average relative distance.
[0071] Step S403: Based on the preset distance interval to which the average relative distance belongs, determine the allowable fluctuation parameter of each curve. Based on the allowable fluctuation parameter of each curve, perform several random reconstructions on the curves corresponding to the mapping parameter range group to obtain the reconstructed curves. Combine the reconstructed curves of different types to obtain the first battery application state parameter group.
[0072] Step S403 is based on determining the fluctuation parameter based on the average relative distance, generating a reconstructed curve that conforms to the mapping parameter range, and constructing the first battery application state parameter group. Specifically, the method involves determining the allowable fluctuation parameter (e.g., ±5W) based on the preset distance interval to which the average relative distance (e.g., 8W) belongs, representing the variable range of the curve near the reference curve. Then, within the mapping parameter range (e.g., [120W, 140W] at the 5th minute), a new curve is generated by randomly reconstructing the value using the reference curve as the center and combining it with the fluctuation parameter. For example, if the reference value at the 5th minute is 130W and the fluctuation parameter is ±5W, the randomly generated value is within [125W, 135W], and connecting the values at each time point forms the reconstructed curve. Combining different types of reconstructed curves (e.g., charging power, discharging temperature) constitutes the first battery application state parameter group. This process ensures that the generated data closely matches the regularity of the coverage area. If the overall matching parameter is low, it indicates that the reconstructed curve is distorted, and the mapping parameter range needs to be narrowed (e.g., [125W, 135W]) to constrain the fluctuation range and improve the authenticity and consistency of the data.
[0073] In the embodiments disclosed in this invention, the method for randomly reconstructing the curves corresponding to the mapping parameter range group several times based on the allowable fluctuation parameter of each curve includes:
[0074] Step S4031: Determine the length of the longitudinal mapping line corresponding to different time nodes in the curve coverage area, and determine the reconstruction difference at different time nodes in combination with the allowable fluctuation parameter.
[0075] Step S4032: Based on the reconstruction difference, construct reconstruction curve nodes above and below the curve nodes of the corresponding time nodes, and connect the reconstruction curve nodes to obtain the reconstruction curve.
[0076] In the embodiments disclosed in this invention, the method for determining the amount of reconstruction difference includes:
[0077] Step S40311: Based on the preset fluctuation parameter range to which the allowed fluctuation parameter belongs, determine the corresponding length ratio coefficient, calculate the product of the length ratio coefficient and the length of the longitudinal mapping line to obtain the maximum reconstruction difference, and randomly generate several reconstruction difference values within the range of values less than or equal to the maximum reconstruction difference.
[0078] In the embodiments disclosed in this invention, the method for analyzing the comprehensive matching parameters of the retrieved matching battery application state parameter set and the verification battery application state parameter set includes:
[0079] Step S501: Compare the corresponding curves in the matching battery application status parameter group and the verification battery application status parameter group. If the coverage area between each corresponding curve is less than or equal to a preset value, it is determined that the parameters between the matching battery application status parameter group and the verification battery application status parameter group are matched, and the health difference between the two is calculated.
[0080] Step S502: Determine the comprehensive matching parameters based on the number of verification comparisons and the corresponding health difference for each comparison.
[0081] .
[0082] Where W is the overall matching parameter, K is the matching sub-parameter adjustment coefficient, b is the matching sub-parameter adjustment constant, and n is the total number of verification comparisons. Let be the measure of health difference during the i-th verification comparison. This is a function for judging differences in health status. based on The corresponding health difference interval is given, and the corresponding matching parameter is output.
[0083] In the embodiments disclosed in this invention, the method for narrowing the range of mapping parameters in the mapping parameter range group corresponding to the matching battery application state parameter group based on comprehensive matching parameters includes:
[0084] Step S503: Based on the preset matching parameter interval to which the comprehensive matching parameter belongs, determine the narrowing ratio of the mapping parameter range, and based on the narrowing ratio of the mapping parameter range, push the upper and lower boundaries of the mapping parameter range inward to narrow it.
[0085] In the embodiments disclosed in this invention, a UPS battery health prediction system based on multi-source data is also disclosed, comprising:
[0086] The first module is used to construct a battery health comparison experiment and determine several health correspondence groups that are related to battery health and battery application state parameters.
[0087] The second module is used to classify health correspondence groups based on battery health, obtain several health correspondence sets, perform feature analysis on the battery application state parameters in each health correspondence group in the health correspondence set, and further classify health correspondence groups with equivalent feature performance to obtain several health correspondence subsets.
[0088] The third module is used to determine the mapping parameter range of each type of battery application state parameter in each health status correspondence subset, and to combine the mapping parameter ranges of different types of battery application state parameters into a mapping parameter range group.
[0089] The fourth module is used to perform data increment based on several mapping parameter range groups corresponding to each health status correspondence set, construct several first battery application status parameter groups that conform to the mapping parameter range groups, and construct several new battery health status comparison experiments to obtain several verification health status correspondence groups.
[0090] The fifth module is used to construct a battery application state parameter library by combining the first battery application state parameter group, and to verify and test the battery application state parameter library using several verification battery application state parameter groups. It analyzes the comprehensive matching parameters between the retrieved matching battery application state parameter group and the verification battery application state parameter group, and based on the comprehensive matching parameters, narrows the mapping parameter range in the mapping parameter range group corresponding to the matching battery application state parameter group, removes the corresponding first battery application state parameter group, and updates the battery application state parameter library.
[0091] The sixth module is used to analyze the real-time battery application status parameter group using the updated battery application status parameter library to determine the real-time battery health.
[0092] This invention proposes a method and system for predicting UPS battery health based on multi-source data, belonging to the field of battery health prediction technology. It establishes a correspondence between battery health and multi-source parameters (such as charging power and discharge temperature curves) through a battery health comparison experiment. Based on health classification and feature analysis, a subset of health correspondences is formed, and the mapping parameter ranges for each parameter are determined and combined into range groups. Data increment is performed using these range groups to generate a first set of battery application status parameters, and a parameter library is constructed through verification experiments. The parameter library is optimized based on comprehensive matching parameters, narrowing the mapping parameter ranges and eliminating distorted data. Finally, the updated parameter library is used to analyze real-time parameters and predict battery health. This invention improves prediction accuracy and robustness through multi-source data integration, feature extraction, and dynamic optimization, providing reliable support for UPS system maintenance.
[0093] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.
[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.
Claims
1. A UPS battery health prediction method based on multi-source data, characterized in that, include: Step S100: Construct a battery health comparison experiment to determine several health correspondence groups that are related to battery health and battery application state parameters; Step S200: Based on battery health, classify the health correspondence groups to obtain several health correspondence sets. Perform feature analysis on the battery application state parameters in each health correspondence group in the health correspondence set. Further classify the health correspondence groups with equivalent feature performance to obtain several health correspondence subsets. Step S300: Determine the mapping parameter range for each type of battery application state parameter in each health status correspondence subset, and combine the mapping parameter ranges of different types of battery application state parameters into a mapping parameter range group. Step S400: Based on the several mapping parameter range groups corresponding to each health status correspondence set, perform data increment, construct several first battery application status parameter groups that conform to the mapping parameter range groups, and construct several new battery health status comparison experiments to obtain several verification health status correspondence groups. Step S500: The first battery application state parameter group is combined to construct a battery application state parameter library, and the battery application state parameter library is verified and tested using several verification battery application state parameter groups. The comprehensive matching parameters between the retrieved matching battery application state parameter group and the verification battery application state parameter group are analyzed. Based on the comprehensive matching parameters, the mapping parameter range in the mapping parameter range group corresponding to the matching battery application state parameter group is narrowed, and the corresponding first battery application state parameter group is removed, thereby updating the battery application state parameter library. Step S600: Analyze the real-time battery application status parameter group using the updated battery application status parameter library to determine the real-time battery health.
2. The UPS battery health prediction method based on multi-source data according to claim 1, characterized in that, The types of battery application status parameters include: charging power curve, discharging power curve, charging temperature change curve, and discharging temperature change curve.
3. The UPS battery health prediction method based on multi-source data according to claim 2, characterized in that, Methods for feature analysis of battery application state parameters in each health correspondence group of the health correspondence set include: Step S201: Align the charging power curve and the charging temperature change curve according to the time correspondence principle, and align the discharging power curve and the discharging temperature change curve. Step S202: Determine the trend of each curve, and cut off the curve segment whose upward trend lasts for more than a preset value and whose average upward amplitude is greater than or equal to the preset value, and record it as an upward curve segment. Cut off the curve segment whose downward trend lasts for more than a preset value and whose average downward amplitude is greater than or equal to the preset value, and record it as a downward curve segment. Cut off the remaining curve segments and record them as stable curve segments. Step S203: Determine the time length and average ordinate value of the rising curve segment, falling curve segment, and stationary curve segment, and sort them according to their size to obtain the curve segment parameter group. Step S204: Based on the consistency between the parameter groups of the curve segments, the health correspondence groups are classified to obtain several subsets of health correspondences.
4. The UPS battery health prediction method based on multi-source data according to claim 2, characterized in that, Methods for determining the mapping parameter range for each type of battery application state parameter in each health status correspondence subset include: Step S301: Overlap and compare curves of the same type in the health correspondence subset to determine the curve coverage area between the curves, and identify the curve coverage area as the mapping parameter range.
5. The UPS battery health prediction method based on multi-source data according to claim 4, characterized in that, Methods for incremental data processing based on several mapping parameter ranges corresponding to each health level correspondence set include: Step S401: Evenly set several reference time points on the horizontal axis of the time covered by the curve, determine the vertical mapping line of the curve covered by each reference time point, and determine the center point of the vertical mapping line, which is recorded as the vertical center point. Step S402: Connect the longitudinal center points to obtain the reference curve, and analyze the boundary distance characteristics of the area covered by the curve relative to the reference curve. The boundary distance characteristics are the relative distances of the boundary nodes of different areas relative to the reference curve. Sort the relative distances from large to small, select a few relative distances with the highest order, and calculate the average value between them, which is recorded as the average relative distance. Step S403: Based on the preset distance interval to which the average relative distance belongs, determine the allowable fluctuation parameter of each curve. Based on the allowable fluctuation parameter of each curve, perform several random reconstructions on the curves corresponding to the mapping parameter range group to obtain the reconstructed curves. Combine the reconstructed curves of different types to obtain the first battery application state parameter group.
6. The UPS battery health prediction method based on multi-source data according to claim 5, characterized in that, Based on the allowable fluctuation parameter of each curve, the method of randomly reconstructing the curves corresponding to the mapping parameter range group several times includes: Step S4031: Determine the length of the longitudinal mapping line corresponding to different time nodes in the curve coverage area, and determine the reconstruction difference at different time nodes in combination with the allowable fluctuation parameter. Step S4032: Based on the reconstruction difference, construct reconstruction curve nodes above and below the curve nodes of the corresponding time nodes, and connect the reconstruction curve nodes to obtain the reconstruction curve.
7. The UPS battery health prediction method based on multi-source data according to claim 6, characterized in that, Methods for determining the amount of reconstruction difference include: Step S40311: Based on the preset fluctuation parameter range to which the allowed fluctuation parameter belongs, determine the corresponding length ratio coefficient, calculate the product of the length ratio coefficient and the length of the longitudinal mapping line to obtain the maximum reconstruction difference, and randomly generate several reconstruction difference values within the range of values less than or equal to the maximum reconstruction difference.
8. The UPS battery health prediction method based on multi-source data according to claim 2, characterized in that, Methods for analyzing the combined fit parameters of the retrieved state parameter sets for matching and verification battery applications include: Step S501: Compare the corresponding curves in the matching battery application status parameter group and the verification battery application status parameter group. If the coverage area between each corresponding curve is less than or equal to a preset value, it is determined that the parameters between the matching battery application status parameter group and the verification battery application status parameter group are matched, and the health difference between the two is calculated. Step S502: Determine the comprehensive matching parameters based on the number of verification comparisons and the corresponding health difference for each comparison. ； Where W is the overall matching parameter, K is the matching sub-parameter adjustment coefficient, b is the matching sub-parameter adjustment constant, and n is the total number of verification comparisons. Let be the measure of health difference during the i-th verification comparison. This is a function for judging differences in health status. based on The corresponding health difference interval is given, and the corresponding matching parameter is output.
9. The UPS battery health prediction method based on multi-source data according to claim 1, characterized in that, Based on comprehensive matching parameters, methods for narrowing the range of mapping parameters in the mapping parameter range group corresponding to the battery application state parameter group for matching include: Step S503: Based on the preset matching parameter interval to which the comprehensive matching parameter belongs, determine the narrowing ratio of the mapping parameter range, and based on the narrowing ratio of the mapping parameter range, push the upper and lower boundaries of the mapping parameter range inward to narrow it.
10. A UPS battery health prediction system based on multi-source data, characterized in that, The UPS battery health prediction method for performing any one of claims 1-9 includes: The first module is used to construct a battery health comparison experiment and determine several health correspondence groups that are related to battery health and battery application state parameters. The second module is used to classify health correspondence groups based on battery health, obtain several health correspondence sets, perform feature analysis on the battery application state parameters in each health correspondence group in the health correspondence set, and further classify health correspondence groups with equivalent feature performance to obtain several health correspondence subsets. The third module is used to determine the mapping parameter range of each type of battery application state parameter in each health status correspondence subset, and to combine the mapping parameter ranges of different types of battery application state parameters into a mapping parameter range group. The fourth module is used to perform data increment based on several mapping parameter range groups corresponding to each health status correspondence set, construct several first battery application status parameter groups that conform to the mapping parameter range groups, and construct several new battery health status comparison experiments to obtain several verification health status correspondence groups. The fifth module is used to construct a battery application state parameter library by combining the first battery application state parameter group, and to verify and test the battery application state parameter library using several verification battery application state parameter groups. It analyzes the comprehensive matching parameters between the retrieved matching battery application state parameter group and the verification battery application state parameter group, and based on the comprehensive matching parameters, narrows the mapping parameter range in the mapping parameter range group corresponding to the matching battery application state parameter group, removes the corresponding first battery application state parameter group, and updates the battery application state parameter library. The sixth module is used to analyze the real-time battery application status parameter group using the updated battery application status parameter library to determine the real-time battery health.
Citation Information
Patent Citations
Battery state evaluation system
CN118914865A
Lithium battery health state monitoring method and system
CN119355557A