Intelligent charging management system and method applied to lithium battery

Abstract

The invention discloses an intelligent charging management system and method applied to a lithium battery, relates to the technical field of big data analysis, and realizes accurate monitoring through multi-dimensional analysis. The current and timestamp data of the constant-voltage charging stage of the lithium battery are collected to construct a real-time current change curve chart, and a time sliding window is utilized to fit a standard current change curve chart, so that accidental data fluctuation is effectively removed, and a stable data basis is provided for evaluating the charging duration and the performance state; obtaining charging data of lithium batteries with the same initial electric quantity and specification and model, constructing and quantifying a transverse current change curve graph, comparing current change rules of different lithium batteries, and providing scientific support for performance evaluation and strategy optimization; according to the method, the charging time deviations of different dimensions are subjected to weighted fusion, a comprehensive charging time deviation sequence diagram is constructed, prediction monitoring is carried out by means of a machine learning algorithm, the change trend is presented in a visual mode, and dynamic monitoring and early warning in the lithium battery charging process are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to an intelligent charging management system and method for lithium batteries. Background Art

[0002] As new lithium-ion batteries are widely used in various electronic devices and energy storage systems, their charging process faces significant challenges. Even lithium-ion batteries of the same specifications and model can experience varying degrees of degradation after the same number of charge cycles. This leads to differences in the actual rate of capacity loss and the current-voltage characteristics during charging. These cumulative differences complicate the estimation of battery life degradation. Furthermore, over long-term use, the same lithium-ion battery exhibits nonlinear aging characteristics, influenced by factors such as charge-discharge cycles, operating temperature, and usage habits, further exacerbating the uncertainty of life degradation. These uncontrollable factors in the charging process not only reduce the device user experience and the accuracy of battery life planning, but can also pose safety risks due to overcharging or undercharging. Furthermore, significant differences in lifespan between lithium-ion batteries make it difficult to effectively assess their service life and performance status, significantly hindering their large-scale application and stable device operation. Therefore, there is currently a lack of a technical solution that can effectively assess the service life and performance changes of lithium-ion batteries during charging. Summary of the Invention

[0003] The object of the present invention is to provide an intelligent charging management system and method for lithium batteries to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent charging management method for lithium batteries, the intelligent charging management method comprising the following steps:

[0005] Step S1, collecting current data of the rechargeable lithium battery in the constant voltage charging stage, and constructing a current change curve graph based on the timestamp data corresponding to the current data, which is recorded as a real-time current change curve graph;

[0006] The current data of the rechargeable lithium battery during the constant voltage charging stage is collected by a current sensor; the x-axis is constructed using the timestamp data corresponding to the collected current data as the horizontal coordinate, and the y-axis is constructed using the current data as the vertical coordinate. A current change curve graph of the rechargeable lithium battery is constructed based on the x-axis and y-axis, which is recorded as a real-time current change curve graph; after the rechargeable lithium battery completes one charge, a real-time current change curve graph is generated based on the current data collected during the charging process and the corresponding timestamp data;

[0007] The current sensor accurately collects the current and timestamp data of the constant voltage charging stage, and constructs a curve with intuitive horizontal and vertical coordinates to fully record the current changes in a single charging process, providing an original and reliable data basis for subsequent analysis of lithium battery charging characteristics, establishment of standard models, and evaluation of charging status.

[0008] Step S2: Setting a time sliding window to obtain multiple sets of historical current change curves of the rechargeable lithium battery, and fitting the historical current change curves within the time sliding window to obtain a standard current change curve of the rechargeable lithium battery;

[0009] Step S2-1: The size of the time sliding window is set to a fixed value N, and the number of times the rechargeable lithium battery is charged is used as the sliding step. When charging is completed, the real-time current change curve within the time sliding window is updated according to the sliding step;

[0010] Step S2-2: Obtain a historical current change curve graph within the time sliding window, perform statistics based on the corresponding current data on the same horizontal axis, calculate the average current value at the corresponding horizontal axis point to form a new set of rechargeable lithium battery current change curve graphs, recorded as standard current change curve graphs;

[0011] By setting a fixed-size time sliding window with the number of charge cycles as the step length, dynamically updating the historical current change curve, and performing statistical fitting on the data within the window, it is possible to effectively integrate the current change characteristics of the lithium battery during multiple charging processes, remove accidental fluctuations in single charging data, and form a representative standard current change curve, providing a stable and reliable reference standard for accurately evaluating the charging time and performance status of lithium batteries.

[0012] Step S3: extracting the standard charging time required for the rechargeable lithium battery based on the standard current change curve analysis, and calculating the charging time decay deviation of the lithium battery in combination with the real-time current change curve analysis, which is recorded as the first time decay deviation;

[0013] Step S3-2: using a machine learning algorithm to predict the time required to complete charging of the rechargeable lithium battery based on the real-time current change curve, and recording the predicted value of the time required to complete charging as the charging time prediction value;

[0014] The machine learning algorithm uses the following formula to predict how long it will take for a rechargeable lithium battery to complete charging:

[0015]

[0016] Where, I set It represents the current value set when the lithium battery is fully charged; n represents the number of training samples; a i Expressed as Lagrange multiplier; y iRepresented as the current value in the i-th training data; K(x i , t) is expressed as kernel function; x i represents the data input for the i-th training sample; t represents the time required for the rechargeable lithium battery to complete charging; b represents the bias term;

[0017] Step S3-3, calculating the difference between the standard charging time value and the predicted charging time value to obtain a charging time decay deviation of the rechargeable lithium battery, which is recorded as a first time decay deviation;

[0018] The standard charging time is extracted from the standard current change curve as a reference benchmark, and the real-time current change curve is analyzed with a machine learning algorithm to predict the charging time. The first-time attenuation deviation is obtained by calculating the difference between the standard value and the predicted value. This can timely detect the difference between the actual charging time of the lithium battery and the standard state, providing a quantitative basis for evaluating the performance attenuation of the lithium battery and optimizing the charging strategy.

[0019] Step S4, obtaining charging data of a lithium battery having the same charge as that of a rechargeable lithium battery during initial charging, analyzing and extracting current data during the constant voltage charging stage to construct multiple sets of current change curves, recorded as transverse current change curves; and quantizing the transverse current change curves to obtain quantized current change curves;

[0020] Step S4-1: Obtaining the power data of the rechargeable lithium battery during initial charging through the intelligent battery management system, using the power data as a search index, and searching for charging data of lithium batteries with the same power and the same specifications and models during initial charging based on a time sliding window mechanism;

[0021] Step S4-2: Analyze and extract the current data during the constant voltage charging stage, and construct multiple sets of current change curves for the found lithium batteries according to the current change curve construction method of step S1, which are recorded as transverse current change curves;

[0022] Step S4-3: extracting the current change slope characteristics from each transverse current change curve graph, and obtaining the quantized maximum current change slope and minimum current change slope through normalization processing according to the maximum current change slope and the minimum current change slope;

[0023] Step S4-4, calculating the area under each transverse current change curve and the x-axis using the trapezoidal rule, and calculating the average of the areas under each transverse current change curve and the x-axis as the standard area of the transverse current change curve;

[0024] The trapezoidal rule calculation formula is as follows:

[0025]

[0026] Where a is the lower limit of the integral interval; b is the upper limit of the integral interval; h = (ba) ÷ n, which is the width of the subinterval, and n is the number of subintervals into which the interval [a, b] is divided; x i =a+ih, i=0, 1, ..., n, represents the endpoints of each subinterval; f(x) represents the integrand; f(x i ) is expressed as the function f(x) at x i The value of

[0027] Step S4-5, calculating the average value of the current data corresponding to the abscissa of each transverse current change curve graph to construct an average transverse current change curve graph, using the maximum slope of the current change and the minimum slope of the current change after quantization and the standard area as restriction conditions, adjusting the average transverse current change curve graph to obtain a quantized current change curve graph, the specific process of the adjustment is: the minimum slope of the quantized current change curve graph satisfies not less than the minimum slope after quantization, the maximum slope of the quantized current change curve graph satisfies not greater than the maximum slope after quantization, and the area formed by the curve under the quantized current change curve graph and the x-axis is calculated to be equal to the standard area;

[0028] By acquiring charging data of lithium batteries with the same initial charge and specifications, constructing and quantifying a transverse current change curve, we can effectively compare and analyze the current change patterns of different lithium batteries during the constant voltage charging stage. By extracting slope characteristics, calculating standard areas, and constructing average curves for adjustment, we can accurately reflect the commonalities and differences of lithium batteries at this stage, providing scientific and accurate data support for lithium battery performance evaluation, charging strategy optimization, and anomaly detection.

[0029] Step S5: Analyze and extract the quantized charging time required for the rechargeable lithium battery according to the quantized current change curve, and record it as the quantized charging time value; calculate the quantized charging time deviation of the rechargeable lithium battery according to the quantized charging time value, and record it as the second time decay deviation;

[0030] The maximum value of the horizontal axis of the quantized current change curve is read as the time required for the rechargeable lithium battery to complete charging after quantization, which is recorded as the quantized charging time value; the difference between the quantized charging time value and the predicted charging time value is calculated to obtain the quantized charging time deviation of the rechargeable lithium battery, which is recorded as the second time decay deviation;

[0031] By extracting the maximum value of the horizontal axis of the quantized current change curve to obtain the quantized value of the charging time, and calculating the difference between it and the predicted value of the charging time, the second time attenuation deviation is obtained. The charging time of the lithium battery can be presented as quantitative data, which intuitively reflects the difference between the actual charging time and the predicted time. It provides quantitative indicators for evaluating the charging performance of the lithium battery, optimizing the charging algorithm, and estimating the battery aging status, thereby improving the accuracy and reliability of controlling the lithium battery charging process.

[0032] Step S6: Obtaining a comprehensive charging time deviation of the rechargeable lithium battery by weighted fusion based on the first time decay deviation and the second time decay deviation, constructing a comprehensive charging time deviation time sequence diagram based on multiple historical sets of comprehensive charging time deviations of the rechargeable lithium battery, and analyzing the comprehensive charging time deviation time sequence diagram to monitor the rechargeable lithium battery;

[0033] Step S6-1, obtaining a comprehensive charging time deviation of the rechargeable lithium battery by weighted fusion based on the first time decay deviation and the second time decay deviation, constructing a y-axis based on the historical multiple sets of comprehensive charging time deviations of the rechargeable lithium battery as the vertical coordinate, constructing an x-axis based on the number of times corresponding to the comprehensive charging time deviations of the rechargeable lithium battery as the horizontal coordinate, and constructing a comprehensive charging time deviation time sequence diagram based on the x-axis and y-axis;

[0034] The weighted fusion calculation uses the following formula:

[0035] T all =w1×t1+w2×t2;

[0036] Where, T all It represents the comprehensive charging time deviation of the rechargeable lithium battery; w1 represents the weight coefficient of the first time decay deviation; t1 represents the first time decay deviation; w2 represents the weight coefficient of the second time decay deviation; t2 represents the second time decay deviation;

[0037] Step S6-2: Obtain comprehensive charging time deviation timing diagrams of multiple lithium batteries according to the model and specifications of the rechargeable lithium batteries to construct a time deviation set, traverse and read the comprehensive charging time deviations in each comprehensive charging time deviation timing diagram in the time deviation set, and select the maximum value as the time deviation threshold;

[0038] Step S6-3: Use a machine learning algorithm to predict the comprehensive charging time deviation corresponding to the next horizontal axis of the comprehensive charging time deviation timing diagram of the rechargeable lithium battery, and monitor the rechargeable lithium battery according to the time deviation threshold:

[0039] When the predicted comprehensive charging time deviation exceeds the time deviation threshold, an early warning signal is issued;

[0040] When the predicted comprehensive charging time deviation does not exceed the time deviation threshold, continue to monitor the charging lithium battery;

[0041] The first and second time decay deviations are weighted and fused to obtain a comprehensive charging time deviation, which is then used to construct a time series diagram. The time deviation threshold is then determined by combining the time series diagrams of similar lithium-ion batteries. A machine learning algorithm is then used to predict subsequent deviations, and monitoring and early warning are performed based on the threshold. This method integrates multi-dimensional charging time deviation information, visualizing the changing trend with a time series diagram. It also uses population data to set reasonable thresholds, enabling dynamic monitoring of the lithium-ion battery charging status through prediction, allowing for timely detection of anomalies and ensuring lithium-ion battery charging safety and stable performance.

[0042] Furthermore, an intelligent charging management system for lithium batteries includes a real-time curve generation module, a standard curve fitting module, a charging deviation calculation module, a transverse curve quantification module, a quantification deviation calculation module, and a battery monitoring and warning module;

[0043] The real-time curve generation module is used to collect constant-voltage charging data of lithium batteries and construct a real-time current change curve graph; the standard curve fitting module is used to fit the historical current change curve graph through a time sliding window to obtain a standard curve; the charging deviation calculation module is used to extract the standard charging time from the standard curve and calculate the charging time attenuation deviation in combination with the real-time curve; the horizontal curve quantization module is used to obtain data of lithium batteries with the same power and construct a quantized current change curve graph; the quantization deviation calculation module is used to obtain the quantized charging time from the quantization curve and calculate the quantized charging time deviation; the battery monitoring and warning module is used to fuse the deviation data to construct a time sequence diagram and monitor and warn the charging status of the lithium battery;

[0044] The output end of the real-time curve generation module is electrically connected to the input end of the standard curve fitting module; the output end of the standard curve fitting module is electrically connected to the input end of the charging deviation calculation module; the output end of the charging deviation calculation module is electrically connected to the input end of the horizontal curve quantization module; the output end of the horizontal curve quantization module is electrically connected to the input end of the quantization deviation calculation module; the output end of the quantization deviation calculation module is electrically connected to the input end of the battery monitoring and early warning module;

[0045] The real-time curve generation module includes a current data acquisition unit and a real-time curve generation unit; the current data acquisition unit is used to obtain the current data and corresponding timestamp of the rechargeable lithium battery in the constant voltage charging stage; the real-time curve generation unit is used to construct a real-time current change curve graph based on the collected data;

[0046] The standard curve fitting module includes a window data updating unit and a standard curve calculation unit; the window data updating unit is used to update the real-time current change curve graph within the time sliding window according to the number of charging times; the standard curve calculation unit is used to calculate the standard current change curve graph based on the statistical calculation of the curve within the window;

[0047] The charging deviation calculation module includes a standard time reading unit and a deviation value calculation unit; the standard time reading unit is used to read the charging time standard value in the standard current change curve; the deviation value calculation unit is used to predict the real-time curve charging time through machine learning and calculate the first time decay deviation;

[0048] The transverse curve quantization module includes a data search and construction unit and a curve quantization adjustment unit; the data search and construction unit is used to search and construct a transverse current change curve diagram based on the power data; the curve quantization adjustment unit is used to extract curve features and obtain a quantized current change curve diagram through adjustment under limited conditions;

[0049] The quantization deviation calculation module includes a quantization time reading unit and a quantization deviation calculation unit; the quantization time reading unit is used to read the quantization value of the charging time in the quantization current change curve diagram; the quantization deviation calculation unit is used to calculate the difference between the quantization value of the charging time and the predicted value to obtain a second time decay deviation;

[0050] The battery monitoring and early warning module includes a timing diagram construction unit and a battery status monitoring unit; the timing diagram construction unit is used to construct a comprehensive charging time deviation timing diagram by weighted fusion deviation data; the battery status monitoring unit is used to set a threshold and monitor and warn the charging status of the lithium battery by predicting the deviation value;

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. The present invention collects the current and timestamp data of the constant-voltage charging stage of the lithium battery to construct a real-time current change curve, and sets a time sliding window to fit the standard current change curve, so as to achieve accurate recording and integration of the current change characteristics during the lithium battery charging process, eliminate accidental data fluctuations, and provide a stable and reliable data basis and reference standard for evaluating the charging time and performance status of the lithium battery, thereby solving the problem of unstable single charging data.

[0053] 2. The present invention obtains charging data of lithium batteries with the same initial power and specifications to construct a horizontal current change curve and conduct quantitative analysis, thereby comparing the current change patterns of different lithium batteries in the constant voltage charging stage, accurately reflecting the commonalities and differences of lithium batteries in this stage, and providing scientific and accurate data support for lithium battery performance evaluation, charging strategy optimization and anomaly detection, effectively solving the problem of difficulty in evaluating performance differences between lithium batteries.

[0054] 3. The present invention constructs a comprehensive charging time deviation time series diagram by weighted fusion of charging time deviations in different dimensions, and determines the threshold value in combination with the data of lithium batteries of the same model. It uses machine learning algorithms for prediction and monitoring to achieve dynamic monitoring and early warning of the charging status of lithium batteries, integrates multi-source information to visualize the changing trend, detects anomalies in time, ensures the safety of lithium battery charging and stable performance, and solves the problems of uncontrollable lithium battery charging process and difficulty in estimating life attenuation. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic flow chart of an intelligent charging management method for lithium batteries according to the present invention;

[0056] Figure 2 The figure is a structural diagram of an intelligent charging management system for lithium batteries according to the present invention. DETAILED DESCRIPTION

[0057] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0058] Example 1: Figure 1 As shown, the present invention provides a technical solution, an intelligent charging management method for lithium batteries, and the intelligent charging management method includes the following steps:

[0059] Step S1, collecting current data of the rechargeable lithium battery in the constant voltage charging stage, and constructing a current change curve graph based on the timestamp data corresponding to the current data, which is recorded as a real-time current change curve graph;

[0060] The current data of the rechargeable lithium battery during the constant voltage charging stage is collected by a current sensor; the x-axis is constructed using the timestamp data corresponding to the collected current data as the horizontal coordinate, and the y-axis is constructed using the current data as the vertical coordinate. A current change curve graph of the rechargeable lithium battery is constructed based on the x-axis and y-axis, which is recorded as a real-time current change curve graph; after the rechargeable lithium battery completes one charge, a real-time current change curve graph is generated based on the current data collected during the charging process and the corresponding timestamp data;

[0061] In practical implementation, taking the charging of lithium batteries in new energy vehicles as an example, a high-precision Hall effect sensor can be used as the current sensor to accurately collect current data during the constant-voltage charging phase. This time-stamped data can be accurately recorded by the vehicle's electronic control system. The principle is to visualize current versus time data to intuitively illustrate the current variation pattern of the lithium battery during the constant-voltage charging phase, facilitating subsequent analysis of the charging status.

[0062] Step S2: Setting a time sliding window to obtain multiple sets of historical current change curves of the rechargeable lithium battery, and fitting the historical current change curves within the time sliding window to obtain a standard current change curve of the rechargeable lithium battery;

[0063] Step S2-1: The size of the time sliding window is set to a fixed value N, and the number of times the rechargeable lithium battery is charged is used as the sliding step. When charging is completed, the real-time current change curve within the time sliding window is updated according to the sliding step;

[0064] Step S2-2: Obtain a historical current change curve graph within the time sliding window, perform statistics based on the corresponding current data on the same horizontal axis, calculate the average current value at the corresponding horizontal axis point to form a new set of rechargeable lithium battery current change curve graphs, recorded as standard current change curve graphs;

[0065] In specific implementation, using a smartphone lithium battery as an example, assume the sliding window size N is set to 10 charge cycles. With each charge, the data within the sliding window is automatically updated. By fitting multiple sets of historical current change curves within the window, a more representative standard current change curve can be obtained. This approach uses statistical methods to eliminate the occasional fluctuations of a single charge cycle and determine the current change pattern of a normal lithium battery charge.

[0066] Step S3: extracting the standard charging time required for the rechargeable lithium battery based on the standard current change curve analysis, and calculating the charging time decay deviation of the lithium battery in combination with the real-time current change curve analysis, which is recorded as the first time decay deviation;

[0067] Step S3-2: using a machine learning algorithm to predict the time required to complete charging of the rechargeable lithium battery based on the real-time current change curve, and recording the predicted value of the time required to complete charging as the charging time prediction value;

[0068] Step S3-3, calculating the difference between the standard charging time value and the predicted charging time value to obtain a charging time decay deviation of the rechargeable lithium battery, which is recorded as a first time decay deviation;

[0069] In specific implementation, taking electric bicycle lithium batteries as an example, the standard charging time can be obtained from the lithium battery's product manual or a large amount of previous test data. A neural network algorithm can be used as a machine learning algorithm, which can accurately predict the charging time based on the characteristics of the real-time current change curve. The principle of calculating the first time decay deviation is to compare the standard charging time with the predicted charging time to evaluate the difference between the lithium battery's current charging performance and the standard performance.

[0070] Step S4, obtaining charging data of a lithium battery having the same charge as that of a rechargeable lithium battery during initial charging, analyzing and extracting current data during the constant voltage charging stage to construct multiple sets of current change curves, recorded as transverse current change curves; and quantizing the transverse current change curves to obtain quantized current change curves;

[0071] Step S4-1: Obtaining the power data of the rechargeable lithium battery during initial charging through the intelligent battery management system, using the power data as a search index, and searching for charging data of lithium batteries with the same power and the same specifications and models during initial charging based on a time sliding window mechanism;

[0072] Step S4-2: Analyze and extract the current data during the constant voltage charging stage, and construct multiple sets of current change curves for the found lithium batteries according to the current change curve construction method of step S1, which are recorded as transverse current change curves;

[0073] Step S4-3: extracting the current change slope characteristics from each transverse current change curve graph, and obtaining the quantized maximum current change slope and minimum current change slope through normalization processing according to the maximum current change slope and the minimum current change slope;

[0074] Step S4-4, calculating the area under each transverse current change curve and the x-axis using the trapezoidal rule, and calculating the average of the areas under each transverse current change curve and the x-axis as the standard area of the transverse current change curve;

[0075] Step S4-5, calculating the average value of the current data corresponding to the abscissa of each transverse current change curve graph to construct an average transverse current change curve graph, using the maximum slope of the current change and the minimum slope of the current change after quantization and the standard area as restriction conditions, adjusting the average transverse current change curve graph to obtain a quantized current change curve graph, the specific process of the adjustment is: the minimum slope of the quantized current change curve graph satisfies not less than the minimum slope after quantization, the maximum slope of the quantized current change curve graph satisfies not greater than the maximum slope after quantization, and the area formed by the curve under the quantized current change curve graph and the x-axis is calculated to be equal to the standard area;

[0076] In practical implementation, taking the lithium-ion battery packs of energy storage power stations as an example, the intelligent battery management system can use database query functionality, using the initial charge level as an index, to quickly locate charging data for lithium-ion batteries of the same capacity and specifications. Quantifying the transverse current variation curve establishes a unified comparison standard across different lithium-ion batteries. By extracting slope characteristics and calculating the area under the curve, complex current variation curves can be converted into quantifiable indicators.

[0077] Step S5: Analyze and extract the quantized charging time required for the rechargeable lithium battery according to the quantized current change curve, and record it as the quantized charging time value; calculate the quantized charging time deviation of the rechargeable lithium battery according to the quantized charging time value, and record it as the second time decay deviation;

[0078] The maximum value of the horizontal axis of the quantized current change curve is read as the time required for the rechargeable lithium battery to complete charging after quantization, which is recorded as the quantized charging time value; the difference between the quantized charging time value and the predicted charging time value is calculated to obtain the quantized charging time deviation of the rechargeable lithium battery, which is recorded as the second time decay deviation;

[0079] In practice, the maximum value on the horizontal axis of the quantized current change curve is read as the quantized charging time value. Calculating the second time decay deviation can further assess changes in lithium battery charging performance. The principle is to compare the quantized charging time with the predicted charging time to identify charging time differences between lithium batteries in horizontal comparisons.

[0080] Step S6: Obtaining a comprehensive charging time deviation of the rechargeable lithium battery by weighted fusion based on the first time decay deviation and the second time decay deviation; constructing a comprehensive charging time deviation time sequence diagram based on multiple historical sets of comprehensive charging time deviations of the rechargeable lithium battery; and analyzing the comprehensive charging time deviation time sequence diagram to monitor the rechargeable lithium battery;

[0081] Step S6-1, obtaining a comprehensive charging time deviation of the rechargeable lithium battery by weighted fusion based on the first time decay deviation and the second time decay deviation, constructing a y-axis based on the historical multiple sets of comprehensive charging time deviations of the rechargeable lithium battery as the vertical coordinate, constructing an x-axis based on the number of times corresponding to the comprehensive charging time deviations of the rechargeable lithium battery as the horizontal coordinate, and constructing a comprehensive charging time deviation time sequence diagram based on the x-axis and y-axis;

[0082] Step S6-2: Obtain comprehensive charging time deviation timing diagrams of multiple lithium batteries according to the model and specifications of the rechargeable lithium batteries to construct a time deviation set, traverse and read the comprehensive charging time deviations in each comprehensive charging time deviation timing diagram in the time deviation set, and select the maximum value as the time deviation threshold;

[0083] Step S6-3: Use a machine learning algorithm to predict the comprehensive charging time deviation corresponding to the next horizontal axis of the comprehensive charging time deviation timing diagram of the rechargeable lithium battery, and monitor the rechargeable lithium battery according to the time deviation threshold:

[0084] When the predicted comprehensive charging time deviation exceeds the time deviation threshold, an early warning signal is issued;

[0085] When the predicted comprehensive charging time deviation does not exceed the time deviation threshold, continue to monitor the charging lithium battery;

[0086] In specific implementations, the weighted fusion of the first and second time decay deviations comprehensively considers the lithium battery's own historical charging performance changes and horizontal comparisons with similar lithium batteries. After constructing a comprehensive charging time deviation time series diagram, a machine learning algorithm is used to predict the comprehensive charging time deviation for the next charging cycle. This is then compared with the time deviation threshold to promptly detect abnormal charging conditions in the lithium battery.

[0087] Example 2, as Figure 2 As shown, the present invention provides an intelligent charging management system for lithium batteries, which includes a real-time curve generation module, a standard curve fitting module, a charging deviation calculation module, a horizontal curve quantification module, a quantification deviation calculation module and a battery monitoring and early warning module;

[0088] The real-time curve generation module is used to collect constant-voltage charging data of lithium batteries and construct a real-time current change curve graph; the standard curve fitting module is used to fit the historical current change curve graph through a time sliding window to obtain a standard curve; the charging deviation calculation module is used to extract the standard charging time from the standard curve and calculate the charging time attenuation deviation in combination with the real-time curve; the horizontal curve quantization module is used to obtain data of lithium batteries with the same power and construct a quantized current change curve graph; the quantization deviation calculation module is used to obtain the quantized charging time from the quantization curve and calculate the quantized charging time deviation; the battery monitoring and warning module is used to fuse the deviation data to construct a time sequence diagram and monitor and warn the charging status of the lithium battery;

[0089] The output end of the real-time curve generation module is electrically connected to the input end of the standard curve fitting module; the output end of the standard curve fitting module is electrically connected to the input end of the charging deviation calculation module; the output end of the charging deviation calculation module is electrically connected to the input end of the horizontal curve quantization module; the output end of the horizontal curve quantization module is electrically connected to the input end of the quantization deviation calculation module; the output end of the quantization deviation calculation module is electrically connected to the input end of the battery monitoring and early warning module;

[0090] The real-time curve generation module includes a current data acquisition unit and a real-time curve generation unit; the current data acquisition unit is used to obtain the current data and corresponding timestamp of the rechargeable lithium battery in the constant voltage charging stage; the real-time curve generation unit is used to construct a real-time current change curve graph based on the collected data;

[0091] The standard curve fitting module includes a window data updating unit and a standard curve calculation unit; the window data updating unit is used to update the real-time current change curve graph within the time sliding window according to the number of charging times; the standard curve calculation unit is used to calculate the standard current change curve graph based on the statistical calculation of the curve within the window;

[0092] The charging deviation calculation module includes a standard time reading unit and a deviation value calculation unit; the standard time reading unit is used to read the charging time standard value in the standard current change curve; the deviation value calculation unit is used to predict the real-time curve charging time through machine learning and calculate the first time decay deviation;

[0093] The transverse curve quantization module includes a data search and construction unit and a curve quantization adjustment unit; the data search and construction unit is used to search and construct a transverse current change curve diagram based on the power data; the curve quantization adjustment unit is used to extract curve features and obtain a quantized current change curve diagram through adjustment under limited conditions;

[0094] The quantization deviation calculation module includes a quantization time reading unit and a quantization deviation calculation unit; the quantization time reading unit is used to read the quantization value of the charging time in the quantization current change curve diagram; the quantization deviation calculation unit is used to calculate the difference between the quantization value of the charging time and the predicted value to obtain a second time decay deviation;

[0095] The battery monitoring and early warning module includes a timing diagram construction unit and a battery status monitoring unit; the timing diagram construction unit is used to construct a comprehensive charging time deviation timing diagram by weighted fusion deviation data; the battery status monitoring unit is used to set a threshold and monitor and warn the charging status of the lithium battery by predicting the deviation value;

[0096] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An intelligent charging management method for lithium batteries, characterized by: The intelligent charging management method comprises the following steps: Step S1, collecting current data of the rechargeable lithium battery in the constant voltage charging stage, and constructing a current change curve graph based on the timestamp data corresponding to the current data, which is recorded as a real-time current change curve graph; Step S2: Setting a time sliding window to obtain multiple sets of historical current change curves of the rechargeable lithium battery, and fitting the historical current change curves within the time sliding window to obtain a standard current change curve of the rechargeable lithium battery; Step S3: extracting the standard charging time required for the rechargeable lithium battery based on the standard current change curve analysis, and calculating the charging time decay deviation of the lithium battery in combination with the real-time current change curve analysis, which is recorded as the first time decay deviation; Step S4, obtaining charging data of a lithium battery having the same charge as that of a rechargeable lithium battery during initial charging, analyzing and extracting current data during the constant voltage charging stage to construct multiple sets of current change curves, recorded as transverse current change curves; and quantizing the transverse current change curves to obtain quantized current change curves; Step S5: Analyze and extract the quantized charging time required for the rechargeable lithium battery according to the quantized current change curve, and record it as the quantized charging time value; calculate the quantized charging time deviation of the rechargeable lithium battery according to the quantized charging time value, and record it as the second time decay deviation; Step S6: Obtain a comprehensive charging time deviation of the rechargeable lithium battery by weighted fusion based on the first time decay deviation and the second time decay deviation; construct a comprehensive charging time deviation timing diagram based on multiple groups of historical comprehensive charging time deviations of the rechargeable lithium battery; and monitor the rechargeable lithium battery by analyzing the comprehensive charging time deviation timing diagram.

2. The intelligent charging management method for lithium batteries according to claim 1, characterized in that: In step S1, current data of the rechargeable lithium battery in the constant voltage charging stage is collected by a current sensor; The timestamp data corresponding to the collected current data is used as the horizontal axis to construct the x-axis, and the current data is used as the vertical axis to construct the y-axis. The current change curve of the rechargeable lithium battery constructed based on the x-axis and y-axis is recorded as the real-time current change curve. After the rechargeable lithium battery completes one charge, a real-time current change curve is generated based on the current data collected during the charging process and the corresponding timestamp data.

3. The intelligent charging management method for lithium batteries according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: The size of the time sliding window is set to a fixed value N, and the number of times the rechargeable lithium battery is charged is used as the sliding step. When charging is completed, the real-time current change curve within the time sliding window is updated according to the sliding step; Step S2-2: Obtain the historical current change curve within the time sliding window, perform statistics based on the corresponding current data on the same horizontal axis, calculate the average current value at the corresponding horizontal axis point to form a new set of current change curves of the rechargeable lithium battery, which are recorded as standard current change curves.

4. The intelligent charging management method for lithium batteries according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1, reading the horizontal axis of the standard current change curve as the standard charging time required for the rechargeable lithium battery, and recording it as the standard charging time value; Step S3-2: using a machine learning algorithm to predict the time required to complete charging of the rechargeable lithium battery based on the real-time current change curve, and recording the predicted value of the time required to complete charging as the charging time prediction value; Step S3-3: Calculate the difference between the standard charging time value and the predicted charging time value to obtain a charging time decay deviation of the rechargeable lithium battery, which is recorded as a first time decay deviation.

5. The intelligent charging management method for lithium batteries according to claim 4, characterized in that: The specific steps of step S4 are as follows: Step S4-1: Obtaining the power data of the rechargeable lithium battery during initial charging through the intelligent battery management system, using the power data as a search index, and searching for charging data of lithium batteries with the same power and the same specifications and models during initial charging based on a time sliding window mechanism; Step S4-2: Analyze and extract the current data during the constant voltage charging stage, and construct multiple sets of current change curves for the found lithium batteries according to the current change curve construction method of step S1, which are recorded as transverse current change curves; Step S4-3: extracting the current change slope characteristics from each transverse current change curve graph, and obtaining the quantized maximum current change slope and minimum current change slope through normalization processing according to the maximum current change slope and the minimum current change slope; Step S4-4, calculating the area under each transverse current change curve and the x-axis using the trapezoidal rule, and calculating the average of the areas under each transverse current change curve and the x-axis as the standard area of the transverse current change curve; Step S4-5, calculate the average value of the current data corresponding to the horizontal coordinate of each horizontal current change curve to construct an average horizontal current change curve, use the maximum slope of the current change after quantization and the minimum slope of the current change and the standard area as constraints, adjust the average horizontal current change curve to obtain a quantized current change curve, and the specific process of the adjustment is: the minimum slope of the quantized current change curve must be no less than the minimum slope after quantization, the maximum slope of the quantized current change curve must be no greater than the maximum slope after quantization, and the area formed under the curve of the quantized current change curve and the x-axis is calculated to be equal to the standard area.

6. The intelligent charging management method for lithium batteries according to claim 5, characterized in that: In step S5, the maximum value of the horizontal axis of the quantized current change curve is read as the time required for the charging of the rechargeable lithium battery after quantization, which is recorded as the quantized charging time value; the difference between the quantized charging time value and the predicted charging time value is calculated to obtain the quantized charging time deviation of the rechargeable lithium battery, which is recorded as the second time attenuation deviation.

7. The intelligent charging management method for lithium batteries according to claim 6, characterized in that: The specific steps of step S6 are as follows: Step S6-1, obtaining a comprehensive charging time deviation of the rechargeable lithium battery by weighted fusion based on the first time decay deviation and the second time decay deviation, constructing a y-axis based on the historical multiple sets of comprehensive charging time deviations of the rechargeable lithium battery as the vertical coordinate, constructing an x-axis based on the number of times corresponding to the comprehensive charging time deviations of the rechargeable lithium battery as the horizontal coordinate, and constructing a comprehensive charging time deviation time sequence diagram based on the x-axis and y-axis; Step S6-2: Obtain comprehensive charging time deviation timing diagrams of multiple lithium batteries according to the model and specifications of the rechargeable lithium batteries to construct a time deviation set, traverse and read the comprehensive charging time deviations in each comprehensive charging time deviation timing diagram in the time deviation set, and select the maximum value as the time deviation threshold; Step S6-3: Use a machine learning algorithm to predict the comprehensive charging time deviation corresponding to the next horizontal axis of the comprehensive charging time deviation timing diagram of the rechargeable lithium battery, and monitor the rechargeable lithium battery according to the time deviation threshold: When the predicted comprehensive charging time deviation exceeds the time deviation threshold, an early warning signal is issued; When the predicted comprehensive charging time deviation does not exceed the time deviation threshold, the charging lithium battery continues to be monitored.

8. An intelligent charging management system for lithium batteries, which is applied to the intelligent charging management method for lithium batteries according to any one of claims 1 to 7, characterized in that: The intelligent charging management system includes a real-time curve generation module, a standard curve fitting module, a charging deviation calculation module, a horizontal curve quantification module, a quantification deviation calculation module and a battery monitoring and early warning module; The real-time curve generation module is used to collect constant-voltage charging data of lithium batteries and construct a real-time current change curve graph; the standard curve fitting module is used to fit the historical current change curve graph through a time sliding window to obtain a standard curve; the charging deviation calculation module is used to extract the standard charging time from the standard curve and calculate the charging time attenuation deviation in combination with the real-time curve; the horizontal curve quantization module is used to obtain data of lithium batteries with the same power and construct a quantized current change curve graph; the quantization deviation calculation module is used to obtain the quantized charging time from the quantization curve and calculate the quantized charging time deviation; the battery monitoring and warning module is used to fuse the deviation data to construct a time sequence diagram and monitor and warn the charging status of the lithium battery; The output end of the real-time curve generation module is electrically connected to the input end of the standard curve fitting module; the output end of the standard curve fitting module is electrically connected to the input end of the charging deviation calculation module; the output end of the charging deviation calculation module is electrically connected to the input end of the horizontal curve quantization module; the output end of the horizontal curve quantization module is electrically connected to the input end of the quantization deviation calculation module; the output end of the quantization deviation calculation module is electrically connected to the input end of the battery monitoring and early warning module.

9. The intelligent charging management system for lithium batteries according to claim 8, characterized in that: The real-time curve generation module includes a current data acquisition unit and a real-time curve generation unit; the current data acquisition unit is used to obtain the current data and corresponding timestamp of the rechargeable lithium battery in the constant voltage charging stage; the real-time curve generation unit is used to construct a real-time current change curve graph based on the collected data; The standard curve fitting module includes a window data updating unit and a standard curve calculation unit; the window data updating unit is used to update the real-time current change curve graph within the time sliding window according to the number of charging times; the standard curve calculation unit is used to calculate the standard current change curve graph based on the statistical calculation of the curve within the window; The charging deviation calculation module includes a standard time reading unit and a deviation value calculation unit; the standard time reading unit is used to read the charging time standard value in the standard current change curve diagram; the deviation value calculation unit is used to predict the real-time curve charging time through machine learning and calculate the first time attenuation deviation.

10. The intelligent charging management system for lithium batteries according to claim 8, characterized in that: The transverse curve quantization module includes a data search and construction unit and a curve quantization adjustment unit; the data search and construction unit is used to search and construct a transverse current change curve diagram based on the power data; the curve quantization adjustment unit is used to extract curve features and obtain a quantized current change curve diagram through adjustment under limited conditions; The quantization deviation calculation module includes a quantization time reading unit and a quantization deviation calculation unit; the quantization time reading unit is used to read the charging time quantization value in the quantization current change curve diagram; The quantization deviation calculation unit is used to calculate the difference between the quantized value of the charging time and the predicted value to obtain a second time decay deviation; The battery monitoring and early warning module includes a timing diagram construction unit and a battery status monitoring unit; the timing diagram construction unit is used to construct a comprehensive charging time deviation timing diagram by weighted fusion deviation data; the battery status monitoring unit is used to set a threshold and monitor and warn the charging status of the lithium battery by predicting the deviation value.

Images