Lithium battery low temperature charging system and charging method

Abstract

The present invention belongs to the technical field of battery charging management, and discloses a lithium battery low-temperature charging system and a charging method thereof; the method comprises: collecting battery characteristic data and battery environment data of a lithium battery in real time; processing the battery characteristic data to obtain battery processing data; analyzing the battery characteristic data and the battery processing data to determine whether to generate an early warning instruction, wherein the early warning instruction comprises an abnormal instruction, a semi-abnormal instruction and a temperature abnormal instruction; if a semi-abnormal instruction is generated, charging is interrupted, and the lithium battery is preheated, and after the preheating is completed, the battery characteristic data is recollected and the battery processing data is obtained, and it is determined again whether an abnormal instruction is generated; the present invention realizes full intelligent management of the low-temperature charging process of the lithium battery, thereby effectively preventing safety hazards from occurring during the low-temperature charging process, improving the charging efficiency and service life of the lithium battery in a low-temperature environment, and realizing more efficient and safe low-temperature charging.

Description

Technical Field

[0001] The present invention relates to the technical field of battery charging management, and more particularly to a lithium battery low-temperature charging system and a charging method thereof. Background Art

[0002] With the continuous development of power electronics technology, lithium batteries are increasingly used, especially in applications that require long-term operation, such as electric vehicles and drones. However, when charging lithium batteries in low-temperature environments, the following problems may arise:

[0003] Dendrite growth: At low temperatures, lithium ions are unevenly deposited on the electrode surface, which easily forms lithium dendrites and causes battery short circuits. Internal resistance increases: At low temperatures, the viscosity of the electrolyte increases, and the migration rate of lithium ions decreases, resulting in increased internal resistance and decreased charging efficiency. Capacity decay: Repeated low-temperature charge and discharge cycles will lead to irreversible loss of active materials in the battery, accelerating battery capacity decay. Safety risks: Uneven deposition of lithium ions and increased internal resistance at low temperatures will cause the battery to heat up, posing a safety hazard. Excessive charging time: The charging rate is significantly reduced at low temperatures, resulting in a significantly extended charging time and reduced practicality. These phenomena will cause battery performance to deteriorate, and in severe cases, may even lead to battery failure or even safety accidents. Therefore, how to effectively charge lithium batteries in low-temperature environments has become an important research direction in the field of battery technology today.

[0004] Chinese patent publication number CN108075542A discloses a low-temperature charging system and method for lithium batteries. The system includes a lithium battery pack, a heating and insulation sleeve, a charging power supply, a single-chip microcomputer, and a temperature sensor. The lithium battery pack and the heating and insulation sleeve are both connected to the charging power supply, which is connected to the single-chip microcomputer, which is connected to the temperature sensor. Under low-temperature conditions, the charging power supply first heats the lithium battery in the heating and insulation sleeve. The heating and insulation sleeve has self-control temperature and heat insulation functions. The temperature sensor only detects the battery temperature. When the temperature of the lithium battery reaches the normal charging temperature, the charging power supply then charges the lithium battery. This ensures normal charging of the lithium battery under low-temperature conditions, ensuring normal charging of the lithium battery and solving the problem of difficulty in charging the lithium battery under low-temperature conditions.

[0005] The above technology determines whether to heat and charge the lithium battery based on the battery temperature. Although the battery temperature has been adjusted, if the battery is still charged in a low-temperature environment, the above-described problems may still occur. Therefore, it is also necessary to adaptively adjust the relevant parameters of the lithium battery to avoid the above-mentioned situation and protect the charging efficiency and battery life of the lithium battery. At the same time, in order to avoid safety hazards such as battery explosion and fire, it is also necessary to monitor the charging status of the lithium battery and promptly detect abnormal conditions during the charging process.

[0006] In view of this, the present invention proposes a lithium battery low-temperature charging system and a charging method thereof to solve the above-mentioned problems. Summary of the Invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a low-temperature charging method for a lithium battery, comprising:

[0008] Real-time collection of lithium battery characteristic data and battery environment data;

[0009] Processing battery characteristic data to obtain battery processing data;

[0010] Analyze battery characteristic data and battery processing data to determine whether to generate warning instructions, including abnormal instructions, semi-abnormal instructions, and temperature abnormal instructions;

[0011] If a semi-abnormal instruction is generated, charging is interrupted and the lithium battery is preheated. After preheating is completed, the battery characteristic data and battery processing data are collected again to determine whether an abnormal instruction is generated. If no abnormal instruction is generated, charging is resumed, and the battery characteristic data, battery environment data, and battery processing data are analyzed to generate a set of adjustment quantities, and the battery characteristic data is adjusted according to the set of adjustment quantities.

[0012] If a temperature abnormality instruction is generated, an adjustment amount set is also generated, the battery characteristic data is adjusted according to the adjustment amount set, and the temperature adjustment power is calculated. The lithium battery heating component operates according to the temperature adjustment power.

[0013] Furthermore, the battery characteristic data includes charging current, charging voltage, battery temperature, battery internal resistance, battery capacity, current battery charge and minimum battery charge; the battery environment data includes ambient temperature, ambient humidity and ambient air pressure;

[0014] The method for obtaining battery processing data includes:

[0015] Battery processing data includes power percentage, capacity change rate, and internal resistance change rate;

[0016] The expression for the battery percentage is: In the formula, SOC is the percentage of power, S is the current power of the battery, S min is the minimum battery capacity, S max is the battery capacity;

[0017] The expression of capacity change rate is: Where, is the capacity change rate, S′ max The historical battery capacity is the battery capacity collected last time. The battery characteristic data is collected according to the preset collection frequency.

[0018] The expression of internal resistance change rate is: Where, is the internal resistance change rate, R is the battery internal resistance, and R′ is the historical battery internal resistance; the historical battery internal resistance is the battery internal resistance collected last time.

[0019] Furthermore, the method for determining whether to generate a warning instruction includes:

[0020] The capacity change rate and internal resistance change rate in the battery processing data, as well as the battery temperature in the battery characteristic data, are used as analysis data. The analysis data is input into the trained state analysis model, and the corresponding judgment results are obtained based on the output judgment labels. The judgment results include normal, abnormal temperature, abnormal capacity or internal resistance, abnormal temperature and internal resistance, abnormal temperature and capacity, and abnormal.

[0021] If the judgment result is all normal, no warning instruction will be generated;

[0022] If the judgment result is that the temperature is abnormal, a temperature abnormality instruction is generated;

[0023] If the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated;

[0024] If the judgment result is not completely normal, the temperature is abnormal, and the capacity or internal resistance is abnormal, a semi-abnormal instruction is generated;

[0025] The training process of the state analysis model includes:

[0026] Pre-set the corresponding judgment results for group a of analysis data, where a is an integer greater than 1, and set different digital labels for different judgment results;

[0027] The digital labels of the judgment results are marked as judgment labels, and the analysis data and the corresponding judgment labels are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of the state analysis model, and the state analysis model uses a set of predicted judgment labels corresponding to each set of analysis data as output, and the actual judgment labels corresponding to each set of analysis data as prediction targets. The actual judgment labels are the pre-collected digital labels of the judgment results corresponding to the analysis data; minimizing the sum of the prediction errors of all analysis data is used as the training goal; the state analysis model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the state analysis model is a deep neural network model.

[0028] Furthermore, the method of determining again whether an abnormal instruction is generated includes:

[0029] A normal temperature range is preset; the lithium battery is preheated through the built-in heating element of the lithium battery or an external heating device, and the battery characteristic data is continuously collected according to the collection frequency; when the battery temperature reaches the normal temperature range, the preheating ends, and the battery processing data is acquired again, and the corresponding judgment result is obtained again through the state analysis model; if the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated again; if the judgment result is all normal, no abnormal instruction is generated.

[0030] Furthermore, the adjustment amount set includes a current adjustment amount and a voltage adjustment amount; the charging current and charging voltage in the battery characteristic data are adjusted according to the adjustment amount set; and the step of generating the adjustment amount set includes:

[0031] Step 1: Calculate the data adjustment range;

[0032] Step 2: Randomly generate an initial antibody population X(0) of size q = {X1, X2, X3, ... X q}; preset iteration threshold; where X q is the qth antibody;

[0033] Step 3: Determine the fitness function;

[0034] Step 4: Calculate the fitness of each antibody in the initial antibody population X(0), select r antibodies with the largest fitness, and construct the memory population X m (t);

[0035] Step 5: Memory population X m Each antibody in (t) is cloned according to the calculated clone size to generate a clone antibody population Among them, is the qth cloned antibody in the clonal antibody population;

[0036] Step 6: Cloning Antibody Group X c Each antibody in (t) is subjected to cloud adaptive mutation using the cloud adaptive mutation operator to generate a mutant antibody group X d (t);

[0037] Step 7: Cloning Antibodies from Group X c In (t), w antibodies are randomly selected as parent antibodies for antibody recombination to generate progeny antibodies S c , until the cloned antibody group X c When the number of unselected antibodies in (t) is less than w, antibody recombination is stopped and a recombinant antibody group X is generated. r (t);

[0038] Step 8: Merge variant antibody group X d (t) and recombinant antibody group X r(t), generate a combined antibody population X L (t);

[0039] Step 9: Randomly generate n new antibodies and merge the antibody group X L The n antibodies with the smallest fitness in (t) are replaced by n randomly generated new antibodies;

[0040] Step 10: If the number of iterations t is less than the iteration threshold, set t = t + 1 and select the merged antibody group X L The r antibodies with the largest fitness in (t) form a new memory population X m (t+1), loop steps 4 to 9; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group X is obtained L The set of adjustment amounts corresponding to the antibody with the largest fitness in (t).

[0041] Furthermore, in step 1, the method for calculating the data adjustment amount range includes:

[0042] Obtain the maximum charging current, minimum charging current, maximum charging voltage, and minimum charging voltage of the lithium battery; subtract the real-time collected charging current from the maximum charging current and minimum charging current to obtain the current adjustment range; subtract the real-time collected charging voltage from the maximum charging voltage and minimum charging voltage to obtain the voltage adjustment range;

[0043] In step 2, the antibodies in the initial antibody group X(0) correspond one-to-one to the adjustment amount sets in the data set, and the range of the antibodies is the data set; wherein each adjustment amount set includes a value in the current adjustment range and a value in the voltage adjustment range, that is, the adjustment amount set is constructed by randomly selecting a value from each of the current adjustment range and the voltage adjustment range; the data set includes all adjustment amount sets, that is, m adjustment amount sets, m=m1×m2, m1 is the number of values ​​in the current adjustment range, and m2 is the number of values ​​in the voltage adjustment range;

[0044] In step 5, the calculation method of the clone size includes:

[0045]

[0046] Φ i =min(exp(||X i -X j ||));

[0047] Where R i is the clone size of the i-th antibody, Int is the rounding function, q is the initial antibody group size, Φ iis the affinity of the ith antibody, min is the minimum function, exp is the natural exponential function, ||X i -X j || is the Euclidean distance between the i-th antibody and the j-th antibody, i≠j, i∈[1,q], j∈[1,q].

[0048] Furthermore, in step 3, the fitness function is expressed as:

[0049] f i =ψ1×XL i +ψ2×AP i ;

[0050] Where, f i is the fitness of the i-th antibody, XL i is the charging efficiency of the ith antibody, AP i is the safety score of the i-th antibody, ψ1 and ψ2 are preset weight factors;

[0051] The methods for obtaining charging efficiency and safety scores include:

[0052] The current regulation amount corresponding to the antibody is added to the charging current collected in real time to obtain the regulation current; the voltage regulation amount corresponding to the antibody is added to the charging voltage collected in real time to obtain the regulation voltage; the battery characteristic data, battery environment data, battery temperature, regulation current and regulation voltage are used as test data; the test data is input into the trained efficiency prediction model to predict the corresponding charging efficiency; the test data is input into the trained score prediction model to predict the corresponding safety score; wherein, the training process of the efficiency prediction model and the score analysis model are consistent with the training process of the state analysis model, and both are deep neural network models, each set of feature vectors input in the training process of the efficiency prediction model is the test data and the charging efficiency corresponding to the test data, and each set of feature vectors input in the training process of the score prediction model is the test data and the safety score corresponding to the test data;

[0053] In step 7, the progeny antibody S c The calculation methods include:

[0054]

[0055] Where X′ w is the w-th parent antibody, s w is the wth scaling factor, which is a randomly generated real number and not all w scaling factors are 0.

[0056] Furthermore, the step of generating the set of adjustment variables when generating the temperature abnormality instruction is consistent with the step of generating the set of adjustment variables when only the semi-abnormality instruction is generated;

[0057] Methods for calculating thermostat power include:

[0058] Subtract the battery temperature in the battery characteristic data from the minimum value of the normal temperature range to obtain the temperature difference; obtain the battery heat capacity; obtain the battery heat by multiplying the battery heat capacity and the temperature difference; the expression for battery heat is: Q = C × T; where Q is the battery heat, C is the battery heat capacity, and T is the temperature difference; preset the heating time, and obtain the temperature control power by the ratio of the battery heat and the heating time; the expression for the temperature control power is Where P is the temperature control power and t' is the heating time.

[0059] Furthermore, the method further comprises: regulating the heating time of the lithium battery heating component;

[0060] The method for regulating the heating time of the lithium battery heating component includes:

[0061] The preset heating time is divided into N time periods according to the acquisition frequency, where N is an integer greater than 1. The end time point of each time period is marked as the end point, and the start time point of each time period is marked as the start point. The end point of the previous time period is the start point of the next time period. The temperature control power is multiplied by the sampling frequency to obtain the change heat, and the change heat is divided by the battery heat capacity to obtain the change temperature. The change temperature and the ambient temperature are used as evaluation data, and the evaluation data are input into the change prediction model to predict the actual change temperature. The actual change temperature is the actual change in battery temperature under the influence of ambient temperature. The training process of the change prediction model is consistent with the training process of the efficiency prediction model, and both are deep neural network models. According to the time sequence, the actual change temperature is added to the battery temperature of each starting point to obtain the battery temperature of each end point.

[0062] Obtain the number of end points between each end point and the first start point respectively, add one to the number of end points corresponding to each end point and multiply it by the acquisition frequency to obtain the charging time corresponding to each end point;

[0063] The ambient temperature, battery internal resistance, battery capacity, and the battery temperature and charging time of an end point are taken as a group of research data, and the research data correspond to the end points one by one; each group of research data is input into the trained internal resistance prediction model to predict the battery internal resistance corresponding to each end point; each group of research data is input into the trained capacity prediction model to predict the battery capacity corresponding to each end point; based on the battery internal resistance in the battery characteristic data and the battery internal resistance corresponding to each end point, the internal resistance change rate corresponding to each end point is calculated; based on the battery capacity in the battery characteristic data and the battery capacity corresponding to each end point, the capacity change rate corresponding to each end point is calculated; based on the internal resistance change rate and capacity change rate corresponding to each end point, whether to generate an abnormal instruction is determined;

[0064] If an abnormal instruction is generated, the end point closest to the first starting point among the end points corresponding to the abnormal instruction is marked as an adjustment point, and the charging time corresponding to the adjustment point is subtracted from the acquisition frequency to obtain the optimal charging time; the heating time of the lithium battery heating component is adjusted to the optimal charging time;

[0065] If no abnormal instruction is generated, the heating time of the lithium battery heating component is not regulated.

[0066] A lithium battery low-temperature charging system implements the lithium battery low-temperature charging method, comprising:

[0067] Data acquisition module, used to collect battery characteristic data and battery environment data of lithium batteries in real time;

[0068] A data processing module is used to process battery characteristic data and obtain battery processing data;

[0069] A data analysis module is used to analyze battery characteristic data and battery processing data to determine whether to generate a warning instruction. The warning instructions include abnormal instructions, semi-abnormal instructions, and temperature abnormal instructions.

[0070] The secondary analysis module is used to interrupt charging and preheat the lithium battery if a semi-abnormal instruction is generated. After the preheating is completed, the battery characteristic data is recollected and the battery processing data is obtained to determine again whether an abnormal instruction is generated;

[0071] a first adjustment module, configured to resume charging if no abnormal instruction is generated, analyze battery characteristic data, battery environment data, and battery processing data, generate an adjustment value set, and adjust the battery characteristic data according to the adjustment value set;

[0072] The second adjustment module is used to generate an adjustment amount set if a temperature abnormality instruction is generated, adjust the battery characteristic data according to the adjustment amount set, and calculate the temperature adjustment power. The lithium battery heating component operates according to the temperature adjustment power.

[0073] The technical effects and advantages of the lithium battery low-temperature charging system and charging method of the present invention are as follows:

[0074] 1. By collecting and processing battery characteristic data in real time, using deep learning technology to intelligently identify abnormal conditions during the lithium battery charging process, taking appropriate charging adjustment or preheating measures based on the abnormal conditions, and then optimizing charging parameters based on the clonal selection algorithm, the entire low-temperature charging process of the lithium battery can be intelligently managed. This effectively prevents safety hazards during low-temperature charging, improves the charging efficiency and service life of the lithium battery in low-temperature environments, and further improves charging safety and reliability, achieving more efficient and safe low-temperature charging.

[0075] 2. Realize real-time intelligent control of lithium battery heating time; subdivide the preset heating time into multiple time periods, and use deep learning technology to obtain a more accurate battery temperature in each time period in combination with the ambient temperature. The corresponding charging time, internal resistance change rate and capacity change rate are calculated to determine whether there are any abnormalities in the lithium battery charging process; if an abnormality occurs, find the most recent abnormal time point and adjust the heating time of the lithium battery heating component in real time according to the corresponding charging time; further realize the full intelligent management of the lithium battery low-temperature charging process, avoid the lithium battery charging in a low-temperature state for a long time, effectively solve the shortcomings of fixed time settings, better ensure the safe charging of lithium batteries, and thus improve the charging efficiency and battery life of lithium batteries. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 This is a schematic diagram of a lithium battery low-temperature charging system according to Example 1 of the present invention;

[0077] Figure 2 This is a schematic diagram of a lithium battery low-temperature charging system according to Example 2 of the present invention;

[0078] Figure 3 This is a flow chart of a low-temperature charging method for a lithium battery according to Example 3 of the present invention;

[0079] Figure 4 This is a schematic diagram of an electronic device according to embodiment 4 of the present invention;

[0080] Figure 5 This is a schematic diagram of the storage medium of Example 5 of the present invention. DETAILED DESCRIPTION

[0081] 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.

[0082] Example 1

[0083] See also Figure 1 As shown, the lithium battery low-temperature charging system described in this embodiment includes a data acquisition module, a data processing module, a data analysis module, a secondary analysis module, a first adjustment module, and a second adjustment module; each module is connected by wired and / or wireless means to achieve data transmission between modules;

[0084] The data acquisition module is used to collect battery characteristic data and battery environment data of the lithium battery in real time.

[0085] Battery characteristic data includes charging current, charging voltage, battery temperature, battery internal resistance, battery capacity, battery current charge, and battery minimum charge;

[0086] Battery temperature, battery internal resistance, battery capacity, battery current charge and battery minimum charge can all be used to judge various abnormal conditions when charging a lithium battery in a low-temperature environment. For example, battery internal resistance is used to reflect the degree of battery aging and internal status; battery capacity is used to reflect the degree of battery degradation and usage status, etc. According to battery temperature, battery internal resistance, battery capacity, battery current charge and battery minimum charge, the charging current and charging voltage can be adjusted in time to avoid various problems that may be encountered when charging a lithium battery in a low-temperature environment.

[0087] Battery environmental data includes ambient temperature, ambient humidity and ambient air pressure;

[0088] The lower the ambient temperature, the slower the chemical reaction rate inside the lithium battery, and the lower the ion transfer efficiency. The charging current needs to be reduced to avoid uneven charging caused by the lower temperature. At the same time, the charging voltage needs to be appropriately increased to overcome the increase in battery internal resistance caused by the lower temperature and ensure charging efficiency. The opposite is true.

[0089] The higher the ambient humidity, the faster the corrosion rate of the internal components of the lithium battery, resulting in reduced insulation performance. The charging current needs to be reduced to avoid the risk of leakage and short circuit in high humidity environments. At the same time, the charging voltage needs to be increased to ensure that charging can still be completed effectively under high humidity conditions, and vice versa.

[0090] The lower the ambient air pressure, the lower the ionic conductivity of the electrolyte in the lithium battery, affecting the battery's charge and discharge efficiency. The charging current needs to be reduced to avoid uneven charging caused by the decrease in ionic conductivity. At the same time, the charging voltage needs to be increased to overcome the increase in battery internal resistance caused by low air pressure and ensure charging quality. The opposite is true.

[0091] The data processing module is used to process the battery characteristic data and obtain battery processing data.

[0092] Methods for obtaining battery processing data include:

[0093] Battery processing data includes power percentage, capacity change rate, and internal resistance change rate.

[0094] The expression for the battery percentage is: In the formula, SOC is the percentage of power, S is the current power of the battery, S min is the minimum battery capacity, S max is the battery capacity.

[0095] The expression of capacity change rate is: Where, is the capacity change rate, S′ max The historical battery capacity is the battery capacity collected last time. The battery characteristic data is collected according to a preset collection frequency. The collection frequency is preset by technicians in this field according to the accuracy of lithium battery charging management.

[0096] The expression of internal resistance change rate is: Where, is the internal resistance change rate, R is the battery internal resistance, and R′ is the historical battery internal resistance; the historical battery internal resistance is the battery internal resistance collected last time.

[0097] The data analysis module is used to analyze the battery characteristic data and battery processing data to determine whether to generate an early warning instruction. The early warning instruction includes an abnormal instruction, a semi-abnormal instruction, and a temperature abnormal instruction.

[0098] Methods for determining whether to generate an early warning instruction include:

[0099] The capacity change rate and internal resistance change rate in the battery processing data and the battery temperature in the battery characteristic data are used as analysis data. The analysis data are input into the trained state analysis model, and the corresponding judgment result is obtained according to the output judgment label.

[0100] The specific training process of the state analysis model includes:

[0101] A corresponding judgment result is set in advance for a group of analysis data, and the judgment results include normal, abnormal temperature, abnormal capacity or internal resistance, abnormal temperature and internal resistance, abnormal temperature and capacity, and abnormal. a is an integer greater than 1, and different digital labels are set for different judgment results. For example, the digital label is set to 1 for normal temperature, the digital label is set to 2 for abnormal temperature, and the digital label is set to 3 for abnormal capacity or internal resistance, etc.; the judgment results corresponding to the analysis data are collected by a person skilled in the art during the historical low-temperature charging process of lithium batteries. The person skilled in the art judges in turn whether the data in the analysis data is abnormal under the conditions of each group of analysis data based on actual experience and the product specifications or technical parameters of the lithium battery, and sets corresponding judgment results for the different analysis data in group a in turn;

[0102] The digital labels of the judgment results are marked as judgment labels, and the analysis data and the corresponding judgment labels are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of the state analysis model, and the state analysis model uses a set of predicted judgment labels corresponding to each set of analysis data as output, and the actual judgment labels corresponding to each set of analysis data as the prediction target. The actual judgment labels are the digital labels of the judgment results corresponding to the analysis data collected in advance; the training goal is to minimize the sum of the prediction errors of all analysis data; wherein, the calculation formula of the prediction error is Z K =(α K -μ K ) 2 , where Z K is the prediction error, K is the group number of the eigenvector corresponding to the analysis data, α K is the predicted judgment label corresponding to the Kth group of analysis data, μ K is the actual judgment label corresponding to the Kth group of analysis data; the state analysis model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the state analysis model is specifically a deep neural network model.

[0103] If the judgment result is all normal, no warning instruction is generated; this means that there is no abnormality during the lithium battery charging process;

[0104] If the judgment result is that the temperature is abnormal, a temperature abnormality instruction is generated, indicating that the battery temperature of the lithium battery is too low and needs to be heated;

[0105] If the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated; this indicates that there is an abnormal change in the capacity or internal resistance during the lithium battery charging process, but it is not caused by the battery temperature being too low. The lithium battery needs to be repaired in time to avoid safety hazards such as battery explosion and fire;

[0106] If the judgment result is not completely normal, the temperature is abnormal, and the capacity or internal resistance is abnormal, a semi-abnormal instruction is generated; this indicates that the abnormal change in capacity or internal resistance during lithium battery charging may be caused by the battery temperature being too low, and further analysis and determination are required.

[0107] The secondary analysis module is used to interrupt charging and preheat the lithium battery if a semi-abnormal instruction is generated. After the preheating is completed, the battery characteristic data is re-collected and the battery processing data is obtained to determine again whether an abnormal instruction is generated.

[0108] The method for determining again whether an abnormal instruction is generated includes:

[0109] A normal temperature range is preset, and the normal temperature range is pre-set by technical personnel in this field based on actual experience and lithium battery technical specifications; the lithium battery is preheated by a lithium battery heating component, and battery characteristic data is continuously collected according to the collection frequency. The lithium battery heating component is a built-in heating element (such as an integrated battery heating device, a PTC heater, etc.) or an external heating device (an electric heating film, a hot air blower, etc.); when the battery temperature reaches the normal temperature range, the preheating ends, and the battery processing data is acquired again, and the corresponding judgment result is acquired again through the state analysis model; if the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated again, indicating that there is still abnormal capacity or internal resistance change during the charging process of the lithium battery; if the judgment result is all normal, no abnormal instruction is generated, indicating that the abnormal change in capacity or internal resistance during the charging process of the lithium battery is caused by the battery temperature being too low.

[0110] It should be noted that, since the battery temperature has reached the normal temperature range after preheating, there will be no temperature abnormality in the judgment result, that is, no temperature abnormality instruction or semi-abnormal instruction will be generated.

[0111] If an abnormal instruction is generated, an audible and visual alarm will be used to remind relevant staff to conduct timely maintenance.

[0112] The first adjustment module is used to resume charging if no abnormal instruction is generated, analyze battery characteristic data, battery environment data and battery processing data, generate an adjustment amount set, and adjust the battery characteristic data according to the adjustment amount set.

[0113] The regulation amount set includes a current regulation amount and a voltage regulation amount; the charging current and charging voltage in the battery characteristic data are regulated according to the regulation amount set; and the steps of generating the regulation amount set include:

[0114] Step 1: Calculate the data adjustment range.

[0115] Step 2: Randomly generate an initial antibody population X(0) of size q = {X1, X2, X3, ... X q}; Preset iteration threshold, which is set in advance according to the algorithm accuracy; where X q is the qth antibody.

[0116] Step 3: Determine the fitness function.

[0117] Step 4: Calculate the fitness of each antibody in the initial antibody population X(0), select r antibodies with the largest fitness, and construct the memory population X m (t).

[0118] Step 5: Memory population X m Each antibody in (t) is cloned according to the calculated clone size to generate a clone antibody population Among them, It is the qth cloned antibody in the clonal antibody population.

[0119] Step 6: Cloning Antibody Group X c Each antibody in (t) is subjected to cloud adaptive mutation using the cloud adaptive mutation operator to generate a mutant antibody group X d (t); The cloud adaptive mutation operator is an existing technology and will not be described in detail here.

[0120] Step 7: Cloning Antibodies from Group X c In (t), w antibodies are randomly selected as parent antibodies for antibody recombination to generate progeny antibodies S c , until the cloned antibody group X c When the number of unselected antibodies in (t) is less than w, antibody recombination is stopped and a recombinant antibody group X is generated. r (t); In this embodiment, w is preferably 3; recombinant antibody group X r (t) includes multiple progeny antibodies S c .

[0121] Step 8: Merge variant antibody group X d (t) and recombinant antibody group X r (t), generate a combined antibody population X L (t).

[0122] Step 9: Randomly generate n new antibodies and merge the antibody group X L The n antibodies with the smallest fitness in (t) are replaced by n randomly generated new antibodies.

[0123] Step 10: If the number of iterations t is less than the iteration threshold, set t = t + 1 and select the merged antibody group X L The r antibodies with the largest fitness in (t) form a new memory population X m (t+1), loop steps 4 to 9; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group X is obtained L The set of adjustment values ​​corresponding to the antibody with the largest fitness in (t); wherein, t=t+1 is a computer language, that is, the value of the iteration number t is added by 1 and then assigned to the iteration number t.

[0124] In step 1 above, the method for calculating the data adjustment amount range includes:

[0125] Obtain the maximum charging current, minimum charging current, maximum charging voltage and minimum charging voltage of the lithium battery; the maximum charging current, minimum charging current, maximum charging voltage and minimum charging voltage of the lithium battery are all obtained by technical personnel in this field from the product specifications or technical parameters of the lithium battery; subtract the real-time collected charging current from the maximum charging current and the minimum charging current to obtain the current adjustment range; subtract the real-time collected charging voltage from the maximum charging voltage and the minimum charging voltage to obtain the voltage adjustment range.

[0126] In the above step 2, the antibodies in the initial antibody group X(0) correspond one-to-one to the adjustment amount sets in the data set, and the range of the antibodies is the data set; wherein, each adjustment amount set includes a value in the current adjustment range and a value in the voltage adjustment range, that is, the adjustment amount set is constructed by randomly selecting a value from each of the current adjustment range and the voltage adjustment range; the data set includes all adjustment amount sets, that is, m adjustment amount sets, m=m1×m2, m1 is the number of values ​​in the current adjustment range, and m2 is the number of values ​​in the voltage adjustment range.

[0127] In step 3 above, the fitness function is expressed as: f i =ψ1×XL i +ψ2×AP i ;

[0128] Where, f i is the fitness of the i-th antibody, XL i is the charging efficiency of the ith antibody, AP i is the safety score of the i-th antibody, and ψ1 and ψ2 are preset weight factors.

[0129] It should be noted that the specific value of the weight factor in the formula can be set according to actual conditions. The weight factor reflects the importance of charging efficiency and safety score. Those skilled in the art can preset the corresponding weight coefficient according to the actual importance of charging efficiency and safety score to accurately evaluate the charging performance of the lithium battery after the charging current and charging voltage are adjusted during charging.

[0130] The methods for obtaining charging efficiency and safety scores include:

[0131] The current regulation amount corresponding to the antibody is added to the charging current collected in real time to obtain the regulated current; the voltage regulation amount corresponding to the antibody is added to the charging voltage collected in real time to obtain the regulated voltage; the battery characteristic data, battery environmental data, battery temperature, regulated current and regulated voltage are used as test data; the test data is input into the trained efficiency prediction model to predict the corresponding charging efficiency; the test data is input into the trained score prediction model to predict the corresponding safety score; among them, the specific training process of the efficiency prediction model and the score analysis model are consistent with the training process of the state analysis model, and both are deep neural network models. Each set of feature vectors input during the training process of the efficiency prediction model is the test data and the charging efficiency corresponding to the test data. Each set of feature vectors input during the training process of the score prediction model is the test data and the safety score corresponding to the test data.

[0132] It should be noted that the charging efficiency and safety score corresponding to the test data are collected by technical personnel in this field during the historical lithium battery charging process. B groups of different test data are collected, and charging tests are performed on the lithium battery under the conditions of each group of test data. The corresponding charging efficiency and safety score are collected based on actual experience, and the corresponding charging efficiency and safety score are set for the different test data in group B in turn.

[0133] It should be understood that the real-time collected charging current and charging voltage and the battery temperature in the test data are all obtained from the battery characteristic data collected after the lithium battery is preheated, and the battery characteristic data for adjustment is also the battery characteristic data collected after the lithium battery is preheated.

[0134] In step 5 above, the calculation method of the clone size includes:

[0135]

[0136] Φ i =min(exp(||X i -X j ||));

[0137] Where R i is the clone size of the i-th antibody, Int is the rounding function, q is the initial antibody group size, Φ i is the affinity of the ith antibody, min is the minimum function, exp is the natural exponential function, ||X i -X j || is the Euclidean distance between the i-th antibody and the j-th antibody, i≠j, i∈[1,q], j∈[1,q].

[0138] It should be noted that when calculating the Euclidean distance between the i-th antibody and the j-th antibody, normalization is performed so that 0≤||Xi -X j ||≤1.

[0139] In the above step 7, the progeny antibody S c The calculation methods include:

[0140]

[0141] Where X′ w is the w-th parent antibody, s w is the wth scaling factor, which is a randomly generated real number and not all w scaling factors are 0.

[0142] The second adjustment module is used to generate an adjustment amount set if a temperature abnormality instruction is generated, adjust the battery characteristic data according to the adjustment amount set, and calculate the temperature adjustment power. The lithium battery heating component operates according to the temperature adjustment power.

[0143] The steps for generating the set of adjustment variables when generating a temperature abnormality instruction are the same as the steps for generating the set of adjustment variables when only a semi-abnormality instruction is generated.

[0144] Methods for calculating thermostat power include:

[0145] Subtract the battery temperature in the battery characteristic data from the minimum value of the normal temperature range to obtain the temperature difference; obtain the battery heat capacity, which is obtained from the product specifications or technical parameters of the lithium battery; obtain the battery heat by multiplying the battery heat capacity and the temperature difference; the expression of battery heat is: Q = C × T; where Q is the battery heat, C is the battery heat capacity, and T is the temperature difference; preset the heating time, and obtain the temperature control power by the ratio of the battery heat and the heating time; the expression of temperature control power is Wherein, P is the temperature control power, and t′ is the heating time. The heating time is calculated by technicians in this field by collecting the operating time of the heating components multiple times during the low-temperature charging process of the historical lithium battery, and taking the average of the operating time of multiple heating components as the heating time.

[0146] This embodiment collects and processes battery characteristic data in real time, uses deep learning technology to intelligently identify abnormal conditions during the lithium battery charging process, takes corresponding charging adjustment or preheating measures based on the abnormal conditions, and then optimizes charging parameters based on a clonal selection algorithm to achieve full intelligent management of the lithium battery's low-temperature charging process. This effectively prevents safety hazards during low-temperature charging, improves the charging efficiency and service life of lithium batteries in low-temperature environments, and further improves charging safety and reliability, achieving more efficient and safe low-temperature charging.

[0147] Example 2

[0148] See also Figure 2As shown, this embodiment further improves the design based on Example 1. In Example 1, when heating the lithium battery, the heating time is a preset time. However, in actual situations, if the heating time of the lithium battery is too long, that is, the lithium battery is charged at a low battery temperature for a long time, it may still cause dendrite growth, increased internal resistance, decreased charging efficiency, etc. Therefore, this embodiment provides a lithium battery low-temperature charging system, which also includes a time control module;

[0149] The time control module is used to control the heating time of the lithium battery heating component.

[0150] The method for regulating the heating time of the lithium battery heating component includes:

[0151] The preset heating time is divided into N time periods according to the sampling frequency, where N is an integer greater than 1. The end time point of each time period is marked as the end point, and the start time point of each time period is marked as the start point. The end point of the previous time period is the start point of the next time period. The temperature control power is multiplied by the sampling frequency to obtain the change heat, and the change heat is divided by the battery heat capacity to obtain the change temperature. The change temperature and the ambient temperature are used as evaluation data, and the evaluation data are input into the change prediction model to predict the actual change temperature. The actual change temperature is the actual change in battery temperature under the influence of ambient temperature. The specific training process of the change prediction model is consistent with the efficiency The training process of the prediction model is consistent, and both are deep neural network models; according to the time sequence, the actual change temperature is added to the battery temperature of each starting point in turn to obtain the battery temperature of each end point; illustratively, the battery temperature of the first starting point is the battery temperature in the battery characteristic data, and the battery temperature in the battery characteristic data is added to the actual change temperature to obtain the battery temperature of the first end point; the battery temperature of the first end point is the battery temperature of the second starting point, and the battery temperature of the second starting point is added to the actual change temperature to obtain the battery temperature of the second end point; and so on, the battery temperature of the Nth end point is obtained.

[0152] Obtain the number of end points between each end point and the first start point respectively, add one to the number of end points corresponding to each end point and multiply it by the acquisition frequency to obtain the charging time corresponding to each end point;

[0153] The ambient temperature, battery internal resistance, battery capacity, and the battery temperature and charging time of an end point are taken as a group of research data, and the research data correspond to the end points one by one; each group of research data is input into the trained internal resistance prediction model to predict the battery internal resistance corresponding to each end point; each group of research data is input into the trained capacity prediction model to predict the battery capacity corresponding to each end point; based on the battery internal resistance in the battery characteristic data and the battery internal resistance corresponding to each end point, the internal resistance change rate corresponding to each end point is calculated; based on the battery capacity in the battery characteristic data and the battery capacity corresponding to each end point, the capacity change rate corresponding to each end point is calculated; based on the internal resistance change rate and capacity change rate corresponding to each end point, whether to generate an abnormal instruction is determined;

[0154] If an abnormal instruction is generated, the end point closest to the first starting point among the end points corresponding to the abnormal instruction is marked as an adjustment point, and the charging time corresponding to the adjustment point is subtracted from the acquisition frequency to obtain the optimal charging time; the heating time of the lithium battery heating component is adjusted to the optimal charging time;

[0155] If no abnormal instruction is generated, the heating time of the lithium battery heating component is not regulated.

[0156] It should be noted that the method of determining whether to generate an abnormal instruction based on the internal resistance change rate and capacity change rate corresponding to each end point is consistent with the method of determining whether to generate an abnormal instruction again in the secondary analysis module.

[0157] This embodiment realizes real-time intelligent control of the heating time of the lithium battery; the preset heating time is subdivided into multiple time periods, and each time period is combined with the ambient temperature using deep learning technology to obtain a more accurate battery temperature, and the corresponding charging time, internal resistance change rate and capacity change rate are calculated to determine whether there are any abnormal conditions during the charging process of the lithium battery; if an abnormality occurs, the most recent abnormal time point is found, and the heating time of the lithium battery heating component is adjusted in real time according to the corresponding charging time; further realizing the full intelligent management of the low-temperature charging process of the lithium battery, avoiding the lithium battery from being charged in a low-temperature state for a long time, effectively solving the shortcomings of fixed time settings, better ensuring the safe charging of the lithium battery, and thus improving the charging efficiency and battery life of the lithium battery.

[0158] Example 3

[0159] See also Figure 3 As shown, for the parts not described in detail in this embodiment, please refer to the description of Example 1 and Example 2. A method for charging a lithium battery at low temperature is provided, and the method includes:

[0160] Real-time collection of lithium battery characteristic data and battery environment data;

[0161] Processing battery characteristic data to obtain battery processing data;

[0162] Analyze battery characteristic data and battery processing data to determine whether to generate warning instructions, including abnormal instructions, semi-abnormal instructions, and temperature abnormal instructions;

[0163] If a semi-abnormal instruction is generated, charging is interrupted and the lithium battery is preheated. After preheating is completed, the battery characteristic data and battery processing data are collected again to determine whether an abnormal instruction is generated. If no abnormal instruction is generated, charging is resumed, and the battery characteristic data, battery environment data, and battery processing data are analyzed to generate a set of adjustment quantities, and the battery characteristic data is adjusted according to the set of adjustment quantities.

[0164] If a temperature abnormality instruction is generated, an adjustment amount set is also generated, the battery characteristic data is adjusted according to the adjustment amount set, and the temperature adjustment power is calculated. The lithium battery heating component operates according to the temperature adjustment power.

[0165] Furthermore, the battery characteristic data includes charging current, charging voltage, battery temperature, battery internal resistance, battery capacity, current battery charge and minimum battery charge; the battery environment data includes ambient temperature, ambient humidity and ambient air pressure;

[0166] The method for obtaining battery processing data includes:

[0167] Battery processing data includes power percentage, capacity change rate, and internal resistance change rate;

[0168] The expression for the battery percentage is: In the formula, SOC is the percentage of power, S is the current power of the battery, S min is the minimum battery capacity, S max is the battery capacity;

[0169] The expression of capacity change rate is: Where, is the capacity change rate, S′ max The historical battery capacity is the battery capacity collected last time. The battery characteristic data is collected according to the preset collection frequency.

[0170] The expression of internal resistance change rate is: Where, is the internal resistance change rate, R is the battery internal resistance, and R′ is the historical battery internal resistance; the historical battery internal resistance is the battery internal resistance collected last time.

[0171] Furthermore, the method for determining whether to generate a warning instruction includes:

[0172] The capacity change rate and internal resistance change rate in the battery processing data, as well as the battery temperature in the battery characteristic data, are used as analysis data. The analysis data is input into the trained state analysis model, and the corresponding judgment results are obtained based on the output judgment labels. The judgment results include normal, abnormal temperature, abnormal capacity or internal resistance, abnormal temperature and internal resistance, abnormal temperature and capacity, and abnormal.

[0173] If the judgment result is all normal, no warning instruction will be generated;

[0174] If the judgment result is that the temperature is abnormal, a temperature abnormality instruction is generated;

[0175] If the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated;

[0176] If the judgment result is not completely normal, the temperature is abnormal, and the capacity or internal resistance is abnormal, a semi-abnormal instruction is generated;

[0177] The training process of the state analysis model includes:

[0178] Pre-set the corresponding judgment results for group a of analysis data, where a is an integer greater than 1, and set different digital labels for different judgment results;

[0179] The digital labels of the judgment results are marked as judgment labels, and the analysis data and the corresponding judgment labels are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of the state analysis model, and the state analysis model uses a set of predicted judgment labels corresponding to each set of analysis data as output, and the actual judgment labels corresponding to each set of analysis data as prediction targets. The actual judgment labels are the pre-collected digital labels of the judgment results corresponding to the analysis data; minimizing the sum of the prediction errors of all analysis data is used as the training goal; the state analysis model is trained until the sum of the prediction errors reaches convergence and the training is stopped; the state analysis model is a deep neural network model.

[0180] Furthermore, the method of determining again whether an abnormal instruction is generated includes:

[0181] A normal temperature range is preset; the lithium battery is preheated through the built-in heating element of the lithium battery or an external heating device, and the battery characteristic data is continuously collected according to the collection frequency; when the battery temperature reaches the normal temperature range, the preheating ends, and the battery processing data is acquired again, and the corresponding judgment result is obtained again through the state analysis model; if the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated again; if the judgment result is all normal, no abnormal instruction is generated.

[0182] Furthermore, the adjustment amount set includes a current adjustment amount and a voltage adjustment amount; the charging current and charging voltage in the battery characteristic data are adjusted according to the adjustment amount set; and the step of generating the adjustment amount set includes:

[0183] Step 1: Calculate the data adjustment range;

[0184] Step 2: Randomly generate an initial antibody population X(0) of size q = {X1, X2, X3, ... X q}; preset iteration threshold; where X q is the qth antibody;

[0185] Step 3: Determine the fitness function;

[0186] Step 4: Calculate the fitness of each antibody in the initial antibody population X(0), select r antibodies with the largest fitness, and construct the memory population X m (t);

[0187] Step 5: Memory population X m Each antibody in (t) is cloned according to the calculated clone size to generate a clone antibody population Among them, is the qth cloned antibody in the clonal antibody population;

[0188] Step 6: Cloning Antibody Group X c Each antibody in (t) is subjected to cloud adaptive mutation using the cloud adaptive mutation operator to generate a mutant antibody group X d (t);

[0189] Step 7: Cloning Antibodies from Group X c In (t), w antibodies are randomly selected as parent antibodies for antibody recombination to generate progeny antibodies S c , until the cloned antibody group X c When the number of unselected antibodies in (t) is less than w, antibody recombination is stopped and a recombinant antibody group X is generated. r (t);

[0190] Step 8: Merge variant antibody group X d (t) and recombinant antibody group X r (t), generate a combined antibody population X L (t);

[0191] Step 9: Randomly generate n new antibodies and merge the antibody group X L The n antibodies with the smallest fitness in (t) are replaced by n randomly generated new antibodies;

[0192] Step 10: If the number of iterations t is less than the iteration threshold, set t = t + 1 and select the merged antibody group XL The r antibodies with the largest fitness in (t) form a new memory population X m (t+1), loop steps 4 to 9; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group X is obtained L The set of adjustment amounts corresponding to the antibody with the largest fitness in (t).

[0193] Furthermore, in step 1, the method for calculating the data adjustment amount range includes:

[0194] Obtain the maximum charging current, minimum charging current, maximum charging voltage, and minimum charging voltage of the lithium battery; subtract the real-time collected charging current from the maximum charging current and minimum charging current to obtain the current adjustment range; subtract the real-time collected charging voltage from the maximum charging voltage and minimum charging voltage to obtain the voltage adjustment range;

[0195] In step 2, the antibodies in the initial antibody group X(0) correspond one-to-one to the adjustment amount sets in the data set, and the range of the antibodies is the data set; wherein each adjustment amount set includes a value in the current adjustment range and a value in the voltage adjustment range, that is, the adjustment amount set is constructed by randomly selecting a value from each of the current adjustment range and the voltage adjustment range; the data set includes all adjustment amount sets, that is, m adjustment amount sets, m=m1×m2, m1 is the number of values ​​in the current adjustment range, and m2 is the number of values ​​in the voltage adjustment range;

[0196] In step 5, the calculation method of the clone size includes:

[0197]

[0198] Φ i =min(exp(||X i -X j ||));

[0199] Where R i is the clone size of the i-th antibody, Int is the rounding function, q is the initial antibody group size, Φ i is the affinity of the ith antibody, min is the minimum function, exp is the natural exponential function, ||X i -X j || is the Euclidean distance between the i-th antibody and the j-th antibody, i≠j, i∈[1,q], j∈[1,q].

[0200] Furthermore, in step 3, the fitness function is expressed as: f i =ψ1×XL i +ψ2×AP i;

[0201] Where, f i is the fitness of the i-th antibody, XL i is the charging efficiency of the ith antibody, AP i is the safety score of the i-th antibody, ψ1 and ψ2 are preset weight factors;

[0202] The methods for obtaining charging efficiency and safety scores include:

[0203] The current regulation amount corresponding to the antibody is added to the charging current collected in real time to obtain the regulation current; the voltage regulation amount corresponding to the antibody is added to the charging voltage collected in real time to obtain the regulation voltage; the battery characteristic data, battery environment data, battery temperature, regulation current and regulation voltage are used as test data; the test data is input into the trained efficiency prediction model to predict the corresponding charging efficiency; the test data is input into the trained score prediction model to predict the corresponding safety score; wherein, the training process of the efficiency prediction model and the score analysis model are consistent with the training process of the state analysis model, and both are deep neural network models, each set of feature vectors input in the training process of the efficiency prediction model is the test data and the charging efficiency corresponding to the test data, and each set of feature vectors input in the training process of the score prediction model is the test data and the safety score corresponding to the test data;

[0204] In step 7, the progeny antibody S c The calculation methods include:

[0205]

[0206] Where X′ w is the w-th parent antibody, s w is the wth scaling factor, which is a randomly generated real number and not all w scaling factors are 0.

[0207] Furthermore, the step of generating the set of adjustment variables when generating the temperature abnormality instruction is consistent with the step of generating the set of adjustment variables when only the semi-abnormality instruction is generated;

[0208] Methods for calculating thermostat power include:

[0209] Subtract the battery temperature in the battery characteristic data from the minimum value of the normal temperature range to obtain the temperature difference; obtain the battery heat capacity; obtain the battery heat by multiplying the battery heat capacity and the temperature difference; the expression for battery heat is: Q = C × T; where Q is the battery heat, C is the battery heat capacity, and T is the temperature difference; preset the heating time, and obtain the temperature control power by the ratio of the battery heat and the heating time; the expression for the temperature control power is Where P is the temperature control power and t' is the heating time.

[0210] Furthermore, the method further comprises: regulating the heating time of the lithium battery heating component;

[0211] The method for regulating the heating time of the lithium battery heating component includes:

[0212] The preset heating time is divided into N time periods according to the acquisition frequency, where N is an integer greater than 1. The end time point of each time period is marked as the end point, and the start time point of each time period is marked as the start point. The end point of the previous time period is the start point of the next time period. The temperature control power is multiplied by the sampling frequency to obtain the change heat, and the change heat is divided by the battery heat capacity to obtain the change temperature. The change temperature and the ambient temperature are used as evaluation data, and the evaluation data are input into the change prediction model to predict the actual change temperature. The actual change temperature is the actual change in battery temperature under the influence of ambient temperature. The training process of the change prediction model is consistent with the training process of the efficiency prediction model, and both are deep neural network models. According to the time sequence, the actual change temperature is added to the battery temperature of each starting point to obtain the battery temperature of each end point.

[0213] Obtain the number of end points between each end point and the first start point respectively, add one to the number of end points corresponding to each end point and multiply it by the acquisition frequency to obtain the charging time corresponding to each end point;

[0214] The ambient temperature, battery internal resistance, battery capacity, and the battery temperature and charging time of an end point are taken as a group of research data, and the research data correspond to the end points one by one; each group of research data is input into the trained internal resistance prediction model to predict the battery internal resistance corresponding to each end point; each group of research data is input into the trained capacity prediction model to predict the battery capacity corresponding to each end point; based on the battery internal resistance in the battery characteristic data and the battery internal resistance corresponding to each end point, the internal resistance change rate corresponding to each end point is calculated; based on the battery capacity in the battery characteristic data and the battery capacity corresponding to each end point, the capacity change rate corresponding to each end point is calculated; based on the internal resistance change rate and capacity change rate corresponding to each end point, whether to generate an abnormal instruction is determined;

[0215] If an abnormal instruction is generated, the end point closest to the first starting point among the end points corresponding to the abnormal instruction is marked as an adjustment point, and the charging time corresponding to the adjustment point is subtracted from the acquisition frequency to obtain the optimal charging time; the heating time of the lithium battery heating component is adjusted to the optimal charging time;

[0216] If no abnormal instruction is generated, the heating time of the lithium battery heating component is not regulated.

[0217] Example 4

[0218] See also Figure 4 As shown, the present application also provides an electronic device 500. The electronic device 500 may include one or more processors and one or more memories. The memories may store computer-readable code, which, when executed by the one or more processors, may execute the lithium battery low-temperature charging method described above.

[0219] The method or system according to the embodiment of the present application can also be used by Figure 4 The electronic device architecture shown in FIG. Figure 4 As shown, the electronic device 500 may include a bus 501, one or more CPUs 502, ROM 503, RAM 504, a communication port 505 connected to a network, an input / output 506, a hard disk 507, etc. The storage device in the electronic device 500, such as ROM 503 or hard disk 507, may store the lithium battery low-temperature charging method provided in this application. Furthermore, the electronic device 500 may also include a user interface 508. Of course, Figure 4 The architecture shown is only exemplary and can be omitted according to actual needs when implementing different devices. Figure 4 One or more components of an electronic device are shown.

[0220] Example 5

[0221] See also Figure 5 As shown, a computer-readable storage medium 600 according to one embodiment of the present application is shown. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are executed by the processor, the low-temperature charging method for lithium batteries according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0222] In addition, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the present application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, such as a method for low-temperature charging of lithium batteries. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the method of the present application are performed.

[0223] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

[0224] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A lithium battery low temperature charging method, characterized in that: include: Real-time collection of battery characteristic data and battery environment data of lithium batteries; Processing the battery characteristic data to obtain battery processing data; Analyze the battery characteristic data and battery processing data to determine whether to generate a warning instruction, which includes an abnormal instruction, a semi-abnormal instruction, and a temperature abnormal instruction; The method for determining whether to generate an early warning instruction includes: The capacity change rate and internal resistance change rate in the battery processing data and the battery temperature in the battery characteristic data are used as analysis data, and the analysis data are input into the trained state analysis model. According to the output judgment label, the corresponding judgment result is obtained; the judgment results include all normal, temperature abnormal, capacity or internal resistance abnormal, temperature and internal resistance abnormal, temperature and capacity abnormal, and all abnormal; If the judgment result is all normal, no warning instruction will be generated; If the judgment result is that the temperature is abnormal, a temperature abnormality instruction is generated; If the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated; If the judgment result is not completely normal, the temperature is abnormal, and the capacity or internal resistance is abnormal, a semi-abnormal instruction is generated; If a semi-abnormal instruction is generated, charging is interrupted and the lithium battery is preheated. After the preheating is completed, the battery characteristic data is collected again and the battery processing data is obtained, and it is determined again whether an abnormal instruction is generated; if no abnormal instruction is generated, charging is resumed, and the battery characteristic data, battery environment data and battery processing data are analyzed to generate a set of adjustment quantities, and the battery characteristic data is adjusted according to the set of adjustment quantities; If a temperature abnormality instruction is generated, an adjustment quantity set is also generated, the battery characteristic data is adjusted according to the adjustment quantity set, and the temperature adjustment power is calculated, and the lithium battery heating component operates according to the temperature adjustment power.

2. The low temperature charging method for lithium batteries according to claim 1, characterized in that: The battery characteristic data includes charging current, charging voltage, battery temperature, battery internal resistance, battery capacity, current battery power and minimum battery power; the battery environment data includes ambient temperature, ambient humidity and ambient air pressure; The method for obtaining battery processing data comprises: Battery processing data includes power percentage, capacity change rate, and internal resistance change rate; The expression for the power percentage is: In the formula, SOC is the percentage of power, S is the current power of the battery, and S min is the minimum battery charge, S max is the battery capacity; The expression of capacity change rate is: In the formula, is the capacity change rate, S′ max The historical battery capacity is the battery capacity collected last time. The battery characteristic data is collected according to the preset collection frequency. The expression of internal resistance change rate is: In the formula, is the internal resistance change rate, R is the battery internal resistance, and R′ is the historical battery internal resistance; the historical battery internal resistance is the battery internal resistance collected last time.

3. The low temperature charging method for lithium batteries according to claim 2, characterized in that: The training process of the state analysis model includes: Pre-set corresponding judgment results for group a of analysis data, where a is an integer greater than 1, and set different digital labels for different judgment results; The digital labels of the judgment results are marked as judgment labels, and the analysis data and the corresponding judgment labels are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of the state analysis model, and the state analysis model uses a set of predicted judgment labels corresponding to each set of analysis data as output, and uses the actual judgment labels corresponding to each set of analysis data as prediction targets, where the actual judgment labels are pre-collected digital labels of the judgment results corresponding to the analysis data; minimizing the sum of prediction errors of all analysis data is used as the training target; the state analysis model is trained until the sum of prediction errors converges and the training is stopped; the state analysis model is a deep neural network model.

4. The low temperature charging method for lithium batteries according to claim 3, characterized in that: The method for determining again whether an abnormal instruction is generated comprises: Preset the normal temperature range; preheat the lithium battery through the lithium battery's built-in heating element or an external heating device, and continuously collect battery characteristic data according to the collection frequency; when the battery temperature reaches the normal temperature range, the preheating ends, and the battery processing data is acquired again, and the corresponding judgment result is acquired again through the state analysis model; if the judgment result is that the capacity or internal resistance is abnormal, an abnormal instruction is generated again; if the judgment result is completely normal, no abnormal instruction is generated.

5. The low temperature charging method for lithium batteries according to claim 4, characterized in that: The adjustment amount set includes a current adjustment amount and a voltage adjustment amount; the charging current and the charging voltage in the battery characteristic data are adjusted according to the adjustment amount set; the step of generating the adjustment amount set includes: Step 1: Calculate the data adjustment range; Step 2: Randomly generate an initial antibody population X(0) of size q = {X1, X2, X3, ... X q }; preset iteration threshold; where X q is the qth antibody; Step 3: Determine the fitness function; Step 4: Calculate the fitness of each antibody in the initial antibody population X(0), select r antibodies with the largest fitness, and construct a memory population X m (t); Step 5: Memory population X m Each antibody in (t) is cloned according to the calculated clone size to generate a clone antibody population Among them, is the qth cloned antibody in the clonal antibody group; Step 6: Cloning Antibody Group X c Each antibody in (t) is subjected to cloud adaptive mutation using a cloud adaptive mutation operator to generate a mutant antibody group X d (t); Step 7: Cloning Antibodies from Population X c (t) randomly selects w antibodies as parent antibodies for antibody recombination to generate offspring antibodies Sc until the cloned antibody population X c When the number of unselected antibodies in (t) is less than w, antibody recombination is stopped and a recombinant antibody group X is generated. r (t); Step 8: Merge variant antibody group X d (t) and recombinant antibody group X r (t), generate a combined antibody population X L (t); Step 9: Randomly generate n new antibodies and merge antibody group X L The n antibodies with the smallest fitness in (t) are replaced by n randomly generated new antibodies; Step 10: If the number of iterations t is less than the iteration threshold, set t = t + 1 and select the merged antibody group X L The r antibodies with the largest fitness in (t) form a new memory population X m (t+1), loop steps 4 to 9; if the number of iterations t is equal to the iteration threshold, the loop ends and the combined antibody group X is obtained. L The set of adjustment values ​​corresponding to the antibody with the largest fitness in (t).

6. The low temperature charging method for lithium batteries according to claim 5, characterized in that: In step 1, the method for calculating the data adjustment amount range includes: Obtain the maximum charging current, minimum charging current, maximum charging voltage and minimum charging voltage of the lithium battery; respectively subtract the charging current collected in real time from the maximum charging current and the minimum charging current to obtain the current adjustment range; respectively subtract the charging voltage collected in real time from the maximum charging voltage and the minimum charging voltage to obtain the voltage adjustment range; In step 2, the antibodies in the initial antibody group X(0) correspond to the adjustment amount sets in the data set one by one, and the range of the antibodies is the data set; wherein each adjustment amount set includes a value in the current adjustment range and a value in the voltage adjustment range, that is, the adjustment amount set is constructed by randomly selecting a value from each of the current adjustment range and the voltage adjustment range; the data set includes all adjustment amount sets, that is, m adjustment amount sets, m=m1×m2, m1 is the number of values ​​in the current adjustment range, and m2 is the number of values ​​in the voltage adjustment range; In step 5, the method for calculating the clone size includes: F i =min(exp(||X i -X j ||)); In the formula, R i is the clone size of the i-th antibody, Int is the upward rounding function, q is the initial antibody population size, Φ i is the affinity of the ith antibody, min is the minimum function, exp is the natural exponential function, ||X i -X j || is the Euclidean distance between the i-th antibody and the j-th antibody, i≠j, i∈[1,q], j∈[1,q].

7. The low temperature charging method for lithium batteries according to claim 6, characterized in that: In step 3, the fitness function is expressed as: i =ψ1×XL i +ψ2×AP i ; In the formula, f i is the fitness of the ith antibody, XL i is the charging efficiency of the ith antibody, AP i is the safety score of the ith antibody, ψ1 and ψ2 are preset weight factors; The methods for obtaining charging efficiency and safety scores include: The current regulation amount corresponding to the antibody is added to the charging current collected in real time to obtain the regulation current; the voltage regulation amount corresponding to the antibody is added to the charging voltage collected in real time to obtain the regulation voltage; the battery characteristic data, battery environment data, battery temperature, regulation current and regulation voltage are used as test data; the test data is input into the trained efficiency prediction model to predict the corresponding charging efficiency; the test data is input into the trained score prediction model to predict the corresponding safety score; wherein, the training process of the efficiency prediction model and the score analysis model are consistent with the training process of the state analysis model, and both are deep neural network models, each set of feature vectors input in the training process of the efficiency prediction model is the test data and the charging efficiency corresponding to the test data, and each set of feature vectors input in the training process of the score prediction model is the test data and the safety score corresponding to the test data; In step 7, the progeny antibody S c The calculation methods include: In the formula, X′ w is the w-th parent antibody, s w is the w-th scaling factor, which is a randomly generated real number and not all w scaling factors are 0.

8. The low temperature charging method for lithium batteries according to claim 7, characterized in that: The steps of generating the set of adjustment quantities when generating the temperature abnormality instruction are consistent with the steps of generating the set of adjustment quantities when only the semi-abnormality instruction is generated; Methods for calculating thermostatic power include: Subtract the battery temperature in the battery characteristic data from the minimum value of the normal temperature range to obtain the temperature difference; obtain the battery heat capacity; obtain the battery heat by multiplying the battery heat capacity and the temperature difference; the expression of the battery heat is: Q = C × T; where Q is the battery heat, C is the battery heat capacity, and T is the temperature difference; preset the heating time, and obtain the temperature control power by the ratio of the battery heat and the heating time; the expression of the temperature control power is Where P is the temperature control power and t' is the heating time.

9. The low temperature charging method for lithium batteries according to claim 8, characterized in that: The method further comprises: regulating the heating time of the lithium battery heating component; The method for regulating the heating time of the lithium battery heating component includes: The preset heating time is divided into N time periods according to the acquisition frequency, where N is an integer greater than 1, and the end time point of each time period is marked as the end point, and the start time point of each time period is marked as the start point. The end point of the previous time period is the start point of the next time period; the temperature control power is multiplied by the sampling frequency to obtain the change heat, and the change heat is divided by the battery heat capacity to obtain the change temperature; the change temperature and the ambient temperature are used as evaluation data, and the evaluation data are input into the change prediction model to predict the actual change temperature. The actual change temperature is the actual change in battery temperature under the influence of ambient temperature; the training process of the change prediction model is consistent with the training process of the efficiency prediction model, and both are deep neural network models; according to the time sequence, the actual change temperature is added to the battery temperature of each starting point in turn to obtain the battery temperature of each end point; The number of end points between each end point and the first start point is obtained respectively, the number of end points corresponding to each end point is added by one and then multiplied by the acquisition frequency to obtain the charging time corresponding to each end point; The ambient temperature, battery internal resistance, battery capacity, and the battery temperature and charging time of an end point are taken as a group of research data, and the research data corresponds to the end point one by one; each group of research data is input into the trained internal resistance prediction model to predict the battery internal resistance corresponding to each end point; each group of research data is input into the trained capacity prediction model to predict the battery capacity corresponding to each end point; the internal resistance change rate corresponding to each end point is calculated based on the battery internal resistance in the battery characteristic data and the battery internal resistance corresponding to each end point; the capacity change rate corresponding to each end point is calculated based on the battery capacity in the battery characteristic data and the battery capacity corresponding to each end point; based on the internal resistance change rate and capacity change rate corresponding to each end point, it is determined whether to generate an abnormal instruction; If an abnormal instruction is generated, the end point closest to the first start point among the end points corresponding to the abnormal instruction is marked as an adjustment point, and the charging time corresponding to the adjustment point is subtracted from the acquisition frequency to obtain the optimal charging time; the heating time of the lithium battery heating component is adjusted to the optimal charging time; If no abnormal instruction is generated, the heating time of the lithium battery heating component is not regulated.

10. A lithium battery low temperature charging system, implementing the lithium battery low temperature charging method according to any one of claims 1 to 9, characterized in that: include: Data acquisition module, used to collect battery characteristic data and battery environment data of lithium batteries in real time; A data processing module, used for processing battery characteristic data and obtaining battery processing data; A data analysis module is used to analyze the battery characteristic data and battery processing data to determine whether to generate a warning instruction, which includes an abnormal instruction, a semi-abnormal instruction, and a temperature abnormal instruction; The secondary analysis module is used to interrupt charging and preheat the lithium battery if a semi-abnormal instruction is generated. After the preheating is completed, the battery characteristic data is recollected and the battery processing data is obtained to determine again whether an abnormal instruction is generated; A first adjustment module, configured to resume charging if no abnormal instruction is generated, analyze battery characteristic data, battery environment data, and battery processing data, generate an adjustment amount set, and adjust the battery characteristic data according to the adjustment amount set; The second regulating module is used to generate a regulating quantity set if a temperature abnormality instruction is generated, adjust the battery characteristic data according to the regulating quantity set, and calculate the temperature regulating power. The lithium battery heating component operates according to the temperature regulating power.

Images