A dual-mode switching control method for a charging adapter

Abstract

The application belongs to the technical field of charging control, and discloses a dual-mode switching control method for a charging adapter. The method comprises the following steps: identifying the battery type, automatically matching a switching parameter set; continuously collecting charging state data, forming a state data set, and performing trend analysis and calculation to obtain a change rate set; dynamically evaluating the battery charging state based on the change rate set and real-time state data in the state data set; dynamically adjusting the switching parameter set to generate a switching threshold set; fusing the change rate set, the battery charging state and the real-time state data, and performing multi-dimensional comparative analysis with the switching threshold set to determine whether to switch modes; if yes, the optimal transition curve is calculated, and the charging current in the charging state data is adjusted step by step. The application can effectively improve the applicability and performance of the charging adapter in the battery charging scene, thereby prolonging the battery life and improving the charging experience.

Description

Technical Field

[0001] This invention relates to the field of charging control technology, and more specifically, to a dual-mode switching control method for a charging adapter. Background Technology

[0002] As mobile devices, wearable devices, and power tools place higher demands on charging performance in terms of speed, safety, and efficiency, intelligent control of charging adapters has gradually become a key focus of technological research and engineering applications. In existing technologies, charging adapters generally adopt a dual-mode charging strategy combining constant current and constant voltage. The switching between the two modes is usually based on whether the battery terminal voltage reaches a preset threshold. That is, when the battery voltage is detected to reach the switching threshold from constant current to constant voltage during the charging process, the charging control logic immediately completes the mode switch. For example, patent CN108365654B discloses a wireless charger suitable for any lithium battery; another example is patent CN116169754A, which discloses a control method, device, and computer equipment for a dual active bridge converter.

[0003] This control method based on a single voltage condition, while simple to implement and easy to integrate, has the following problems in practical applications:

[0004] (1) Fixed switching point, lack of dynamic adjustment capability: Different types of batteries (such as lithium-ion, lithium iron phosphate, and high-voltage cells) have different requirements for charging strategies, and fixed voltage thresholds cannot adapt to diverse battery characteristics.

[0005] (2) Highly susceptible to external environmental interference: Factors such as battery temperature, load status, and voltage fluctuations may affect the accuracy of voltage detection, leading to incorrect switching or switching lag.

[0006] (3) Lack of comprehensive judgment on multi-dimensional parameters (such as current trend, charging efficiency, etc.): Relying solely on voltage as the basis for switching can easily lead to premature switching to the constant voltage stage before the battery reaches its optimal charging state, thus affecting charging efficiency.

[0007] Therefore, there is an urgent need for an intelligent dual-mode switching control method to achieve better matching for different charging scenarios and battery types, thereby improving charging efficiency, reliability, and safety.

[0008] In view of this, the present invention proposes a dual-mode switching control method for a charging adapter to solve the above problems. Summary of the Invention

[0009] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: a dual-mode switching control method for a charging adapter, comprising:

[0010] Step S1: Use the pulse charging test method to obtain the battery response characteristic curve, identify the battery type based on the battery response characteristic curve, and automatically match and switch the parameter set based on the battery type;

[0011] Step S2: Continuously collect charging status data according to the preset sampling interval to form a status dataset;

[0012] Step S3: Perform trend analysis calculations on the state dataset to obtain a set of rates of change, which includes the rate of change of voltage and the rate of change of current.

[0013] Step S4: Construct a nonlinear weighted fusion model to dynamically evaluate the battery charging status based on real-time state data in the rate of change set and state dataset;

[0014] Step S5: Obtain the ambient temperature and, in conjunction with real-time status data, dynamically adjust the set of switching parameters to generate a set of switching thresholds;

[0015] Step S6: Combine current change rate, battery charging status and real-time status data, and perform multi-dimensional comparative analysis with the switching threshold set to determine whether to switch modes.

[0016] Step S7: If a mode switch is to be performed, the optimal transition curve is calculated using a gradient transition strategy. Based on the optimal transition curve, the charging current in the charging status data is gradually adjusted to complete the charging mode switch.

[0017] Furthermore, methods for identifying battery types based on battery response characteristic curves include:

[0018] Based on the battery response characteristic curve, a multi-physical quantity coupling model is constructed. This model includes *a* sub-models, where *a* represents the number of curves in the battery response characteristic curve. The battery response characteristic curve is input into the multi-physical quantity coupling model, and a multi-objective optimization method is used to fit the model to obtain the model's back-calculated parameters. A pre-defined type mapping matrix is ​​used, containing parameter ranges corresponding to different battery types. These parameter ranges include the numerical ranges corresponding to each parameter in the model's back-calculated parameters. The model's back-calculated parameters are compared with each parameter range in the type mapping matrix. Parameter ranges where each parameter in the model's back-calculated parameters falls within its corresponding numerical range are selected and marked as matching ranges. The battery type corresponding to each matching range is then obtained.

[0019] Furthermore, methods for obtaining the rate of change of current include:

[0020] The charging currents in the state dataset are sorted in ascending order according to the corresponding acquisition time to generate a current sequence. The adjacent rate of change between every two adjacent charging currents in the current sequence is calculated sequentially, and a rate of change sequence is generated. The window length is dynamically set, and the rate of change sequence is divided into c rate of change windows based on the window length. Each rate of change window includes d adjacent rates of change, where d is the window length. For the adjacent rates of change within each rate of change window, a linear regression method is used to fit the trend of change, and the corresponding current characteristic line is obtained. The slope of the current characteristic line is taken as the current rate of change.

[0021] The method for obtaining the rate of change of voltage is the same as the method for obtaining the rate of change of current.

[0022] Furthermore, the steps for dynamically setting the window length include:

[0023] Step S301: Preset a type factor mapping table, which includes different battery types and their corresponding response speeds; based on the identified battery type, obtain the corresponding response speed from the type factor mapping table;

[0024] Step S302: Calculate the standard deviation of all adjacent rates of change and mark it as the degree of volatility;

[0025] Step S303: Construct multiple fuzzy sets for response speed and fluctuation degree respectively;

[0026] Step S304: Convert the response speed and fluctuation degree into the membership degree of each corresponding fuzzy set using fuzzification technology;

[0027] Step S305: Input all membership degrees into the trained membership analysis model to predict the corresponding membership set; wherein, the membership set includes the membership degree of each window level;

[0028] Step S306: Set corresponding level intervals for each window level and calculate the corresponding interval mean; based on the membership degree and interval mean of each window level, use the centroid method to dynamically calculate the window length.

[0029] Furthermore, methods for dynamically assessing battery state of charge include:

[0030] Different numerical labels are set for different battery types and marked as type labels; ambient temperature and charging time are obtained, and the type labels, ambient temperature and charging time are used as analysis data. The analysis data is input into the trained weight allocation model to predict the corresponding weight set; where the weight set includes the weight coefficient corresponding to each data in the battery state data, and the battery state data includes voltage change rate, current change rate and charging state data.

[0031] Real-time state data is obtained from the state dataset, which represents the charging state data at the current moment. The voltage and current change rates corresponding to the real-time state data are obtained from the change rate set and used together with the real-time state data. The corresponding real-time change rate is calculated for each data point in the real-time data, and the total change rate is summed to obtain the total change rate value. Based on the real-time change rate and the total change rate value, the weight coefficients corresponding to each data point in the real-time data are corrected to obtain the corrected weight coefficients. A weight fusion function is preset, and the real-time data and the corresponding corrected weight coefficients are substituted into the weight fusion function in sequence to calculate the battery charging state.

[0032] Furthermore, the steps for generating the set of switching thresholds include:

[0033] Step S501: Preset threshold range, and calculate parameter adjustment range based on threshold range and switching parameter set;

[0034] Step S502: Construct n adjustment sets based on the parameter adjustment range, where n is an integer greater than 1;

[0035] Step S503: Randomly select m adjustment sets from n adjustment sets, and treat all m adjustment sets as particles to construct a particle population, and set the number of iterations to 0;

[0036] Step S504: Construct the potential function and calculate the potential value corresponding to each particle in the particle population;

[0037] Step S505: Based on the potential value, divide the particle population into multiple particle subpopulations;

[0038] Step S506: Set a corresponding update mechanism for each particle subpopulation;

[0039] Step S507: Update the particles in each particle subpopulation according to the update mechanism;

[0040] Step S508: Compare the number of iterations with the preset iteration threshold. If the number of iterations is less than the iteration threshold, merge the particles in all particle subpopulations into a single particle population, increment the number of iterations by one, and return to step S504. If the number of iterations is greater than or equal to the iteration threshold, proceed to step S509.

[0041] Step S509: Obtain the optimal particle in the particle swarm, mark the adjustment set corresponding to the optimal particle as the optimal set, dynamically adjust the switching parameter set based on the optimal set, and generate a switching threshold set.

[0042] Furthermore, in step S505, the method for dividing the particle population into multiple particle subpopulations includes:

[0043] The particles in the particle population are sorted from largest to smallest according to their corresponding potential values ​​to generate a particle sequence. A preset division standard is used, which includes two incremental division coefficients. The products of the two division coefficients and m are used as the first and second division quantities, respectively. Each particle is assigned a corresponding number according to the ascending order of the particle sequence. The first and second particles are determined according to the number of each particle. The particles before the first particle and the first particle itself are classified as the superior population. The particles before the second particle and after the first particle, as well as the second particle itself, are classified as the ordinary population. The particles after the second particle are classified as the inferior population.

[0044] Furthermore, the methods for determining whether to switch modes include:

[0045] The rate of change of current in the real-time data is marked as the real-time rate of change. The real-time rate of change is compared with the rate of change threshold in the switching threshold set. If the real-time rate of change is less than the rate of change threshold, the change duration is calculated according to the rate of change set. If the real-time rate of change is greater than or equal to the rate of change threshold, no mode switching is performed. The change duration, battery charging status, and charging voltage and battery surface temperature in the real-time status data are used as judgment data. Each data in the judgment data is compared with the corresponding switching threshold in the switching threshold set. If the battery surface temperature is less than the corresponding switching threshold, and each of the remaining data in the judgment data is greater than or equal to the corresponding switching threshold, a mode switch is performed. If the battery surface temperature is greater than or equal to the corresponding switching threshold, or if there are any remaining data in the judgment data that are less than the corresponding switching threshold, no mode switching is performed.

[0046] Furthermore, methods for statistically analyzing the duration of change based on the set of rates of change include:

[0047] Based on the ascending order of the rate of change sequence, the current rate of change in the rate of change set is sorted to generate an electrical change sequence; the real-time electrical change rate in the electrical change sequence is selected as the starting point, and the electrical change sequence is traversed in reverse; if the current rate of change is less than the rate of change threshold, it is added to the pre-constructed pass set; if the current rate of change is greater than or equal to the rate of change threshold, the reverse traversal of the electrical change sequence is terminated; the number of current rate of change in the pass set is counted and marked as the number of electrical changes; the change duration is calculated based on the number of electrical changes, the window length d, and the sampling interval.

[0048] Furthermore, methods for calculating the optimal transition curve include:

[0049] Define the set of parameters, which includes the starting current, ending current, transition time, time step, current change rate limit, and current range limit.

[0050] A smoothing objective function is constructed. Based on the set of parameters and the smoothing objective function, the gradient descent method is used to dynamically adjust the charging current at each current control point to obtain the optimal transition curve.

[0051] In the process of gradually adjusting the charging current in the charging state data based on the optimal transition curve, sliding mode control is used to track and control the charging current.

[0052] The technical effects and advantages of the dual-mode switching control method for a charging adapter of the present invention are as follows:

[0053] By analyzing battery response characteristics using a pulse charging test method, the charging needs of different battery types can be accurately identified, and the switching parameter set can be automatically matched to achieve intelligent adaptation to different battery types. During the charging process, multi-dimensional state data such as charging current, charging voltage, and battery surface temperature are continuously collected. Based on dynamic trend analysis and a nonlinear weight fusion model, the real-time charging status of the battery is accurately evaluated, providing a reliable basis for subsequent mode switching. By incorporating external factors such as ambient temperature, the switching parameter set is dynamically adjusted through an improved swarm intelligence optimization algorithm to generate a switching threshold set, which significantly improves the dynamics and adaptability of the switching strategy compared to the traditional single voltage threshold. The optimal transition curve is calculated using a gradient transition strategy, and the charging current is tracked and adjusted in real time based on sliding mode control to achieve smooth switching of charging modes, avoiding damage to the battery from sudden changes in charging current, and improving charging efficiency and safety. This effectively solves the problems of fixed parameters, response lag, and poor safety in traditional charging adapter switching control, improves the applicability and performance of the charging adapter in battery charging scenarios, ensures the optimal charging state for different types of batteries, thereby extending battery life and improving the charging experience. Attached Figure Description

[0054] Figure 1 This is a flowchart of a dual-mode switching control method for a charging adapter according to Embodiment 1 of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Example 1

[0057] Please see Figure 1 As shown in this embodiment, a dual-mode switching control method for a charging adapter includes:

[0058] Step S1: Obtain the battery response characteristic curve using the pulse charging test method, identify the battery type based on the battery response characteristic curve, and automatically match and switch the parameter set based on the battery type.

[0059] The pulse charging test method is a discontinuous charging technique, and its specific steps include:

[0060] Step S101: Those skilled in the art set the pulse sequence based on practical experience; the pulse sequence includes pulse current, charging time, and rest time;

[0061] Step S102: Based on the pulse sequence, apply current pulses to the battery and record the battery response data in each pulse cycle; the battery response data includes battery terminal voltage, battery surface temperature, etc.

[0062] Step S103: Construct a battery response characteristic curve based on the battery response data within each pulse cycle; the battery response characteristic curve includes time-voltage curve, time-battery surface temperature curve, etc.; that is, the battery response characteristic curve is a complete dynamic dataset of battery response data such as voltage and battery surface temperature recorded within each pulse cycle, which changes over time and is the basis for identifying battery type and matching switching parameters.

[0063] Methods for identifying battery type based on battery response characteristic curves include:

[0064] Based on the battery response characteristic curve, a multi-physical quantity coupling model is constructed. This model includes *a* sub-models, where *a* represents the number of curves in the battery response characteristic curve. Each sub-model corresponds one-to-one with a curve in the battery response characteristic curve. Examples of sub-models include equivalent circuit models and equivalent thermal network models. The battery response characteristic curve is input into the multi-physical quantity coupling model. A multi-objective optimization method (i.e., simultaneously minimizing the fitting error of each curve in the battery response characteristic curve, ensuring that the difference between the output curve of the multi-physical quantity coupling model and the battery response characteristic curve obtained using the pulse charging test method is minimized) is used to fit the multi-physical quantity coupling model and obtain the model back-inference parameters. Examples of model back-inference parameters include ohmic internal resistance, capacitance, heat capacity, and thermal resistance.

[0065] A preset type mapping matrix is ​​used, which includes parameter ranges corresponding to different battery types. The parameter ranges include the numerical range of each parameter in the model back-derived parameters. The type mapping matrix is ​​preset by those skilled in the art according to the actual situation. Battery types include lithium-ion, lithium iron phosphate, lithium manganese oxide, etc. The model back-derived parameters are compared with each parameter range in the type mapping matrix. The parameter ranges in which each parameter in the model back-derived parameters is within the corresponding numerical range are selected and marked as matching ranges. The battery type corresponding to the matching range is obtained.

[0066] Methods for automatically matching and switching parameter sets based on battery type include:

[0067] A preset switching parameter set is provided, which includes b sets of switching parameters, where b is the number of battery types. Each switching parameter set corresponds one-to-one with a battery type. The switching parameter sets are preset by those skilled in the art based on actual conditions. The switching parameter sets include charging state threshold, rate of change threshold, duration threshold, voltage threshold, and temperature threshold. The corresponding switching parameter set is automatically matched from the switching parameter set based on the identified battery type.

[0068] It should be noted that different types of batteries have significant differences in chemical composition, internal structure, and electrical characteristics, and their corresponding charging requirements also vary. Therefore, it is necessary to identify the battery type and automatically match the corresponding set of switching parameters to ensure the safety and efficiency of the charging process, optimize battery life, prevent overcharging and damage, and thus achieve intelligent adaptation of the charging adapter to multiple battery types.

[0069] Step S2: Continuously collect charging status data according to the preset sampling interval to form a status dataset.

[0070] The charging status data reflects the current charging process and status of the battery, and is used to describe the real-time status of the battery during the charging process. The charging status data includes, but is not limited to, charging voltage, charging current, battery surface temperature, and battery internal resistance. The charging voltage and charging current are obtained by voltage sensors and current sensors installed on the charging adapter, respectively, and the battery surface temperature is obtained by a thermistor fixed to the surface of the battery casing. The battery internal resistance is obtained by inputting the real-time collected charging voltage, charging current, and battery surface temperature into the equivalent circuit model in sequence, and using online parameter estimation algorithms such as Kalman filtering and recursive least squares to fit the equivalent circuit model to obtain the battery internal resistance. The sampling interval is preset by those skilled in the art based on the actual charging situation of the battery.

[0071] Step S3: Perform trend analysis calculations on the state dataset to obtain a set of rates of change, which includes the rate of change of voltage and the rate of change of current.

[0072] Methods for obtaining the rate of change of current include:

[0073] The charging currents in the state dataset are sorted in ascending order according to their corresponding acquisition time (i.e., from earliest to latest) to generate a current sequence. The adjacent rate of change between every two adjacent charging currents in the current sequence is calculated sequentially, generating a rate of change sequence. The method for calculating the adjacent rate of change is as follows: for two adjacent currents in the current sequence, the current following the sequence is subtracted from the current preceding the sequence, and then divided by a preset sampling interval to obtain the adjacent rate of change. The window length is dynamically set, and the rate of change sequence is divided into c rate of change windows based on the window length. Each rate of change window includes d adjacent rates of change, where c is an integer greater than 1 and d is the window length. For each rate of change window, a linear regression method is used to fit the trend of the adjacent rates of change, obtaining the corresponding current characteristic line. The slope of the current characteristic line is taken as the current rate of change. The linear regression method is an existing technology, and the specific process will not be elaborated here.

[0074] The method for obtaining the rate of change of voltage is the same as the method for obtaining the rate of change of current.

[0075] The steps to dynamically set the window length include:

[0076] Step S301: Preset a type factor mapping table, which includes different battery types and their corresponding response speeds; wherein, the response speed is how quickly the battery responds to changes in charging current, that is, the rate at which the battery current changes when an external charging action is applied; the type factor mapping table is preset by those skilled in the art according to the actual situation; according to the identified battery type, the corresponding response speed is obtained from the type factor mapping table;

[0077] Step S302: Calculate the corresponding standard deviation based on all adjacent rates of change and mark it as the degree of volatility; the method for calculating the standard deviation is existing technology, and the specific calculation process will not be elaborated here.

[0078] Step S303: Construct multiple fuzzy sets for response speed and fluctuation degree respectively; for example, the fuzzy set corresponding to response speed is fast, medium speed, slow speed, etc.; the fuzzy set corresponding to fluctuation degree is large standard deviation, medium standard deviation, low standard deviation, etc.

[0079] Step S304: Convert the response speed and fluctuation level into the membership degree of each corresponding fuzzy set using fuzzification technology; fuzzification technology is the process of converting precise values ​​into the membership degree of the corresponding fuzzy set. Fuzzification technology includes, for example, triangular membership function, trapezoidal membership function, etc.; for example, if the response speed value is high, it is inferred that the membership degree of fast is 0.9, the membership degree of medium speed is 0.3, and the membership degree of slow speed is 0.

[0080] Step S305: Input all membership degrees into the trained membership analysis model to predict the corresponding membership set; wherein, the membership analysis model is a deep neural network model, which is an existing technology, and the specific training process will not be described in detail here; the membership set includes the membership degree of each window level, such as large window, medium window, small window, etc.

[0081] Step S306: Set corresponding level intervals for each window level and calculate the corresponding interval mean; wherein, the interval mean is the average of the maximum and minimum values ​​of each level interval; based on the membership degree of each window level and the interval mean, use the centroid method to dynamically calculate the window length.

[0082] The expression for window length is:

[0083] In the formula, l is the window length, ω g Let k be the membership degree of the g-th window level. g Let g be the window mean of the g-th window level, where g∈[1,h] and h is the number of window levels.

[0084] It should be noted that the reason for dynamically setting the window length based on response speed and fluctuation level is as follows: Response speed reflects how quickly the battery reacts to changes in charging current; the faster the response speed, the smaller the window is suitable for quickly capturing changes; the slower the response speed, the larger the window is needed to cover its lag effect; while fluctuation level reflects the drastic change in charging current; the greater the fluctuation level, the smaller the window length needs to be to improve response sensitivity; the smaller the fluctuation level, the larger the window should be to smooth noise; by dynamically setting the window length by combining these two factors, trend changes can be captured more accurately, improving the intelligence and stability of the charging strategy.

[0085] Step S4: Construct a nonlinear weighted fusion model to dynamically evaluate the battery charging status based on real-time state data in the rate of change set and state dataset.

[0086] The nonlinear weight fusion model is a mathematical model that dynamically assigns weights to different input data and combines them using a nonlinear function. The specific construction process includes determining the input variables, designing the weight allocation mechanism, and selecting the weight fusion function, which can be specifically constructed by those skilled in the art according to the actual situation.

[0087] Methods for dynamically assessing battery state of charge include:

[0088] Different digital labels are assigned to different battery types and marked as type labels. Ambient temperature and charging time are acquired; ambient temperature is obtained through a temperature sensor, and charging time is obtained through a timer. Ambient temperature refers to the external air temperature of the space where the charging adapter is located, reflecting the thermal conditions of the current working environment. Charging time is the total time elapsed from the start of battery charging to the current moment. The type labels, ambient temperature, and charging time are used as analysis data and input into a trained weight allocation model to predict the corresponding weight set. The weight allocation model is a deep neural network model, and the weight set includes the weight coefficients corresponding to each data point in the battery state data, which includes voltage change rate, current change rate, and charging state data.

[0089] Real-time state data is obtained from the state dataset, which represents the charging state data at the current moment. The voltage and current change rates corresponding to the real-time state data are obtained from the change rate set and used together with the real-time state data. The corresponding real-time change rate is calculated for each data point in the real-time data, and the total change rate is summed. Based on the real-time change rate and the total change rate, the weight coefficients corresponding to each data point in the real-time data are corrected to obtain corrected weight coefficients. Those skilled in the art, considering battery characteristics and actual operating experience, pre-define a weight fusion function. The real-time data and the corresponding corrected weight coefficients are sequentially substituted into the weight fusion function to calculate the battery charging state.

[0090] Among them, the battery charging status is a comprehensive indicator of the battery's current charging process, reflecting the battery's health and performance during charging; the real-time change rate is the rate of change of each data point in the real-time data compared to the previous moment; the total change rate is the sum of all real-time change rates; the expression for the corrected weighting coefficient is: In the formula, δ new To correct the weighting coefficients, δ is the weighting coefficient, α is the adjustment coefficient used to control the correction magnitude, and it is set by those skilled in the art according to the actual situation. p is the ratio of the real-time rate of change to the total rate of change. The average percentage is the mean of the ratios of all real-time change rates to the total change rate.

[0091] Step S5: Obtain the ambient temperature and, in conjunction with real-time status data, dynamically adjust the set of switching parameters to generate a set of switching thresholds.

[0092] The steps for generating the set of switching thresholds include:

[0093] Step S501: Preset threshold range, and calculate parameter adjustment range based on threshold range and switching parameter set; wherein, threshold range is preset by those skilled in the art according to actual situation, threshold range includes value range corresponding to each switching parameter, and parameter adjustment range includes adjustment range corresponding to each switching parameter;

[0094] Step S502: Construct n adjustment sets based on the parameter adjustment range, where n is an integer greater than 1;

[0095] Step S503: Randomly select m adjustment sets from n adjustment sets, and treat all m adjustment sets as particles to construct a particle population, and set the number of iterations to 0;

[0096] Step S504: Construct the potential function and calculate the potential value corresponding to each particle in the particle population;

[0097] Step S505: Based on the potential value, divide the particle population into multiple particle subpopulations;

[0098] Step S506: Set a corresponding update mechanism for each particle subpopulation;

[0099] Step S507: Update the particles in each particle subpopulation according to the update mechanism;

[0100] Step S508: Compare the number of iterations with the preset iteration threshold. If the number of iterations is less than the iteration threshold, merge the particles in all particle subpopulations into a single particle population, increment the number of iterations by one, and return to step S504. If the number of iterations is greater than or equal to the iteration threshold, proceed to step S509.

[0101] Step S509: Obtain the optimal particle in the particle swarm, mark the adjustment set corresponding to the optimal particle as the optimal set, dynamically adjust the switching parameter set based on the optimal set, and generate a switching threshold set.

[0102] In step S501 above, the method for calculating the parameter adjustment range includes:

[0103] Subtract the maximum and minimum values ​​in the value range of each switching parameter in the threshold range from the corresponding switching parameter in the switching parameter set to obtain the adjustment range for each switching parameter, and form the parameter adjustment range; for example, the threshold range is [5,10], the switching parameter is 6, so the adjustment range is [-1,4].

[0104] In step S502 above, the method for constructing n sets of adjustment sets based on the parameter adjustment range includes:

[0105] Randomly select a value from each adjustment range and construct an adjustment set; continue in this manner to construct a total of n adjustment sets, all of which are different.

[0106] In step S504 above, the expression for the potential function is:

[0107]

[0108] In the formula, V i Let lb be the electric potential value of the i-th particle. i yb represents the degree of current variation of the i-th particle. i cd represents the voltage variation of the i-th particle. i Let be the charging efficiency of the i-th particle, i∈[1,m]. These are all proportional coefficients pre-set by those skilled in the art based on actual circumstances.

[0109] Methods for obtaining the degree of current variation, voltage variation, and charging efficiency include:

[0110] Based on the adjustment set corresponding to the particle, the switching parameter set is dynamically adjusted to generate a candidate threshold set; the ambient temperature, real-time status data and candidate threshold set are sequentially input into the trained index prediction model to predict the corresponding index set; the index prediction model is a deep neural network model, and the index set includes the degree of current variation, the degree of voltage variation and charging efficiency.

[0111] Among them, the degree of current variation and the degree of voltage variation are the fluctuations of charging current and charging voltage after the mode switching is performed based on the adjustment set corresponding to the i-th particle and the switching parameter set is dynamically adjusted according to the candidate threshold set; the charging efficiency is the time required for the battery to be fully charged after the mode switching is performed according to the candidate threshold set.

[0112] In step S505 above, the method for dividing the particle population into multiple particle subpopulations includes:

[0113] The particles in the particle population are sorted from largest to smallest according to their corresponding potential values ​​to generate a particle sequence. A preset division standard is used, which includes two incremental division coefficients, both of which take values ​​in the range (0,1). The division standard is preset by those skilled in the art based on the actual situation. The products of the two division coefficients and m are used as the first division quantity and the second division quantity, respectively. That is, the product of the larger division coefficient and m is the first division quantity, and the product of the smaller division coefficient and m is the second division quantity. According to the ascending order of the particle sequence, each particle is assigned a corresponding sequence number, with the sequence number ranging from [1,m]. According to the sequence number of each particle, the first particle and the second particle are determined. The particles before the first particle and the first particle are classified as the superior population. The particles before the second particle and after the first particle and the second particle are classified as the ordinary population. The particles after the second particle are classified as the inferior population.

[0114] The method for determining the first particle and the second particle is as follows: subtract the first division number and the second division number from the index of each particle, and take the absolute value of each, to obtain the difference between the first index and the second index for each particle; the particle with the smallest difference between the first index and the particle with the smallest difference between the second index and the particle with the smallest difference between the second index and the particle are taken as the first particle.

[0115] Step S506 above: The method for setting a corresponding update mechanism for each particle subpopulation includes:

[0116] When the particle subpopulation is a high-quality population, the corresponding update mechanism is as follows:

[0117]

[0118] In the formula, x i (t+1) represents the updated i-th particle, x i (t) represents the i-th particle, x i (t-1) represents the i-th particle before the last update. η is the gradient of the potential function, η is the convergence factor used to control the convergence speed, and β is the momentum factor used to avoid oscillations. Both the convergence factor and the momentum factor are preset by those skilled in the art according to the actual situation. The gradient of the potential function is obtained by using the numerical difference method on the potential function. The numerical difference method is an existing technology, and the specific process will not be described in detail here.

[0119] When the particle subpopulation is a normal population, the corresponding update mechanism is as follows:

[0120]

[0121] In the formula, F i Let F be the resultant electric field force of the i-th particle.i ‖ represents the magnitude of the resultant electric field force of the i-th particle. γ is the normalized direction vector of the i-th particle, used to ensure that the movement direction is correct and the step size is the same each time. γ is the step size factor, σ is the perturbation factor, and N(0,1) is a random number in the standard normal distribution. The step size factor and perturbation factor are preset by those skilled in the art according to the actual situation.

[0122] The resultant force F of the electric field i The expression is:

[0123] In the formula, θ ij V is the weighting factor between the i-th particle and the j-th particle. j Let x be the electric potential of the j-th particle. j -x i ‖ represents the Euclidean distance between the i-th particle and the j-th particle, x j (t) represents the j-th particle;

[0124] Weighting factor θ ij The expression is:

[0125] In the formula, e is the natural constant, and ε is the decay factor, which is used to control the decay rate and determines the strength of the influence of the Euclidean distance on the weighting factor.

[0126] When the particle subpopulation is a low-quality population, the corresponding update mechanism is as follows:

[0127] x i (t+1)=x i (t)+λ×Levy(q);

[0128] In the formula, λ is the scaling factor, Levy(q) is the Levy distribution, q is the stability index used to control the tail thickness of the Levy distribution, q∈(0,2], and the scaling factor is preset by those skilled in the art according to the actual situation.

[0129] In step S509 above, the optimal particle in the particle swarm is the particle with the largest potential value in the particle swarm; the method for dynamically adjusting the switching parameter set based on the optimal set is as follows: add the corresponding switching parameter in the switching parameter set to each value in the optimal set, obtain the switching threshold corresponding to each switching parameter, and form a switching threshold set.

[0130] It should be noted that the purpose of using the improved swarm intelligence optimization algorithm to dynamically adjust the switching parameter set is as follows: By constructing a population containing multiple particles, the particles are comprehensively evaluated using the potential function, and the particles are divided into high-quality, ordinary, and low-quality subpopulations based on their performance. For different subpopulations, various update mechanisms such as gradient descent, electric field driving, and stable distribution perturbation are adopted to fully leverage the combination of global search and local optimization capabilities. This not only effectively finds the optimal adjustment set and improves the accuracy of the switching threshold, but also enhances the robustness and anti-disturbance capability of the swarm intelligence optimization algorithm, adapts to the dynamic changes in ambient temperature and real-time charging status, ensures a smooth and stable charging mode switching process, avoids excessive current and voltage fluctuations and reduced charging efficiency, and ultimately achieves safe, efficient, and intelligent charging process management.

[0131] Step S6: Combine current change rate, battery charging status and real-time status data, and perform multi-dimensional comparative analysis with the switching threshold set to determine whether to switch modes.

[0132] Methods for determining whether to switch modes include:

[0133] The rate of change of current in the real-time data is marked as the real-time rate of change. The real-time rate of change is compared with the rate of change threshold in the switching threshold set. If the real-time rate of change is less than the rate of change threshold, the change duration is calculated according to the rate of change set. If the real-time rate of change is greater than or equal to the rate of change threshold, no mode switching is performed. The change duration, battery charging status, and charging voltage and battery surface temperature in the real-time status data are used as judgment data. Each data in the judgment data is compared with the corresponding switching threshold in the switching threshold set. If the battery surface temperature is less than the corresponding switching threshold, and each of the remaining data in the judgment data is greater than or equal to the corresponding switching threshold, a mode switch is performed. If the battery surface temperature is greater than or equal to the corresponding switching threshold, or if there are any remaining data in the judgment data that are less than the corresponding switching threshold, no mode switching is performed.

[0134] Methods for calculating the duration of change based on a set of rates of change include:

[0135] Based on the ascending order of the rate of change sequence, the current rate of change in the rate of change set is sorted to generate an electrical change sequence; the real-time electrical change rate in the electrical change sequence is selected as the starting point, and the electrical change sequence is traversed in reverse; if the current rate of change is less than the rate of change threshold, it is added to the pre-constructed pass set; if the current rate of change is greater than or equal to the rate of change threshold, the reverse traversal of the electrical change sequence is terminated; the number of current rate of change in the pass set is counted and marked as the number of electrical changes; the number of electrical changes, the window length d, and the sampling interval are multiplied in sequence to obtain the change duration.

[0136] Step S7: If a mode switch is to be performed, the optimal transition curve is calculated using a gradient transition strategy. Based on the optimal transition curve, the charging current in the charging status data is gradually adjusted to complete the charging mode switch.

[0137] Methods for calculating the optimal transition curve include:

[0138] A set of parameters is defined, including the starting current, ending current, transition time, time step (the time interval between two current control points after discretizing continuous time, indicating how often the current is controlled), current change rate limit (the maximum value that the current change rate can reach), and current range limit (the maximum and minimum values ​​that the charging current can reach). Among them, the starting current is the charging current in the real-time status data, and the ending current, transition time, current change rate limit, and current range limit are all preset by those skilled in the art according to the actual situation.

[0139] A smoothing objective function is constructed to optimize the charging current at each current control point while satisfying a set of parameters, so that the transition process is as smooth as possible; the expression of the smoothing objective function is:

[0140]

[0141] In the formula, Y is the smoothing objective function value, used to measure the smoothness of the current change curve. The smaller the value, the smoother the current change. N is the number of current control points, that is, how many current control points the entire transition time is divided into according to the time step. For example, if the transition time is 5s and the time step is 1s, then there are 6 current control points. K Let Δt be the charging current at the Kth current control point, and Δt be the time step.

[0142] Based on the set of parameters and the smoothing objective function, the gradient descent method is used to dynamically adjust the charging current at each current control point to obtain the optimal transition curve that satisfies the set of parameters and minimizes the smoothing objective function. The gradient descent method is an existing technology, and the specific process will not be described in detail here.

[0143] During the process of gradually adjusting the charging current in the charging state data based on the optimal transition curve, sliding mode control is used to accurately track and robustly control the charging current, thereby effectively improving the stability and anti-disturbance capability of the charging process, preventing current surges, ensuring battery safety, extending battery life, and achieving smooth switching of charging modes, improving charging efficiency and overall performance. Sliding mode control is an existing technology, and the specific process will not be elaborated on here.

[0144] This embodiment analyzes battery response characteristics using a pulse charging test method, accurately identifying the charging needs of different battery types and automatically matching switching parameter sets to achieve intelligent adaptation to different battery types. During charging, it continuously collects multi-dimensional state data such as charging current, charging voltage, and battery surface temperature, and accurately assesses the real-time charging status of the battery based on dynamic trend analysis and a nonlinear weighted fusion model, providing a reliable basis for subsequent mode switching. By incorporating external factors such as ambient temperature, an improved swarm intelligence optimization algorithm dynamically adjusts the switching parameter set to generate a switching threshold set, significantly improving the dynamics and adaptability of the switching strategy compared to traditional single voltage thresholds. A gradient transition strategy is used to calculate the optimal transition curve, and sliding mode control is used to track and adjust the charging current in real time, achieving smooth switching of charging modes, avoiding damage to the battery from sudden changes in charging current, and improving charging efficiency and safety. This effectively solves the problems of fixed parameters, lag response, and poor safety in traditional charging adapter switching control, improving the applicability and performance of the charging adapter in battery charging scenarios, ensuring the optimal charging state for different battery types, thereby extending battery life and improving the charging experience.

[0145] Example 2

[0146] This application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable code that, when executed by the one or more processors, can perform a dual-mode switching control method for a charging adapter as described above.

[0147] The method or system according to the embodiments of this application can also be implemented using the architecture of the electronic device shown in this application. The electronic device may include a bus, one or more CPUs, ROM, RAM, a communication port connected to a network, input / output, a hard disk, etc. The storage device in the electronic device, such as a ROM or hard disk, may store a dual-mode switching control method for a charging adapter provided in this application. Furthermore, the electronic device may also include a user interface. Of course, the architecture shown in this application is merely exemplary; when implementing different devices, one or more components of the electronic device shown in this application may be omitted according to actual needs.

[0148] Example 3

[0149] Please refer to the accompanying drawings. One embodiment of this application discloses a computer-readable storage medium. The computer-readable storage medium stores computer-readable instructions. When the computer-readable instructions are executed by a processor, a dual-mode switching control method for a charging adapter according to an embodiment of this application, as described above, can be performed. The storage medium 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.

[0150] Furthermore, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, this application provides a non-transitory machine-readable storage medium storing machine-readable instructions that can be executed by a processor to perform instructions corresponding to the method steps provided in this application, such as a dual-mode switching control method for a charging adapter. When this computer program is executed by a central processing unit (CPU), it performs the functions defined in the method of this application.

[0151] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0152] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0153] In the description of this invention, it should be understood that the terms "first," "second," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0154] In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0155] In the description of this invention, "several" means one or more, and "a large number" means two or more.

[0156] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0157] All formulas in this manual are dimensionless and calculated numerically. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0158] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A dual-mode switching control method for a charging adapter, characterized in that, include: Step S1: Use the pulse charging test method to obtain the battery response characteristic curve, identify the battery type based on the battery response characteristic curve, and automatically match and switch the parameter set based on the battery type; Step S2: Continuously collect charging status data according to the preset sampling interval to form a status dataset; Step S3: Perform trend analysis calculations on the state dataset to obtain a set of rates of change, which includes the rate of change of voltage and the rate of change of current. Step S4: Construct a nonlinear weighted fusion model to dynamically evaluate the battery charging status based on real-time state data in the rate of change set and state dataset; Step S5: Obtain the ambient temperature and, in conjunction with real-time status data, dynamically adjust the set of switching parameters to generate a set of switching thresholds; The steps for generating the set of switching thresholds include: Step S501: Preset threshold range, and calculate parameter adjustment range based on threshold range and switching parameter set; Step S502: Construct based on parameter adjustment range Group adjustment set, It is an integer greater than 1; Step S503: From Randomly selected from the group adjustment set Group adjustment set, and All elements in the adjusted set are treated as particles to construct a particle population, and the number of iterations is set to 0. Step S504: Construct the potential function and calculate the potential value corresponding to each particle in the particle swarm; the expression of the potential function is: ； In the formula, For the first The electric potential of each particle, For the first The degree of change in the current of each particle For the first The degree of voltage variation of each particle For the first Charging efficiency of individual particles. , , , All of these are proportional coefficients preset by those skilled in the art based on actual circumstances; Step S505: Based on the potential value, divide the particle population into multiple particle subpopulations; Step S506: Set a corresponding update mechanism for each particle subpopulation; Step S507: Update the particles in each particle subpopulation according to the update mechanism; Step S508: Compare the number of iterations with the preset iteration threshold. If the number of iterations is less than the iteration threshold, merge the particles in all particle subpopulations into a single particle population, increment the number of iterations by one, and return to step S504. If the number of iterations is greater than or equal to the iteration threshold, proceed to step S509. Step S509: Obtain the optimal particle in the particle swarm, mark the adjustment set corresponding to the optimal particle as the optimal set, dynamically adjust the switching parameter set based on the optimal set, and generate a switching threshold set. Step S6: Combine current change rate, battery charging status and real-time status data, and perform multi-dimensional comparative analysis with the switching threshold set to determine whether to switch modes. Step S7: If a mode switch is to be performed, the optimal transition curve is calculated using a gradient transition strategy. Based on the optimal transition curve, the charging current in the charging status data is gradually adjusted to complete the charging mode switch.

2. The dual-mode switching control method for a charging adapter according to claim 1, characterized in that, Methods for identifying battery type based on battery response characteristic curves include: Based on the battery response characteristic curve, a multi-physical quantity coupling model is constructed; the multi-physical quantity coupling model includes Sub-model, The number of curves in the battery response characteristic curve is specified. The battery response characteristic curve is input into a multi-physical quantity coupling model, and a multi-objective optimization method is used to fit the multi-physical quantity coupling model to obtain the model back-inference parameters. A type mapping matrix is ​​preset, which includes the parameter ranges corresponding to different battery types. The parameter ranges include the numerical ranges corresponding to each parameter in the model back-inference parameters. The model back-inference parameters are compared with each parameter range in the type mapping matrix, and the parameter ranges in which each parameter in the model back-inference parameters is within the corresponding numerical range are selected and marked as matching ranges. The battery type corresponding to the matching range is obtained.

3. The dual-mode switching control method for a charging adapter according to claim 2, characterized in that, Methods for obtaining the rate of change of current include: The charging currents in the state dataset are sorted in ascending order according to their corresponding acquisition times to generate a current sequence. The rate of change between every two adjacent charging currents in the current sequence is calculated sequentially, and a rate of change sequence is generated. The window length is dynamically set, and the rate of change sequence is divided into windows based on this window length to obtain... Each rate of change window includes [number] rate of change windows. Adjacent rates of change The window length is defined as follows: For adjacent rates of change within each rate of change window, a linear regression method is used to fit the trend and obtain the corresponding current characteristic line. The slope of the current characteristic line is then used as the current rate of change. The method for obtaining the rate of change of voltage is the same as the method for obtaining the rate of change of current.

4. The dual-mode switching control method for a charging adapter according to claim 3, characterized in that, The steps to dynamically set the window length include: Step S301: Preset a type factor mapping table, which includes different battery types and their corresponding response speeds; based on the identified battery type, obtain the corresponding response speed from the type factor mapping table; Step S302: Calculate the standard deviation of all adjacent rates of change and mark it as the degree of volatility; Step S303: Construct multiple fuzzy sets for response speed and fluctuation degree respectively; Step S304: Convert the response speed and fluctuation degree into the membership degree of each corresponding fuzzy set using fuzzification technology; Step S305: Input all membership degrees into the trained membership analysis model to predict the corresponding membership set; wherein, the membership set includes the membership degree of each window level; Step S306: Set corresponding level intervals for each window level and calculate the corresponding interval mean; based on the membership degree and interval mean of each window level, use the centroid method to dynamically calculate the window length.

5. The dual-mode switching control method for a charging adapter according to claim 4, characterized in that, Methods for dynamically assessing battery state of charge include: Different numerical labels are set for different battery types and marked as type labels; ambient temperature and charging time are obtained, and the type labels, ambient temperature and charging time are used as analysis data. The analysis data is input into the trained weight allocation model to predict the corresponding weight set; where the weight set includes the weight coefficient corresponding to each data in the battery state data, and the battery state data includes voltage change rate, current change rate and charging state data. Real-time state data is obtained from the state dataset, which represents the charging state data at the current moment. The voltage and current change rates corresponding to the real-time state data are obtained from the change rate set and used together with the real-time state data. The corresponding real-time change rate is calculated for each data point in the real-time data, and the total change rate is summed to obtain the total change rate value. Based on the real-time change rate and the total change rate value, the weight coefficients corresponding to each data point in the real-time data are corrected to obtain the corrected weight coefficients. A weight fusion function is preset, and the real-time data and the corresponding corrected weight coefficients are substituted into the weight fusion function in sequence to calculate the battery charging state.

6. The dual-mode switching control method for a charging adapter according to claim 5, characterized in that, Methods for determining whether to switch modes include: The rate of change of current in the real-time data is marked as the real-time rate of change. The real-time rate of change is compared with the rate of change threshold in the switching threshold set. If the real-time rate of change is less than the rate of change threshold, the change duration is calculated according to the rate of change set. If the real-time rate of change is greater than or equal to the rate of change threshold, no mode switching is performed. The change duration, battery charging status, and charging voltage and battery surface temperature in the real-time status data are used as judgment data. Each data in the judgment data is compared with the corresponding switching threshold in the switching threshold set. If the battery surface temperature is less than the corresponding switching threshold, and each of the remaining data in the judgment data is greater than or equal to the corresponding switching threshold, a mode switch is performed. If the battery surface temperature is greater than or equal to the corresponding switching threshold, or if there are any remaining data in the judgment data that are less than the corresponding switching threshold, no mode switching is performed.

7. The dual-mode switching control method for a charging adapter according to claim 6, characterized in that, Methods for calculating the duration of change based on a set of rates of change include: Based on the ascending order of the rate of change sequence, the current rate of change in the rate of change set is sorted to generate an electrical transformer sequence. The real-time electrical rate of change in the electrical transformer sequence is selected as the starting point, and the sequence is traversed in reverse. If the current rate of change is less than a rate of change threshold, it is added to a pre-built pass set; if the current rate of change is greater than or equal to the rate of change threshold, the reverse traversal of the electrical transformer sequence is terminated. The number of current rates of change in the pass set is counted and marked as the number of electrical transformers. Based on the number of electrical transformers and the window length... And the sampling interval, and calculate the duration of change.

8. The dual-mode switching control method for a charging adapter according to claim 7, characterized in that, Methods for calculating the optimal transition curve include: Define the set of parameters, which includes the starting current, ending current, transition time, time step, current change rate limit, and current range limit. A smoothing objective function is constructed. Based on the set of parameters and the smoothing objective function, the gradient descent method is used to dynamically adjust the charging current at each current control point to obtain the optimal transition curve. In the process of gradually adjusting the charging current in the charging state data based on the optimal transition curve, sliding mode control is used to track and control the charging current.

Images