A method for optimizing charging parameters of lead-acid storage batteries under a parked air-conditioning working condition and application thereof

Abstract

The application relates to a method for optimizing charging parameters of a lead-acid storage battery under a parked air conditioner working condition and application, and the method comprises the following steps: mapping measurable parameters in actual operation working conditions of a vehicle into controllable factors in experiments, constructing a multi-round full-factor and response surface experiment, collecting time length data of discharging of the battery to a specified voltage under the Nth cycle; establishing a quadratic regression mathematical model with the factors as input and the discharging time length as response, wherein the model comprises interaction terms between key factors; solving the optimal factor combination for maximizing the discharging time length through model effect analysis and a response optimizer; and adopting a two-factor two-level center point DOE design to quantitatively evaluate the battery water loss risk under the optimized combination. The application solves the problems of insufficient charging and fast life attenuation of the parked air conditioner battery under a real use scene, and significantly improves the deep cycle available capacity and service life of the battery.

Description

Technical Field

[0001] This invention belongs to the field of lead-acid battery application technology, specifically relating to an optimization method for the charging parameters of lead-acid batteries for commercial vehicles (such as trucks and RVs) equipped with parking air conditioning, and the application of adjusting vehicle charging strategies based on this method. Background Technology

[0002] With the rapid development of the logistics and transportation industry and RV self-driving tours, parking air conditioners are becoming increasingly common in vehicles such as trucks and RVs. Parking air conditioners rely on the vehicle's battery for power when the engine is off, which places much higher demands on the battery's deep-cycle discharge capability than ordinary starter batteries.

[0003] Traditional vehicle charging strategies are designed to meet the short-term, high-current discharge demands of engine starting, typically employing a fixed voltage regulation (e.g., 14.1V). This strategy suffers from the following technical drawbacks when dealing with deep-cycle discharge conditions caused by parking air conditioning:

[0004] (1) Charging parameters are disconnected from actual driving conditions: Existing technologies often test in the laboratory with fixed charging time and fixed voltage and current, without effectively linking the charging time with the actual driving time of the vehicle, the charging voltage and current with the actual output characteristics of the generator, and the discharge cut-off voltage with the air conditioning protection voltage. This makes it difficult for the experimental optimization results to directly guide the adjustment of actual vehicle parameters.

[0005] (2) Lack of systematic analysis of multi-factor interaction: In existing studies, the effects of charging voltage, charging current, charging time and cutoff voltage on battery discharge time are often analyzed in isolation, ignoring the interaction effects between factors (such as the nonlinear effect of voltage and current on charging efficiency), which leads to deviations in optimization strategies.

[0006] (3) Lack of quantitative basis for balancing performance improvement and side effects: Although simply increasing the charging voltage can increase battery capacity, it will accelerate battery water loss and shorten lifespan. Existing technologies lack quantitative assessment models for the risk of water loss after increasing voltage, making it difficult to find the optimal balance between performance and lifespan.

[0007] (4) Existing patents (such as CN114861527A, CN116893343A, etc.) mainly involve lithium battery life prediction or general testing methods. They do not provide a complete DOE analysis and response surface modeling scheme for the optimization of charging parameters of lead-acid batteries under parking air conditioning conditions. Moreover, their technical paths mostly use complex algorithms such as neural networks and time series, which are not convenient for engineering applications. Summary of the Invention

[0008] The purpose of this invention is to solve the problems existing in the prior art and provide a method for optimizing the charging parameters of lead-acid batteries under parking air conditioning conditions. The method aims to map experimental factors with actual vehicle operating parameters, establish a high-precision prediction model using the Response Surface (DOE) method, solve for the optimal combination of charging parameters that maximizes the usable time of the parking air conditioning, and assess its water loss risk, thereby effectively improving the user experience and cycle life of the battery under deep cycle conditions.

[0009] The present invention also aims to provide an application of a method for adjusting vehicle charging strategies based on the optimization of lead-acid battery charging parameters under parking air conditioning conditions.

[0010] To achieve the above objectives, the present invention provides the following technical solution:

[0011] A method for optimizing the charging parameters of a lead-acid battery under parking air conditioning conditions includes the following steps:

[0012] Step S1: Operating Condition Factor Mapping and Experimental Design

[0013] Measurable parameters in the actual operating conditions of the vehicle are mapped to experimental controllable factors. Based on the experimental controllable factors, experimental parameters are designed in DOE to simulate the constant current discharge condition of the parking air conditioner. Multiple rounds of tests are conducted according to the designed parameters to obtain the response data of the lead-acid battery after cyclic charging and discharging. The response data includes the first duration of the battery discharging to the first intermediate voltage at a constant current in the Nth cycle, and the second duration of discharging to the undervoltage protection threshold in the same cycle.

[0014] Step S2: Mathematical Model Construction and Effect Analysis

[0015] Based on the response data, regression analysis was used to establish response surface mathematical models with the first duration and the second duration as response variables and the experimental controllable factors as input variables, and the main effect value and interaction effect value of each experimental controllable factor were calculated to identify key factors that have statistical significance for discharge duration.

[0016] Step S3: Optimal operating condition parameter optimization

[0017] With the goal of maximizing the first duration and the second duration, a response optimizer is used to numerically optimize the mathematical model of the response surface, and the optimal combination of operating conditions that maximizes the discharge duration is output.

[0018] Step S4: Supplementary assessment of water loss risk

[0019] Based on the optimal combination of operating conditions, the cumulative water loss of the battery under a specified number of cycles is calculated using a pre-established water loss rate model, and it is determined whether the water loss is below a preset water loss threshold.

[0020] In step S1, the controllable factors of the experiment include equivalent driving time, generator output voltage, generator output current, and air conditioning compressor undervoltage protection threshold.

[0021] The equivalent driving time corresponds to the actual time spent charging the battery during the vehicle's engine operation. Its value is calibrated by dividing the product of the vehicle's actual driving mileage and the preset charging efficiency coefficient by the average vehicle speed. The calibration formula is: t_charge = (D / v) × η, where D is the vehicle's single driving mileage, v is the average vehicle speed, and η is the charging efficiency coefficient.

[0022] The generator output voltage and generator output current correspond to the charging voltage value and the output current value at typical operating speed of the vehicle's alternator regulator, respectively.

[0023] The air conditioning compressor undervoltage protection threshold corresponds to the shutdown voltage value set by the parking air conditioning system to prevent over-discharge of the battery.

[0024] In step S1, the experimental parameter design includes full factorial design and response surface design.

[0025] In step S1, the constant current is 25-30A; N in the Nth cycle is 45-55; the first intermediate voltage is 11.8-12.2V, used to characterize the degree of battery degradation; the undervoltage protection threshold voltage is determined according to the actual setting of the vehicle air conditioning system, and is 10.8-12.2V.

[0026] In step S2, regression analysis is used to establish response surface mathematical models with the first duration and the second duration as response variables and the experimental controllable factor as input variables. These models include a first duration response surface mathematical model with the first duration as the response variable and the experimental controllable factor as the input variable, and a second response surface mathematical model with the second duration as the response variable and the experimental controllable factor as the input variable.

[0027] The response surface mathematical model is a quadratic polynomial regression equation. This equation includes at least quadratic terms for each factor, as well as interaction terms between generator output voltage and generator output current, and between generator output voltage and equivalent driving time. The first-time response surface mathematical model is expressed as follows:

[0028] T1_discharge = A9 + A10·V_charge + A11·I_charge + A12·t_charge +A13·V_charge² + A14·I_charge 2 ;

[0029] Where T1_discharge is the first duration in minutes; V_charge is the generator output voltage in volts; I_charge is the generator output current in amperes; t_charge is the equivalent driving time in hours; A9~A14 are model coefficients determined through regression analysis;

[0030] The mathematical model of the second response surface is expressed as follows:

[0031] T_discharge = A0 + A1·V_cutoff + A2·V_charge + A3·I_charge + A4·t_charge + A5·V_cutoff² + A6·V_cutoff·V_charge + A7·V_charge·I_charge +A8·V_charge·t_charge;

[0032] Where T_discharge is the second duration in minutes; V_cutoff is the undervoltage protection threshold in volts; V_charge is the generator output voltage in volts; I_charge is the generator output current in amperes; t_charge is the equivalent driving time in hours; and A0~A8 are model coefficients determined through regression analysis.

[0033] In step S2, the main effect value and interaction effect value of each experimental controllable factor are calculated, and key factors with statistical significance for discharge duration are identified, specifically:

[0034] The effect size and p-value of each controllable factor in the experiment were calculated by analysis of variance. The absolute value of the effect size was used to characterize the influence of each factor on the discharge duration. Factors with p-values ​​less than 0.05 were identified as key factors with statistical significance.

[0035] In step S3, with the optimization objective of maximizing the first duration and the second duration, a response optimizer is used to numerically optimize the response surface mathematical model, specifically including:

[0036] Set constraints on the range of values ​​for each controllable factor in the experiment, use the gradient descent algorithm to search for the factor combination that maximizes the predicted values ​​of the first and second durations within the constraint space, output the theoretical optimal solution with a satisfactoryness reaching a preset threshold (e.g., above 0.95), and make robust adjustments to the theoretical optimal solution based on the engineering margin to finally obtain the optimal combination of working condition factors.

[0037] The water loss rate model in step S4 is as follows:

[0038] W_total=b0+b1·t_charge+b2·V_charge+b12·(t_charge·V_charge);

[0039] Where W_total is the cumulative water loss, t_charge is the equivalent driving time, V_charge is the generator output voltage, and b0, b1, b2, b12 are the regression coefficients calibrated by a two-factor, two-level, center-point DOE experiment.

[0040] A method for adjusting a vehicle charging strategy, which applies the optimal combination of operating condition factors obtained by the optimization method described in the above technical solution, and performs at least one of the following adjustments based on the combination:

[0041] Based on the optimal generator output voltage value, the voltage regulator of the vehicle-mounted generator is recalibrated;

[0042] Based on the optimal generator output current value, configure the generator excitation current or the output current of the external DC-DC converter;

[0043] Set the undervoltage protection threshold of the vehicle air conditioning system to the value in the optimal operating condition factor combination or keep the original factory setting.

[0044] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for optimizing the charging parameters of a lead-acid battery under parking air conditioning conditions as described in the above technical solution.

[0045] Beneficial effects

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] (1) Engineering practicality of working condition mapping: For the first time, a one-to-one mapping relationship is established between the charging parameters (voltage, current, time, cut-off voltage) in the laboratory and the measurable parameters (generator output voltage / current, driving time, air conditioning protection voltage) in the actual driving conditions of the vehicle, and a quantitative calibration formula (including the charging efficiency coefficient η) is given, so that the optimization results can be directly used for vehicle generator regulator calibration and air conditioning system parameter setting, avoiding the disconnect between laboratory optimization and actual vehicle application.

[0048] (2) High-precision response surface model: Through a two-stage experimental design of full factor + response surface, a quadratic regression model of discharge duration was established. The model includes key interaction terms (voltage × current, voltage × time), which can accurately predict the available time of parking air conditioner under different driving durations and generator output characteristics, providing a quantitative design basis for vehicle manufacturers.

[0049] (3) Systematic identification of multi-factor interaction: Not only were the main effects of single factors identified, but the interaction effects between factors were also quantitatively given, revealing the limitations of increasing voltage or current alone, and guiding the adoption of a strategy of co-optimization of voltage and time.

[0050] (4) Quantification of performance and lifespan balance: A water loss rate model was established using a two-factor, two-level, and center-point DOE design. The increase in water loss rate (2.44%) caused by each 0.1V increase in charging voltage was quantitatively given. This proves that the water loss rate under the optimized high-voltage charging strategy (14.4V~14.7V) still meets the requirements of users for more than 2 years of normal use, and achieves an acceptable balance between performance improvement and lifespan loss.

[0051] (5) Engineering simplicity by avoiding complex algorithms: This invention adopts the classic DOE and response surface method, which does not rely on black box models such as neural networks and deep learning. It has a small amount of computation and is easy to implement in vehicle controllers or calibration tools, which facilitates engineering promotion and application. Attached Figure Description

[0052] Figure 1 The Pareto effect diagram shows the time it takes to discharge to 12V.

[0053] Figure 2 Pareto effect diagram showing the time it takes for the discharge to reach the undervoltage protection threshold;

[0054] Figure 3 This is a diagram of the response optimizer interface for discharging to the first duration.

[0055] Figure 4 The interface diagram of the response optimizer is shown for the second discharge duration;

[0056] Figure 5 Pareto effect diagram of water loss (based on a two-factor, two-level, center-point DOE experiment);

[0057] Figure 6 This is a surface plot showing how the second duration changes with the charging voltage and discharge cutoff voltage after fixing the charging time and charging current.

[0058] Figure 7 This is a surface plot showing the variation of the second duration with charging current and discharge cutoff voltage after fixing the charging voltage and charging time.

[0059] Figure 8 This is a surface plot showing the change of the second duration with charging time and discharge cutoff voltage after fixing the charging voltage and charging current.

[0060] Figure 9 This is a surface plot showing the variation of the second duration with charging voltage and charging current after fixing the charging time and discharge cutoff voltage.

[0061] Figure 10 This is a surface plot showing the change of the second duration with charging voltage and charging time after fixing the charging current and discharge cutoff voltage.

[0062] Figure 11 This is a surface plot showing the variation of the second duration with charging current and charging time after fixing the charging voltage and discharge cutoff voltage. Detailed Implementation

[0063] This invention provides a method for optimizing the charging parameters of a lead-acid battery under parking air conditioning conditions, comprising the following steps:

[0064] S1: Operating Condition Factor Mapping and Experimental Design

[0065] First, the measurable parameters in the actual operating conditions of the vehicle are mapped to experimentally controllable factors. The specific mapping relationship (Table 1) is as follows:

[0066] (1) Equivalent driving time (t_charge): The actual charging time for the battery during the vehicle's engine operation, expressed in hours. Its calibration formula is: t_charge = (D / v) × η, where D is the vehicle's single-trip mileage (km), v is the average vehicle speed (km / h), and η is the charging efficiency coefficient, ranging from 0.7 to 0.9. For example, when D = 200km, v = 80km / h, and η = 0.8, the equivalent driving time is 2 hours.

[0067] (2) Generator output voltage (V_charge): This corresponds to the charging voltage output by the vehicle's alternator regulator (the charging voltage of the battery during vehicle engine operation), measured in volts. Its typical range is between 13.8V and 15.1V.

[0068] (3) Generator output current (I_charge): This corresponds to the current value that the generator can continuously output at a typical cruising speed (e.g., 1500 rpm) (the typical cruising speed of the vehicle engine is the charging current provided by the battery), in amperes. Its typical range is between 40A and 140A. In practical applications, the maximum charging current can be limited to this optimized value by configuring a DC-DC converter, or the characteristic curve can be calibrated by using a generator voltage regulator.

[0069] (4) Air conditioning compressor undervoltage protection threshold (V_cutoff): This corresponds to the shutdown voltage value (i.e., discharge cutoff voltage) set by the parking air conditioning system to prevent over-discharge of the battery, in volts. Common settings are 11.0V and 11.5V.

[0070] Table 1: Mapping Relationship of Operating Condition Factors

[0071]

[0072] As shown in Table 1, this invention converts measurable parameters from actual vehicle operating conditions into experimentally controllable factors through mapping relationships. Specifically, the equivalent driving time is calibrated using mileage, average vehicle speed, and the charging efficiency coefficient η; the generator output voltage / current directly corresponds to the actual vehicle regulator output value and generator output capacity; and the undervoltage protection threshold directly corresponds to the air conditioning system setting. η ranges from 0.7 to 0.9 and is calibrated through actual measurements.

[0073] Based on the controllable factors of the experiment, multiple rounds of experimental parameters were designed in DOE to simulate the constant current discharge condition of the parking air conditioner and obtain the response data of the lead-acid battery after cyclic charging and discharging.

[0074] The response data includes the first duration for which the battery discharges at a constant current to the first intermediate voltage in the Nth cycle, and the second duration for which it continues to discharge to the undervoltage protection threshold in the same cycle;

[0075] Discharge time to the first intermediate voltage (first duration): The time required for the battery voltage to drop from a full charge (approximately 12.8V or higher) to 11.8-12.2V when discharged at a constant current of 25-30A. This metric characterizes the degree of battery degradation during deep cycle use; a shorter duration indicates more severe degradation.

[0076] Discharge time to undervoltage protection threshold (second duration): Discharge at a constant current of 25-30A and record the total time required for the battery voltage to drop from full charge to the set undervoltage protection threshold (e.g., 10.8-12.2V). This metric directly corresponds to the actual continuous duration of the user's use of the parking air conditioner.

[0077] The experimental parameters were designed in two stages:

[0078] Phase 1: Full factorial experimental design, used for preliminary screening of key factors and exploration of linear response regions.

[0079] The second stage involves response surface design (such as central composite design or Box-Behnken design) to fit the secondary and interaction effects of the factors and find the extreme regions of the response.

[0080] The experiment is conducted in the following cycle: discharge at a constant current of 27A to the specified undervoltage protection threshold (first record the first duration when the voltage reaches the first intermediate voltage, then record the second duration when the voltage reaches the undervoltage protection threshold) → charge the generator according to the generator output voltage (charging voltage), generator output current (charging current), and equivalent driving time set in the experimental setup (charging time) → repeat the above process until the 50th cycle, and record the response data of the 50th cycle.

[0081] Phase 1: Detailed design and results of the full-factor experiment

[0082]

[0083] Detailed experimental scheme and results of the second-stage response surface design phase.

[0084]

[0085] S2: Mathematical Model Construction and Effect Analysis

[0086] Regression analysis was performed on the response data of the 50th cycle after the response surface design experiment using statistical software (such as Minitab). Response surface mathematical models were established with the first and second durations as response variables and V_cutoff, V_charge, I_charge, and t_charge as input variables. The response surface mathematical models were quadratic polynomial regression models.

[0087] The mathematical model of the first-duration response surface, with the first duration (T1_discharge) as the response variable and V_cutoff, V_charge, I_charge, and t_charge as input variables, is as follows:

[0088] T1_discharge = A9 + A10·V_charge + A11·I_charge + A12·t_charge +A13·V_charge² + A14·I_charge 2 ;

[0089] Where T1_discharge is the first duration and T_discharge is the second duration, both in minutes; V_cutoff is the undervoltage protection threshold, in volts; V_charge is the generator output voltage, in volts; I_charge is the generator output current, in amperes; t_charge is the equivalent driving time, in hours; and A0~A14 are model coefficients determined through regression analysis.

[0090] The mathematical model of the first-duration response surface, with the second duration (T_discharge) as the response variable and V_cutoff, V_charge, I_charge, and t_charge as input variables, is as follows:

[0091] T_discharge = A0 + A1·V_cutoff + A2·V_charge + A3·I_charge + A4·t_charge + A5·V_cutoff² + A6·V_cutoff·V_charge + A7·V_charge·I_charge +A8·V_charge·t_charge;

[0092] The effect size and p-value of each controllable factor in the experiment were calculated using analysis of variance (ANOVA). The absolute value of the effect size characterizes the degree of influence of each factor on the discharge duration. The p-value is the probability of the observed value or a more extreme value of the test statistic, assuming the null hypothesis (H0) is true. Factors with p-values ​​less than 0.05 were identified as key factors with statistical significance. Typical results show that the equivalent driving time (t_charge) has the largest absolute effect size, followed by the undervoltage protection threshold (V_cutoff), generator output current (I_charge), and generator output voltage (V_charge).

[0093] S3: Optimal operating condition parameter optimization

[0094] The optimization objective is set as "maximizing the discharge time to the first intermediate voltage (first duration) and the undervoltage protection threshold time (second duration)". The response optimizer is used to optimize the mathematical model of the response surface within the range of experimentally controllable factors. Specifically:

[0095] Set constraints on the range of values ​​for each controllable factor in the experiment, use the gradient descent algorithm to search for the factor combination that maximizes the predicted values ​​of the first and second durations within the constraint space, output the theoretical optimal solution with a satisfactoryness reaching a preset threshold (e.g., above 0.95), and make robust adjustments to the theoretical optimal solution based on the engineering margin to finally obtain the optimal combination of working condition factors.

[0096] The response optimizer synthesizes multiple objective responses using a desirability function, outputting a combination of factors that maximizes the predicted response value. A typical optimal solution is: equivalent driving time 4.5~5.5h, generator output voltage 14.4V~14.7V, generator output current 90A~130A, and undervoltage protection threshold set to 11V.

[0097] S4: Supplementary Assessment of Water Loss Risk

[0098] Based on the optimal combination of operating conditions, the cumulative water loss of the battery under a specified number of cycles is calculated using a pre-established water loss rate model, and it is determined whether the water loss is lower than the preset water loss threshold (the electrolyte level is not lower than the plate height).

[0099] To assess the water loss side effects of increasing voltage, a water loss rate model was established:

[0100] A two-factor, two-level, center-point DOE design was adopted:

[0101] Fixed conditions: undervoltage protection threshold 11V, generator output current 80A;

[0102] Charging time factor: 2h for low level, 6h for high level, and 4h for center point;

[0103] Charging voltage factor: 14V for low level, 14.8V for high level, and 14.4V for center point;

[0104] Weigh the battery every 50 cycles, calculate the cumulative water loss, and establish a water loss rate model (the obtained water loss data is input into Minitab software, and the regression equation for the water loss rate model is as follows):

[0105] Wtotal=b0+b1·t_charge+b2·V_charge+b12·(t_charge·V_charge);;

[0106] Where Wtotal is the cumulative water loss, t_charge is the equivalent driving time, V_charge is the generator output voltage, and b0, b1, b2, b12 are the regression coefficients calibrated by a two-factor, two-level, center-point DOE experiment.

[0107] The optimal factor combination obtained from S3 was input into the water loss rate model analysis results, which show:

[0108] With a fixed charging time of 6 hours, a charging current of 80A, and 150 cycles, the water loss rate increases by approximately 2.44% for every 0.1V increase in charging voltage.

[0109] Charging at 14.4V: Average water loss is 0.59g per hour, and water loss per cell is 0.098g / H;

[0110] The total amount of electrolyte (from the liquid level to the upper edge of the plate) is 1150g.

[0111] Based on 14 hours of vehicle operation per day, the battery can last for 27 months; similarly, it can last for 30 months with 14V charging and 25 months with 14.8V charging.

[0112] In summary, the cumulative water loss under the optimal combination of operating conditions is lower than the preset water loss threshold, proving that the water loss risk is controllable and meets the user's requirement of normal use for more than 2 years.

[0113] The present invention also provides a method for adjusting a vehicle charging strategy, which applies the optimal combination of operating condition factors obtained by the optimization method described above, and performs at least one of the following adjustments based on the combination:

[0114] Based on the optimal generator output voltage value, the voltage regulator of the vehicle-mounted generator is recalibrated;

[0115] Based on the optimal generator output current value, configure the generator excitation current or the output current of the external DC-DC converter;

[0116] Set the undervoltage protection threshold of the vehicle air conditioning system to the value in the optimal operating condition factor combination or keep the original factory setting.

[0117] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the above technical solution.

[0118] Example 1

[0119] Experimental Design and Data Acquisition

[0120] A 12V 220Ah flooded lead-acid starting battery of a certain model was selected as the test object. The 27A constant current discharge condition of the parking air conditioner was simulated. The experimental factors and their levels were set according to common actual vehicle parameters as follows:

[0121]

[0122] The experiment was conducted using full factorial design (resolution V) and response surface design (central composite design). Each round of the experiment was repeated 50 times. The duration of discharge to 12V (first duration) and the duration of discharge to undervoltage protection threshold (second duration) were recorded in the 50th cycle.

[0123] Model building and effect analysis

[0124] Using "duration of discharge to 12V" (first duration) as the response variable and V_charge, I_charge, and t_charge as independent variables, a response surface regression analysis was performed. The resulting regression equation, expressed in uncoded units (typical example):

[0125] T1_discharge = -13837 + 1883·V_charge + 1.757·I_charge + 16.49·t_charge - 64·V_charge² - 0.00921·I_charge 2

[0126] like Figure 1The Pareto effect plot for discharge to 12V was obtained. Analysis of variance showed that the model p-value was <0.001, R-sq(adjusted) = 70.67%, and R-sq(predicted) = 60.59%, indicating model validity. Effects analysis showed that the standardized effect absolute values ​​were ranked as follows: t_charge (5.54) > V_charge (4.40) > I_charge (2.55) > V_charge² (2.43) > I_charge 2 (2.24).

[0127] Using "duration of discharge to undervoltage protection threshold" (second duration) as the response variable and V_cutoff, V_charge, I_charge, and t_charge as independent variables, response surface regression analysis was performed. The resulting regression equation, expressed in uncoded units (typical example):

[0128] T_discharge = -32255 + 4480·V_cutoff + 932·V_charge + 0.556·I_charge + 24.81·t_charge - 148.9·V_cutoff² - 78.0·V_cutoff·V_charge +15.2·V_charge·I_charge + 8.7·V_charge·t_charge

[0129] Analysis of variance showed that the model's p-value was <0.001, R-sq (adjusted) = 77.18%, and R-sq (predicted) = 66.36%, indicating that the model is effective. Figure 2 This is the Pareto effect diagram. The length of the bars represents the magnitude of each effect; the larger the effect, the greater its impact on discharge duration. Effect analysis shows that the order of the absolute values ​​of the standardized effects is: t_charge (5.61) > V_cutoff (5.08) > V_cutoff² (3.88) > I_charge (3.35) > V_cutoff·V_charge (2.64) > V_charge (2.62).

[0130] Optimal operating condition parameter optimization

[0131] The optimization objective is set to maximize T1_discharge. Solving using the response optimizer yields a set of optimal solutions with a composite desirability of 1.0: V_cutoff = 11V, V_charge = 14.7V, I_charge = 95A, t_charge = 5.1h, and a predicted T1_discharge time of 179.548 minutes. Figure 3This is the response optimizer interface diagram for the first discharge duration.

[0132] Maximize T_discharge. Using the response optimizer, a set of optimal solutions with a composite desirability of 1.0 is obtained: V_cutoff = 11.09V, V_charge = 15.1V, I_charge = 136A, t_charge = 5.1h, and the predicted T_discharge time is 328.8 minutes. Figure 4 This is the response optimizer interface diagram for the second discharge duration. Figure 6-11 The graph is a surface plot of the discharge duration. The graph is based on two of the following parameters: V_cutoff, V_charge, I_charge, and t_charge. The changes in the other two parameters significantly affect the discharge duration.

[0133] Considering model bending and actual engineering margins, the recommended more robust parameters are: V_charge=14.7V, I_charge=95A, t_charge=5.1h.

[0134] Water loss risk assessment

[0135] With fixed V_cutoff = 11V and I_charge = 80A, a two-factor, two-level, center-point DOE design is employed.

[0136]

[0137] Weigh the battery every 50 cycles, and calculate the cumulative water loss after 150 cycles. Establish a regression model for water loss.

[0138] Experimental results:

[0139] like Figure 3 The Pareto effect plot of water loss shows that the normalized absolute values ​​of the effects are ranked as follows: t_charge (573.45) > V_charge (127.98). Furthermore, there is a significant interaction between t_charge and V_charge. Figure 5 As shown.

[0140] Perform response surface regression analysis. The resulting regression equation (typical example) expressed in uncoded units is: Wtotal = -1182 - 29·t_charge + 77·V_charge + 9.1·(t_charge·V_charge);

[0141] With a fixed charging time of 6 hours and 150 cycles, the water loss rate increases by approximately 2.44% for every 0.1V increase in charging voltage.

[0142] Charging at 14.4V: Average water loss is 0.59g per hour, and water loss per cell is 0.098g / H;

[0143] Total electrolyte volume (from liquid level to the top edge of the electrode plate): 1150g;

[0144] Based on 14 hours of vehicle operation per day:

[0145] 14.4V charging: can be used for 27 months;

[0146] 14.0V charging: can be used for 30 months;

[0147] 14.8V charging: can be used for 25 months;

[0148] In summary, under the recommended optimal operating parameters (V_charge = 14.4V~14.7V), the battery water loss rate meets the requirements of users for normal use for more than 2 years, and the risk of water loss is controllable.

[0149] The method proposed in this invention can be directly applied to:

[0150] (1) Vehicle manufacturer charging strategy calibration: Based on the optimal generator output voltage range (14.4V~14.7V) determined according to the present invention, the voltage regulator parameters of the vehicle generator are recalibrated.

[0151] (2) Aftermarket modification market: For vehicles with low generator voltage, an intelligent DC-DC boost charging module with the optimization strategy of this invention can be installed to achieve precise charging of the parking air conditioner battery.

[0152] (3) Battery Management System (BMS): As the core control logic of the intelligent charging algorithm, it is embedded in the lead-acid battery management system with communication function.

[0153] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing the charging parameters of a lead-acid battery under parking air conditioning conditions, characterized in that, Includes the following steps: Step S1: Operating Condition Factor Mapping and Experimental Design Measurable parameters in the actual operating conditions of the vehicle are mapped to experimental controllable factors. Based on the experimental controllable factors, experimental parameters are designed in DOE to simulate the constant current discharge condition of the parking air conditioner. Multiple rounds of tests are conducted according to the designed parameters to obtain the response data of the lead-acid battery after cyclic charging and discharging. The response data includes the first duration of the battery discharging to the first intermediate voltage at a constant current in the Nth cycle, and the second duration of discharging to the undervoltage protection threshold in the same cycle. Step S2: Mathematical Model Construction and Effect Analysis Based on the response data, regression analysis was used to establish response surface mathematical models with the first duration and the second duration as response variables and the experimental controllable factors as input variables, and the main effect value and interaction effect value of each experimental controllable factor were calculated to identify key factors that have statistical significance for discharge duration. Step S3: Optimal operating condition parameter optimization With the goal of maximizing the first duration and the second duration, a response optimizer is used to numerically optimize the mathematical model of the response surface, and the optimal combination of operating conditions that maximizes the discharge duration is output. Step S4: Supplementary assessment of water loss risk Based on the optimal combination of operating conditions, the cumulative water loss of the battery under a specified number of cycles is calculated using a pre-established water loss rate model, and it is determined whether the water loss is below a preset water loss threshold.

2. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, In step S1, the controllable factors of the experiment include equivalent driving time, generator output voltage, generator output current, and air conditioning compressor undervoltage protection threshold. The equivalent driving time corresponds to the actual time spent charging the battery during the vehicle's engine operation. Its value is calibrated by dividing the product of the vehicle's actual driving mileage and the preset charging efficiency coefficient by the average vehicle speed. The calibration formula is: t_charge = (D / v) × η, where D is the vehicle's single driving mileage, v is the average vehicle speed, and η is the charging efficiency coefficient. The generator output voltage and generator output current correspond to the charging voltage value and the output current value at typical operating speed of the vehicle's alternator regulator, respectively. The air conditioning compressor undervoltage protection threshold corresponds to the shutdown voltage value set by the parking air conditioning system to prevent over-discharge of the battery.

3. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, In step S1, the experimental parameter design includes full factorial design and response surface design.

4. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, In step S1, the constant current is 25-30A; N in the Nth cycle is 45-55; the first intermediate voltage is 11.8-12.2V, used to characterize the degree of battery degradation; the undervoltage protection threshold voltage is determined according to the actual setting of the vehicle air conditioning system, and is 10.8-12.2V.

5. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, In step S2, regression analysis is used to establish response surface mathematical models with the first duration and the second duration as response variables and the experimental controllable factor as input variables. These models include a first duration response surface mathematical model with the first duration as the response variable and the experimental controllable factor as the input variable, and a second response surface mathematical model with the second duration as the response variable and the experimental controllable factor as the input variable. The response surface mathematical model is a quadratic polynomial regression equation. This equation includes at least quadratic terms for each factor, as well as interaction terms between generator output voltage and generator output current, and between generator output voltage and equivalent driving time. The first-time response surface mathematical model is expressed as follows: T1_discharge = A9 + A10·V_charge + A11·I_charge + A12·t_charge + A13·V_charge² + A14·I_charge 2 ; Where T1_discharge is the first duration in minutes; V_charge is the generator output voltage in volts; I_charge is the generator output current in amperes; t_charge is the equivalent driving time in hours; A9~A14 are model coefficients determined through regression analysis; The mathematical model of the second response surface is expressed as follows: T_discharge = A0 + A1·V_cutoff + A2·V_charge + A3·I_charge + A4·t_charge + A5·V_cutoff² + A6·V_cutoff·V_charge + A7·V_charge·I_charge +A8·V_charge·t_charge; Where T_discharge is the second duration in minutes; V_cutoff is the undervoltage protection threshold in volts; V_charge is the generator output voltage in volts; I_charge is the generator output current in amperes; t_charge is the equivalent driving time in hours; and A0~A8 are model coefficients determined through regression analysis.

6. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, In step S2, the main effect value and interaction effect value of each experimental controllable factor are calculated, and key factors with statistical significance for discharge duration are identified, specifically: The effect size and p-value of each controllable factor in the experiment were calculated by analysis of variance. The absolute value of the effect size was used to characterize the influence of each factor on the discharge duration. Factors with p-values ​​less than 0.05 were identified as key factors with statistical significance.

7. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, In step S3, with the optimization objective of maximizing the first duration and the second duration, a response optimizer is used to numerically optimize the response surface mathematical model, specifically including: Set constraints on the range of values ​​for each controllable factor in the experiment, use the gradient descent algorithm to search for the factor combination that maximizes the predicted values ​​of the first and second durations within the constraint space, output the theoretical optimal solution with a satisfactoryness reaching a preset threshold (e.g., above 0.95), and make robust adjustments to the theoretical optimal solution based on the engineering margin to finally obtain the optimal combination of working condition factors.

8. The method for optimizing lead-acid battery charging parameters under parking air conditioning conditions according to claim 1, characterized in that, The water loss rate model in step S4 is as follows: W_total=b0+b1·t_charge+b2·V_charge+b12·(t_charge·V_charge); Where W_total is the cumulative water loss, t_charge is the equivalent driving time, V_charge is the generator output voltage, and b0, b1, b2, b12 are the regression coefficients calibrated by a two-factor, two-level, center-point DOE experiment.

9. A method for adjusting a vehicle charging strategy, characterized in that, The optimal combination of operating conditions factors is obtained by applying the optimization method according to any one of claims 1 to 8, and at least one of the following adjustments is performed based on this combination: Based on the optimal generator output voltage value, the voltage regulator of the vehicle-mounted generator is recalibrated; Based on the optimal generator output current value, configure the generator excitation current or the output current of the external DC-DC converter; Set the undervoltage protection threshold of the vehicle air conditioning system to the value in the optimal operating condition factor combination or keep the original factory setting.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1 to 8.

Images