Intelligent reproduction control method of lead-acid battery pack based on voltage identification

Abstract

The invention discloses an intelligent reproduction control method of a lead-acid battery pack based on voltage identification, which comprises the following steps: S1, temperature acquisition and constant-current charging: acquiring the temperature T of the lead-acid battery pack in real time through a temperature sensor, and applying constant-current charging of rated current I0 to the lead-acid battery pack for a first time window Tc; s2, current order reduction maintenance: the charging current is reduced to I0 / 3, and a second time window Ts is maintained; s3, applying a discharge pulse, discharging the lead-acid battery pack with a discharge pulse current Id = k.I0, and continuing for a third time window Td; and S4, standing and voltage acquisition: standing and waiting for a fourth time window Tw, and acquiring the terminal voltage Vm of the lead-acid battery pack. According to the method, the polarization characteristic and the aging rule of the lead-acid battery pack can be combined with an intelligent control algorithm, the defects of an existing fixed parameter control method in the aspects of precision, adaptability and safety are overcome, and dynamic optimization and safety guarantee in the battery charging process are achieved.

Description

Technical Field

[0001] The present invention relates to the field of battery control, and in particular to an intelligent reproduction control method for a lead-acid battery pack based on voltage identification. Background Art

[0002] Lead-acid battery packs are widely used in energy storage, backup power supply and other scenarios due to their low cost and high reliability. However, their charging and discharging processes have problems such as significant polarization effects, characteristic drift after aging, and strict safety thresholds, and they urgently need dynamic adaptive control methods.

[0003] Existing reproducible control methods mostly use fixed parameter modes (such as constant current charging time and discharge current coefficient), which have the following problems:

[0004] Static polarization compensation: does not take into account the dynamic changes of polarization voltage with battery aging and discharge pulses, resulting in voltage measurement errors;

[0005] Single specification matching: Only relies on static voltage to match battery specifications, without considering dynamic characteristics such as voltage change rate, resulting in low matching accuracy;

[0006] Safety monitoring lag: Lack of real-time verification of voltage fluctuations and polarization offsets can easily lead to battery overcharging or failure under abnormal operating conditions.

[0007] Therefore, an intelligent reproducible control method for lead-acid battery packs based on voltage identification is proposed. Summary of the Invention

[0008] The present invention solves the above technical problems through the following technical solutions, which include the following steps:

[0009] S1: Temperature acquisition and constant current charging: The temperature T of the lead-acid battery pack is collected in real time through the temperature sensor, and a constant current charge of the rated current I0 is applied to the lead-acid battery pack for the first time window Tc.

[0010] S2: Current reduction maintenance, reducing the charging current to I0 / 3, maintaining the second time window Ts;

[0011] S3: discharge pulse is applied with discharge pulse current I d =k·I0 to discharge the lead-acid battery pack for a third time window Td, where k is the discharge current coefficient, k∈(0,0.5), and I0 is the rated charging current of the lead-acid battery pack;

[0012] S4: Standstill and voltage acquisition: Standstill and wait for the fourth time window Tw, and acquire the terminal voltage Vm of the lead-acid battery pack;

[0013] S5: Calculation of actual voltage: V r =V m +α·ΔVpolar

[0014] Where α is the preset polarization compensation coefficient, ΔV polar is the polarization voltage correction calculated based on the discharge pulse parameters, Vm is the terminal voltage collected at rest;

[0015] S6: Specification type matching, matching the actual voltage Vr with the battery specification database to determine the specification type of the lead-acid battery pack;

[0016] S7: Adaptive charging execution, selecting charging parameters based on specification type and executing adaptive charging;

[0017] S8: Cycle data recording, recording the number of cycle charging times Ncycle of the lead-acid battery pack and storing the historical data of the charging process in the battery management system.

[0018] Furthermore, the polarization voltage correction amount ΔV in step S5 is polar The calculation process is as follows:

[0019]

[0020] Where β is the battery aging factor, Td is the discharge pulse duration, and τ is the polarization time constant, which is dynamically updated using the following formula:

[0021] τ=τ0·(1+γ·N cycle );

[0022] τ0 is the initial polarization constant, γ is the attenuation coefficient, and Ncycle is read from the battery management system.

[0023] Furthermore, the calculation process of the battery aging factor β is:

[0024]

[0025] Where βbase is the base aging factor, Vocv is the open circuit voltage mapped by the temperature sensor through the preset temperature-voltage compensation table, and Vnom is the nominal voltage of the lead-acid battery pack.

[0026] The mapping relationship between Vocv and the real voltage Vr is determined by a preset temperature-voltage compensation table in the historical data.

[0027] Furthermore, the matching process of step S6 includes the following process:

[0028] First, calculate the specification matching degree. The specific calculation process is as follows:

[0029]

[0030] Among them, Vref,i is the reference voltage of the i-th specification in the database, is the voltage change rate calculated from the voltage sampling values ​​collected synchronously during the discharge pulse, w1 and w2 are weight coefficients;

[0031] Then, the matching result is selected, and the specification type with the smallest specification matching degree Si is selected as the matching result.

[0032] Furthermore, the value of the weight coefficient w2 is negatively correlated with the polarization time constant τ, specifically: w2 = η·exp(-λ·τ);

[0033] Where η and λ are preset adjustment parameters.

[0034] Furthermore, the adaptive charging process of step S7 includes:

[0035] Parameter determination: determine the maximum charging voltage Vmax and temperature protection threshold Tlim according to the specification type;

[0036] Then the charging current is dynamically adjusted, specifically:

[0037]

[0038] Where T is the real-time temperature, T0 is the reference temperature, and the function g(Vr) is the current derating factor based on the real voltage.

[0039] Furthermore, the expression of the function g(Vr) is:

[0040]

[0041] Where ζ is the voltage sensitivity factor, Vmin is the minimum allowable voltage of the specification type

[0042] Furthermore, the S6 also includes a database self-learning process, specifically:

[0043] Data recording, each time charging is completed, that is, after step S7 is completed, record the triplet (V r ,ΔV polar ,τ);

[0044] Then the reference voltage is updated as follows:

[0045]

[0046] Among them, θ is the learning rate, and its size is inversely proportional to the confidence of the matching degree Si. The confidence is defined as Confidence = 1 / (Si + ε). is the original reference voltage.

[0047] Furthermore, during the discharge pulse in step S3:

[0048] Synchronous voltage acquisition: synchronously acquire multiple groups of voltage sampling values ​​Vi;

[0049] Data validity verification: Calculate the voltage fluctuation coefficient. The specific process is as follows:

[0050] Where N is the number of voltage sampling values; Vi is the voltage sampling value of the i-th group; is the average value of voltage sampling;

[0051] If σv>σth, the re-discharge sequence is triggered and steps S2-S4 are repeated until the data is valid.

[0052] Furthermore, when the matching in step S6 fails, the security protocol is executed, and the specific process is as follows:

[0053] Use the benchmark charging parameters, Isafe = 0.1I0, Vsafe = 0.9Vnom to charge;

[0054] Polarization offset monitoring is then performed, that is, the polarization offset is continuously calculated during the charging process:

[0055]

[0056] Where Vcharge is the actual voltage during charging, and Vpred is the predicted voltage based on the historical charging data (voltage-time curve) of batteries of the same specification and type stored in the battery management system;

[0057] If δ>δth, the charging is terminated and the battery is marked as failed.

[0058] Compared with the existing technology, the present invention has the following advantages: the intelligent reproducible control method of the lead-acid battery pack based on voltage identification can dynamically adjust the charging parameters according to the specification type of the lead-acid battery pack through voltage identification and database matching, realize personalized charging strategy, improve charging efficiency and battery adaptability, utilize polarization voltage correction amount calculation and compensation coefficient, combine battery aging factor and cycle number to dynamically update the polarization time constant, effectively offset the polarization effect during discharge, improve voltage measurement accuracy, and avoid misjudgment, update the reference voltage based on historical charging data through the database self-learning process, so that the system gradually improves the specification matching accuracy over time and adapts to battery aging or parameter drift, and sets a voltage fluctuation coefficient verification and re-discharge mechanism to ensure data validity; execute a safe charging protocol when matching fails, and terminate abnormal charging in real time in combination with polarization offset monitoring to prevent battery overcharging or failure and enhance system safety. The weight coefficient is negatively correlated with the polarization time constant, and the charging current is dynamically derated with temperature and voltage. The design enables the charging process to be flexibly adjusted according to the real-time status, optimizes the battery charging and discharging performance, and extends the service life. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is the overall structural diagram of the present invention. DETAILED DESCRIPTION

[0060] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process. However, the protection scope of the present invention is not limited to the following embodiment.

[0061] like Figure 1 As shown, this embodiment provides a technical solution: an intelligent recurrence control method for a lead-acid battery pack based on voltage identification, comprising the following steps:

[0062] S1: Temperature acquisition and constant current charging: The temperature T of the lead-acid battery pack is collected in real time through the temperature sensor, and a constant current charge of the rated current I0 is applied to the lead-acid battery pack for the first time window Tc.

[0063] S2: Current reduction maintenance, reducing the charging current to I0 / 3, maintaining the second time window Ts;

[0064] S3: discharge pulse is applied with discharge pulse current I d =k·I0 to discharge the lead-acid battery pack for a third time window Td, where k is the discharge current coefficient, k∈(0,0.5), and I0 is the rated charging current of the lead-acid battery pack;

[0065] S4: Standstill and voltage acquisition: Standstill and wait for the fourth time window Tw, and acquire the terminal voltage Vm of the lead-acid battery pack;

[0066] S5: Calculation of actual voltage: V r =V m +α·ΔV polar

[0067] Where α is the preset polarization compensation coefficient, ΔV polar is the polarization voltage correction calculated based on the discharge pulse parameters, Vm is the terminal voltage collected at rest;

[0068] S6: Specification type matching, matching the actual voltage Vr with the battery specification database to determine the specification type of the lead-acid battery pack;

[0069] S7: Adaptive charging execution, selecting charging parameters based on specification type and executing adaptive charging;

[0070] S8: Cycle data recording, recording the number of cycle charging times Ncycle of the lead-acid battery pack and storing the historical data of the charging process in the battery management system;

[0071] For example, the charging of a lead-acid battery pack with a rated charging current of I0=10A is controlled.

[0072] Step S1: temperature acquisition and constant current charging;

[0073] The temperature sensor measures the real-time temperature as 25°C;

[0074] Charging is performed at a constant current of 10 A for a first time window Tc = 20 min.

[0075] Step S2: current reduction and maintenance;

[0076] The charging current is reduced to I0 / 3≈3.33A, and the second time window Ts=10min is maintained.

[0077] Step S3: applying a discharge pulse;

[0078] Discharge current coefficient k = 0.3, then the discharge pulse current Id = 0.3 × 10 = 3A;

[0079] The discharge lasts for a third time window Td=15s.

[0080] Step S4: standing and voltage acquisition;

[0081] The system waits for a fourth time window Tw=3 min, and the voltage at the acquisition terminal Vm=12.5 V.

[0082] Step S5: Calculate the actual voltage;

[0083] The preset polarization compensation coefficient α is 1.0;

[0084] Calculate the polarization voltage correction amount ΔVpolar (assuming that ΔVpolar = 0.2 V is obtained according to the formula in claim 2);

[0085] The real voltage Vr = 12.5 + 1.0 × 0.2 = 12.7V.

[0086] Step S6: specification type matching;

[0087] Vr=12.7V is matched with the database to determine that the battery specification is "12V / 50Ah" type.

[0088] Step S7: Adaptive charging execution

[0089] Select charging parameters according to the specification type: maximum charging voltage Vmax=14.8V, temperature protection threshold ℃;

[0090] Dynamically adjust the charging current.

[0091] Step S8: Loop data recording

[0092] Record the number of cycle charging Ncycle=50 and store the charging data in the battery management system.

[0093] , the polarization voltage correction amount ΔV in step S5 polar The calculation process is as follows:

[0094]

[0095] Where β is the battery aging factor, Td is the discharge pulse duration, and τ is the polarization time constant, which is dynamically updated using the following formula:

[0096] τ=τ0·(1+γ·N cycle );

[0097] τ0 is the initial polarization constant, γ is the attenuation coefficient, and Ncycle is read from the battery management system;

[0098] By calculating the polarization voltage correction amount ΔVpolar, the voltage measurement deviation caused by polarization during discharge can be offset in real time, the calculation accuracy of the real voltage Vr can be improved, and specification matching errors caused by polarization errors can be avoided. The polarization time constant τ is dynamically updated with the number of cycles Ncycle to reflect the influence of battery aging on the polarization effect (such as the increase of the polarization time constant of aging batteries), so that the compensation model is more in line with the actual state of the battery. Combined with the correlation calculation of the battery aging factor β and the open circuit voltage Vocv, real-time parameters such as temperature and voltage are integrated into the polarization compensation logic to enhance the adaptability of the system under different working conditions.

[0099] For example, a lead-acid battery pack with a rated current of I0 = 10A, a discharge current coefficient of k = 0.3, and a discharge time of Td = 15s has a known cycle number of Ncycle = 50.

[0100] Calculation process of polarization voltage correction value ΔVpolar in step S5

[0101] Calculate the polarization time constant τ:

[0102] Assume that the initial polarization constant τ0 = 20s, the attenuation coefficient γ = 0.01,

[0103] Then τ = τ0 × (1 + γ × Ncycle) = 20 × (1 + 0.01 × 50) = 20 × 1.5 = 30s.

[0104] Calculate the battery aging factor β:

[0105] Assume that the base aging factor βbase = 0.05, the open circuit voltage Vocv = 12.8V, and the nominal voltage Vnom = 12V.

[0106] Then β==0.05×12.8 / 12≈0.053.

[0107] Calculate the polarization voltage correction ΔVpolar:

[0108] Discharge pulse current Id=k·I0=0.3×10=3A,

[0109] Substitute into the formula: ΔVpolar=β·Id·(1-e -T / τ )=0.053×3×(1-e -15 / 30 )≈0.159×(1-0.6065)≈0.062V;

[0110] By quantifying the polarization effect, ΔVpolar = 0.062 V, the real voltage Vr = Vm + α·ΔVpolar = 12.5 + 1.0 × 0.062 = 12.562 V, which is closer to the actual state of the battery than when it is not compensated;

[0111] The dynamically updated τ(30s) increases with battery aging (Ncycle=50), reflecting the adaptive compensation for battery characteristic degradation and avoiding the error accumulation caused by fixed polarization parameters.

[0112] The calculation process of the battery aging factor β is:

[0113]

[0114] Where βbase is the base aging factor, Vocv is the open circuit voltage mapped by the temperature sensor through the preset temperature-voltage compensation table, and Vnom is the nominal voltage of the lead-acid battery pack.

[0115] The mapping relationship between Vocv and the real voltage Vr is determined by the preset temperature-voltage compensation table in the historical data;

[0116] Dynamically quantify battery aging: The aging factor β is dynamically calculated using the ratio of the open-circuit voltage (Vocv) to the nominal voltage (Vnom). This allows β to directly reflect the actual aging state of the battery, avoiding compensation deviations caused by fixed parameters. For example, the open-circuit voltage of an aging battery decreases due to increased internal resistance.

[0117] The temperature is measured using a temperature sensor, and the open-circuit voltage Vocv is mapped using a preset temperature-voltage compensation table. This allows the calculation of β to simultaneously consider the impact of temperature on the battery voltage, improving the accuracy of polarization compensation under different operating conditions.

[0118] By coupling battery aging with real-time voltage parameters, the calculation of polarization voltage correction can be automatically adjusted with the battery aging degree and ambient temperature, avoiding control strategy failure caused by battery performance degradation.

[0119] For example, for a lead-acid battery pack with a rated current of I0 = 10A and a number of cycles Ncycle = 50, and a current temperature of 25°C, the open circuit voltage Vocv is determined using the temperature-voltage compensation table.

[0120] Calculation process of battery aging factor β:

[0121] Parameter settings:

[0122] Base aging factor βbase = 0.06 (preset according to battery type);

[0123] Nominal voltage Vnom=12V;

[0124] The temperature sensor measures 25°C, and the open circuit voltage Vocv = 12.6V is obtained from the temperature-voltage compensation table (this value has taken into account the effect of temperature on voltage).

[0125] Calculate the β value: β = 0.06 × 1212.6 = 0.06 × 1.05 = 0.063

[0126] Impact on subsequent steps:

[0127] Substitute β=0.063 into the ΔVpola calculation formula: ΔVpolar=β·Id·(1-e -T / τ )=0.063×3×(1-e -15 / 30 )≈0.189×(1-0.6065)≈0.074V

[0128] Compared with β=0.053 in the example of claim 2, β in this example increases due to the increase in Voc, indicating that the battery aging is less severe (or the voltage is closer to the nominal value after temperature compensation), ultimately making the ΔVpolar calculation result more consistent with the current state of the battery.

[0129] The matching process of step S6 includes the following steps:

[0130] First, calculate the specification matching degree. The specific calculation process is as follows:

[0131]

[0132] Among them, V ref,i is the reference voltage of the i-th specification in the database, is the voltage change rate calculated from the voltage sampling values ​​collected synchronously during the discharge pulse, w1 and w2 are weight coefficients;

[0133] Then, the matching result is selected, and the specification type with the smallest specification matching degree Si is selected as the matching result;

[0134] By simultaneously calculating the difference between the real voltage Vr and the reference voltage Vref,i, and the voltage change rate The difference from the reference change rate is used to match battery specifications from two dimensions: static voltage and dynamic change, avoiding misjudgment caused by single-dimensional matching (for example, batteries of different specifications may have similar voltages but significantly different charge and discharge characteristics).

[0135] Adjustable weights enhance adaptability: The weight coefficients w1 and w2 can be adjusted according to actual needs. For example, the priority is given to the absolute value difference or change trend of the voltage, making the matching logic more suitable for different application scenarios. For example, in the fast charging scenario, more attention is paid to the voltage change rate.

[0136] Combined with the database self-learning process, the reference voltage is updated through historical data, so that the matching model is gradually optimized as the battery ages or the specification library expands, avoiding the decline in matching accuracy due to fixed parameters.

[0137] The personnel know that the real voltage Vr = 12.7V, and the voltage change rate during the discharge pulse Two battery specifications need to be matched from the database:

[0138] Specification A: Reference voltage Vref,1=12.8V, reference voltage change rate

[0139] Specification B: Reference voltage Vref,2=12.5V, reference voltage change rate

[0140] Let the weight coefficient \(w1 = 0.6\) and \(w2 = 0.4\).

[0141] Calculate the matching degree \(S1\) of specification A: \(S1 = 0.6\cdot|12.7 - 12.8|+0.4\cdot|-0.1-(-0.12)| = 0.6×0.1 + 0.4×0.02 = 0.06 + 0.008 = 0.068\)

[0142] Calculate the matching degree \(S2\) of specification B: \(S2 = 0.6\cdot|12.7 - 12.5|+0.4\cdot|-0.1-(-0.08)| = 0.6×0.2 + 0.4×0.02 = 0.12 + 0.008 = 0.128\)

[0143] Matching result:

[0144] Since \(S1 = 0.068 < S2 = 0.128\), select specification A as the matching result.

[0145] The value of the weight coefficient \(w2\) is negatively correlated with the polarization time constant \(\tau\), specifically: \(w2=\eta\cdot\exp(-\lambda\cdot\tau)\);

[0146] where \(\eta\) and \(\lambda\) are preset adjustment parameters;

[0147] By associating the weight coefficient \(w2\) with the polarization time constant \(\tau\) negatively, the matching weight of the voltage change rate is dynamically adjusted according to the battery aging degree. When the polarization effect of a new battery is weak, \(w2\) increases, and more attention is paid to the voltage change rate characteristics; when the polarization effect of an aging battery is strong, \(w2\) decreases, and more dependence is on the voltage absolute value matching, improving the specification matching accuracy under different working conditions.

[0148] The polarization time constant \(\tau\) reflects the polarization response speed inside the battery. For an aging battery, \(\tau\) increases, and the voltage change rate measurement is easily interfered by polarization. By the decay of \(w2\) with \(\tau\), the weight of the interference factor can be reduced, avoiding misjudging the battery specification due to the polarization effect.

[0149] There is no need for manual parameter adjustment. The system automatically dynamically allocates the matching dimension weights according to the real-time state of the battery (reflected by \(\tau\)), reducing the dependence on fixed parameters and enhancing the self-adaptability of the algorithm.

[0150] For example, for two known states of batteries, it is necessary to calculate the weight coefficient \(w2\) and explain the influence on the matching degree:

[0151] New battery: The polarization time constant \(\tau1 = 20s\);

[0152] Aging battery: The polarization time constant \(\tau2 = 40s\);

[0153] Let the preset adjustment parameters \(\eta = 0.5\) and \(\lambda = 0.05\).

[0154] Calculation process of the weight coefficient;

[0155] Calculation of w2 for new battery: w21=0.5·exp(-0.05×20)=0.5·e-1≈0.5×0.3679≈0.184

[0156] Calculation of aged battery w2: w22 = 0.5 exp(-0.05 × 40) = 0.5 e-2 ≈ 0.5 × 0.1353 ≈ 0.068;

[0157] Assuming the actual voltage of the two batteries Vr = 12.7V, the voltage change rate Vref,i=12.8V for specification A in the database. Weight w1 = 0.8 (fixed):

[0158] New battery (w2 = 0.184): Si = 0.8·|12.7-12.8|+0.184·|-0.1-(-0.12)|=0.8×0.1+0.184×0.02≈0.08+0.0037≈0.0837

[0159] Aged battery (w2 = 0.068): Si = 0.8·|12.7-12.8|+0.068·|-0.1-(-0.12)|

[0160] =0.8×0.1+0.068×0.02≈0.08+0.0014≈0.0814

[0161] For new batteries, τ is small and w2 is large (0.184), so the voltage change rate deviation contributes 0.0037 to the matching degree. For aged batteries, τ is large and w2 is reduced (0.068), and the change rate contribution drops to 0.0014, indicating that the system automatically reduces its dependence on the change rate parameters of aged batteries, avoiding matching errors caused by polarization interference.

[0162] This mechanism enables new batteries to focus more on dynamic characteristics (rate of change) and aging batteries to focus more on static voltage, adapting to the characteristic differences of different battery states and improving the flexibility of the matching strategy.

[0163] The adaptive charging process of step S7 includes:

[0164] Parameter determination: determine the maximum charging voltage Vmax and temperature protection threshold Tlim according to the specification type;

[0165] Then the charging current is dynamically adjusted, specifically:

[0166]

[0167] Where T is the real-time temperature, T0 is the reference temperature, and the function g(Vr) is the current derating factor based on the real voltage.

[0168] By combining the temperature protection threshold Tlim with the actual voltage Vr, the charging current is adjusted synchronously with the real-time temperature and voltage status of the battery, avoiding overcharging in high temperature environments or excessive current when the voltage is close to the upper limit, thereby improving charging safety and battery life.

[0169] Adaptive temperature protection mechanism: using Function: When the battery temperature T approaches the protection threshold Tlim, the charging current is automatically attenuated proportionally (for example, the current drops to 0 when T=Tlim) to prevent safety risks caused by temperature runaway.

[0170] Compatible with different battery parameters: By linking the reference charging current Ibase with specification type parameters (Tlim, Vmax, etc.), the same control logic can adapt to lead-acid batteries of different capacities and voltage levels, enhancing system versatility.

[0171] The expression of the function g(Vr) is:

[0172]

[0173] Where ζ is the voltage sensitivity factor, Vmin is the minimum allowable voltage of the specification type;

[0174] The current derating factor g(Vr) in the form of a quadratic function automatically reduces the charging current as the actual voltage Vr approaches the maximum allowable voltage Vmax, preventing battery overcharging. The closer Vr is to Vmax, the smaller g(Vr) is, and the lower the charging current is, improving charging safety.

[0175] The nonlinear characteristics of the quadratic function make the current derating process slow at first and then rapid (slow derating when the voltage is low, and rapid derating when approaching the upper limit), which meets the constant current and constant voltage charging curve requirements of lead-acid batteries, reducing charging time while ensuring safety.

[0176] Adapting to the characteristics of batteries of different specifications: The voltage sensitivity factor ζ can be used to adjust the sensitivity of the current to voltage response (for example, a large-capacity battery can be set to a smaller ζ, allowing a wider voltage fluctuation range), making the same control logic applicable to batteries of different specifications and enhancing system versatility.

[0177] If the parameters of a certain specification of lead-acid battery are known to be:

[0178] Maximum allowable voltage Vmax = 14.8V, minimum allowable voltage Vmin = 12.0V;

[0179] Voltage sensitivity factor ζ = 0.5;

[0180] The actual voltage Vr=14.0V (close to the upper limit of charging).

[0181] Current derating factor calculation process

[0182] Substitute into the formula to calculate g(Vr): g(Vr) = 1-0.5·((14-12) / (14.8-12.0)) 2 ;

[0183] Step-by-step calculation: Numerator difference: 14.0-12.0=2.0V;

[0184] Denominator difference: 14.8-12.0=2.8V;

[0185] Ratio squared: (2 / 2.8) 2 ≈(0.714) 2 ≈0.510;

[0186] Derating factor: g(Vr)=1-0.5×0.510≈1-0.255=0.745.

[0187] If the base charging current Ibase = 10A and the temperature factor does not limit the current, the actual charging current is: Icharge = 10A × 1 × 0.745 = 7.45A

[0188] When Vr increases from 12.0V to 14.0V, g(Vr) decreases from 1 to 0.745, and the charging current decays accordingly, avoiding the risk of overcharging due to excessive current when the voltage approaches Vmax.

[0189] The S6 also includes a database self-learning process, specifically:

[0190] Data recording, each time charging is completed, that is, after step S7 is completed, record the triplet (V r ,ΔV polar ,τ);

[0191] Then the reference voltage is updated as follows:

[0192]

[0193] Among them, θ is the learning rate, and its size is inversely proportional to the confidence of the matching degree Si. The confidence is defined as Confidence = 1 / (Si + ε). is the original reference voltage;

[0194] The reference voltage in the database is continuously updated through the self-learning process, allowing the system to adapt to long-term changes such as battery aging and parameter drift (such as voltage characteristic shift caused by a decrease in electrolyte concentration), avoiding the degradation of specification matching accuracy caused by fixed reference data.

[0195] The design of the learning rate θ being inversely proportional to the matching confidence enables the system to use small step updates (small θ) for high-confidence matching (small Si) and large step updates (large θ) for low-confidence matching (large Si), achieving adaptive optimization with more corrections as the accuracy increases, and gradually reducing the deviation between the reference voltage and the actual voltage.

[0196] There is no need for manual regular updates to the database. The system automatically iterates and optimizes based on historical charging data, reducing maintenance costs. It is especially suitable for long-term battery management scenarios (such as energy storage power stations and backup power systems).

[0197] For example, when a lead-acid battery pack is charged for the first time, it is matched with specification A (reference ), after three charges, the reference voltage is updated by self-learning, and the learning rate θ is set to 0.2 (assuming that θ increases when the confidence is low).

[0198] Self-learning update process:

[0199] Record triplet after first charge:

[0200] The true voltage Vr = 12.7 V, the polarization voltage correction ΔVpolar = 0.062 V, and the polarization time constant τ = 30 s.

[0201] Matching degree Si = 0.068, confidence = 1 / (0.068+10 -6 )≈14.7, and because of the high confidence level, θ is taken as 0.1.

[0202] Reference voltage after first update:

[0203] Update after second charge:

[0204] New triplet Vr = 12.6 V, ΔVpolar = 0.07 V, τ = 32 s;

[0205] Matching degree Si = 0.075, confidence level ≈ 13.3, θ is still 0.1;

[0206] Updated reference voltage:

[0207] Update after the third charge:

[0208] Triplet Vr = 12.5 V, ΔVpolar = 0.08 V, τ = 35 s;

[0209] Matching degree Si = 0.08, confidence level ≈ 12.5, θ increases to 0.2 (confidence level decreases);

[0210] Updated reference voltage:

[0211] Through three self-learning steps, the reference voltage is gradually adjusted from the initial 12.8V to 11.871V, which is closer to the actual battery voltage trend and reflects the adaptive correction of battery aging or parameter changes.

[0212] The learning rate θ increases as the confidence level decreases, which enables the system to accelerate the update of the reference voltage when the matching accuracy decreases to avoid error accumulation. For example, the third update step size (0.2) is larger than the previous two, which quickly corrects the reference value deviation.

[0213] Furthermore, during the discharge pulse in step S3:

[0214] Synchronous voltage acquisition: synchronously acquire multiple groups of voltage sampling values ​​Vi;

[0215] Data validity verification: Calculate the voltage fluctuation coefficient. The specific process is as follows:

[0216] Where N is the number of voltage sampling values; Vi is the voltage sampling value of the i-th group; is the average value of voltage sampling;

[0217] If σv>σth, the re-discharge sequence is triggered and steps S2-S4 are repeated until the data is valid, where σth is the preset threshold;

[0218] The stability of the voltage sampling value during the discharge pulse is verified by calculating the voltage fluctuation coefficient σv. When the fluctuation exceeds the threshold, a re-discharge is triggered. This prevents abnormal data caused by factors such as electromagnetic interference and poor contact from participating in subsequent calculations, ensures the accuracy of Vr calculations, filters out voltage data with abnormal fluctuations, and prevents specification matching misjudgments caused by data noise (such as false voltage peaks that cause matching results to deviate from the actual specifications). This enhances the system's adaptability to complex working conditions. Without manual intervention, the system automatically eliminates the impact of transient interference through a re-discharge sequence. This is particularly suitable for scenarios with frequent electromagnetic interference in industrial environments, reducing control strategy failures caused by data mis-sampling.

[0219] For example, during the discharge pulse period Td=15s, 5 groups of voltage sampling values ​​Vi are collected synchronously, and the voltage fluctuation threshold σth is set to be 10mV.

[0220] Voltage fluctuation coefficient calculation and verification process:

[0221] Voltage sampling values: V1=12.35, V2=12.38, V3=12.45, V4=12.36, V5=12.37;

[0222] Calculate the average voltage

[0223] Calculate the voltage fluctuation coefficient σv:

[0224] σv=[(12.35-12.382) 2 +(12.38-12.382) 2 +(12.45-12.382) 2 +(12.36-

[0225] 12.382) 2 +(12.37-12.382)2] / 5;

[0226] Step-by-step calculation:

[0227] (12.35-12.382) 2 =(-0.032) 2 =0.001024;

[0228] (12.38-12.382) 2 =(-0.002) 2 =0.000004;

[0229] (12.45-12.382) 2 =(0.068) 2 =0.004624;

[0230] (12.36-12.382) 2 =(-0.022) 2 =0.000484;

[0231] (12.37-12.382) 2 =(-0.012) 2 =0.000144;

[0232] Sum: 0.001024 + 0.000004 + 0.004624 + 0.000484 + 0.000144 = 0.00628;

[0233] σv=0.00628 / 5=0.001256V2, which is approximately 35.4mV after square root.

[0234] Since σv=35.4mV>σth=10mV, the re-discharge sequence is triggered and steps S2-S4 are repeated until σv≤10mV.

[0235] When the match fails in step S6, the security protocol is executed. The specific process is as follows:

[0236] Use the benchmark charging parameters, Isafe = 0.1I0, Vsafe = 0.9Vnom to charge;

[0237] Polarization offset monitoring is then performed, that is, the polarization offset is continuously calculated during the charging process:

[0238]

[0239] Where Vcharge is the actual voltage during charging, and Vpred is the predicted voltage based on the historical charging data (voltage-time curve) of batteries of the same specification and type stored in the battery management system;

[0240] If δ>δth, terminate charging and mark the battery as failed;

[0241] When specification matching fails, the system automatically switches to baseline charging parameters to avoid risks such as overcharging and overcurrent caused by specification misjudgment. This provides basic protection for batteries with unknown specifications or abnormalities. The system monitors the deviation between the charging voltage and the predicted value through the polarization offset δ, promptly identifying internal polarization anomalies in the battery (such as plate sulfation and electrolyte deterioration), and avoiding safety accidents caused by continued charging of faulty batteries. There is no need to preset battery specifications, and the voltage is predicted based on historical data. This is suitable for aging batteries or special models not covered by the specification library, enhancing the system's fault tolerance to unknown working conditions.

[0242] If the specifications of a lead-acid battery pack fail to match, the safety protocol is triggered. The known parameters are as follows:

[0243] Rated charging current I0 = 10A, nominal voltage Vnom = 12V;

[0244] Benchmark charging parameters: Isafe = 0.1I0 = 1A, Vsafe = 0.9Vnom = 10.8V;

[0245] Polarization offset threshold δth=0.1.

[0246] Security protocol execution process:

[0247] Baseline charging stage:

[0248] Charge with a current of 1A and collect the actual voltage Vcharge in real time.

[0249] Polarization offset calculation:

[0250] Based on historical data of similar batteries in the battery management system, the voltage Vpred is predicted to be 11.5V after 10 minutes of charging.

[0251] The actual measurement Vcharge = 12.3V, substitute into the formula:

[0252] First judgment:

[0253] δ=0.0696<δth=0.1, continue charging.

[0254] When charging for 20 minutes:

[0255] Predicted voltage Vpred = 12.0 V, actual voltage Vcharge = 13.2 V;

[0256] calculate Reached the threshold.

[0257] Safe Termination:

[0258] Immediately terminate charging and mark the battery as failed to avoid overcharging due to abnormal polarization (such as internal short circuit or severe aging).

[0259] Through benchmark charging and dynamic monitoring mechanisms, charging safety can still be guaranteed even when specification matching fails: charging is allowed to continue when δ does not exceed the threshold for the first time, and is terminated in time when δ = 0.1 is detected for the second time, preventing battery damage in an unknown state.

[0260] This logic is applicable to batteries not covered by the specification library (such as non-standard products) or severely aged batteries, and achieves safety control without specification dependence through historical data prediction and real-time deviation monitoring.

[0261] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0262] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0263] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. An intelligent reproducible control method for a lead-acid battery pack based on voltage identification, characterized in that: The following steps are involved: S1: Temperature acquisition and constant current charging: The temperature T of the lead-acid battery pack is collected in real time through the temperature sensor, and a constant current charge of the rated current I0 is applied to the lead-acid battery pack for the first time window Tc. S2: Current reduction maintenance, reducing the charging current to I0 / 3, maintaining the second time window Ts; S3: discharge pulse is applied with discharge pulse current I d =k·I0 discharges the lead-acid battery pack for a third time window Td; S4: Standstill and voltage acquisition: Standstill and wait for the fourth time window Tw, and acquire the terminal voltage Vm of the lead-acid battery pack; S5: Calculate the real voltage Vr; S6: Specification type matching, matching the actual voltage Vr with the battery specification database to determine the specification type of the lead-acid battery pack; S7: Adaptive charging execution, selecting charging parameters based on specification type and executing adaptive charging; S8: Cycle data recording, recording the number of cycle charging times Ncycle of the lead-acid battery pack and storing the historical data of the charging process in the battery management system.

2. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 1, characterized in that: The polarization voltage correction amount ΔV in step S5 polar The calculation process is as follows: Where β is the battery aging factor, Td is the discharge pulse duration, and τ is the polarization time constant, which is dynamically updated using the following formula: τ=τ0·(1+γ·N cycle ); τ0 is the initial polarization constant, γ is the attenuation coefficient, and Ncycle is read from the battery management system.

3. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 2, characterized in that: The calculation process of the battery aging factor β is: Where βbase is the base aging factor, Vocv is the open circuit voltage mapped by the temperature sensor through the preset temperature-voltage compensation table, and Vnom is the nominal voltage of the lead-acid battery pack. The mapping relationship between Vocv and the real voltage Vr is determined by a preset temperature-voltage compensation table in the historical data.

4. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 3, characterized in that: The matching process of step S6 includes the following steps: First, calculate the specification matching degree. The specific calculation process is as follows: Among them, V ref,i is the reference voltage of the i-th specification in the database, is the voltage change rate calculated from the voltage sampling values ​​collected synchronously during the discharge pulse, w1 and w2 are weight coefficients; Then, the matching result is selected, and the specification type with the smallest specification matching degree Si is selected as the matching result.

5. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 4, characterized in that: The value of the weight coefficient w2 is negatively correlated with the polarization time constant τ, specifically: w2 = η·exp(-λ·τ); Where η and λ are preset adjustment parameters.

6. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 1, characterized in that: The adaptive charging process of step S7 includes: Parameter determination: determine the maximum charging voltage Vmax and temperature protection threshold Tlim according to the specification type; Then the charging current is dynamically adjusted, specifically: Where T is the real-time temperature, T0 is the reference temperature, and the function g(Vr) is the current derating factor based on the real voltage.

7. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 1, characterized in that: The expression of the function g(Vr) is: Where ζ is the voltage sensitivity factor, and Vmin is the minimum allowable voltage for this specification type.

8. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 4, characterized in that: The S6 also includes a database self-learning process, specifically: Data recording, each time charging is completed, that is, after step S7 is completed, record the triplet (V r ,ΔV polar ,τ); Then the reference voltage is updated as follows: Among them, θ is the learning rate, and its size is inversely proportional to the confidence of the matching degree Si. The confidence is defined as Confidence = 1 / (Si + ε). is the original reference voltage.

9. The intelligent recurrence control method for a lead-acid battery pack based on voltage identification according to claim 1, characterized in that: During the discharge pulse of step S3: Synchronous voltage acquisition: synchronously acquire multiple groups of voltage sampling values ​​Vi; Data validity verification: Calculate the voltage fluctuation coefficient. The specific process is as follows: Where N is the number of voltage sampling values; Vi is the voltage sampling value of the i-th group; is the average value of voltage sampling; If σv>σth, the re-discharge sequence is triggered and steps S2-S4 are repeated until the data is valid.

10. The intelligent recurrence control method of a lead-acid battery pack based on voltage identification according to claim 1, characterized in that: When the match fails in step S6, the security protocol is executed. The specific process is as follows: Use the benchmark charging parameters, Isafe = 0.1I0, Vsafe = 0.9Vnom to charge; Polarization offset monitoring is then performed, that is, the polarization offset is continuously calculated during the charging process: Where Vcharge is the actual voltage during charging, and Vpred is the predicted voltage based on the historical charging data (voltage-time curve) of batteries of the same specification and type stored in the battery management system; If δ>δth, the charging is terminated and the battery is marked as failed.

Images