Method for determining activation time of liquid storage type reserve battery

Abstract

The invention discloses a method for determining activation time of a liquid storage type reserve battery, which comprises the following steps: S10, acquiring parameters of a standard-size battery: selecting the standard-size battery as a pilot sample and an independent verification sample, and acquiring an electrolyte infiltration quality-time curve and actual activation time; s20, prediction parameter acquisition: calculating an initial time scaling factor and an initial infiltration quality critical threshold value of the pilot sample according to the electrolyte infiltration quality-time curve and the actual activation time of the pilot sample, and performing error verification by using independent verification sample parameters to obtain a prediction time scaling factor and a prediction infiltration quality critical threshold value; and S30, target battery activation time prediction: according to a target battery electrolyte infiltration quality-time curve, obtaining the time required for the target battery infiltration quality to reach the prediction critical threshold, and calculating the activation time of the target battery in combination with the prediction time scaling factor. The activation time determination method is high in efficiency and good in precision.

Description

Technical Field

[0001] The invention relates to the technical field of numerical simulation of liquid storage type reserve batteries, and in particular to a method for determining activation time of a liquid storage type reserve battery. Background Art

[0002] Liquid-storage backup batteries offer an exceptionally long storage life (>10 years), making them an optimal choice for electromechanical system power supplies in applications requiring extremely high lifespan and reliability. Liquid-storage backup batteries encapsulate liquid electrolyte in a bottle isolated from the electrodes. When overloads during launch or flight exceed a preset threshold (such as launch acceleration and impact loads), a mechanical structure ruptures the bottle, allowing the electrolyte to flow to the electrode interface and soak the electrodes, activating the battery and discharging it for external power supply. The entire battery activation process, involving complex multi-physics responses, can be divided into three sequential, cascaded, and mutually coupled sub-processes: liquid infusion, soaking, and discharge.

[0003] Some specialized equipment requires rapid battery activation to ensure timely startup of electromechanical systems. For example, for certain aerospace equipment, delayed battery activation can lead to the loss of critical data, jeopardizing the execution of high-precision, high-transient operations such as orbital adjustment and rendezvous and docking. Existing research often uses an exhaustive trial-and-error method to analyze activation time. This involves first conducting a large number of experiments to obtain data, and then inferring the impact of dimensions based on the data. This is costly, inefficient, and subject to large errors. Research on the activation time of liquid storage backup batteries through mathematical modeling and digital simulation has significant application value in solving the problem of rapid activation of liquid storage backup batteries.

[0004] However, due to the extremely strong transient and coupled nature of the activation process of liquid storage batteries, there are significant limitations in the current modeling of the activation process of liquid storage batteries. First, the activation process of liquid storage batteries involves highly dynamic transient material transport and reaction processes such as electrolyte infiltration of porous electrodes and electrochemical reactions; second, the activation time scale of liquid storage batteries is relatively low, the coupling of multiple physical fields is stronger, and dynamic modeling and simulation calculations are extremely difficult. Among them, in the transient process of electrolyte infiltration of porous electrodes, the front of the dynamic electrolyte continuously changes the effective reaction interface, directly affecting the external power supply voltage of the battery in the initial stage of the discharge process, making it difficult to accurately predict the activation time. Therefore, it is crucial to determine the battery activation time efficiently and accurately.

[0005] Currently, existing research lacks a method to accurately determine the activation time of liquid storage reserve batteries. The specific reasons are as follows.

[0006] When the porous electrode is first wetted by the electrolyte, the battery does not immediately possess full external power supply capability. This is because the lithium anode and the thionyl chloride electrolyte are in direct contact at this point, with chemical reactions occurring at the interface rather than electrochemical reactions. As the chemical reaction continues, the lithium chloride product forms a SEI film on the surface of the lithium anode, preventing electrons from passing through it to react directly with the electrolyte. Consequently, the lithium anode and the thionyl chloride electrolyte each undergo their own electrochemical reactions, transferring electrons through an external circuit. As a result, the short-circuit effect between the positive and negative electrodes gradually disappears, and the battery begins to possess external power supply capability, with its output voltage gradually increasing from 0V to its maximum voltage. The time-dependent relationship between the electrolyte wetting mass within the porous electrode (indicating the integrity of the SEI film growth) and the battery output voltage-time curve (indicating the progress of the electrochemical reaction) exhibit a proportional constraint on the time scale, with their ratio being the time scaling factor. This is essentially the time-domain synergistic effect of fluid transport and electrochemical polarization in the porous medium.

[0007] However, existing research generally ignores the SEI formation stage. Furthermore, due to the complexity of the reaction, voltage can only be determined by simultaneously solving multi-field coupled equations for wetting, diffusion, and electrochemical reactions. This results in reduced computational efficiency and insufficient accuracy, making it impractical for engineering applications. These shortcomings collectively render existing models incapable of predicting activation time, severely hindering the optimization of fast-response battery designs.

[0008] In summary, the problems existing in the prior art are that the determination of the activation time of the liquid storage type reserve battery is inefficient and has poor accuracy. Summary of the Invention

[0009] The object of the present invention is to provide a method for determining the activation time of a liquid storage type reserve battery, which has high efficiency and good precision.

[0010] The technical solution for achieving the purpose of the present invention is:

[0011] A method for determining activation time of a liquid storage type reserve battery comprises the following steps:

[0012] S10, obtaining parameters of standard-sized batteries: selecting standard-sized batteries as pilot samples and independent verification samples, obtaining electrolyte infiltration mass-time curves of the pilot samples and the independent verification samples through simulation, and obtaining actual activation times of the pilot samples and the independent verification samples through experiments;

[0013] S20, obtaining prediction parameters: calculating the initial time scaling factor and the initial infiltration quality critical threshold of the lead sample based on the electrolyte infiltration quality-time curve of the lead sample and the actual activation time, performing error checking using the parameters of the independent verification sample to obtain the predicted time scaling factor and the predicted infiltration quality critical threshold;

[0014] S30, target battery activation time prediction: according to the target battery electrolyte infiltration quality-time curve, the time required for the target battery infiltration quality to reach the predicted critical threshold is obtained, and the activation time of the target battery is calculated in combination with the predicted time scaling factor.

[0015] Compared with the prior art, the present invention has the following significant advantages:

[0016] 1. High Accuracy: This method precisely establishes dynamic correlation rules between electrolyte wetting quality and battery activation time, quantifying the temporal coupling relationship between the wetting process and voltage response. Furthermore, by combining parameter migration with a closed-loop verification mechanism, it dynamically corrects model deviations, significantly reducing activation time prediction errors and significantly improving the accuracy and reliability of cross-architecture battery performance assessments.

[0017] 2. High efficiency: This invention addresses the computational complexity of multi-field coupling problems such as porous media seepage and dynamic electrochemical reactions. It adopts a step-by-step decoupling strategy, driven by experimental data and modeling, to avoid the calculation of simultaneous multi-physics field strong coupling simulation. At the same time, it avoids the large amount of cost consumption brought by the traditional exhaustive trial and error method. While ensuring accuracy, this invention improves the computational efficiency of the entire process and provides technical support for the rapid evaluation of battery activation time. It solves the problem that existing models cannot predict activation time, which seriously restricts the optimization design of battery rapid response characteristics, and constructs a universal prediction model. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is the main flow chart of the method for determining activation time of a liquid storage type backup battery of the present invention.

[0019] Figure 2 yes Figure 1 Flowchart of the steps for obtaining standard battery parameters.

[0020] Figure 3 yes Figure 1 Flowchart of the steps for obtaining prediction parameters.

[0021] Figure 4 yes Figure 3 Flowchart of the error checking steps. DETAILED DESCRIPTION

[0022] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0023] like Figure 1 As shown, the method for determining the activation time of a liquid storage type reserve battery of the present invention comprises the following steps:

[0024] S10, obtaining parameters of standard-sized batteries: selecting standard-sized batteries as pilot samples and independent verification samples, obtaining the electrolyte infiltration mass-time curves of the pilot samples and independent verification samples through simulation, and obtaining the actual activation time of the pilot samples and independent verification samples through experiments.

[0025] like Figure 2 As shown, the step S10, obtaining standard battery parameters includes:

[0026] S11, standard battery sample selection: select the first and second standard size batteries from the standard size batteries as the lead samples, and select the third standard size battery as the independent verification sample;

[0027] S12, obtaining a lead sample curve: obtaining, through simulation, an electrolyte infiltration mass-time curve of each lead sample and an independent verification sample;

[0028] S13, obtaining the activation time of the lead sample: through the activation experiment, the actual activation time of each lead sample and the independent verification sample is obtained respectively.

[0029] Specifically, taking the pilot sample as an example, the electrolyte infiltration quality-time curve is shown in Equation (1) and Equation (2). The time parameters for the electrolyte quality to reach the critical threshold are t i and t j The activation times of the two pilot sample batteries are T i and T j .

[0030] y i =M i (t)(1);

[0031] y j =M j (t)(2);

[0032] The liquid immersion time and activation time must satisfy the proportional relationship of physical laws as shown in formula (3). This formula constrains the model to follow the dynamic laws and avoid the risk of overfitting caused by pure data drive.

[0033] T i :T j =t i :t j (3);

[0034] This step uses the data of the microscopic infiltration process and the macroscopic power-up process as prior data to avoid prediction errors caused by using only single process data. At the same time, it ensures the consistency between the activation process predicted by the quantitative model and the actual data obtained, constrains the physical proportional relationship of the activation time under different sizes, ensures that the model conforms to the dynamic laws of the activation process, and guarantees the accuracy of the prediction.

[0035] S20, predictive parameter acquisition: Calculate the initial time scaling factor and the initial infiltration quality critical threshold of the lead sample based on the lead sample electrolyte infiltration quality-time curve and the actual activation time, perform error checking using independent verification sample parameters, and obtain the predicted time scaling factor and the predicted infiltration quality critical threshold.

[0036] like Figure 3 As shown, the step S20, obtaining prediction parameters, includes:

[0037] S21, inverse function model construction: based on the electrolyte infiltration mass-time curve of the lead sample, construct inverse function models of the electrolyte infiltration mass-time curve of two lead samples respectively;

[0038] S22, constructing an objective function: constructing an objective function according to the inverse function model of the electrolyte infiltration mass-time curve, as shown in the following formula;

[0039]

[0040] Where, are the inverse function models of the electrolyte wetting mass-time curves of the first and second standard size batteries of the pilot sample, T i and T j are the actual activation times of the first and second standard-sized batteries of the pilot sample, respectively;

[0041] S23, calculation of the initial parameter set of the pilot sample: by minimizing the objective function, the initial time scaling factor and the critical threshold of the infiltration quality of the pilot sample are calculated;

[0042] S24, error check: use the independent verification sample parameters and the initial time scaling factor of the pilot sample and the critical threshold of the infiltration quality to calculate the prediction error; if the prediction error exceeds the error threshold, return to S22, the objective function construction step, and recalculate the scaling factor, otherwise go to the next step.

[0043] like Figure 4 As shown, the error checking step S24 includes:

[0044] S241, independent verification sample activation time prediction: the independent verification sample predicted activation time is calculated according to the following formula:

[0045] Tpredict =k·t stmulatton (6);

[0046] Where, T predict Predict activation time for independent validation samples, t simulation is the time when the lead sample infiltration mass reaches the critical threshold according to the electrolyte infiltration mass-time curve of the independent verification sample, and k is the initial time scaling factor of the lead sample;

[0047] S242, prediction error calculation: Based on the independent verification sample prediction activation time and the independent verification sample actual activation time, the prediction error is calculated as follows:

[0048]

[0049] Where ε is the prediction error, T real To independently verify the actual activation time of the sample;

[0050] S243, calculation convergence judgment: Based on the prediction error and error threshold, judge the convergence of the standard battery time scaling factor calculation. When the calculation converges, go to S25, the prediction parameter group determination step; otherwise, go to S22, the objective function construction step.

[0051] S25, determining a prediction parameter group: setting the initial time scaling factor and the initial infiltration quality critical threshold of the lead sample as the prediction time scaling factor and the prediction infiltration quality critical threshold.

[0052] The above steps can obtain a combination of two parameters: the critical threshold of the infiltration mass and the time scaling factor. The time scaling factor normalizes the activation time of different samples to the same scale, eliminating the problem of inaccurate activation time prediction caused by differences in sample size. By forcing the infiltration rate (characterized by the inverse function) of different samples to satisfy an inverse proportional relationship with the activation time, that is, when the electrode volume increases, the time required for electrolyte infiltration is synchronously extended, and the activation time must be linearly related to this. This constraint essentially converts the physical laws of mass transport rate and electrochemical reaction time scale in fluid dynamics into mathematical expressions to ensure that the model prediction strictly follows the actual physical process.

[0053] S30, target battery activation time prediction: according to the target battery electrolyte infiltration quality-time curve, the time required for the target battery infiltration quality to reach the predicted critical threshold is obtained, and the activation time of the target battery is calculated in combination with the predicted time scaling factor.

[0054] The present invention quantifies the temporal coupling relationship between the electrolyte infiltration-time curve and the voltage-time curve by determining the time scaling factor and the critical threshold of the infiltration quality. Furthermore, by combining parameter migration with a closed-loop verification mechanism, the model deviation is dynamically corrected, significantly reducing the activation time prediction error and significantly improving the accuracy of cross-architecture battery performance evaluation. At the same time, a step-by-step decoupling strategy is adopted to avoid the calculation of multiple physical field strong coupling simulations simultaneously. While ensuring the accuracy of activation time prediction, the present invention improves computational efficiency and provides a high-precision solution tool for the fast activation design of liquid storage backup batteries.

Claims

1. A method for determining activation time of a liquid storage type reserve battery, characterized in that: The following steps are involved: S10, obtaining parameters of standard-sized batteries: selecting standard-sized batteries as pilot samples and independent verification samples, obtaining electrolyte infiltration mass-time curves of the pilot samples and the independent verification samples through simulation, and obtaining actual activation times of the pilot samples and the independent verification samples through experiments; S20, obtaining prediction parameters: calculating the initial time scaling factor and the initial infiltration quality critical threshold of the lead sample based on the electrolyte infiltration quality-time curve of the lead sample and the actual activation time, performing error checking using the parameters of the independent verification sample to obtain the predicted time scaling factor and the predicted infiltration quality critical threshold; S30, target battery activation time prediction: according to the target battery electrolyte infiltration quality-time curve, the time required for the target battery infiltration quality to reach the predicted critical threshold is obtained, and the activation time of the target battery is calculated in combination with the predicted time scaling factor.

2. The method for determining activation time of a liquid storage type reserve battery according to claim 1, wherein: The step of obtaining standard battery parameters in step S10 includes: S11, standard battery sample selection: select the first and second standard size batteries from the standard size batteries as the lead samples, and select the third standard size battery as the independent verification sample; S12, obtaining a lead sample curve: obtaining, through simulation, an electrolyte infiltration mass-time curve of each lead sample and an independent verification sample; S13, obtaining the activation time of the lead sample: through the activation experiment, the actual activation time of each lead sample and the independent verification sample is obtained respectively.

3. The method for determining activation time of a liquid storage type reserve battery according to claim 2, wherein: described S20, the prediction parameter acquisition step includes: S21, inverse function model construction: based on the electrolyte infiltration mass-time curve of the lead sample, construct inverse function models of the electrolyte infiltration mass-time curve of two lead samples respectively; S22, constructing an objective function: constructing an objective function according to the inverse function model of the electrolyte infiltration mass-time curve, as shown in the following formula; Where, are the inverse function models of the electrolyte wetting mass-time curves of the first and second standard size batteries of the pilot sample, T i and T j are the actual activation times of the first and second standard-sized batteries of the pilot sample, respectively; S23, calculation of the initial parameter set of the pilot sample: by minimizing the objective function, the initial time scaling factor and the critical threshold of the infiltration quality of the pilot sample are calculated; S24, error checking: using the independent verification sample parameters and the initial time scaling factor of the lead sample and the critical infiltration quality threshold, calculate the prediction error; if the prediction error exceeds the error threshold, return to S22, the objective function construction step, and recalculate the scaling factor; otherwise, go to the next step; S25, determining a prediction parameter group: setting the initial time scaling factor and the initial infiltration quality critical threshold of the lead sample as the prediction time scaling factor and the prediction infiltration quality critical threshold.

4. The method for determining activation time of a liquid storage type reserve battery according to claim 3, wherein: The error checking step S24 includes: S241, independent verification sample activation time prediction: the independent verification sample predicted activation time is calculated according to the following formula: Where, T predict Predict activation time for independent validation samples, t simulation is the time when the lead sample infiltration mass reaches the critical threshold according to the electrolyte infiltration mass-time curve of the independent verification sample, and k is the initial time scaling factor of the lead sample; S242, prediction error calculation: Based on the independent verification sample prediction activation time and the independent verification sample actual activation time, the prediction error is calculated as follows: Where ε is the prediction error, T real To independently verify the actual activation time of the sample; S243, calculation convergence judgment: Based on the prediction error and error threshold, judge the convergence of the standard battery time scaling factor calculation. When the calculation converges, go to S25, the prediction parameter group determination step; otherwise, go to S22, the objective function construction step.

Images