A battery life prediction model generation, prediction method, device and electronic device

Abstract

The present invention discloses a battery life prediction model generation, prediction method, device and electronic device, wherein the battery life prediction model generation method includes: deriving the mapping relationship between the time parameter in the battery chemical reaction mechanism and the battery capacity attenuation; performing equivalent conversion on the mapping relationship to generate a battery capacity attenuation model; using the historical data of the battery capacity to identify the model parameters of the battery capacity attenuation model, and generating a battery life prediction model based on the identification results of the model parameters. Compared with the P2D model of the prior art, the technical solution provided by the present invention has the advantages of fewer parameters, simple model and fast calculation speed. Under the premise of ensuring the accuracy of battery life prediction, the model price is reduced.

Description

Technical Field

[0001] The present invention relates to the field of algorithm design, and in particular to a battery life prediction model generation, prediction method, device and electronic equipment. Background Art

[0002] Renewable energy will play a vital role in future power systems, but its large-scale integration into the power grid will reduce its reliability and stability. Currently, renewable energy coupled with large-scale energy storage systems is considered a mainstream trend in future power systems, with battery storage being the primary method of energy storage. Battery energy storage systems are broadly categorized as traditional, non-reconfigurable battery energy storage networks and reconfigurable battery energy storage networks that incorporate flexible power electronic switch arrays. Battery energy storage systems are often composed of a large number of battery cells. Due to the low voltage and capacity of battery cells, different battery cells must be connected in series and parallel to meet the voltage and power requirements on the load side.

[0003] A reliable and efficient lithium-ion battery model is the basis for life prediction and fault diagnosis of battery management systems. Existing technologies often use a pseudo-two-dimensional (P2D) mechanism model described by partial differential equations. Commercial lithium-ion batteries are full-cell structures with a closed internal environment. Their internal physical and chemical behaviors are severely affected by temperature, rate, time, etc. The liquid phase diffusion coefficient, solid phase diffusion coefficient, and migration coefficient of the P2D model are strongly coupled in the model equation and are unmeasurable on-site. On the other hand, the strict P2D model has many parameters, is not easy to implement online calculation, and consumes a lot of computing resources, making it difficult to apply it to BMS. Therefore, it is an urgent problem to propose a new battery life prediction and related indicator evaluation method with a relatively simplified model that can guarantee accuracy. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a battery life prediction model generation, prediction method, device and electronic device, which reduce the complexity of the battery life prediction model.

[0005] According to a first aspect, an embodiment of the present invention provides a method for generating a battery life prediction model, the method comprising: deriving a mapping relationship between a time parameter in a battery chemical reaction mechanism and a battery capacity attenuation amount; performing an equivalent conversion on the mapping relationship to generate a battery capacity attenuation model; identifying model parameters of the battery capacity attenuation model using historical data on battery capacity, and generating a battery life prediction model based on the identification results of the model parameters.

[0006] Optionally, the battery chemical reaction mechanism includes at least solid electrolyte interface membrane reaction, loss of positive and negative electrode active materials and lithium plating phenomenon, and the mapping relationship between the time parameter and the battery capacity attenuation in the battery chemical reaction mechanism is derived, including: based on the generation rate of the solid electrolyte interface membrane, deriving the first sub-mapping relationship between the time parameter and the battery capacity attenuation under the solid electrolyte interface membrane reaction condition; based on the loss rate of the positive and negative electrode active materials, deriving the second sub-mapping relationship between the time parameter and the battery capacity attenuation under the positive and negative electrode active material loss condition; based on the lithium plating reaction current and the lithium plating reflection overpotential of the lithium plating phenomenon, deriving the third sub-mapping relationship between the time parameter and the battery capacity attenuation under the lithium plating phenomenon condition; combining the first sub-mapping relationship, the second sub-mapping relationship and the third sub-mapping relationship to obtain the mapping relationship.

[0007] Optionally, the model parameters of the battery capacity attenuation model are identified using the historical data of the battery capacity, including: generating a prior distribution of the model parameters using the historical data of the battery capacity and the initial parameters of the battery; calculating the posterior distribution of the prior distribution of the model parameters, and identifying the model parameters through the posterior distribution.

[0008] Optionally, before generating a battery life prediction model based on the identification results of the model parameters, the method also includes: updating parameters between the mapping relationship and the battery capacity attenuation model based on the comparison between the parameter items in the mapping relationship and the model parameters, and using the battery capacity attenuation model after the parameter update as the battery life prediction model.

[0009] Optionally, based on the comparison between the parameter items in the mapping relationship and the model parameters, parameters are updated between the mapping relationship and the battery capacity attenuation model, including: adjusting the corresponding parameter items of the battery chemical reaction mechanism in the mapping relationship based on the identified model parameters; and updating at least the decay ratio parameters in the battery capacity attenuation model through the corresponding parameter items to generate a battery life prediction model.

[0010] According to the second aspect, an embodiment of the present invention provides a battery life prediction method, which includes: inputting the current time parameter into a battery life prediction model generated by any method described in the first aspect to output the attenuation of the current battery capacity; and calculating the remaining battery usage time based on the attenuation of the current battery capacity.

[0011] According to the third aspect, an embodiment of the present invention provides a battery life prediction model generation device, which includes: a mechanism analysis unit, which is used to derive the mapping relationship between the time parameter in the battery chemical reaction mechanism and the battery capacity attenuation; a modeling unit, which is used to perform equivalent conversion on the mapping relationship to generate a battery capacity attenuation model; and a parameter identification unit, which is used to identify the model parameters of the battery capacity attenuation model using historical data on the battery capacity, and generate a battery life prediction model based on the identification results of the model parameters.

[0012] According to the fourth aspect, an embodiment of the present invention provides a battery life prediction device, which includes: an input unit for inputting the current time parameter into the battery life prediction model generated by the method described in any one of the first aspects to output the attenuation of the current battery capacity; an output unit for calculating the remaining battery usage time based on the attenuation of the current battery capacity.

[0013] According to the fifth aspect, an embodiment of the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the method described in the first aspect or any optional embodiment of the first aspect by executing the computer instructions.

[0014] According to the sixth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method described in the first aspect or any optional embodiment of the first aspect.

[0015] The technical solution provided by this application has the following advantages:

[0016] The technical solution provided by this application first derives the mapping relationship between the time parameter and the battery capacity attenuation through the electrochemical reaction mechanism of the battery. Then, based on the trend between the time parameter and the battery capacity attenuation presented by the mapping relationship, a battery capacity attenuation model of the battery capacity attenuation over time is derived. Then, the model parameters of the battery capacity attenuation model are identified using the historical data of the battery capacity, and a battery life prediction model is generated based on the identification results of the model parameters. Thus, based on the obtained software model for predicting the battery life, the battery life prediction result corresponding to the target time parameter is output. Compared with the P2D model of the prior art, this model has the advantages of fewer parameters, simple model and fast calculation speed. Under the premise of ensuring the accuracy of the battery life prediction, the model has been reduced in price.

[0017] In addition, the battery capacity decay model of the embodiment of the present invention is derived based on the electrochemical reaction mechanisms of solid electrolyte interface membrane reaction, loss of positive and negative electrode active materials, and lithium precipitation. Based on the characteristics of the three chemical mechanisms, the initial decay stage of battery capacity is dominated by solid electrolyte interface membrane reaction, the mid-term is dominated by loss of positive and negative electrode active materials, and the late stage is dominated by lithium precipitation. This fully covers the main chemical reactions in the battery life cycle, further improving the accuracy of battery life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The features and advantages of the present invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present invention in any way. In the accompanying drawings:

[0019] Figure 1 A schematic diagram showing the steps of a method for generating a battery life prediction model in one embodiment of the present invention is shown;

[0020] Figure 2 A schematic flow chart of a method for generating a battery life prediction model in one embodiment of the present invention is shown;

[0021] Figure 3 A schematic diagram showing the steps of a battery life prediction method according to one embodiment of the present invention is shown;

[0022] Figure 4 A schematic structural diagram of a battery life prediction model generating device according to one embodiment of the present invention is shown;

[0023] Figure 5 A schematic structural diagram of a battery life prediction device according to one embodiment of the present invention is shown;

[0024] Figure 6 A schematic structural diagram of an electronic device in one embodiment of the present invention is shown. DETAILED DESCRIPTION

[0025] To make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0026] Renewable energy will play a crucial role in future power systems, but its large-scale integration into the power grid can reduce its reliability and stability. Currently, renewable energy coupled with large-scale energy storage systems is considered a mainstream trend in future power systems, with battery storage being the predominant form of energy storage. Battery energy storage systems can be broadly categorized as traditional non-reconfigurable battery energy storage networks and reconfigurable battery energy storage networks that incorporate flexible power electronic switch arrays. Battery energy storage systems typically consist of a large number of battery cells. Due to their low voltage and capacity, they must be connected in series and parallel to meet load-side voltage and power requirements. Reliable and efficient lithium-ion battery models are fundamental to state estimation and fault diagnosis in battery management systems. The P2D model, constructed by Doyle et al. and Fuller et al., is an electrochemical model based on the microscopic structure of anode-separator-cathode in lithium-ion batteries. The P2D model primarily simulates the migration and diffusion of lithium ions in the electrodes and electrolyte during the battery's charge and discharge processes. During ion transport, the three conservation conditions of mass conservation, charge conservation, and electrochemical kinetic equilibrium are met. Fick's second law in polar coordinates is used to describe the diffusion behavior of lithium ions in the electrode solid phase. In the liquid phases of the electrodes and separators, the diffusion and migration of lithium ions are described using concentrated solution theory and Fick's second law. Charge balance in the solid, liquid, and separator phases is described using Ohm's law. The liquid phase potential equilibrium is affected not only by the diffusion and migration behavior of lithium ions, but also by the power supply or load current. Electrochemical reactions occur at the interface between the solid active particle surface and the electrolyte solution, and the Butler-Volmer equation is used to describe the overpotential as a function of current density. However, commercial lithium-ion batteries are full-cell structures with a closed internal environment. Their internal physicochemical behavior is significantly affected by temperature, rate, and time. The P2D model's liquid-phase diffusion coefficient, solid-phase diffusion coefficient, and migration coefficient are strongly coupled in the model equations, making them unmeasurable in the field and thus limiting the accuracy of battery life predictions. On the other hand, the strict P2D model is not easy to implement online calculation, and the high consumption of computing resources makes it difficult to be applied on BMS.

[0027] Therefore, please see Figure 1 and Figure 2 In one embodiment, the present invention provides a method for generating a battery life prediction model, which specifically includes the following steps:

[0028] Step S101: deriving a mapping relationship between a time parameter in a battery chemical reaction mechanism and a battery capacity attenuation amount.

[0029] Step S102: performing equivalent conversion on the mapping relationship to generate a battery capacity attenuation model.

[0030] Step S103: using the historical data of the battery capacity to identify the model parameters of the battery capacity attenuation model, and generating a battery life prediction model based on the identification results of the model parameters.

[0031] Specifically, this embodiment first derives a mapping relationship between time parameters and battery capacity decay based on the impact of battery chemical reactions on battery capacity. For example, the change in battery capacity decay as the time parameter changes at a given time point or time period. Battery chemical reaction mechanisms include, but are not limited to, degradation of active materials, aging and deterioration of motor coatings such as conductive agents and adhesives, and film formation and decomposition of electrolytes. Based on the detailed parameter values ​​between the time parameter and battery capacity decay, the specific parameters that determine the specific proportional relationship between the time parameter and battery capacity decay are determined. A mathematical equation is then created based on these parameters to form a battery capacity decay model. Finally, the model parameters of the battery capacity decay model are identified using real-world historical data representing battery capacity changes to determine the correct parameters within the model, thereby generating a battery life prediction model. Based on the resulting software model for predicting battery life, a battery life prediction result corresponding to the target time parameters is output. Compared to existing P2D models, this model offers advantages in terms of fewer parameters, simpler modeling, and faster computational speed. While maintaining the accuracy of battery life prediction, the model has been streamlined.

[0032] Specifically, in one embodiment, the above step S101 specifically includes the following steps:

[0033] Step 1: Based on the generation rate of the solid electrolyte interface film, derive the first sub-mapping relationship between the time parameter and the battery capacity attenuation under the solid electrolyte interface film reaction conditions.

[0034] Step 2: Based on the loss rate of the positive and negative electrode active materials, derive the second sub-mapping relationship between the time parameter and the battery capacity attenuation under the condition of positive and negative electrode active material loss.

[0035] Step 3: Based on the lithium deposition reaction current and lithium deposition reflection overpotential of the lithium deposition phenomenon, derive the third sub-mapping relationship between the time parameter and the battery capacity attenuation under the conditions of the lithium deposition phenomenon.

[0036] Step 4: Combine the first sub-mapping relationship, the second sub-mapping relationship, and the third sub-mapping relationship to obtain a mapping relationship.

[0037] Specifically, in this embodiment, the battery chemical reaction mechanism uses at least the formation of a solid electrolyte interface (SEI), loss of active materials (LAM) at the positive and negative electrodes, and lithium plating (LP) to create a corresponding mapping relationship. During the initial charge and discharge process of a typical lithium-ion battery, the electrode material and the electrolyte react at the solid-liquid interface to form a passivation layer covering the electrode surface, namely the solid electrolyte interface. This passivation layer has the characteristics of a solid electrolyte and is an electronic insulator but an excellent conductor of lithium ions. Lithium ions can freely embed and de-embedding through the SEI film. The formation of the SEI film consumes some lithium ions and also causes a reduction in battery capacity. The loss of active materials at the positive and negative electrodes is mainly caused by reasons such as electrode material loss, current collector deformation, and transition metal dissolution. Active material loss occurs during daily static and charge-discharge cycles. Lithium plating occurs when lithium ions de-embedding at the positive electrode cannot embed into the negative electrode. The lithium ions are forced to precipitate on the negative electrode surface, forming a layer of gray material. Through actual experiments, we know that the above-mentioned battery chemical reactions can divide the battery capacity decay process into three stages. The chemical reaction that causes the fastest battery capacity decay in the initial stage is the formation of a solid electrolyte interface membrane. In the middle stage, the chemical reaction that causes the fastest capacity decay is the loss of active materials (LAM) at the positive and negative electrodes. In the later stage, if the battery exhibits lithium plating (LP), the available capacity of the battery will decay sharply, and the battery life usually reaches its end at this stage.

[0038] 1. The side reaction equation of SEI film growth on the electrode surface can be expressed as follows:

[0039]

[0040] Among them, S is the reactant, P is the reduction product, Li is the lithium ion, and e is the electron. During the battery aging process, the lithium battery capacity attenuation Q caused by the growth and thickening of the SEI film is SEI The equivalent side reaction current density i s The reaction process is directly proportional to the reaction time.

[0041]

[0042] Where A is the reaction contact area.

[0043] This example makes the following assumptions about the battery reaction process: assuming that there is only one representative particle in the anode and cathode of the battery, and that parameters such as concentration and current density are calculated using average values, the capacity loss expression can be expressed as follows:

[0044]

[0045] In the above formula, k SEI is the SEI film reaction rate, E SEI is the activation energy of SEI film formation, R is the ideal gas constant, T is the Kelvin temperature, λ and θ are intermediate parameters. After laboratory calibration, λ is set to 86995, and θ can be taken as 2.32 at 25°C. Solving the above formula, we can see that after the integration is completed, the first mapping relationship between battery capacity attenuation and time parameters is obtained:

[0046]

[0047] Through the above first mapping relationship, it can be obtained that the battery capacity attenuation Q SEI Proportional to the 1 / 2 power of the reaction time.

[0048] 2. Model the mechanism of the loss of active materials at the positive and negative electrodes. During the charge and discharge process, lithium ions are freely inserted and removed. During this period, the material is affected by different mechanical stresses, and the structure and composition change slightly. Some active materials are separated, resulting in lithium ion loss. The relationship between the loss rate of active materials at the positive and negative electrodes and the battery capacity attenuation can be described by the following formula:

[0049]

[0050] Where N represents the total number of equivalent cycles, p and n represent the positive and negative electrodes respectively. It represents the loss rate of active substances, which is affected by factors such as temperature and activation energy of side reactions. Its expression is as follows:

[0051]

[0052] in, is the coefficient factor, is the side reaction activation energy, is the power law coefficient, R is the ideal gas constant, and T is the Kelvin temperature.

[0053] In this embodiment, in order to simplify the calculation of the above formula, it is considered that the loss rate ratio of the positive and negative active materials of the lithium battery is constant, and the equivalent power law Thus, the following relationship exists:

[0054]

[0055]

[0056] That is, the battery capacity attenuation caused by the loss of positive and negative active materials is in a negative exponential relationship, which is proportional to and In practical applications, it can be approximately equivalent to the battery aging decay rate f d , where f d It has a functional relationship with the time parameter and is an intermediate quantity used to determine the battery capacity degradation rate. It decomposes the degradation process into calendar aging and cycle aging. At the battery cell level, the battery degradation ratio f d It can be expressed as follows:

[0057]

[0058] Among them, f t is the calendar aging rate, f c is the cycle aging ratio, δ i is the discharge depth of the i-th cycle, σ i is the SOC of the i-th cycle, is the average SOC, is the average temperature, n i Represents the number of stress cycles, and N represents the total number of cycles. Through the above formula, the second mapping relationship between time parameter and battery capacity attenuation is obtained:

[0059] 3. Lithium plating is the most serious side reaction in the aging process of lithium-ion batteries. It will lead to accelerated capacity decay. When the lithium plating phenomenon develops to a certain extent, lithium dendrites will grow inside the lithium battery. After the lithium dendrites pierce the diaphragm, it will cause internal short circuit of the battery and even thermal runaway and combustion. The analysis of the lithium plating phenomenon can be described by the Butler-Volmer equation, resulting in the third mapping relationship:

[0060]

[0061] where i 0,Li is the lithium deposition reaction current, α a and α c is the charge transfer coefficient, η Li is the overpotential of the lithium deposition reaction, F is the Faraday constant, R is the ideal gas constant, and T is the Kelvin temperature. The above equation shows that during lithium deposition, the battery capacity decay caused by lithium deposition exhibits a negative exponential relationship. The mapping relationship for battery capacity decay is the sum of the first, second, and third mapping relationships described above.

[0062] In this embodiment, considering the complex factors of lithium battery aging and the many side reactions, this patent divides the full life cycle of lithium batteries into three stages. The first stage is the SEI film growth period of the lithium battery, during which the capacity decay of the lithium battery is dominated by the SEI film growth process and is accompanied by other side reactions. The second stage is the cyclic charge and discharge degradation stage, during which the capacity decay is mainly dominated by the loss of active materials near the electrode. The third stage is the diving stage, at which time the available capacity of the lithium battery drops rapidly, and its capacity decay is dominated by the lithium precipitation reaction. Therefore, based on the above mapping relationship, an equivalent conversion is performed to create a battery capacity decay model.

[0063] In actual operation, the degradation rate of lithium batteries over their entire life cycle is nonlinear. During the overall aging process, the degradation rate is related to the number of remaining active lithium ions in the battery and the stability of different structural materials. Assuming that the degradation rate within each small cycle of the lithium battery is linear, the degradation rate of each charge and discharge cycle is expressed as follows, where L is a normalized value between 0 and 1, representing the overall battery capacity decay, with 0 representing a new battery. Battery life is usually defined as the battery can only provide 80% of its rated maximum capacity, that is, L = 0.2. Battery capacity decay rate f L It represents the derivative of the battery capacity with respect to the number of charge and discharge cycles. The derivative characterizes the battery capacity degradation rate, and the degradation rate can be expressed using the degradation ratio. The schematic formula is as follows, and the detailed formula is not repeated here.

[0064] dL / d(Cycle)=f L (L,f d )

[0065] An important factor that leads to the nonlinearity of the degradation rate is that the rate is proportional to the number of active lithium ions remaining in the battery. That is, the more the remaining number, the more lithium ions participate in the reaction. Correspondingly, the degree of various degradation side reactions is high and the degradation rate is fast. Therefore, the overall situation is fast at first and then slow, and the active lithium ions are more at first and then less. This can be expressed as f L (L,f d )=(1-L)f d (t,δ,σ,T), that is, the more remaining lithium ions, the smaller L, the larger 1-L, and f L The bigger.

[0066] Performing integral processing can obtain Therefore, it is further explained that the cycle decay characteristics of lithium batteries show exponential function characteristics.

[0067] According to the three stages of the full life decay process, the present invention performs equivalent conversion based on the above mapping relationship and proposes the following lithium battery capacity decay model:

[0068]

[0069] Where L is the overall battery capacity attenuation, α sei , α normal is the coefficient of each stage of customization, M is the custom rate coefficient, β r is the lithium deposition reaction rate coefficient. Through parameter fitting, it can be found that the battery capacity attenuation in the first stage is α sei t 0.5 The middle part is dominated by In the diving stage, according to the historical operation data and attenuation rate of lithium ion, its decline rate is faster. The decline is accelerated, forming an upward convex shape. In the initial stage of battery capacity decay, solid electrolyte interface reactions dominate, in the middle stage, the loss of positive and negative electrode active materials dominates, and in the later stage, lithium deposition dominates. This comprehensive coverage of the major chemical reactions in the battery life cycle further improves the accuracy of battery life prediction and simplifies the complexity of battery capacity prediction models.

[0070] Specifically, in one embodiment, the above step S103 specifically includes the following steps:

[0071] Step 5: Generate the prior distribution of model parameters using historical data of battery capacity and initial battery parameters.

[0072] Step 6: Calculate the posterior distribution of the prior distribution of the model parameters and identify the model parameters through the posterior distribution.

[0073] Specifically, in this embodiment, considering that the battery needs to judge the battery status in real time during actual operation, it is necessary to add a link for real-time parameter acquisition in the battery management system, that is, to obtain the α SEI , α normal , M and other parameters, this patent adopts the Bayesian-Monte Carlo algorithm to fit the model parameters. Considering that the structure and materials of batteries are very similar during the manufacturing process, the initial values ​​of the parameters and the prior distribution of the parameters can be obtained through the historical data of the same type of batteries. After obtaining the initial values ​​and historical data, the posterior probability distribution P(x k |SOH 1:k ), where x k is the state parameter set in the battery capacity decay model, which includes α SEI , α normal etc. Among them, SOH 1:k-1 A historical data sequence of battery health status defined as:

[0074] SOH 1:k-1=[SOH(1),SOH(2),...,SOH(k-1)]

[0075] In the algorithm framework, let’s assume that the P(x k-1 |SOH 1:k-1 ) is known, then

[0076] P(x k |SOH 1:k-1 )=∫P(x k |x k-1 )P(x k-1 |SOH 1:k-1 )dx k-1

[0077] When entering the kth period, SOH(k) can be obtained by combining the ampere-hour integration with the battery capacity attenuation model. At this time, the following relationship can be obtained through the Bayesian formula:

[0078]

[0079] Considering that the calculation of the integral term in the above formula is too complicated, the Monte Carlo method is introduced into the calculation process to simplify the calculation and obtain the following formula:

[0080]

[0081] where N s is the random sample size, is from P(x k |SOH 1:k ), δ(·) is the Dirac function, Every The corresponding Bayesian weight, its iterative calculation formula can be expressed as follows:

[0082]

[0083] In the above series of formulas, x k The initial value can be selected as x k =x k-1 +w, w is the noise, after substituting it into the formula, we get the probability distribution, and finally determine the final x by the expected value of the posterior distribution k In this embodiment, the α obtained after the above parameter identification is SEI , α normalBy identifying parameters such as SOH, an accurate battery capacity decay model can be obtained. The resulting battery capacity decay model can then be used as a battery life prediction model. This invention introduces the Bayesian-Monte Carlo method for online parameter identification. This algorithm leverages historical data to obtain the probability distribution of the SOH under the current state and updates the model parameters in real time. The updated parameters can then be used in the next iteration to predict the SOH for the next period. This further improves the accuracy of the battery life prediction model.

[0084] Specifically, in one embodiment, after the model parameters are identified in the above steps, the method for generating a battery life prediction model provided by this embodiment further includes the following steps:

[0085] Step 7: Based on the comparison between the parameter items in the mapping relationship and the model parameters, update the parameters between the mapping relationship and the battery capacity attenuation model, and use the battery capacity attenuation model after the parameter update as the battery life prediction model.

[0086] Specifically, in this embodiment, since the model parameters in the battery life prediction model are obtained by equivalent conversion based on the corresponding parameter items in the mapping relationship, based on the identified model parameters, the real-time external characteristics of the battery (including voltage, current, SOH evaluation value) are substituted into the corresponding parameter items in the mapping relationship, and the parameters of the mapping relationship are updated, and then the model parameters in the battery life prediction model are iteratively updated, thereby further improving the accuracy of the battery life prediction model. In this embodiment, taking the mapping relationship of steps 1 to 4 above as an example, the parameter update step includes:

[0087] 1. Adjust the corresponding parameter items of the battery chemical reaction mechanism within the mapping relationship based on the identified model parameters.

[0088] 2. Update at least the decay ratio parameter in the battery capacity decay model through the corresponding parameter item to generate a battery life prediction model.

[0089] For example: After identifying α sei , α normal After the parameters such as , M are obtained, the obtained parameters are compared with the parameter items in the mapping relationship. For example, for the parameter α sei .

[0090] α sei With the formula in (hereinafter referred to as the target part) is corresponding.

[0091] Thus according to α seiAfter adjusting the parameters in the target part, the battery capacity attenuation model is introduced based on the adjusted parameters to update f d A new battery life prediction model is obtained. At this time, the battery life prediction model is used to calculate the new battery capacity, so that the aging degree of the battery capacity is more accurate. In addition, in this embodiment, the real-time external characteristics of the battery, namely the voltage, current, and SOH evaluation value are substituted into the detailed formula of the above mapping relationship to calculate the aging degree of the battery under different side reactions, and the internal aging degree of the battery can also be evaluated.

[0092] See also Figure 3 In one embodiment, the present invention provides a battery life prediction method, which specifically includes the following steps:

[0093] Step S201: inputting the current time parameter into the battery life prediction model generated by the above-mentioned model generation method to output the attenuation of the current battery capacity.

[0094] Step S202: Calculate the remaining battery life based on the attenuation of the current battery capacity.

[0095] Specifically, the time parameters such as the current time point or time period are input into the battery life prediction model generated by the above-mentioned model generation method, so as to quickly and accurately obtain the attenuation of the current battery capacity. The calculation process within the model refers to the relevant description of the above-mentioned model generation method embodiment and will not be repeated here. The attenuation of the current battery capacity is then compared with the preset life threshold to obtain the difference between the life threshold and the current capacity attenuation. The time parameter corresponding to the difference is then inferred based on the battery life prediction model to calculate the accurate remaining battery life.

[0096] Through the above steps, the technical solution provided by this application first derives the mapping relationship between the time parameter and the battery capacity attenuation through the electrochemical reaction mechanism of the battery. Then, based on the trend between the time parameter and the battery capacity attenuation presented by the mapping relationship, a battery capacity attenuation model of the battery capacity attenuation over time is derived, and then the model parameters of the battery capacity attenuation model are identified using the historical data of the battery capacity, and a battery life prediction model is generated based on the identification results of the model parameters. Thus, based on the obtained software model for predicting battery life, the battery life prediction result corresponding to the target time parameter is output. Compared with the P2D model of the prior art, this model has the advantages of fewer parameters, simple model and fast calculation speed. Under the premise of ensuring the accuracy of battery life prediction, the model price is reduced.

[0097] In addition, the battery capacity decay model of the embodiment of the present invention is derived based on the electrochemical reaction mechanisms of solid electrolyte interface membrane reaction, loss of positive and negative electrode active materials, and lithium precipitation. Based on the characteristics of the three chemical mechanisms, the initial decay stage of battery capacity is dominated by solid electrolyte interface membrane reaction, the mid-term is dominated by loss of positive and negative electrode active materials, and the late stage is dominated by lithium precipitation. This fully covers the main chemical reactions in the battery life cycle, further improving the accuracy of battery life prediction.

[0098] like Figure 4 As shown, this embodiment also provides a battery life prediction model generation device, which includes:

[0099] The mechanism analysis unit 101 is used to derive the mapping relationship between the time parameter in the battery chemical reaction mechanism and the battery capacity attenuation. For details, please refer to the relevant description of step S101 in the above method embodiment, which will not be repeated here.

[0100] The modeling unit 102 is used to perform equivalent conversion on the mapping relationship to generate a battery capacity attenuation model. For details, please refer to the relevant description of step S102 in the above method embodiment, which will not be repeated here.

[0101] The parameter identification unit 103 is used to identify the model parameters of the battery capacity degradation model using the historical battery capacity data, and generate a battery life prediction model based on the identification results of the model parameters. For details, please refer to the relevant description of step S103 in the above method embodiment, which will not be repeated here.

[0102] The battery life prediction model generation device provided in an embodiment of the present invention is used to execute the battery life prediction model generation method provided in the above embodiment. Its implementation method and principle are the same. For details, please refer to the relevant description of the above method embodiment and will not be repeated here.

[0103] like Figure 5 As shown, this embodiment also provides a battery life prediction device, which includes:

[0104] Input unit 201 is used to input the current time parameter into the battery life prediction model generated by the above-mentioned battery life prediction model generation method to output the current battery capacity attenuation. For details, please refer to the relevant description of step S201 in the above-mentioned method embodiment, which will not be repeated here.

[0105] The output unit 202 is used to calculate the remaining battery life based on the attenuation of the current battery capacity. For details, please refer to the relevant description of step S202 in the above method embodiment, which will not be repeated here.

[0106] The battery life prediction device provided in an embodiment of the present invention is used to execute the battery life prediction method provided in the above embodiment. Its implementation method and principle are the same. For details, please refer to the relevant description of the above method embodiment and will not be repeated here.

[0107] Through the collaborative cooperation of the above-mentioned components, the technical solution provided by this application first derives the mapping relationship between the time parameter and the battery capacity attenuation through the electrochemical reaction mechanism of the battery. Then, based on the trend between the time parameter and the battery capacity attenuation presented by the mapping relationship, a battery capacity attenuation model of the battery capacity attenuation over time is equivalently derived, and then the model parameters of the battery capacity attenuation model are identified using the historical data of the battery capacity, and a battery life prediction model is generated based on the identification results of the model parameters. Thus, based on the obtained software model for predicting battery life, the battery life prediction result corresponding to the target time parameter is output. Compared with the P2D model of the prior art, this model has the advantages of fewer parameters, simple model and fast calculation speed. Under the premise of ensuring the accuracy of battery life prediction, the model price is reduced.

[0108] In addition, the battery capacity decay model of the embodiment of the present invention is derived based on the electrochemical reaction mechanisms of solid electrolyte interface membrane reaction, loss of positive and negative electrode active materials, and lithium precipitation. Based on the characteristics of the three chemical mechanisms, the initial decay stage of battery capacity is dominated by solid electrolyte interface membrane reaction, the mid-term is dominated by loss of positive and negative electrode active materials, and the late stage is dominated by lithium precipitation. This fully covers the main chemical reactions in the battery life cycle, further improving the accuracy of battery life prediction.

[0109] Figure 6 An electronic device according to an embodiment of the present invention is shown, which includes a processor 901 and a memory 902, which can be connected via a bus or other means. Figure 6 The bus connection is taken as an example.

[0110] The processor 901 may be a central processing unit (CPU). The processor 901 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0111] Memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as the program instructions / modules corresponding to the methods in the above-described method embodiments. Processor 901 executes the non-transitory software programs, instructions, and modules stored in memory 902 to perform various processor functions and data processing, thereby implementing the methods in the above-described method embodiments.

[0112] The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created by the processor 901, etc. In addition, the memory 902 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 902 may optionally include a memory remotely located relative to the processor 901, and these remote memories may be connected to the processor 901 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0113] One or more modules are stored in the memory 902 and, when executed by the processor 901 , perform the method in the above method embodiment.

[0114] The specific details of the above electronic device can be understood by referring to the corresponding descriptions and effects in the above method embodiments, and will not be repeated here.

[0115] Those skilled in the art will appreciate that all or part of the processes in the above-described embodiments can be implemented by instructing related hardware through a computer program. The implemented program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-described embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD). The storage medium can also include a combination of the above-mentioned types of memory.

[0116] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for generating a battery life prediction model, characterized in that: The method comprises: Based on the generation rate of the solid electrolyte interface film, the first sub-mapping relationship between the time parameter and the battery capacity attenuation under the solid electrolyte interface film reaction conditions is derived; Based on the loss rate of positive and negative electrode active materials, the second sub-mapping relationship between the time parameter and the battery capacity attenuation under the condition of positive and negative electrode active material loss is derived; Based on the lithium deposition reaction current and lithium deposition overpotential, the third sub-mapping relationship between the time parameter and the battery capacity attenuation under the lithium deposition phenomenon is derived; Combining the first sub-mapping relationship, the second sub-mapping relationship, and the third sub-mapping relationship to obtain a mapping relationship between a time parameter in a battery chemical reaction mechanism and a battery capacity attenuation amount; Based on the detailed parameter values ​​between the time parameter and the battery capacity attenuation, determine which specific parameters have a specific proportional relationship between the time parameter and the battery capacity attenuation, and then create a mathematical equation based on the relevant parameters as a battery capacity attenuation model; Generate a priori distribution of model parameters using the historical data of battery capacity and initial battery parameters; Calculating a posterior distribution of a prior distribution of the model parameters, and identifying the model parameters through the posterior distribution; A battery life prediction model is generated based on the identification results of the model parameters.

2. The method according to claim 1, characterized in that Before generating the battery life prediction model based on the identification results of the model parameters, the method further includes: According to the comparison between the parameter items in the mapping relationship and the model parameters, parameters are updated between the mapping relationship and the battery capacity attenuation model, and the battery capacity attenuation model after parameter update is used as a battery life prediction model.

3. The method according to claim 2, characterized in that According to a comparison between the parameter items in the mapping relationship and the model parameters, updating parameters between the mapping relationship and the battery capacity attenuation model includes: Adjusting corresponding parameter items of the battery chemical reaction mechanism within the mapping relationship based on the identified model parameters; At least the decay ratio parameter in the battery capacity decay model is updated by using the corresponding parameter item to generate a battery life prediction model.

4. A battery life prediction method, characterized in that: The method comprises: Inputting the current time parameter into the battery life prediction model generated by the method according to any one of claims 1 to 3 to output the attenuation of the current battery capacity; The remaining battery usage time is calculated based on the attenuation of the current battery capacity.

5. A battery life prediction model generation device, characterized in that: The device comprises: A mechanism analysis unit is used to derive a first sub-mapping relationship between the time parameter and the battery capacity attenuation under the condition of a solid electrolyte interface membrane reaction based on the generation rate of the solid electrolyte interface membrane; a second sub-mapping relationship between the time parameter and the battery capacity attenuation under the condition of loss of positive and negative electrode active materials based on the loss rate of the positive and negative electrode active materials; a third sub-mapping relationship between the time parameter and the battery capacity attenuation under the condition of lithium deposition phenomenon based on the lithium deposition reaction current and the lithium deposition reflection overpotential of the lithium deposition phenomenon; the first sub-mapping relationship, the second sub-mapping relationship and the third sub-mapping relationship are combined to obtain a mapping relationship between the time parameter and the battery capacity attenuation in the battery chemical reaction mechanism; a modeling unit is used to determine, based on the detailed parameter values ​​between the time parameter and the battery capacity attenuation, which specific parameters have a specific proportional relationship between the time parameter and the battery capacity attenuation, and then create a mathematical equation based on the relevant parameters as a battery capacity attenuation model; A parameter identification unit is used to generate a prior distribution of model parameters using the historical data of the battery capacity and the initial parameters of the battery; calculate the posterior distribution of the prior distribution of the model parameters and identify the model parameters through the posterior distribution; and generate a battery life prediction model based on the identification results of the model parameters.

6. A battery life prediction device, characterized in that: The device comprises: An input unit, configured to input a current time parameter into a battery life prediction model generated by the method according to any one of claims 1 to 3, so as to output a current battery capacity attenuation; The output unit is configured to calculate the remaining battery life based on the attenuation of the current battery capacity.

7. An electronic device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the method according to any one of claims 1 to 3 or 4 by executing the computer instructions.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method according to any one of claims 1 to 3 or 4.

Images