Cycle life prediction method and system for sodium ion battery

Abstract

The invention provides a cycle life prediction method and system for a sodium ion battery. The method comprises the following steps: modeling a target sodium ion battery to obtain an equivalent circuit of the sodium ion battery; solving the residual electric quantity of the sodium ion battery at each moment based on the equivalent circuit; detecting the residual electric quantity and circuit parameters of the target sodium ion battery under different cycle times; calculating the correlation between the circuit parameters and the residual electric quantity, and screening out the circuit parameters of which the correlation is greater than a set value; inputting the screened circuit parameters and the residual electric quantity into a neural network model for training to obtain a cycle life prediction model; and completing the prediction of the cycle life of the target sodium ion battery by using the cycle life prediction model. According to the method, the cycle life of the sodium ion battery is predicted, so that the performance reduction trend of the battery can be identified in advance, a worker is helped to formulate a use strategy of the battery more scientifically, and the overall efficiency of an energy management system is improved.

Description

Technical Field

[0001] The present invention relates to the field of battery technology, and in particular to a method and system for predicting the cycle life of a sodium ion battery. Background Art

[0002] With the widespread use of renewable energy and the rapid development of electric vehicles, the demand for battery storage technology is growing. Sodium-ion batteries have gradually become an alternative to lithium-ion batteries due to their abundant raw materials, low cost and environmental friendliness, especially in the fields of large-scale energy storage and electric vehicles. However, the cycle life and performance stability of sodium-ion batteries are still important factors limiting their widespread application.

[0003] In practical applications, effective monitoring and maintenance of sodium-ion batteries are crucial. Traditional battery management systems mainly rely on empirical methods and simple mathematical models to predict battery performance, which often cannot accurately reflect the actual state of the battery, resulting in premature retirement or performance degradation of the battery. This not only increases operation and maintenance costs, but may also cause safety hazards. Therefore, developing an effective cycle life prediction method that can provide a scientific basis for the health management of sodium-ion batteries is an important topic in current battery technology research. Summary of the invention

[0004] To solve the above problems, the object of the present invention is to provide a method and system for predicting the cycle life of a sodium ion battery.

[0005] A method for predicting the cycle life of a sodium ion battery, comprising:

[0006] Step 1: Model the target sodium-ion battery to obtain an equivalent circuit of the sodium-ion battery;

[0007] Step 2: Calculate the remaining power of the sodium-ion battery at each time based on the equivalent circuit;

[0008] Step 3: Detect the remaining power and circuit parameters of the target sodium ion battery at different cycle times;

[0009] Step 4: Calculate the correlation between the circuit parameters and the remaining power, and select the circuit parameters whose correlation is greater than the set value;

[0010] Step 5: Input the screened circuit parameters and remaining power into the neural network model for training to obtain a cycle life prediction model;

[0011] Step 6: Use the cycle life prediction model to complete the prediction of the cycle life of the target sodium ion battery.

[0012] Preferably, the step 2: solving the remaining power of the sodium ion battery at each time based on the equivalent circuit includes:

[0013] Step 2.1: Use the second-order RC equivalent circuit model to construct the equivalent circuit of the target sodium-ion battery;

[0014] Step 2.2: Model the second-order RC equivalent circuit model based on Kirchhoff's law to obtain the battery remaining power function;

[0015] Step 2.3: Discretize the battery remaining capacity function to obtain a working state model;

[0016] Step 2.4: Solve the working state model to obtain the remaining power of the sodium ion battery at each moment.

[0017] Preferably, in step 2.2, the battery remaining power function is:

[0018]

[0019] Among them, U L Indicates terminal voltage, U OC represents the open circuit voltage, U1 represents the terminal voltage of the first RC loop, U2 represents the terminal voltage of the second RC loop, I represents the battery operating current, R1 represents the first polarization resistance, R2 represents the second polarization resistance, C1 represents the first polarization capacitance, C2 represents the second polarization capacitance, k represents the current time, k0 represents the initial time, η represents the discharge efficiency, Q c Indicates the rated capacity of the battery, S OC (k) indicates the remaining power at the current moment.

[0020] Preferably, the step 2.3: discretizing the battery remaining capacity function to obtain the working state model includes:

[0021] The system noise is introduced to discretize the battery remaining capacity function to obtain the working state model; wherein the working state model is:

[0022]

[0023] Among them, ω k represents the system noise at time k, and Δk represents the time difference.

[0024] Preferably, the step 2.4: solving the working state model to obtain the remaining power of the sodium ion battery at each moment includes:

[0025] The least squares method is used to identify the parameters of the working state model to obtain the remaining power of the sodium ion battery at each moment; the recursive formula of the least squares method is:

[0026]

[0027] in, represents the parameter identification value at time k, y(k) represents the actual value of the parameter collected at time k, P(k) represents the error covariance matrix at time k, K(k) represents the algorithm gain factor at time k, represents the information matrix at time k, and λ represents the forgetting factor.

[0028] Preferably, the step 4: calculating the correlation between the circuit parameters and the remaining power, and screening out the circuit parameters whose correlation is greater than the set value, includes:

[0029] Using the formula:

[0030]

[0031] Where r represents the correlation, n represents the number of circuit parameter and remaining power data pairs in a measurement cycle, ∑xy represents the product of circuit parameter and remaining power, Σx represents the sum of circuit parameters in a measurement cycle, Σy represents the sum of remaining power in a measurement cycle, Σx 2 represents the sum of squares of circuit parameters within a measurement cycle, Σy 2 Indicates the sum of squares of remaining power in a measurement cycle.

[0032] Preferably, the step 5: inputting the screened circuit parameters and the remaining power into a neural network model for training to obtain a cycle life prediction model comprises:

[0033] Step 5.1: Normalize the screened circuit parameters and remaining power to obtain training samples; the normalization formula is:

[0034]

[0035] Among them, x″ represents the normalized parameter, x′ represents the input value, μ represents the mean of the input value, and σ represents the standard deviation of the input value;

[0036] Step 5.2: Input the training samples into the neural network model and use the cross entropy loss function to optimize the neural network model to obtain a cycle life prediction model; wherein the cross entropy loss function is:

[0037]

[0038] Where n0 represents the number of training samples, c represents the number of cycle life, and y ic Indicates the actual cycle life, represents the cycle life predicted by the neural network.

[0039] The present invention also provides a cycle life prediction system for sodium ion batteries, comprising:

[0040] An equivalent circuit building module is used to model the target sodium-ion battery and obtain the equivalent circuit of the sodium-ion battery;

[0041] The remaining power calculation module is used to solve the remaining power of the sodium ion battery at each time based on the equivalent circuit;

[0042] A detection module, used to detect the remaining power and circuit parameters of the target sodium ion battery at different cycle times;

[0043] A correlation calculation module is used to calculate the correlation between the circuit parameters and the remaining power, and screen out the circuit parameters whose correlation is greater than a set value;

[0044] A training module, used for inputting the screened circuit parameters and the remaining power into the neural network model for training to obtain a cycle life prediction model;

[0045] The cycle life prediction module is used to use the cycle life prediction model to complete the prediction of the cycle life of the target sodium ion battery.

[0046] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned method for predicting the cycle life of a sodium ion battery.

[0047] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for predicting the cycle life of a sodium ion battery are implemented.

[0048] The beneficial effect of a cycle life prediction method and system for sodium ion batteries provided by the present invention is that: compared with the prior art, the present invention can identify the trend of battery performance degradation in advance by predicting the cycle life of sodium ion batteries, help staff to formulate battery usage strategies more scientifically, and improve the overall efficiency of the energy management system.

[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 A flow chart of a cycle life prediction method for a sodium ion battery provided by an embodiment of the present invention is shown;

[0052] Figure 2 An equivalent circuit diagram of a sodium ion battery provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0053] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0054] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0055] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0056] See also Figure 1-2 , a cycle life prediction method for sodium ion batteries, comprising:

[0057] Step 1: Model the target sodium-ion battery to obtain an equivalent circuit of the sodium-ion battery;

[0058] Step 2: Calculate the remaining power of the sodium-ion battery at each time based on the equivalent circuit;

[0059] Furthermore, the step 2 comprises:

[0060] Step 2.1: Use the second-order RC equivalent circuit model to construct the equivalent circuit of the target sodium-ion battery;

[0061] Step 2.2: Model the second-order RC equivalent circuit model based on Kirchhoff's law to obtain the battery remaining power function;

[0062] In step 2.2, the remaining battery power function is:

[0063]

[0064] Among them, U L Indicates terminal voltage, U OC represents the open circuit voltage, U1 represents the terminal voltage of the first RC loop, U2 represents the terminal voltage of the second RC loop, I represents the battery operating current, R1 represents the first polarization resistance, R2 represents the second polarization resistance, C1 represents the first polarization capacitance, C2 represents the second polarization capacitance, k represents the current time, k0 represents the initial time, η represents the discharge efficiency, Q c Indicates the rated capacity of the battery, S OC (k) indicates the remaining power at the current moment.

[0065] Step 2.3: Discretize the battery remaining capacity function to obtain a working state model;

[0066] In step 2.3, the present invention introduces system noise to discretize the battery remaining capacity function to obtain a working state model; wherein the working state model is:

[0067]

[0068] Among them, ω k represents the system noise at time k, and Δk represents the time difference.

[0069] Step 2.4: Solve the working state model to obtain the remaining power of the sodium ion battery at each time;

[0070] In the present invention, the least square method can be used to perform parameter identification on the working state model to obtain the remaining power of the sodium ion battery at each moment; wherein the recursive formula of the least square method is:

[0071]

[0072] in, represents the parameter identification value at time k, y(k) represents the actual value of the parameter collected at time k, P(k) represents the error covariance matrix at time k, K(k) represents the algorithm gain factor at time k, represents the information matrix at time k, and λ represents the forgetting factor.

[0073] As time goes by, the influence of old data on the model gradually weakens, resulting in the limitation of the effect of new data on the update of model parameters. This phenomenon usually leads to distortion of the battery model: old data may no longer represent the current system state, a phenomenon known as data saturation. Therefore, this application introduces a forgetting factor in the least squares method, which can not only weaken the influence of old data, but also enhance the correction effect of new data on parameters.

[0074] Step 3: Detect the remaining power and circuit parameters of the target sodium ion battery at different cycle times;

[0075] Step 4: Calculate the correlation between the circuit parameters and the remaining power, and select the circuit parameters whose correlation is greater than the set value; the circuit parameters include the battery current, open circuit voltage, operating temperature, polarization resistance and polarization capacitance, etc.

[0076] In step 4, the present invention may adopt the formula:

[0077]

[0078] Calculate the correlation between the circuit parameters and the remaining power; where r represents the correlation, n represents the number of circuit parameter and remaining power data pairs in a measurement cycle, ∑xy represents the product of the circuit parameters and the remaining power, ∑x represents the sum of the circuit parameters in a measurement cycle, ∑y represents the sum of the remaining power in a measurement cycle, ∑x 2 represents the sum of squares of circuit parameters within a measurement cycle, Σy 2 Indicates the sum of squares of remaining power in a measurement cycle.

[0079] In actual production activities, there is a close relationship between remaining power and cycle life. This application can reflect the relationship between circuit parameters and battery cycle life by calculating the correlation between circuit parameters and remaining power, thereby achieving accurate prediction of battery cycle life.

[0080] Step 5: Input the screened circuit parameters and remaining power into the neural network model for training to obtain a cycle life prediction model;

[0081] Further, step 5 includes:

[0082] Step 5.1: Normalize the screened circuit parameters and remaining power to obtain training samples; the normalization formula is:

[0083]

[0084] Among them, x″ represents the normalized parameter, x′ represents the input value, μ represents the mean of the input value, and σ represents the standard deviation of the input value;

[0085] Step 5.2: Input the training samples into the neural network model and use the cross entropy loss function to optimize the neural network model to obtain a cycle life prediction model; wherein the cross entropy loss function is:

[0086]

[0087] Where n0 represents the number of training samples, c represents the number of cycle life, and y ic Indicates the actual cycle life, represents the cycle life predicted by the neural network.

[0088] Step 6: Use the cycle life prediction model to complete the prediction of the cycle life of the target sodium ion battery.

[0089] By predicting the cycle life of sodium-ion batteries, the present invention can identify the trend of battery performance degradation in advance, help staff to formulate battery usage strategies more scientifically, and improve the overall efficiency of the energy management system.

[0090] The present invention also provides a cycle life prediction system for sodium ion batteries, comprising:

[0091] An equivalent circuit building module is used to model the target sodium-ion battery and obtain the equivalent circuit of the sodium-ion battery;

[0092] The remaining power calculation module is used to solve the remaining power of the sodium ion battery at each time based on the equivalent circuit;

[0093] A detection module, used to detect the remaining power and circuit parameters of the target sodium ion battery at different cycle times;

[0094] A correlation calculation module is used to calculate the correlation between the circuit parameters and the remaining power, and screen out the circuit parameters whose correlation is greater than a set value;

[0095] A training module, used for inputting the screened circuit parameters and the remaining power into the neural network model for training to obtain a cycle life prediction model;

[0096] The cycle life prediction module is used to use the cycle life prediction model to complete the prediction of the cycle life of the target sodium ion battery.

[0097] The present invention also provides an electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, and wherein the computer program, when executed by the processor, implements the steps in the above-mentioned method for predicting the cycle life of a sodium ion battery. Compared with the prior art, the beneficial effects of the electronic device provided by the present invention are the same as the beneficial effects of the method for predicting the cycle life of a sodium ion battery described in the above-mentioned technical solution, and are not described in detail herein.

[0098] The present invention also provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps in the above-mentioned method for predicting the cycle life of a sodium ion battery are implemented. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present invention are the same as the beneficial effects of the method for predicting the cycle life of a sodium ion battery described in the above-mentioned technical solution, which will not be repeated here.

[0099] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technical solution that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for predicting the cycle life of a sodium ion battery, characterized in that: include: Step 1: Model the target sodium-ion battery to obtain an equivalent circuit of the sodium-ion battery; Step 2: Calculate the remaining power of the sodium-ion battery at each time based on the equivalent circuit; Step 3: Detect the remaining power and circuit parameters of the target sodium ion battery at different cycle times; Step 4: Calculate the correlation between the circuit parameters and the remaining power, and select the circuit parameters whose correlation is greater than the set value; Step 5: Input the screened circuit parameters and remaining power into the neural network model for training to obtain a cycle life prediction model; Step 6: Use the cycle life prediction model to complete the prediction of the cycle life of the target sodium ion battery.

2. A method for predicting the cycle life of a sodium ion battery according to claim 2, characterized in that: The step 2: solving the remaining power of the sodium ion battery at each time based on the equivalent circuit, includes: Step 2.1: Use the second-order RC equivalent circuit model to construct the equivalent circuit of the target sodium-ion battery; Step 2.2: Model the second-order RC equivalent circuit model based on Kirchhoff's law to obtain the battery remaining power function; Step 2.3: Discretize the battery remaining capacity function to obtain a working state model; Step 2.4: Solve the working state model to obtain the remaining power of the sodium ion battery at each moment.

3. A method for predicting the cycle life of a sodium ion battery according to claim 2, characterized in that: In step 2.2, the remaining battery power function is: Among them, U L Indicates terminal voltage, U OC represents the open circuit voltage, U1 represents the terminal voltage of the first RC loop, U2 represents the terminal voltage of the second RC loop, I represents the battery operating current, R1 represents the first polarization resistance, R2 represents the second polarization resistance, C1 represents the first polarization capacitance, C2 represents the second polarization capacitance, k represents the current time, k0 represents the initial time, η represents the discharge efficiency, Q c Indicates the rated capacity of the battery, S OC (k) indicates the remaining power at the current moment.

4. A method for predicting the cycle life of a sodium ion battery according to claim 3, characterized in that: The step 2.3: discretizing the battery remaining capacity function to obtain a working state model includes: The system noise is introduced to discretize the battery remaining capacity function to obtain the working state model; wherein the working state model is: Among them, ω k represents the system noise at time k, and Δk represents the time difference.

5. A method for predicting the cycle life of a sodium ion battery according to claim 4, characterized in that: The step 2.4: solving the working state model to obtain the remaining power of the sodium ion battery at each moment, including: The least squares method is used to identify the parameters of the working state model to obtain the remaining power of the sodium ion battery at each moment; the recursive formula of the least squares method is: in, represents the parameter identification value at time k, y(k) represents the actual value of the parameter collected at time k, P(k) represents the error covariance matrix at time k, K(k) represents the algorithm gain factor at time k, represents the information matrix at time k, and λ represents the forgetting factor.

6. A method for predicting the cycle life of a sodium ion battery according to claim 5, characterized in that: The step 4: calculating the correlation between the circuit parameters and the remaining power, and screening out the circuit parameters whose correlation is greater than the set value, includes: Using the formula: Where r represents the correlation, n represents the number of circuit parameter and remaining power data pairs in a measurement cycle, ∑xy represents the product of circuit parameter and remaining power, ∑x represents the sum of circuit parameters in a measurement cycle, ∑y represents the sum of remaining power in a measurement cycle, ∑x 2 represents the sum of squares of circuit parameters within a measurement cycle, ∑y 2 Indicates the sum of squares of remaining power in a measurement cycle.

7. A method for predicting the cycle life of a sodium ion battery according to claim 6, characterized in that: The step 5: inputting the screened circuit parameters and the remaining power into the neural network model for training to obtain a cycle life prediction model, including: Step 5.1: Normalize the screened circuit parameters and remaining power to obtain training samples; the normalization formula is: Among them, x″ represents the normalized parameter, x′ represents the input value, μ represents the mean of the input value, and σ represents the standard deviation of the input value; Step 5.2: Input the training samples into the neural network model and use the cross entropy loss function to optimize the neural network model to obtain a cycle life prediction model; wherein the cross entropy loss function is: Where n0 represents the number of training samples, c represents the number of cycle life, and y ic Indicates the actual cycle life, represents the cycle life predicted by the neural network.

8. A cycle life prediction system for sodium ion batteries, characterized in that: include: An equivalent circuit building module is used to model the target sodium-ion battery and obtain the equivalent circuit of the sodium-ion battery; The remaining power calculation module is used to solve the remaining power of the sodium ion battery at each time based on the equivalent circuit; A detection module, used to detect the remaining power and circuit parameters of the target sodium ion battery at different cycle times; A correlation calculation module is used to calculate the correlation between the circuit parameters and the remaining power, and screen out the circuit parameters whose correlation is greater than a set value; A training module, used for inputting the screened circuit parameters and the remaining power into the neural network model for training to obtain a cycle life prediction model; The cycle life prediction module is used to use the cycle life prediction model to complete the prediction of the cycle life of the target sodium ion battery.

9. An electronic device, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, wherein: When the computer program is executed by the processor, the steps in the cycle life prediction method for a sodium ion battery according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the cycle life prediction method for a sodium ion battery as claimed in any one of claims 1 to 7 are implemented.

Images