A method for identifying short-circuit faults in batteries based on pre-charge current pulses

Abstract

The present invention discloses a method for identifying internal battery short-circuit faults based on pre-charge current pulses. This method utilizes the current conversion process during the transition from the pre-charge phase to the high-current charging phase of an electric vehicle. Based on the collected voltage response and current information of each cell in a series battery pack, the method first examines the battery terminal voltage outliers during the low-current pre-charge phase, focusing on cells with overall low voltages during the pre-charge process. An equivalent circuit model for the internal battery short-circuit fault is then established for the target battery. The value of the micro-short-circuit resistance within the model is then identified to determine whether the battery cell has an internal short-circuit fault. This method thus enables early detection of internal battery micro-short-circuit faults. The present invention is suitable for battery cell and battery pack applications, such as electric vehicles, energy storage systems, and power tools.

Description

Technical Field

[0001] The present invention relates to a method for identifying a short circuit fault in a battery, and in particular to a method for identifying a short circuit fault in a battery based on a pre-charge current pulse. Background Art

[0002] Due to manufacturing defects and prolonged use, lithium-ion batteries can cause dendrites to form between the positive and negative separators. This can puncture the separators and cause internal micro-short circuits. This malfunction threatens battery safety and, in severe cases, can lead to thermal runaway, fire, and explosion.

[0003] To predict potential internal short-circuit faults in batteries, identification and determination are typically based on the battery terminal voltage. However, battery aging and increased internal resistance can cause the battery terminal voltage to drop, making it difficult to accurately identify and determine whether the problem is caused by an internal short circuit. A search revealed that most existing literature uses a single voltage criterion as a means of identifying internal short-circuit faults, or employs complex parameter identification methods to diagnose internal short-circuit faults. A lack of computationally efficient and simple internal short-circuit fault diagnosis methods is lacking. Summary of the Invention

[0004] The purpose of the present invention is to provide a battery internal short circuit fault identification method based on pre-charge current pulse, which solves the problems such as uncertainty caused by complex model and complex calculation that make it difficult to identify the internal short circuit resistance.

[0005] The technical solution for achieving the purpose of the present invention is: a method for identifying a short circuit fault in a battery based on a pre-charge current pulse, comprising the following steps:

[0006] Step 1: During the pre-charge phase, determine whether any target battery cell has an internal short circuit fault; that is, when charging with a low current, observe the voltage curve of each battery cell and mark the battery cells with outliers in terminal voltage.

[0007] Step 2: For the target battery cell marked during low-current charging in step 1, record the voltage and current status information during the transition from the pre-charge stage to the formal high-current charging;

[0008] Step 3: Establishing an equivalent circuit model of short circuit in the battery cell;

[0009] Step 4: Based on the obtained voltage response during the current switching pulse phase, identify the value of the internal short-circuit resistance in the model;

[0010] Step 5: Differentiate between slight, moderate, and severe internal short circuit faults based on the identified internal short circuit resistance values.

[0011] Compared with the prior art, the present invention has the following advantages: (1) The present invention provides a method for identifying internal short-circuit faults in batteries based on pre-charge current pulses. This method can detect the voltage response and current information of each cell in the series battery pack, focus on the outliers of battery voltage under low current conditions during the pre-charge phase, and examine cells with lower voltages during the charging process. This method establishes a model for identifying internal short-circuit faults in batteries, identifies the value of the micro-short-circuit resistance in the model, and judges whether a cell has an internal short-circuit fault. This enables early detection of micro-short-circuit faults in batteries. (2) The method for identifying internal short-circuit faults in batteries based on pre-charge current pulses proposed by the present invention does not rely on other cells in the battery pack, thus overcoming the problems of uncertainty and high computational complexity caused by complex systems and complex models. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is a single cell voltage curve.

[0013] Figure 2 It is the equivalent model of battery internal short circuit.

[0014] Figure 3 The figure is a flow chart of a method for identifying a short circuit fault in a battery based on a pre-charge current pulse.

[0015] Figure 4 This is the SOC-OCV curve of a single battery.

[0016] Figure 5 This is the short-circuit resistance diagram obtained by identification.

[0017] Figure 6 This is the short-circuit resistance diagram identified near the pulse excitation point. DETAILED DESCRIPTION

[0018] The present invention proposes a method for identifying internal short-circuit faults in batteries based on pre-charge current pulses. Before identifying internal short-circuit faults in batteries, the battery is first pre-charged, the voltage curve of each battery cell is observed, and battery cells with voltage curves below the normal value are marked. Then, high-current charging is performed on the target battery cells, and voltage and current status information is recorded. An equivalent model for internal short circuits in the battery is established. The model parameters are identified using the least squares method. Finally, the obtained internal short-circuit resistance of the battery is compared with a set threshold to determine the degree of the internal short circuit fault. The method has the following steps:

[0019] Step 1: During the pre-charge phase, identify the target battery cells that need to be examined for internal short circuit faults. That is, when charging at a low current rate of 0.01C, observe the voltage curve of each battery cell. The voltage curves of normal batteries are relatively concentrated. Battery cells with voltage curves below the normal level are marked for subsequent attention. Specifically:

[0020] Step 1-1: The battery cell is first discharged at a constant current of 0.5C rate to the lower cut-off voltage of the battery;

[0021] Step 1-2, let it stand for 2 hours;

[0022] Steps 1-3: pre-charge the battery with a low current of 0.01C for one minute;

[0023] Steps 1-4: Record the voltage curves of the battery cells and mark the battery cells with voltage curves lower than the normal ones for subsequent attention.

[0024] Step 2: For the target battery cell marked in step 1, record the voltage and current status information of the transition from the pre-charge stage to the formal high-current charge; specifically:

[0025] After the target battery cell marked in step 1-4 is pre-charged with a small current, the charging current is increased to perform constant current charging. The battery voltage gradually increases with the constant current charging process, and the voltage and current information of the battery cell during high current charging is recorded.

[0026] Step 3: Establish an equivalent circuit model of the short circuit inside the battery cell, specifically:

[0027] Conduct an intermittent discharge-rest experiment, fully charge the battery at a 0.5C rate and rest for 2 hours. Discharge the battery at a 0.5C rate, and rest for 1 minute every time 1% of the charge is discharged. The voltage after rest is the open circuit voltage of the battery. The SOC-OCV relationship expression is determined by fitting:

[0028]

[0029] Among them, U oc is the battery open circuit voltage (OCV), z is the battery SOC, n is the fitting order, c i is the fitting coefficient

[0030] The normal battery model uses the Thevenin model. When the battery is short-circuited, it is equivalent to connecting a resistor in parallel, such as Figure 2 As shown, a mathematical model is established for the internal short-circuit battery, and the KCL and KVL equations are written:

[0031]

[0032]

[0033] I a =I s +I (4)

[0034] U oc =I aR0+U1+U (5)

[0035]

[0036] Where I1 is the current flowing through capacitor C1, I a is the electrochemical current, I s is the short-circuit current, I is the total current, R0 is the ohmic internal resistance, R1 is the polarization internal resistance, R s is the short-circuit internal resistance, C1 is the polarized capacitor, t is the time, U1 is the voltage across the polarized capacitor, and U is the terminal voltage.

[0037] Combining (2)(3)(4)(6) and performing a pull-type transformation, we get

[0038]

[0039] Where τ=R1C1, is the time constant;

[0040] Substituting formula (7) into formula (5), we get

[0041]

[0042] Using the ampere-hour integration method, we can know the battery According to formula (4) and formula (6) and

[0043] Pull transformation shows

[0044]

[0045] Since the open circuit voltage U oc There is a nonlinear functional relationship with the battery SOC. For the convenience of modeling, it is considered to be a linear relationship within a small range, that is,

[0046] U oc =a1z+a2 (10)

[0047] Among them, a1 and a2 are fitting coefficients;

[0048] Combining equations (8), (9), and (10) yields the relationship between U(s) and I(s):

[0049]

[0050] So, the transfer function is

[0051]

[0052] The established battery monomer internal short circuit equivalent model is discretized using the bilinear transformation method, and the discretized transfer function is obtained as follows:

[0053]

[0054] where b i are parameters that need to be estimated by identification method, i=1, 2, ..., 6.

[0055] Step 4: Combine the voltage response obtained during the current switching pulse phase to identify the value of the internal short-circuit resistance in the model. The least squares method is used to estimate the internal short-circuit resistance of the battery in real time, specifically:

[0056] Step 4-1, initialize parameters based on existing information and experience;

[0057] Step 4-2, calculate the estimated error e(k), where y(k) is the voltage matrix and h(k) is the parameter matrix. The parameters to be identified

[0058]

[0059] Step 4-3, calculate the gain matrix K(k), where P(k) is the covariance matrix, λ is the forgetting factor, 0≤λ≤1; in this embodiment, λ=1;

[0060]

[0061] Step 4-4, parameter estimation

[0062]

[0063] Step 4-5, update the covariance matrix P(k), where I0 is the identity matrix

[0064]

[0065] Step 5: Based on the identified internal short-circuit resistance value, distinguish between slight, moderate and severe internal short-circuit faults, specifically:

[0066] The internal short-circuit resistance is estimated using the least squares method. The current single-cell internal short-circuit status is determined based on a comparison of the preset internal short-circuit resistance threshold and the calculated real-time internal short-circuit resistance. A mild internal short-circuit fault is identified when the estimated resistance is greater than 10Ω and less than or equal to 1000Ω. A moderate internal short-circuit fault is identified when the estimated resistance is greater than 0.1Ω and less than or equal to 10Ω. A severe internal short-circuit fault is identified when the estimated resistance is less than or equal to 0.1Ω.

[0067] The present invention will be described in detail below using a certain ternary lithium battery as an example.

[0068] Example

[0069] like Figure 3As shown, this embodiment uses an ISR18650-2.2Ah ternary lithium battery as an experimental object, and performs step 1 of the method of the present invention at room temperature. The specific process is as follows:

[0070] Step 1-1: The battery cell is first discharged at a constant current of 0.5C rate to the lower cut-off voltage of the battery;

[0071] Step 1-2, let it stand for 2 hours;

[0072] Steps 1-3: pre-charge the battery with a small current;

[0073] Steps 1-4: Record the voltage curves of the battery cells and mark the battery cells with voltage curves lower than the normal ones for subsequent attention.

[0074] According to the target battery cell obtained in step 1, after the battery voltage rises to the trickle charge threshold or above, increase the charging current and perform constant current charging. The battery voltage gradually increases with the constant current charging process, and record the voltage and current information of the battery cell during high current charging. Record the corresponding voltage curve and current curve, such as Figure 1 shown.

[0075] Then establish the internal short circuit equivalent model, such as Figure 2 As shown, first establish the functional relationship between SOC and OCV when the battery is not short-circuited, as shown in Figure 4 A 200Ω resistor is connected in parallel to the battery output terminal to simulate the internal short circuit fault of the battery. At room temperature, the U.S. Federal City Operating Conditions (FUDS) are used. The method described in the present invention is used to estimate the internal short circuit resistance of the battery in real time based on the least squares method. The internal short circuit resistance identification result is as follows: Figure 5 As shown, the peak point error is 13%. Further estimation of the data near the pulse excitation point yields the internal short-circuit resistance identification result as shown in Figure 6 As shown, the peak point error is 0.3%.

[0076] The internal short circuit resistance identification result is recorded, and the internal short circuit state of the single battery at the current moment is determined based on the preset internal short circuit resistance value and the calculated real-time internal short circuit resistance. In this embodiment, the internal short circuit is determined to be low risk.

[0077] In summary, this invention utilizes only the measured values ​​of battery load current and terminal voltage, employing the least squares method to online identify the battery's internal short-circuit resistance. This resistance is then used to determine the risk of a battery cell internal short-circuit failure. Compared to traditional internal short-circuit detection methods based on electrical models, this invention is independent of other cells within the battery pack, overcoming the uncertainties and computational complexity inherent in complex systems and models.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for identifying a short circuit fault in a battery based on a pre-charge current pulse, characterized in that: The following steps are involved: Step 1: During the pre-charge phase, determine whether there is a target battery cell with an internal short circuit fault. That is, when charging with a low current, observe the voltage curve of each battery cell and mark the battery cells with outliers in terminal voltage. Specifically: Step 1-1: The battery cell is first discharged at a constant current of 0.5C rate to the lower cut-off voltage of the battery; Step 1-2, let it stand for 2 hours; Steps 1-3: pre-charge the battery with a low current of 0.01C for one minute; Step 1-4, record the voltage curve of the battery cells and mark the battery cells with voltage curves lower than the normal voltage curve; Step 2: For the target battery cell marked during low-current charging in step 1, record the voltage and current status information during the transition from the pre-charge stage to the formal high-current charging; Step 3: Establish an equivalent circuit model of the short circuit inside the battery cell. The specific method is as follows: Step 3-1, conduct an intermittent discharge-rest experiment, fully charge the battery at a 0.5C rate and let it rest for 2 hours; discharge the battery at a 0.5C rate, and let it rest for 1 minute every time 1% of the charge is discharged. The voltage after rest is the open circuit voltage of the battery, and the SOC-OCV relationship expression is determined by fitting: Among them, U oc is the battery open circuit voltage, z is the battery SOC, n is the fitting order, c i is the fitting coefficient; Step 3-2: When the battery is short-circuited, it is equivalent to connecting a resistor in parallel. A mathematical model is established for the short-circuited battery, and the KCL and KVL equations are listed: I a =I s +I (4) IN oc =I a R0+U1+U (5) Where I1 is the current flowing through capacitor C1, I a is the electrochemical current, I s is the short-circuit current, I is the total current, R0 is the ohmic internal resistance, R s is the short-circuit internal resistance, R1 is the polarization internal resistance, C1 is the polarization capacitance, t is the time, U1 is the voltage across the polarization capacitance, and U is the terminal voltage; Combine (2)(3)(4)(6) and perform a pull-type transformation to obtain Where τ=R1C1, is the time constant; Substituting formula (7) into formula (5), we get Using the ampere-hour integration method, we can know the battery According to formula (4) and formula (6) and the pull transformation, we can know Since the open circuit voltage U oc There is a nonlinear functional relationship with the battery SOC. For the convenience of modeling, it is considered to be a linear relationship within a small range, that is, <h2 style=";text-align:left;direction:ltr">U<h2 style=";text-align:left;direction:ltr"> oc <h2 style=";text-align:left;direction:ltr"> =a1z+a2 (10) Among them, a1 and a2 are fitting coefficients; Combining equations (8), (9), and (10) yields the relationship between U(s) and I(s): Therefore, the transfer function G(s) is The bilinear transformation method is used to discretize the established battery monomer internal short circuit equivalent model, and the discretized transfer function is obtained as follows: where b i is the parameter to be estimated by identification method, i=1, 2, ..., 6; Step 4: Based on the obtained voltage response during the current switching pulse phase, identify the value of the internal short-circuit resistance in the model; Step 5: Differentiate between slight, moderate, and severe internal short circuit faults based on the identified internal short circuit resistance values.

2. The method for identifying a short circuit fault in a battery based on a pre-charge current pulse according to claim 1, wherein By establishing an equivalent circuit model of the internal short circuit of the battery cell, the internal short circuit resistance can be identified in real time and the degree of the internal short circuit fault can be determined.

3. The method for identifying a short circuit fault in a battery based on a pre-charge current pulse according to claim 1, wherein Step 2 is as follows: After the marked target battery cell is pre-charged with a small current, the charging current is increased and constant current charging is performed. The battery voltage gradually increases with the constant current charging process, and the voltage and current information of the battery cell during high current charging is recorded.

4. The method for identifying a short circuit fault in a battery based on a pre-charge current pulse according to claim 1, wherein Step 4 uses the least squares method to estimate the short-circuit resistance of the battery in real time, specifically: Step 4-1, initialize parameters based on existing information and experience; Step 4-2, calculate the estimated error e(k): Where y(k) is the voltage matrix, h(k) is the parameter matrix, is the parameter to be identified; Step 4-3, calculate the gain matrix K(k): Where P(k) is the covariance matrix, λ is the forgetting factor, 0≤λ≤1; Step 4-4, parameter estimation Step 4-5, update the covariance matrix P(k), where I0 is the identity matrix 5. The method for identifying a short circuit fault in a battery based on a pre-charge current pulse according to claim 1, wherein: Step 5 is as follows: According to step 4, the internal short-circuit resistance is estimated using the least squares method. The internal short-circuit state of the single cell at the current moment is determined based on the preset internal short-circuit resistance value and the estimated real-time internal short-circuit resistance. When the estimated resistance is greater than 10Ω and less than or equal to 1000Ω, it is determined to be a mild internal short-circuit fault. When the estimated resistance is greater than 0.1Ω and less than or equal to 10Ω, it is determined to be a moderate internal short-circuit fault. When the estimated resistance is less than or equal to 0.1Ω, it is determined to be a severe internal short-circuit fault.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for identifying a short circuit fault in a battery based on a pre-charge current pulse as described in any one of claims 1 to 5 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for identifying a short circuit fault in a battery based on a pre-charge current pulse as described in any one of claims 1 to 5 is implemented.

Images