CN117269768A

CN117269768A - Parallel battery pack short-circuit fault diagnosis method and device based on capacity increment curve

Info

Publication number: CN117269768A
Application number: CN202311243698.5A
Authority: CN
Inventors: 赵秀亮; 王金志; 赵明明; 杨烨; 周家宝; 王丽梅; 汪若尘; 孙洪良
Original assignee: Jiangsu University
Current assignee: Jiangsu University
Priority date: 2023-09-25
Filing date: 2023-09-25
Publication date: 2023-12-22

Abstract

The invention discloses a method and a device for diagnosing a short-circuit fault of a parallel battery pack based on a capacity increment curve, wherein the method comprises the steps of constructing a first-order RC equivalent circuit model; acquiring open-circuit voltage, ohmic internal resistance and polarization parameters; constructing a single battery simulation model; obtaining optimal time domain screening by polarization parameters; constructing a parallel battery pack short-circuit fault simulation model; selecting a characteristic curve for short-circuit fault analysis; acquiring a characteristic peak and a charging multiplying power for short circuit fault analysis; and performing fault diagnosis and fusion fault diagnosis based on the characteristic peaks. Mid-term of charging process through n of IC curve _Ⅰ The peak realizes the rough judgment of the short circuit resistance value, and the later stage of the charging process passes through n of the IC curve _Ⅱ Peak implementationThe short circuit resistance value is quantitatively diagnosed more accurately, and quick and high-precision fault diagnosis is realized. Performing fault diagnosis through different characteristic peaks of the IC curve, and eliminating the need of a complete charging voltage curve; at the same time fuse n _Ⅰ Peak sum n _Ⅱ The fault diagnosis of the peak is more flexible and has stronger adaptability when the charge and discharge states are uncertain under the actual vehicle working condition.

Description

Parallel battery pack short-circuit fault diagnosis method and device based on capacity increment curve

Technical Field

The invention relates to a method and a device for diagnosing a short-circuit fault of a parallel battery pack based on a capacity increment curve, and belongs to the technical field of power batteries.

Background

In order to meet the requirements of electric vehicles on endurance mileage and maximum power, a power battery pack is usually formed by connecting hundreds or thousands of single batteries in series and parallel. When a certain monomer in the battery pack fails, the normal operation of the whole battery pack can be influenced. In order to ensure the safety and reliability of a vehicle, a rapid and efficient battery fault diagnosis method is particularly important.

Current research for battery short circuit fault diagnosis focuses on single cells and series battery packs. In the prior art, an internal short circuit fault test of the battery is designed, and the internal short circuit fault judgment is carried out by analyzing the electrical characteristic and the thermal characteristic of the battery. However, considering the high risk, long time consumption and unrepeatability of battery failure test, more research on short-circuit failure diagnosis is now being conducted by a model-based method. In the single battery fault, a researcher extracts fault characteristics by using an equivalent circuit model and estimated short-circuit current, model terminal voltage and the like, and diagnosis of the battery short-circuit fault is realized.

In the aspect of fault diagnosis research of parallel battery packs, in consideration of the structural specificity thereof, the fault diagnosis research is currently carried out by regarding the parallel battery packs as a whole. Some students train the artificial neural network-back propagation neural network model through a test data set, and voltage information is used for estimating the leakage current of the short-circuit fault battery. In order to realize fault diagnosis of each single battery in the parallel battery, partial scholars explore the characteristics of the parallel battery in a differential state. It is found that the parallel battery pack with cell aging during charging has a higher charging voltage than the normal parallel battery pack.

In summary, currently, for the fault diagnosis research of the parallel battery pack, considering that the parallel branch current is not easy to measure, the parallel branch current is often regarded as a whole for the fault diagnosis research. There are also some partial operators focusing on studying the effect of battery variability on parallel battery characteristics in certain situations. However, the research on the characteristics of the parallel battery packs and the diagnosis of the short circuit faults is focused on the specific parallel battery packs, the comparison analysis of the short circuit faults of the parallel battery packs under different working conditions is lacked, and meanwhile, the influence of the number of the parallel single batteries and the charge and discharge currents of the parallel battery packs on the diagnosis of the faults is not explored. In other words, the quick and accurate diagnosis of the short-circuit faults of the battery packs with different parallel battery numbers under different working conditions is not really realized. In addition, the existing fault diagnosis is often developed based on a certain characteristic quantity, and the charge and discharge states have uncertainty under the actual vehicle condition, so that the adaptability of the fault diagnosis method is reduced.

Disclosure of Invention

The invention aims to: according to the method and the device for diagnosing the short-circuit faults of the parallel battery packs based on the capacity increment curve, the short-circuit fault simulation model of the parallel battery packs is built on the basis of the built single battery simulation model, the degree of the short-circuit faults is quantitatively diagnosed through characteristic peaks of the IC curve, and the fault diagnosis is carried out by fusing the characteristic peaks, so that the rapid and accurate diagnosis of the short-circuit faults of the battery packs with different numbers of parallel batteries under different working conditions is realized.

The technical scheme is as follows: the method for diagnosing the short circuit fault of the parallel battery pack based on the capacity increment curve comprises the following steps:

s1: constructing a first-order RC equivalent circuit model;

s2: carrying out hybrid power pulse test on the battery to obtain the model parameters of the first-order RC equivalent circuit model constructed in the S1, including open-circuit voltage, ohmic internal resistance and polarization parameters;

s3: constructing a single battery simulation model;

s4: obtaining optimal time domain screening by polarization parameters: the polarization parameters comprise polarization internal resistance and polarization capacitance, the polarization parameters acquired in different time domains are input into a single battery simulation model, the difference value between simulation voltage and actually measured voltage in different time domains is compared, the time domain with the smallest difference value is selected as an optimal time domain, and the corresponding polarization parameter is the optimal polarization parameter;

S5: constructing a parallel battery pack short-circuit fault simulation model: according to the single battery simulation model obtained in the step S3 and the model parameters obtained in the step S2 and the step S4, a parallel battery pack short circuit fault simulation model is constructed and verified;

s6: selecting a characteristic curve for short-circuit fault analysis: based on the S5 constructed parallel battery pack short-circuit fault simulation model, respectively carrying out short-circuit fault characteristic analysis on the charging voltage curve and the IC curve, comparing analysis results of the charging voltage curve and the IC curve, and selecting the IC curve for short-circuit fault characteristic analysis;

s7: performing IC curve analysis of short-circuit resistance batteries with different degrees under different charging multiplying powers and IC curve analysis of battery packs with different parallel batteries under different short-circuit resistances according to the IC curves selected in the step S6, and obtaining influences of the charging multiplying powers and the numbers of the parallel batteries on the characteristic peak values of the IC curves of the battery packs with the parallel batteries under short-circuit faults;

s8: according to the influences of the charging multiplying power and the number of the parallel batteries obtained in the S7 on the characteristic peak value of the IC curve of the parallel battery pack under the short circuit fault, the characteristic peak value and the charging multiplying power for short circuit fault analysis are obtained through statistics of the running working conditions of the real vehicle, including statistics of the actual charging multiplying power and the actual charging voltage interval;

s9: and (3) performing fault diagnosis and fusion fault diagnosis on the characteristic peak obtained in the step (S8) to finally obtain the diagnosis method of the short circuit fault of the parallel battery pack.

The invention obtains model parameters corresponding to a constructed first-order RC equivalent circuit model through a hybrid power pulse test, wherein the model parameters comprise open-circuit voltage, ohmic internal resistance and polarization parameters, a single battery simulation model is constructed according to the model parameters, the polarization parameters are screened to obtain optimal polarization parameters with highest precision of the single battery simulation model, namely the polarization parameters obtained under different time domains are sent into the simulation model, the difference value of the simulation voltage and the actually measured voltage is compared, the time domain with the smallest difference value is the optimal time domain, and the corresponding polarization parameters are the optimal polarization parameters. And constructing a parallel battery pack short-circuit fault simulation model according to the single battery simulation model. And carrying out IC curve analysis of the short-circuit resistance batteries with different degrees under different charging multiplying powers and IC curve analysis of the battery packs with different numbers of parallel batteries under different short-circuit resistances based on the parallel battery pack short-circuit fault simulation model. Further, through statistics of actual operation conditions including charging multiplying power and charging voltage intervals, characteristic peaks and charging multiplying power for short circuit fault analysis are obtained. And finally, carrying out fault diagnosis and fusion fault diagnosis based on characteristic peaks of the IC curves, realizing quick and accurate diagnosis of short-circuit faults of the battery packs with different parallel batteries under different working conditions, and improving the adaptability of a fault diagnosis method.

Preferably, the specific steps of S1 are as follows:

based on kirchhoff's law, the state space equation of the first-order RC equivalent circuit model is as follows:

wherein R is ₀ Represents ohmic internal resistance, R ₁ For polarization resistance, C ₁ Representing the polarization capacitance, I is the load current, U represents the battery terminal voltage, U _OCV Is open circuit voltage, U ₁ Representing polarisation voltage, i.e. R ₁ C ₁ The voltage across it.

Preferably, the specific steps of S2 are as follows:

the hybrid power pulse test specifically comprises: performing charge-discharge pulse excitation and standing treatment on batteries under different charge States (SOC) so as to obtain battery characteristic parameters under different SOC;

obtaining an open circuit voltage: in the process of constant-current discharging of the battery, the SOC of the battery is reduced, and then the battery is subjected to long-time standing treatment, so that the open-circuit voltage can reach a stable state, and further corresponding data of different SOCs and open-circuit voltages are obtained;

and (3) obtaining ohmic internal resistance: the ohmic internal resistance is obtained according to a current abrupt change stage, wherein in the hybrid power pulse test, the battery is suddenly excited by a discharge pulse, and the voltage is rapidly changed from U _A Down to U _B Ohmic internal resistance R at this time ₀ The calculation formula is as follows:

wherein R is ₀ Is ohmic impedance, U _A For the voltage of A point, U _B Is the voltage of the point B;

obtaining polarization parameters: in the hybrid power pulse test, there are a zero state response stage and a zero input response stage which can be used for acquiring polarization parameters, and due to the polarization characteristics of the battery, the terminal voltage is controlled from U _B Slowly drop to U _C At this time, the zero state response phase of the RC loop is realized;

the time domain analysis is carried out on the circuit, and the functional relation between the terminal voltage U and the time t at the stage can be obtained as follows:

U＝U _OCV -IR ₀ -IR ₁ [1-exp(-t/t)] (3)

wherein t is a time constant (t=r ₁ C ₁ )。

Preferably, the specific steps of S4 are as follows:

selecting different time domains of a discharging stage in a hybrid power pulse test at different temperatures, screening the time domains for obtaining polarization parameters, analyzing simulation model results under different working conditions, comparing the simulation model results with actual results, screening out the polarization parameter extraction time domain with the minimum difference value as the optimal time domain, and constructing a more accurate single battery equivalent circuit simulation model by taking the corresponding polarization parameters as the optimal polarization parameters.

Preferably, the specific step of S5 is as follows:

constructing a short-circuit fault simulation model of the parallel battery pack: based on the built single battery simulation models, connecting the two single battery models in parallel according to a physical connection mode of a circuit to build a parallel battery pack simulation model; then, on the basis of a parallel battery pack simulation model, a short-circuit resistor is directly added to realize short-circuit fault setting, and meanwhile, a current sensor is added to the connection position of the short-circuit resistor so as to directly obtain leakage current of the short-circuit fault, thereby completing construction of the parallel battery pack short-circuit fault model;

And (3) verifying a short-circuit fault simulation model of the parallel battery pack: and comparing the end voltage simulation of the parallel battery pack with experimental results under the same discharging and charging conditions at the same temperature and under the same short-circuit resistance, judging whether the model precision meets the requirement of the short-circuit fault diagnosis of the battery pack through the difference value between the end voltage simulation and the experimental results, indicating that the simulation model precision is higher as the difference value is smaller, and selecting the simulation model with the highest precision for the short-circuit fault diagnosis of the battery pack.

Preferably, the specific step of S6 is as follows:

and (3) charging voltage curve analysis: as the charging process continues, the battery voltage gradually increases, compared with the normal battery charging time t ₁ A battery having a short-circuit fault requires a longer time t ₂ Can reach the charge cut-off voltage deltat ₁₂ The total leakage amount in the whole charging process can be calculated according to the time difference between the charging current and the charging time for charging the normal single battery and the short-circuit fault battery to the cut-off voltage, but in the actual application process, the normal single battery is difficult to be fully charged due to the difference of the single batteries, so that the situation that the fault diagnosis is difficult to compare exists by directly utilizing the charging voltage curve;

IC curve analysis: obtaining an IC curve by obtaining a first derivative of a charge capacity-voltage (Q-V) curve in a battery charging process, namely, a charge capacity change rate-voltage (dQ/dV-V) curve, replacing dV with a fixed voltage interval Δv, replacing dQ with Δq, wherein Δq is a charge capacity change under a corresponding Δv interval, and when Δv=1 mV, the IC curve is characterized by Δq/dv=dq/dV;

there are clear n characteristic peaks on the IC curve, which are arranged according to the order of appearance during chargingThey are divided into: (n) _Ⅰ Peak, n _Ⅱ Peak, n _Ⅲ Peak, n _Ⅳ Peak, n _i Peak …), selecting characteristic peak n with high peak value _i A peak; comparing the normal single battery with the short-circuit fault single battery, wherein the number of the single batteries is n _Ⅰ Peak sum n _Ⅱ There is a significant difference in the peak-to-peak values.

Preferably, the specific steps of S7 are as follows:

short circuit resistance IC curve analysis of different degrees under different charging multiplying powers: considering that the charging modes of the battery pack are divided into slow charging, pulse charging and the like in the actual charging process, charging currents have differences, the IC curve characteristics of 1 single battery in parallel connection under different short-circuit resistance values of different degrees under different multiplying powers are analyzed based on a parallel battery pack short-circuit fault simulation model, and finally a small-multiplying power charging working condition is used for researching short-circuit faults of the battery;

IC curve analysis under different short circuit resistance values of battery packs with different parallel batteries: the method is characterized in that the real vehicle battery pack is formed by connecting a plurality of single batteries in parallel, so that the influence of the number of the single batteries connected in parallel on the characteristics of the IC curve is analyzed, and the IC curve of a normal battery pack formed by connecting m single batteries in parallel and the IC curve of the battery pack formed by connecting the single batteries in parallel in different degrees under the constant current charging working condition is analyzed based on a parallel battery pack short circuit fault simulation model.

Preferably, the specific steps of S9 are as follows:

based on n _Ⅰ Fault diagnosis of peaks: respectively connecting m single batteries in parallel, setting short-circuit resistors at intervals to simulate under a charging working condition, obtaining an IC curve according to simulated charging data, and obtaining an IC curve n under different short-circuit resistance values by using a charging multiplying power obtained by an actual charging test with a charging current of S8 in a simulation process _Ⅰ Counting peak-to-peak values, constructing functions of characteristic peak values and short circuit resistance values, setting short circuit fault conditions of at least two groups of short circuit resistance values, verifying the constructed functions, and respectively obtaining IC curves n under the short circuit faults _Ⅰ Carrying the peak value of the peak into the constructed fault diagnosis function to solve the short circuit resistance value, and comparing the solved short circuit resistance value with the set short circuit resistance value;

based on n _Ⅱ Fault diagnosis of peaks: respectively connecting m single batteries in parallel, setting short-circuit resistors at intervals to simulate under a charging working condition, obtaining an IC curve according to simulated charging data, and obtaining an IC curve n under different short-circuit resistance values by using a charging multiplying power obtained by an actual charging test with a charging current of S8 in a simulation process _Ⅱ Counting peak-to-peak values, constructing functions of characteristic peak values and short circuit resistance values, setting short circuit fault conditions of at least two groups of short circuit resistance values, verifying the constructed functions, and respectively obtaining IC curves n under the short circuit faults _Ⅱ Carrying the peak value of the peak into the constructed fault diagnosis function to solve the short circuit resistance value, and comparing the solved short circuit resistance value with the set short circuit resistance value;

based on n _Ⅰ Peak sum n _Ⅱ Fusion failure diagnosis of peaks: in the actual charging process, n _Ⅰ The peak voltage interval charging current will be larger, resulting in a large diagnostic error, taking into account the comparison with n _Ⅱ Peak, n _Ⅰ Earlier peak occurrence, favoring earlier fault finding, thus, combining n _Ⅰ Peak sum n _Ⅱ Peak to short circuit fault diagnosis, medium term of charging process through n of IC curve _Ⅰ The peak realizes the primary judgment of the short circuit resistance value, and the later stage of the charging process passes through n of the IC curve _Ⅱ The peak realizes accurate quantitative diagnosis of the short circuit resistance value.

The device for realizing the method for diagnosing the short-circuit faults of the parallel battery packs based on the capacity increment curve comprises an offline calibration module and an online estimation module;

the off-line calibration module is used for constructing a functional relation between the characteristic peak value and the short-circuit resistance value of the IC curve under the off-line condition and mainly comprises n _Ⅰ Off-line calibration function of peak-to-peak value and short circuit resistance value and n _Ⅱ Off-line calibration functions of peak-to-peak values and short circuit resistance values;

the on-line estimation module is used for calculating the capacity change rate (dQ/dV-V) according to the actually measured voltage and current values of the parallel batteries, and judging whether the parallel batteries reach n in real time in the charging process of the parallel batteries _Ⅰ Peak or n _Ⅱ If the peak reaches any peak value, the peak value is brought into a corresponding offline function, the short circuit resistance value of the actual parallel battery pack is calculated,and fault diagnosis is realized.

The off-line calibration module is used for acquiring model open-circuit voltage, ohmic internal resistance and polarization parameters of practically applied parallel batteries, constructing a parallel battery pack short-circuit fault simulation model, simulating IC curves under different short-circuit faults of different numbers of parallel batteries, and then respectively utilizing n on the IC curves _Ⅰ Peak and n _Ⅱ The peak-peak value establishes the functional relation between the characteristic peak value and the short-circuit resistance value under the specific parallel battery number under the actual condition, thereby realizing n _Ⅰ Peak-to-Peak and short Circuit resistance function and n _Ⅱ Off-line calibration of peak value and short circuit resistance function;

the on-line estimation module is used for obtaining the current and the voltage of the parallel battery according to actual measurement, calculating the capacity change rate (dQ/dV-V) of the parallel battery, and judging whether the parallel battery reaches n in real time in the charging process of the parallel battery _Ⅰ Peak and n _Ⅱ Diagnostic range of peak, if n is reached _Ⅰ The peak voltage range is then brought into n _Ⅰ Off-line calibration function of peak-to-peak value and short circuit resistance value, thereby based on n _Ⅰ Calculating a corresponding short circuit resistance value by the peak value; if n is reached _Ⅱ The peak voltage range is then brought into n _Ⅱ Off-line calibration function with short circuit resistance value, thereby based on n _Ⅱ And obtaining the short circuit resistance value by the peak value.

The beneficial effects are that: the invention provides a parallel battery fault diagnosis method integrating characteristic peaks, which passes through n in the middle stage of the charging process _Ⅰ The peak realizes the rough judgment of the short circuit resistance value, and the later stage of the charging process charges the n of the IC curve through a small multiplying power _Ⅱ The peak realizes more accurate quantitative diagnosis of the short circuit resistance value, thereby realizing rapid and high-precision fault diagnosis. The fault diagnosis is carried out through different characteristic peaks of the IC curve, a complete charging voltage curve is not needed, and the diagnosis of the short circuit fault of the battery is relatively more facilitated; at the same time fuse n _Ⅰ Peak sum n _Ⅱ The fault diagnosis of the peak has the characteristics of more flexibility and stronger adaptability when the charge and discharge states are uncertain under the actual vehicle working condition.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.

FIG. 1 is a flow chart of the method of the present invention;

FIG. 2 is a first order RC equivalent circuit model diagram;

FIG. 3 is a graph of the step voltage and current variation for a hybrid pulse test cycle;

FIG. 4 is a comparison of simulation of model parameters extracted by different time domains at 5℃with experimental terminal voltages; (a) And (b) model parameter simulations at 0.5C and 1C at 20s are compared to experimental terminal voltages; (c) And (d) model parameter simulations at 0.5C and 1C at 40s are compared to experimental terminal voltages; (e) And (f) model parameter simulation at 0.5C and 1C at 60s is compared with experimental terminal voltage;

FIG. 5 is a comparison of simulation of model parameters extracted by different time domains at 25deg.C with experimental terminal voltage; (a) And (b) model parameter simulations at 0.5C and 1C at 20s are compared to experimental terminal voltages; (c) And (d) model parameter simulations at 0.5C and 1C at 40s are compared to experimental terminal voltages; (e) And (f) model parameter simulation at 0.5C and 1C at 60s is compared with experimental terminal voltage;

FIG. 6 is a simulation model of a parallel battery pack short circuit fault;

FIG. 7 is a graph showing the comparison of parallel battery simulation and experimental terminal voltage under different short circuit resistance values under a 1C discharging condition; (a) is 50Ω; (b) is 10Ω; (c) is 5Ω; (d) is 2.5 Ω;

FIG. 8 is a comparison of parallel battery simulation and experimental terminal voltage under different short circuit resistance values under a 1C charging condition; (a) is 50Ω; (b) is 10Ω; (c) is 5Ω; (d) is 2.5 Ω;

FIG. 9 is a graph showing the charge voltage of the battery cells at different short circuit resistances;

FIG. 10 is a graph showing capacity increment curves of single batteries under different short circuit resistance values;

FIG. 11 is a graph showing the comparison of IC curves of different levels of short-circuit resistance cells at different charge rates; (a) is 1/2C; (b) is 1/5C; (C) is 1/8C; (d) is 1/20C;

FIG. 12 is a graph showing the comparison of IC curves for different shorting resistances for different parallel cell numbers; (a) 1 single battery is connected in parallel; (b) a battery pack with 2 single cells connected in parallel; (c) a battery pack with 4 single cells connected in parallel; (d) a battery pack with 8 single cells connected in parallel;

fig. 13 is an average charging rate frequency statistic;

fig. 14 is a graph showing statistics of (a) a charge start SOC and (b) a charge cut-off SOC frequency;

FIG. 15 is a graph of III peak-to-peak values versus short circuit resistance for different parallel battery IC curves; (a) 1 single battery is connected in parallel; (b) a battery pack with 2 single cells connected in parallel; (c) a battery pack with 4 single cells connected in parallel; (d) a battery pack with 8 single cells connected in parallel;

FIG. 16 is a graph showing the short circuit resistance estimation error based on peak III for different parallel battery packs; wherein 2p represents a battery pack with 2 single batteries connected in parallel, 4p represents a battery pack with 4 single batteries connected in parallel, and 8p represents a battery pack with 8 single batteries connected in parallel;

FIG. 17 is a graph showing the relationship between the peak-to-peak value and the short circuit resistance of IC curves of different parallel battery packs; (a) 1 single battery is connected in parallel; (b) a battery pack with 2 single cells connected in parallel; (c) a battery pack with 4 single cells connected in parallel; (d) a battery pack with 8 single cells connected in parallel;

FIG. 18 is a graph showing the estimated error of the IV peak-based short circuit resistance for different parallel battery packs; wherein 2p represents a battery pack with 2 single batteries connected in parallel, 4p represents a battery pack with 4 single batteries connected in parallel, and 8p represents a battery pack with 8 single batteries connected in parallel;

fig. 19 is a schematic diagram of a parallel battery pack short-circuit fault diagnosis apparatus based on a capacity increment curve.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate orientations or positional relationships based on the drawings, are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.

In the present invention, unless expressly stated or limited otherwise, a first feature "above" or "below" a second feature may include both the first and second features being in direct contact, as well as the first and second features not being in direct contact but being in contact with each other through additional features therebetween. Moreover, a first feature being "above," "over" and "on" a second feature includes the first feature being directly above and obliquely above the second feature, or simply indicating that the first feature is higher in level than the second feature. The first feature being "under", "below" and "beneath" the second feature includes the first feature being directly under and obliquely below the second feature, or simply means that the first feature is less level than the second feature.

As shown in fig. 1, the method for diagnosing the short-circuit fault of the parallel battery pack based on the capacity increment curve comprises the following steps:

s1: constructing a first-order RC equivalent circuit model;

as shown in fig. 2, based on kirchhoff's law, the state space equation of the first-order RC equivalent circuit model is as follows:

wherein R is ₀ Represents ohmic internal resistance, R ₁ For polarization resistance, C ₁ Representing the polarization capacitance, I is the load current, U represents the battery terminal voltage, U _OCV Is opened toRoad voltage, U ₁ Representing polarisation voltage, i.e. R ₁ C ₁ The voltage across it.

The basic parameters of the battery of this example are shown in table 1:

TABLE 1 Battery performance parameters

Attributes of	Specific specification of
		Battery class	Ternary lithium ion
Rated capacity	3.0Ah
		Nominal voltage	3.6V
Voltage range	3.0～4.2V
		Maximum charging current	For 5.0A, pulse 6.0A
Maximum discharge current	Lower than 80 ℃,30A; higher than 80 ℃ and 15A
		Size of the device	Diameter 18mm, height 65mm

S2: carrying out hybrid power pulse test on the battery to obtain model parameters of the first-order RC equivalent circuit model constructed in the step S1, wherein the model parameters comprise open-circuit voltage, ohmic internal resistance and polarization parameters;

the hybrid power pulse test specifically comprises: and (3) carrying out 1C constant current charging on the battery until the voltage is greater than or equal to 4.2V and is cut off, carrying out 4.2V constant voltage charging until the current is less than or equal to 0.15A and is cut off, carrying out 5C constant current discharging after the battery is kept stand for 5h and is cut off after the current exceeds 10s, carrying out 5C constant current charging after the battery is kept stand for 40s again, carrying out 0.5C constant current charging after the battery is kept stand for 40s and 40s until the current exceeds 10s, and carrying out 5C constant current discharging again after the battery is kept stand for 2h after 6min and is kept stand after the current is kept stand for 10s again, so that the battery is circulated for 20 times or the voltage is less than or equal to 3.0V and is ended. As shown in fig. 3, the current and voltage change curves of the step pulse phase of the hybrid pulse test cycle were obtained when the SOC was 0.6.

Obtaining an open circuit voltage: the hybrid power pulse test has a process of discharging for 6min at constant current of 0.5C, the SOC is reduced by about 5%, then the battery is allowed to stand for 2 hours, so that the battery terminal voltage can reach a stable state, and at this time, the battery terminal voltage can be regarded as an open circuit voltage, and further corresponding data of the SOC and the open circuit voltage are obtained, as shown in table 2,

TABLE 2 open circuit voltage parameter values

And (3) obtaining ohmic internal resistance: as shown in fig. 3, the phases a to B are suddenly excited by a discharge pulse of 10s, the voltage rapidly drops from UA to UB, which is caused by the ohmic internal resistance R0, and according to this phase, the ohmic internal resistance R0 can be identified, and the calculation formula is as follows:

where R0 is the instantaneous impedance, UA is the voltage at point A and UB is the voltage at point B. Ohmic internal resistance parameters are shown in table 3:

TABLE 3 ohm internal resistance parameter

SOC	R0(Ω)	SOC	R0(Ω)
				0	0.029	55	0.021
5	0.026	60	0.021
				10	0.023	65	0.022
15	0.022	70	0.022
				20	0.021	75	0.023
25	0.021	80	0.024
				30	0.021	85	0.026
35	0.021	90	0.028
				40	0.021	95	0.032
45	0.021	100	0.036
				50	0.021	-	-

Obtaining polarization parameters: the 10s discharge pulse excitation and the 40s standing phase respectively correspond to a zero state response phase and a zero input response phase, and can be used for identifying polarization parameters, wherein the 10s discharge pulse excitation is taken as an example, and due to the polarization characteristic of a battery, the terminal voltage slowly drops from UB to UC, and is the zero state response phase of an RC loop;

U＝U _OCV -IR ₀ -IR ₁ [1-exp(-t/t)] (3)

Wherein t is a time constant (t=r ₁ C ₁ )。

S3: constructing a single battery simulation model;

and (3) selecting the first 20s, 40s and 60s time domains of the 6min discharge stage in the mixed power pulse test data at the temperature of 5 ℃ and 25 ℃ to carry out model polarization parameter identification, then carrying out simulation result analysis under different working conditions, and researching the influence of the polarization parameter identification time domains on simulation precision.

The results of simulation of model polarization parameters extracted through different time domains at 5 ℃ and 25 ℃ compared with experimental terminal voltages are shown in fig. 4 and 5. As can be seen from the comparison result of the graph, the accuracy of extracting the corresponding model parameters by adopting the 40s time domain is high. The polarization parameters extracted in the 40s time domain are shown in tables 4 and 5:

TABLE 4 internal resistance parameter of polarization

SOC	R1(Ω)	SOC	R1(Ω)
				0	0.066	55	0.030
5	0.060	60	0.031
				10	0.052	65	0.031
15	0.044	70	0.03
				20	0.038	75	0.028
25	0.033	80	0.025
				30	0.029	85	0.023
35	0.027	90	0.022
				40	0.027	95	0.024
45	0.028	100	0.031
				50	0.029	-	-

TABLE 5 polarization capacitance parameters

Constructing a short-circuit fault simulation model of the parallel battery pack: in the embodiment, a simulation model based on a Simscape in MATLAB/Simulink is selected to build a single battery equivalent circuit model. And connecting the two single battery models in parallel according to a physical connection mode of the circuit to construct a parallel battery pack simulation model. Then, on the basis of a parallel battery pack simulation model, short-circuit fault setting can be realized by directly adding a short-circuit resistor, and meanwhile, a current sensor is added at the connection position of the short-circuit resistor so as to directly obtain leakage current of the short-circuit fault, and the principle is shown in fig. 6. The impedance of the connecting pieces existing in the actual connection structure, namely the connecting pieces 1 and 2 in the figure, is simulated by a resistor module in the Simtape library. In addition, current sensors are directly added on the branches of the two parallel batteries to monitor the branch currents, and the branch currents correspond to the branch current 1 and the branch current 2 respectively. Meanwhile, SOC of each single battery and each parallel battery pack are calculated respectively to correspond to SOC1, SOC2 and SOC respectively, and an ampere-hour integration method is adopted in SOC calculation.

And (3) verifying a short-circuit fault simulation model of the parallel battery pack: the simulation terminal voltage and the experimental terminal voltage of the parallel battery pack are compared under different short circuit resistance values under the working conditions of discharging at 25 ℃ and charging at 1C respectively, and the results are shown in fig. 7 and 8. Under the 1C discharging working condition, under four short circuit resistance values, the maximum absolute error is 60mV, and the corresponding maximum relative error is 1.9%. Under the 1C charging condition, the maximum absolute error is about 28mV and the maximum relative error is about 0.79% under the four short circuit resistance values. The charge and discharge results show that the model precision can meet the requirement of subsequent battery pack short-circuit fault diagnosis.

the embodiment analyzes the charging voltage curve and the IC curve based on the normal state of the single battery and the charging voltage curve and the IC curve when the 100 omega short circuit fault under the 1/8C constant current charging working condition.

And (3) charging voltage curve analysis: as shown in fig. 9, the battery voltage gradually increases as the charging process continues. Compared with the normal single battery charging time t ₁ A battery having a short-circuit fault requires a longer time t ₂ The charge cut-off voltage can be reached. Δt (delta t) ₁₂ For the time difference between the normal single battery and the 100 omega short-circuit fault battery to be charged to the cut-off voltage, the charging current and the charging time differenceThe total leakage amount in the whole charging process can be calculated. However, in the practical application process, the normal single battery is difficult to be fully charged absolutely due to the difference of the single batteries. Therefore, the direct use of the charge voltage time difference for fault diagnosis has a problem of difficult time comparison.

IC curve analysis: fig. 10 shows the IC curves of a normal cell and a 100 Ω short-circuit fault cell under a 1/8C charging condition. There are four distinct characteristic peaks on the IC curve, which can be divided into according to the order in which they occur during charging: peak I, peak II, peak III, peak IV. The III peak and the IV peak are high, and the voltage interval accords with the normal charging condition. Analyzing the IC curve, the IC curve in the whole voltage interval can be divided into 4 peak intervals, and the maximum voltage interval of the peak i and the minimum voltage interval of the peak ii can be seen from fig. 9; the III peak and the IV peak are high and the voltage interval is in the conventional charging interval. The comparison shows that the difference between the III peak and the IV peak is obvious between the normal single battery and the 100 omega short-circuit fault single battery.

short circuit resistance IC curve analysis of different degrees under different charging multiplying powers: considering that the battery pack is charged in actual charging, the charging modes are classified into slow charging, pulse charging and the like, and the charging currents are different. Based on a parallel battery pack short-circuit fault simulation model, the IC curve characteristics of 1 single battery in parallel under the conditions of 1/2C, 1/5C, 1/8C and 1/20C multiplying power and different short-circuit resistance values are researched. As shown in fig. 11, it can be seen that under the small-rate charging condition, the difference between the characteristic peak and peak values of the IC curves of the normal single battery and the single battery with the short-circuit fault is relatively more obvious, which indicates that the small-rate charging condition is more suitable for researching the short-circuit fault of the battery.

IC curve analysis under different short circuit resistance values of battery packs with different parallel batteries: considering that the real vehicle battery pack is formed by connecting a plurality of single batteries in parallel, the influence of the number of the single batteries connected in parallel on the characteristics of the IC curve is analyzed, and based on a parallel battery pack short-circuit fault simulation model, the IC curve of a normal battery pack with 1 single battery connected in parallel, 2 single batteries connected in parallel, 4 single batteries connected in parallel and 8 single batteries connected in parallel and the IC curve of the short-circuit fault battery with different degrees are analyzed, as shown in fig. 12, the result of the IC curve given in the graph shows that the difference between the characteristic peak value and the characteristic peak value of the normal battery and the characteristic peak value with the short-circuit fault is smaller and smaller along with the increase of the number of the parallel batteries, which means that the more the number of the parallel batteries is, the difficulty of diagnosing the short-circuit fault through the characteristic peak is larger.

as shown in fig. 13, the data of the actual charging rate test shows that the charging rate is within 0.2C in most of the charging processes, wherein the charging frequency is 609 times in the interval from 0.05C to 0.1C; from the statistical results, nearly half of the average charge rate was less than 0.1C. Based on the statistical result, an IC curve with a charging multiplying power of 0.075C is selected for subsequent short-circuit fault diagnosis.

As shown in fig. 14, the data of the actual charge voltage interval test is used for statistics, so that the charge start SOC does not have a very significantly fixed SOC interval, the distribution is wide, and the charge cut-off SOC is relatively high, which is higher than 95%. The frequency of charging is 773 times when the SOC is lower than 40%, that is, half of the charging process can cover the interval from 40% to 98% of the SOC, the voltage interval corresponding to the interval of the SOC is approximately 3.74V to 4.15V, and the III peak and the IV peak of the IC curve can be formed basically completely in the voltage interval. Therefore, the short-circuit fault diagnosis of the subsequent battery is carried out, and the III peak and the IV peak of the IC curve are selected for research.

Fault diagnosis based on peak iii: the method specifically comprises the steps of respectively carrying out simulation under a charging working condition on a battery pack with 1 single battery connected in parallel, 2 single batteries connected in parallel, 4 single batteries connected in parallel and 8 single batteries connected in parallel, and setting a short-circuit resistor with an interval of 100 omega between 100 omega and 1000 omega, wherein the charging current is 0.075C in the simulation process. And obtaining an IC curve according to the simulated charging data, and counting the III peak and the peak of the IC curve under different short circuit resistance values.

As shown in fig. 15, the relationship between the peak-to-peak value of III and the short-circuit resistance of the IC curves of different parallel battery packs is shown. As can be seen from fig. 14, as the short circuit resistance increases, the III peak to peak value of the IC curve gradually decreases; as the short circuit resistance increases, the peak change rate also becomes smaller. And establishing a functional relation between the peak value of the peak III of the IC curve and the short circuit resistance value of different parallel battery packs, as shown in a fitting curve in fig. 14.

In order to verify the accuracy of the established function, short circuit faults with the short circuit resistance values of 950 omega, 750 omega, 550 omega, 350 omega and 150 omega are respectively set for verification research. And respectively bringing the III peak and the peak value of the IC curve obtained under the short circuit resistance value into the constructed relation function between the III peak and the peak value and the short circuit resistance value, thereby solving the short circuit resistance value, wherein the result is shown in Table 6:

TABLE 6 diagnosis results of short circuit resistance values of different parallel battery packs

As shown in fig. 16, the short-circuit resistance error chart based on the III peak diagnosis is for the parallel batteries of different parallel battery numbers. From the diagnosis error result, the diagnosis errors of the battery packs of 1 single battery in parallel connection, 2 single batteries in parallel connection and 4 single batteries in parallel connection under the five short circuit resistance values are all less than 1%; for the battery pack with 8 single batteries connected in parallel, the diagnosis error is about 1% under the fault of the short-circuit resistance value of 150Ω, but the diagnosis error is larger and larger along with the increase of the short-circuit resistance value, and the diagnosis error reaches 7% under the short-circuit resistance value of 950 Ω. The analysis is that as the short circuit resistance value increases, the peak-to-peak value difference of fault characteristics becomes smaller and smaller, and the influence of fitting errors increases.

IV peak-based fault diagnosis: and selecting battery packs of which 1 single battery is connected in parallel, 2 single batteries are connected in parallel, 4 single batteries are connected in parallel and 8 single batteries are connected in parallel, and respectively carrying out simulation under the charging working condition of 0.075C at intervals of 100 omega to 1000 omega. And obtaining an IC curve according to the simulated charging data, and counting the III peak and the peak of the IC curve under different short circuit resistance values.

As shown in fig. 17, the relationship between the peak value and the short-circuit resistance of the IC curves iv of different parallel battery packs is shown. As can be seen by comparing fig. 15 and 17, the IV peak to peak value is larger than the III peak to peak value, and the IV peak to peak difference between different short circuit resistances is relatively larger.

In order to verify the short-circuit fault diagnosis function constructed by the IV peak, the short-circuit fault conditions with the short-circuit resistance values of 950 omega, 750 omega, 550 omega, 350 omega and 150 omega are set for verification research. And respectively bringing the IV peak and the peak value of the IC curve obtained under the short circuit resistance value into the constructed relation function of the IV peak and the peak value and the short circuit resistance value, so as to solve the short circuit resistance value, wherein the specific diagnosis result is shown in the table 7:

TABLE 7 diagnosis results of short circuit resistance values of different parallel battery packs

As shown in fig. 18, the short-circuit resistance error chart of the parallel battery pack based on IV peak diagnosis is shown for different numbers of parallel batteries. As can be seen by comparing fig. 16 and 18, the error in the short circuit diagnosis by the iv peak is relatively more stable.

Fusion fault diagnosis based on III peak and IV peak: the short-circuit fault diagnosis results of the III peak and the IV peak are comprehensively compared, and the diagnosis accuracy is higher through the two peaks under the condition that the number of the parallel batteries is small; however, as the number of parallel cells increases, particularly in the case of a battery pack in which 8 unit cells are connected in parallel, the error in diagnosis by the III peak increases significantly. Considering that the IC curve IV peak voltage interval is in the later stage of charging, the trickle charging current is small at the moment, and the IV peak is considered to be more suitable for fault diagnosis by using small-rate current; in the actual charging process, the charging current in the III peak voltage interval may be larger, resulting in a large diagnostic error. But it is considered that the III peak occurs earlier than the IV peak, facilitating earlier fault discovery. Therefore, the III peak and the IV peak can be synthesized to diagnose the short circuit fault, the rough judgment of the short circuit resistance value is realized through the III peak in the middle stage of the charging process, and the more accurate quantitative diagnosis of the short circuit resistance value is realized through the IV peak in the later stage of the charging process.

As shown in fig. 19, the device for implementing the method for diagnosing the short-circuit fault of the parallel battery pack based on the capacity increment curve comprises an offline calibration module and an online estimation module;

the off-line calibration module is used for constructing a functional relation between the characteristic peak value of the IC curve and the short-circuit resistance value under the off-line condition and mainly comprises an off-line calibration function of a III peak value and the short-circuit resistance value and an off-line calibration function of an IV peak value and the short-circuit resistance value;

the online estimation module is used for calculating the capacity change rate (dQ/dV-V) of the parallel battery according to the actually measured voltage and current values of the parallel battery, judging whether the parallel battery reaches a III peak or an IV peak in real time in the charging process of the parallel battery, and if the parallel battery reaches a certain peak, taking the parallel battery into a corresponding offline function, calculating the short circuit resistance of the actually parallel battery, and realizing fault diagnosis.

The off-line calibration module is used for carrying out model parameter identification on actually applied parallel batteries, constructing a parallel battery pack short circuit fault simulation model of the parallel battery pack, simulating IC curves under different short circuit faults of different parallel batteries, and then respectively utilizing III peak and IV peak on the IC curves to establish a functional relation between characteristic peak and short circuit resistance under specific parallel battery numbers, so as to realize off-line calibration of III peak and short circuit resistance functions and IV peak and short circuit resistance functions;

The online estimation module is used for obtaining the current and the voltage of a certain parallel battery according to actual measurement, calculating the capacity change rate (dQ/dV-V) of the parallel battery, judging whether the parallel battery reaches the diagnosis range of a III peak and an IV peak in real time in the charging process of the parallel battery, and if the parallel battery reaches the III peak voltage range, bringing the III peak and the IV peak into an offline calibration function of the short circuit resistance value, so that the corresponding short circuit resistance value is calculated based on the III peak and the IV peak; if the IV peak voltage range is reached, the IV and short circuit resistance off-line calibration function is carried in, so that the more accurate short circuit resistance is obtained based on the IV peak value.

In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The method for diagnosing the short circuit fault of the parallel battery pack based on the capacity increment curve is characterized by comprising the following steps of: the method comprises the following steps:

s1: constructing a first-order RC equivalent circuit model;

s2: carrying out hybrid power pulse test on the battery to obtain parameters of a first-order RC equivalent circuit model constructed in the S1, wherein the parameters comprise open-circuit voltage, ohmic internal resistance and polarization parameters;

s3: constructing a single battery simulation model;

s4: obtaining optimal time domain screening by polarization parameters: the polarization parameters comprise polarization internal resistance and polarization capacitance, the polarization parameters acquired in different time domains are input into a single battery simulation model, the difference value of simulation voltage and actually measured voltage in different time domains is compared, the time domain with the smallest difference value is selected as an optimal time domain, and the corresponding polarization parameter is the optimal polarization parameter;

s8: according to the influences of the charging multiplying power and the number of the parallel batteries obtained in the S7 on the characteristic peak value of the IC curve of the parallel battery pack under the short circuit fault, the characteristic peak value and the charging multiplying power for short circuit fault analysis are obtained through statistics on actual operation conditions including actual charging multiplying power and actual charging voltage intervals;

2. The capacity increment curve-based diagnosis method for short-circuit faults of the parallel battery pack according to claim 1, characterized in that: the specific steps of the S1 are as follows:

3. The capacity increment curve-based diagnosis method for short-circuit faults of the parallel battery pack according to claim 2, characterized in that: the specific steps of the S2 are as follows:

the hybrid power pulse test specifically comprises: performing charge-discharge pulse excitation and standing treatment on batteries under different SOCs to obtain battery characteristic parameters under different SOCs;

U＝U _OCV -IR ₀ -IR ₁ [1-exp(-t/t)] (3)

wherein t is a time constant (t=r ₁ C ₁ )。

4. The capacity increment curve-based diagnosis method for short-circuit fault of parallel battery pack according to claim 3, wherein: the specific steps of the S4 are as follows:

selecting different time domains of a discharge stage in a hybrid power pulse test at different temperatures, screening the time domains for acquiring polarization parameters, analyzing simulation model results under different working conditions, comparing the simulation model results with actual results, screening out the polarization parameter extraction time domain with the minimum difference value as the optimal time domain, and constructing an equivalent circuit simulation model of the single battery by taking the corresponding polarization parameters as the optimal polarization parameters.

5. The capacity increment curve-based diagnosis method for short-circuit faults of the parallel battery pack according to claim 4, characterized in that: the specific steps of the S5 are as follows:

6. The capacity increment curve-based diagnosis method for short-circuit faults of the parallel battery pack according to claim 5, characterized by: the specific steps of the S6 are as follows:

there are clear n characteristic peaks on the IC curve, which are divided into according to the order in which they occur during charging: (n) _Ⅰ Peak, n _Ⅱ Peak, n _Ⅲ Peak, n _Ⅳ Peak, n _i Peak …), selecting characteristic peak n with high peak value _i A peak; comparing the normal single battery with the short-circuit fault single battery, wherein the number of the single batteries is n _Ⅰ Peak sum n _Ⅱ There is a significant difference in the peak-to-peak values.

7. The capacity increment curve-based diagnosis method for short-circuit faults of the parallel battery pack according to claim 6, characterized in that: the specific steps of the S7 are as follows:

short circuit resistance IC curve analysis of different degrees under different charging multiplying powers: considering that the charging modes of the battery pack are divided into slow charging, pulse charging and the like in the actual charging process, charging currents have differences, the IC curve characteristics of 1 single battery in parallel connection under different short-circuit resistance values of different degrees under different multiplying powers are analyzed based on a parallel battery pack short-circuit fault simulation model, and finally a small-multiplying power charging working condition is used for diagnosing the short-circuit faults of the battery;

8. The capacity increment curve-based diagnosis method for short-circuit faults of the parallel battery pack according to claim 7, characterized in that: the specific steps of the S9 are as follows:

Based on n _Ⅱ Fault diagnosis of peaks: respectively connecting m single batteries in parallel, setting short-circuit resistors at intervals to simulate under a charging working condition, obtaining an IC curve according to simulated charging data, and obtaining an IC curve n under different short-circuit resistance values by using a charging multiplying power obtained by an actual charging test with a charging current of S8 in a simulation process _Ⅱ Counting peak-to-peak values, constructing a function of the characteristic peak values and short circuit resistance values, and setting short circuit fault conditions of at least two groups of short circuit resistance values to performVerifying the construction function, namely respectively obtaining IC curves n under the short-circuit fault _Ⅱ Carrying the peak value of the peak into the constructed fault diagnosis function to solve the short circuit resistance value, and comparing the solved short circuit resistance value with the set short circuit resistance value;

9. The device for realizing the method for diagnosing the short-circuit faults of the parallel battery packs based on the capacity increment curve is characterized in that: the system comprises an offline calibration module and an online estimation module;

the on-line estimation module is used for calculating the capacity change rate (dQ/dV-V) according to the actually measured voltage and current values of the parallel batteries, and judging whether the parallel batteries reach n in real time in the charging process of the parallel batteries _Ⅰ Peak or n _Ⅱ And if the peak value reaches any peak value, the peak value is brought into a corresponding offline function, and the short circuit resistance value of the actual parallel battery pack is calculated, so that fault diagnosis is realized.

10. The apparatus for implementing a capacity increment curve-based short circuit fault diagnosis method for parallel battery packs according to claim 9, wherein:

the off-line calibration module is used for obtaining the model opening of the practically applied parallel batteriesObtaining path voltage, ohmic internal resistance and polarization parameters, constructing a parallel battery pack short-circuit fault simulation model, simulating IC curves under different short-circuit faults of different numbers of parallel batteries, and then respectively utilizing n on the IC curves _Ⅰ Peak and n _Ⅱ The peak-peak value establishes the functional relation between the characteristic peak value and the short-circuit resistance value under the specific parallel battery number under the actual condition, thereby realizing n _Ⅰ Peak-to-Peak and short Circuit resistance function and n _Ⅱ Off-line calibration of peak value and short circuit resistance function;