CN115494401A - Power battery cloud end data cleaning method based on information fusion - Google Patents
Power battery cloud end data cleaning method based on information fusion Download PDFInfo
- Publication number
- CN115494401A CN115494401A CN202211417093.9A CN202211417093A CN115494401A CN 115494401 A CN115494401 A CN 115494401A CN 202211417093 A CN202211417093 A CN 202211417093A CN 115494401 A CN115494401 A CN 115494401A
- Authority
- CN
- China
- Prior art keywords
- soc
- data
- matrix
- curve
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
Abstract
The invention relates to a battery state estimation technology, in particular to a power battery cloud end data cleaning method based on information fusion, which comprises the following steps: inputting a SOC sequence and a voltage matrix of a battery; judging whether the current cycle number exceeds a rated value or not, and determining and executing the step; constructing a preliminary SOC-V curve; constructing an SOC-V curve through quadratic function fitting; and combining the current measured value with the SOC-V curve by using an information fusion method to obtain a battery voltage estimated value. The method estimates data which is closer to the actual working condition of the battery by combining historical data of the power battery and measured data of the sensor, weakens the influence of random disturbance and measurement error in the signal transmission process on the data quality of the battery, and provides assistance for the state estimation and fault early warning technology of the power battery.
Description
Technical Field
The invention belongs to the technical field of battery state estimation, and particularly relates to a power battery cloud data cleaning method based on information fusion.
Background
In recent years, the field of new energy automobiles is rapidly developed, and a power battery is taken as one of three cores of the new energy automobiles, and the energy density value of the power battery is also increased year by year. In order to ensure the safe and stable operation of the power battery, a battery state estimation and safety early warning technology based on BMS (battery management system) collected data is continuously developed. However, the real vehicle data of the power battery usually contains large noise, which has a great influence on the accuracy of the battery state estimation and the robustness of the safety early warning technology. Therefore, the data cleaning of the real vehicle data of the power battery and the improvement of the data quality are very critical to the safe use of the power battery system.
At present, the data cleaning technology for the real vehicle data of the power battery mainly uses signal processing methods such as Gaussian filtering, wavelet packet decomposition and the like. In the method, original data is decomposed into a low-frequency component and a high-frequency component by a signal decomposition method, wherein the high-frequency component is used as noise to be processed, and only the low-frequency component is reserved as cleaned data. However, the method has many parameters to be adjusted, and cannot give consideration to various application scenarios, and the optimization is difficult to realize. And simply processing the current cycle data a priori ignores the correlation of the battery's historical data with the current data.
Therefore, the variation information of the battery full-period data needs to be counted from the system level urgently, and information fusion is carried out on the variation information and the measured data to obtain more reliable high-quality battery data, so that assistance is provided for the state estimation and fault early warning technology of the power battery.
Disclosure of Invention
Aiming at the problems in the background art, the invention provides a power battery cloud data cleaning method based on information fusion.
In order to solve the technical problems, the invention adopts the following technical scheme: a power battery cloud data cleaning method based on information fusion comprises the following steps:
step 2, judging whether the current cycle number exceeds a rated value, and determining to execute the step;
step 3, constructing a preliminary SOC-V curve;
and 5, combining the current measured value with the SOC-V curve by using an information fusion method to obtain a battery voltage estimated value.
In the above power battery cloud-end data cleaning method based on information fusion, the method specifically comprises the following steps:
s1, inputting a current cycle battery SOC sequence { S 1 ,…, S i ,…, S T And a voltage matrix V (1,1) ,…,V (i,j) ,…,V (T,N) H, and the current cycle number C; subscripts i and T represent sampling point serial numbers and total sampling length, and subscript N represents the initial cycle number of the battery serial number to be 1;
s2, judging whether the current cycle number C is larger than a threshold value N, and if so, executing S3; if the value is larger than the threshold value N, executing S7; wherein the threshold value N is 10;
s3, constructing a primary SOC-V curve comprising a charging curve and a discharging curve;
s3.1, constructing 2 same SOC empty matrixes c _ (100×N) And SOC disc _ (100×N) In which SOC is c _ (100×N) For recording charging data, SOC disc _ (100×N) For recording discharge data, subscript 100×N Representing a matrix size of 100 rows and N columns;
s3.2, judging the current cycle number C, and if C is an odd number, storing data into a matrix SOC disc _ (100×N) (ii) a If C is even number, data is stored in matrix SOC c _ (100×N) ；
S3.3, values in the SOC sequence are all positive integers of 1-100, and values of the SOC sequence and the voltage matrix are all time sequence data, and the values are in one-to-one correspondence at the same sampling point; the SOC empty matrix recording data rule is as follows:
for voltageMatrix { V } (1,1) ,…,V (i,j) ,…,V (T,N) And SOC sequence S 1 ,…, S i ,…, S T H, voltage V (i,j) The position stored in the empty SOC matrix is S i Row, jth column;
s3.4. Matrix SOC based on complete data recording disc _ (100×N) Sum matrix SOC c _ (100×N) Calculating the median of the data of each position; drawing a curve for each row of data of the SOC matrix, wherein the abscissa is the number of matrix columns, and the ordinate is a data value; obtaining N preliminary SOC-V charging curves and N preliminary SOC-V discharging curves;
s4, for the N primary SOC-V charging curves and the N primary SOC-V discharging curves, fitting a quadratic function to obtain a function set { f (x) 1 ,…, f(x) N ,…, f(x) 2N }；
S5, replacing the part of the preliminary SOC-V charging curve in the S4, of which the vertical coordinate is 0, with a corresponding fitting function to obtain an SOC-V curve;
s6, turning to S1 for circulation, wherein the circulation times C = C +1;
s7, for SOC-V curve and actually measured voltage matrix { V (1,1) ,…,V (i,j) ,…,V (T,N) Information fusion is carried out according to the following rules:
s7.1, through a calculation formula: h _ i = P _ (i-1)/(Q _ (i-1) + P _ (i-1)) to obtain H _ i at the ith sampling point;
wherein P _ (i-1) represents the estimated error value of i-1 samples, and the initial value P 0 The value is 0.5 for adjustable parameters; q _ (i-1) represents the measured error value of i-1 sampling points, and the initial value Q 0 The value is 0.1 for adjustable parameters;
s7.2, calculating a formula: v _ i = V _ (i-1) + H _ i (V) (i,j) V _ (i-1)) to obtain V _ i at the ith sampling point as an estimated voltage value at the ith sampling point;
wherein V _ (i-1) is an estimated voltage value at the i-1 th sampling point, and the initial value V _0 is a voltage value of a corresponding SOC in the jth SOC-V curve;
s7.3, calculating a formula: p _ i = (1-H _ i) × P _ (i-1), obtaining P _ i at the ith sampling point, which is an estimated error value at the ith sampling point;
s7.4, repeating the step 7.1 to the step 7.3, starting to calculate from the 1 st sampling point to the T th sampling point, and obtaining an estimated voltage matrix;
and S8, outputting an estimated voltage matrix.
Compared with the prior art, the invention has the beneficial effects that: the method and the device count the change information of the battery full-period data from the system level, perform information fusion with the measured data to obtain more reliable high-quality battery data, have less required adjusting parameters, and can meet the requirements of battery state estimation and fault early warning technologies. The method estimates data close to the actual working condition of the battery by combining historical data of the power battery and measured data of the sensor, weakens the influence of random disturbance and measurement error in the signal transmission process on the data quality of the battery, and provides assistance for the state estimation and fault early warning technology of the power battery.
And estimating data which is closer to the actual working condition of the battery based on historical data of the power battery and the current sensor measurement data, and weakening the influence of random disturbance and measurement error in the signal transmission process on the data quality of the battery.
Drawings
Fig. 1 is a flowchart of a power battery cloud data cleaning method based on information fusion according to an embodiment of the present invention;
FIG. 2 is a preliminary SOC-V charging curve of a battery cell of example 1 of the present invention;
FIG. 3 is a SOC-V charging curve of the battery cell of example 1.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the following embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive efforts based on the embodiments of the present invention, shall fall within the scope of protection of the present invention.
It should be noted that the embodiments and features of the embodiments of the present invention may be combined with each other without conflict.
The present invention is further illustrated by the following examples, which are not intended to limit the scope of the invention.
In the prior art, a signal processing method is mainly used in a real-vehicle data cleaning technology of a power battery. The method has the advantages that the number of parameters needing to be adjusted is large, various application scenes cannot be considered, and the relevance between the historical data and the current data of the battery is ignored. The embodiment counts the change information of the battery full-period data from the system level, performs information fusion with the measured data to obtain more reliable high-quality battery data, requires fewer adjusting parameters, and can meet the requirements of battery state estimation and fault early warning technologies.
The embodiment is realized through the following technical scheme, as shown in fig. 1, the cloud data cleaning method of the power battery based on information fusion comprises the steps of firstly inputting an SOC sequence and a voltage matrix of the battery, then judging whether the current cycle number exceeds a rated value, then constructing a primary SOC-V curve, constructing a complete SOC-V curve through quadratic function fitting, and finally combining a current measurement value and the SOC-V curve by using the information fusion method to obtain a final battery voltage estimation value.
The technical scheme comprises the following specific implementation steps:
(a) Inputting the SOC sequence { S) of the current cycle battery 1 ,…, S i ,…, S T A sum matrix of voltages V (1,1) ,…,V (i,j) ,…,V (T,N) And the current cycle number C. Where subscripts i and T denote the sample point number and the total sample length and subscript N denotes the battery number initial cycle number C =1. Go to step (b).
(b) It is determined whether the cycle number C is greater than a threshold N, where N is a positive integer and is generally set to 10. If not, go to step (c). And (g) if the threshold value N is larger than the threshold value N, turning to the step (g).
(c) The SOC-V curve comprises two parts, a charging curve and a discharging curve.
Firstly, 2 same SOC empty matrixes SOC are constructed c _ (100×N) And SOC disc _ (100×N) Wherein SOC is c _ (100×N) For recording the number of chargesAccording to the SOC disc _ (100×N) Recording discharge data, subscript 100×N Representing a matrix size of 100 rows and N columns;
then judging the cycle number C, if C is odd number, storing the data into the matrix SOC disc _ (100×N) (ii) a If C is even number, data is stored in SOC c _ (100×N) ；
The values in the SOC sequence are positive integers of 1-100, the values of the SOC sequence and the voltage matrix are time sequence data, and the values at the same sampling point are in one-to-one correspondence. Therefore, the SOC empty matrix record data rule is as follows:
for the voltage matrix V (1,1) ,…,V (i,j) ,…,V (T,N) And SOC sequence S 1 ,…, S i ,…, S T V, voltage V (i,j) The position stored in the empty SOC matrix is S i Row, jth column;
SOC based on recording complete data disc _ (100×N) Matrix and SOC c _ (100×N) The median is calculated for the data at each position. And then drawing a curve for each row of data of the SOC matrix, wherein the abscissa is the number of the matrix columns, and the ordinate is the data value. N preliminary SOC-V charging curves and N preliminary SOC-V discharging curves can be obtained. Go to step (d).
(d) For the N SOC-V charging curves and the N preliminary SOC-V discharging curves, fitting a quadratic function to obtain a function set { f (x) 1 ,…, f(x) N ,…, f(x) 2N }. Go to step (e).
(e) And (f) replacing the part of the SOC-V curve with the ordinate of 0 in the step (d) by a corresponding fitting function to obtain a complete SOC-V curve, and transferring to the step (f).
(f) Cycle number C = C +1. Go to step (a).
(g) For the complete SOC-V curve and the actual measured voltage matrix V (1,1) ,…,V (i,j) ,…,V (T,N) Information fusion is carried out according to the following rules:
firstly, the formula is calculated: h _ i = P _ (i-1)/(Q _ (i-1) + P _ (i-1)) to obtain H _ i at the ith sampling point.
Wherein P _ (i-1) represents the estimated error value of i-1 samples, and the initial value P 0 For adjustable parameters, it is typically set to 0.5.
Q _ (i-1) represents the measured error value of i-1 sampling points, and the initial value Q 0 For adjustable parameters, it is typically set to 0.1.
Then the formula is calculated: v _ i = V _ (i-1) + H _ i (V) (i,j) V _ (i-1)) to obtain V _ i at the ith sampling point, namely the estimated voltage value at the ith sampling point. V _ (i-1) is the estimated voltage value at the i-1 th sampling point, and the initial value V _0 is the voltage value of the corresponding SOC in the jth SOC-V curve. It is noted that a charging SOC-V curve is used for the charging process and a discharging SOC-V curve is used for the discharging process.
And finally, calculating a formula: p _ i = (1-H _ i) × P _ (i-1), which results in P _ i at the ith sample point, i.e., the estimated error value at the ith sample point.
Thus, starting at sample point 1, the calculation ends at sample point T. And obtaining a final estimated voltage matrix.
(h) And outputting an estimated voltage matrix.
Counting the battery full-period data information at the system level, and constructing a complete SOC-V curve which can show the historical working condition characteristics of the battery;
the battery historical information and the current measurement data are combined by using an information fusion method, so that the quality of the battery data can be effectively improved.
In specific implementation, power battery cloud end data acquired in the actual operation process of the electric automobile is used as input, and the input data at least comprises total SOC data of the battery pack, voltage data of each battery monomer in the battery pack, cycle times and sampling point serial numbers. Wherein the accuracy of the SOC data is 1 percent, and the accuracy of the voltage data is 0.001V. The basic data format is shown in table 1.
For convenience of illustration, the number of cells is assumed to be 6 in this example, and the number of cells may vary from tens to a hundred in practical processes. The method is independently carried out on the data calculation of each battery monomer, and the number of the battery monomers can be automatically adjusted according to the actual situation.
TABLE 1 basic data Table
As can be seen from table 1, the current cycle number of the battery is 12, and if the data of the cycle number of 12 needs to be cleaned by using the method, the steps are as follows:
(1) A threshold N =10 and a cycle number C =1 are set. The threshold value N represents the historical circulation times for information fusion, the larger the setting of N is, the more the historical data for information fusion is, the better the cleaning effect is on the historical data with the circulation times larger than N, and the more reliable the result is. However, a larger value of N indicates that the data cleansing method cannot be used for cycles having a cycle number of N or less.
(2) Namely, the SOC data of the C cycle and the voltage data of each battery unit are obtained. The SOC data is a one-dimensional time sequence, and the voltage data of the battery monomer is a matrix with N rows and T columns. Taking table 1 as an example, the SOC data length is 94, and the cell voltage data is a matrix with rows of 94 and columns of 6.
(3) And judging the current cycle number. If C is an odd number, the current cycle is a discharge cycle; if C is an even number, the current cycle is a charging cycle.
(4) And judging whether the cycle number C is greater than a threshold value N. And (5) if the value is not greater than the threshold value N, turning to the step. If the value is larger than the threshold value N, the step (12) is carried out.
(5) And constructing an SOC-V curve. The SOC-V curve comprises two parts, a charging curve and a discharging curve. If the data to be cleaned is a charging cycle, cleaning the data by using a charging curve; and if the data to be cleaned is the discharge cycle, cleaning the data by using the discharge curve. 2 identical SOC empty matrices SOCc (100 XN) and SOCdIsc (100 XN) were constructed, where SOCc (100 XN) was used to record charging data and SOCdIsc (100 XN) was used to record discharging data, with the 100 XN subscript indicating a matrix size of 100 rows and N columns.
(6) Judging the cycle number C, if C is an odd number, storing the data into a matrix SOCdense _ (100 multiplied by N); if C is an even number, the data is stored in SOCc _ (100 XN).
(7) The values in the SOC sequence are positive integers of 1-100, the values of the SOC sequence and the voltage matrix are time sequence data, and the values at the same sampling point are in one-to-one correspondence. Therefore, the SOC null matrix record data rule is as follows:
for the voltage matrix { V (1, 1) \8230;, V (i, j) \8230;, V (T, N) } and SOC sequence { S1, \8230;, si, \8230;, ST }, the location where the voltage V (i, j) is stored in the SOC empty matrix is the jth row and jth column of Si.
(8) The median is determined for the data at each position based on the socdic _ (100 × N) matrix and SOCc _ (100 × N) of the recorded data. And then drawing a curve for each row of data of the SOC matrix, wherein the abscissa is the number of the matrix columns, and the ordinate is the data value. N preliminary SOC-V charging curves and N preliminary SOC-V discharging curves can be obtained. Taking cell number 1 as an example, the initial SOC-V charging curve is shown in fig. 2. The data with the SOC between 1% and 11% is a missing value, and if the SOC data of the next cycle appears between 1% and 11%, the data cannot be cleaned.
(9) And for the N primary SOC-V charging curves and the N primary SOC-V discharging curves, fitting a quadratic function to obtain a function set { f (x) 1, \8230;, f (x) N, \8230;, f (x) 2N }. For the initial SOC-V charging curve of battery cell No. 1, the function y =0.000047611x2+0.0024x +3.5989 is obtained through fitting of a quadratic function, and the function is used to perform data fitting on SOC data between 1% and 11% to obtain the final SOC-V charging curve of battery cell No. 1, as shown in fig. 3.
(10) And (4) replacing the part of the SOC-V curve with the ordinate of 0 in the step (9) by a corresponding fitting function. And obtaining a complete SOC-V curve.
(11) Cycle number C = C +1. And (4) turning to the step (2).
(12) P0 is set to 0.5 and Q0 is set to 0.1. Firstly, the formula is calculated: h _ i = P _ (i-1) \8260: (Q _ (i-1) + P _ (i-1)) gives H _ i at the ith sampling point. Then the formula is calculated: v _ i = V _ (i-1) + H _ i (V (i, j) -V _ (i-1)) to obtain V _ i at the ith sampling point, which is the estimated voltage value at the ith sampling point. And finally calculating a formula: p _ i = (1-H _ i) × P _ (i-1), which results in P _ i at the ith sample point, i.e., the estimated error value at the ith sample point.
Thus, for cell number 12, the calculation starts from sample point 1 and ends at sample point 94. And obtaining a final estimated voltage matrix, and keeping the decimal point back 3 positions of the output data precision. The final estimated voltage matrix is shown in table 2.
TABLE 2 Final estimate Voltage matrix
(13) And outputting an estimated voltage matrix.
The embodiment estimates data which is closer to the actual working condition of the battery based on historical data of the power battery and current sensor measurement data, and reduces the influence of random disturbance and measurement error in the signal transmission process on the quality of the battery data.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (2)
1. A power battery cloud end data cleaning method based on information fusion is characterized by comprising the following steps:
step 1, inputting an SOC sequence and a voltage matrix of a battery;
step 2, judging whether the current cycle number exceeds a rated value or not, and determining to execute the step;
step 3, constructing a preliminary SOC-V curve;
step 4, constructing an SOC-V curve through quadratic function fitting;
and 5, combining the current measured value with the SOC-V curve by using an information fusion method to obtain a battery voltage estimated value.
2. The power battery cloud-end data cleaning method based on information fusion as claimed in claim 1, is characterized by comprising the following specific steps:
s1, inputting a current cycle battery SOC sequence (S) 1 ,…, S i ,…, S T And a voltage matrix V (1,1) ,…,V (i,j) ,…,V (T,N) H, and the current cycle number C; subscripts i and T represent sampling point serial numbers and total sampling length, and subscript N represents the initial cycle number of the battery serial number to be 1;
s2, judging whether the current cycle number C is larger than a threshold value N, and if so, executing S3; if the value is larger than the threshold value N, executing S7; wherein the threshold N is 10;
s3, constructing a preliminary SOC-V curve comprising a charging curve and a discharging curve;
s3.1, constructing 2 same SOC empty matrixes c _ (100×N) And SOC disc _ (100×N) In which SOC is c _ (100×N) For recording charging data, SOC disc _ (100×N) For recording discharge data, subscripts 100×N Representing a matrix size of 100 rows and N columns;
s3.2, judging the current cycle number C, if C is an odd number, storing the data into a matrix SOC disc _ (100×N) (ii) a If C is even number, data is stored in matrix SOC c _ (100×N) ；
S3.3. The values in the SOC sequence are all positive integers from 1 to 100, and the values of the SOC sequence and the voltage matrix are all time sequence data which are in one-to-one correspondence at the same sampling point; the SOC empty matrix recording data rule is as follows:
for the voltage matrix V (1,1) ,…,V (i,j) ,…,V (T,N) And SOC sequence S 1 ,…, S i ,…, S T V, voltage V (i,j) The position stored in the empty SOC matrix is S i Row, jth column;
s3.4. Matrix SOC based on record complete data disc _ (100×N) Sum matrix SOC c _ (100×N) Calculating the median of the data of each position; drawing a curve for each row of data of the SOC matrix, wherein the abscissa is the number of matrix columns, and the ordinate is a data value; obtaining N preliminary SOC-V charging curves and N preliminary SOC-V discharging curves；
S4, fitting a quadratic function for N primary SOC-V charging curves and N primary SOC-V discharging curves to obtain a function set { f (x) 1 ,…, f(x) N ,…, f(x) 2N }；
S5, replacing the part of the preliminary SOC-V charging curve in the S4, of which the vertical coordinate is 0, with a corresponding fitting function to obtain an SOC-V curve;
s6, turning to S1 for circulation, wherein the circulation frequency C = C +1;
s7, for SOC-V curve and actually measured voltage matrix { V (1,1) ,…,V (i,j) ,…,V (T,N) Information fusion is carried out according to the following rules:
s7.1, through a calculation formula: h _ i = P _ (i-1)/(Q _ (i-1) + P _ (i-1)) to obtain H _ i at the ith sampling point;
where P _ (i-1) represents the estimated error value for i-1 sample points, the initial value P 0 The value is 0.5 for adjustable parameters; q _ (i-1) represents the measurement error value of i-1 sampling points, the initial value Q 0 The value is 0.1 for adjustable parameters;
s7.2, calculating a formula: v _ i = V _ (i-1) + H _ i (V) (i,j) V _ (i-1)) to obtain V _ i at the ith sampling point as an estimated voltage value at the ith sampling point;
wherein V _ (i-1) is an estimated voltage value at the i-1 th sampling point, and the initial value V _0 is a voltage value of a corresponding SOC in the jth SOC-V curve;
s7.3, through a calculation formula: p _ i = (1-H _ i) × P _ (i-1), and P _ i at the ith sampling point is an estimated error value at the ith sampling point;
s7.4, repeating the step 7.1 to the step 7.3, starting to calculate from the 1 st sampling point to the T th sampling point, and obtaining an estimated voltage matrix;
and S8, outputting an estimated voltage matrix.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211417093.9A CN115494401B (en) | 2022-11-14 | 2022-11-14 | Power electricity Chi Yun end data cleaning method based on information fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211417093.9A CN115494401B (en) | 2022-11-14 | 2022-11-14 | Power electricity Chi Yun end data cleaning method based on information fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115494401A true CN115494401A (en) | 2022-12-20 |
| CN115494401B CN115494401B (en) | 2023-03-10 |
Family
ID=85115628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211417093.9A Active CN115494401B (en) | 2022-11-14 | 2022-11-14 | Power electricity Chi Yun end data cleaning method based on information fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115494401B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116879763A (en) * | 2023-09-07 | 2023-10-13 | 上海融和元储能源有限公司 | Battery fault early warning method based on Kalman filtering algorithm |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105842627A (en) * | 2016-02-01 | 2016-08-10 | 北京理工大学 | Method for estimating power battery capacity and charge state based on data model fusion |
| CN107422269A (en) * | 2017-06-16 | 2017-12-01 | 上海交通大学 | A kind of online SOC measuring methods of lithium battery |
| CN109946623A (en) * | 2019-03-27 | 2019-06-28 | 吉林大学 | A kind of SOC estimation online method of lithium battery |
| CN111856178A (en) * | 2020-03-31 | 2020-10-30 | 同济大学 | SOC partition estimation method based on electrochemical characteristics of lithium ion capacitor |
| CN111965547A (en) * | 2020-09-27 | 2020-11-20 | 哈尔滨工业大学（威海） | Battery system sensor fault diagnosis method based on parameter identification method |
| CN112557906A (en) * | 2020-12-14 | 2021-03-26 | 湖南大学 | SOC and capacity online joint estimation method in full life cycle of power battery |
| CN113960476A (en) * | 2021-09-23 | 2022-01-21 | 北京理工大学 | Power battery fault monitoring method and system based on information physical fusion technology |
| CN114418465A (en) * | 2022-03-29 | 2022-04-29 | 湖北工业大学 | Data-driven power battery use behavior quantitative evaluation method |
| CN115320387A (en) * | 2022-10-17 | 2022-11-11 | 北京理工大学 | Vehicle fuel cell membrane humidity prediction method and system based on multi-source information fusion |
-
2022
- 2022-11-14 CN CN202211417093.9A patent/CN115494401B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105842627A (en) * | 2016-02-01 | 2016-08-10 | 北京理工大学 | Method for estimating power battery capacity and charge state based on data model fusion |
| CN107422269A (en) * | 2017-06-16 | 2017-12-01 | 上海交通大学 | A kind of online SOC measuring methods of lithium battery |
| CN109946623A (en) * | 2019-03-27 | 2019-06-28 | 吉林大学 | A kind of SOC estimation online method of lithium battery |
| CN111856178A (en) * | 2020-03-31 | 2020-10-30 | 同济大学 | SOC partition estimation method based on electrochemical characteristics of lithium ion capacitor |
| CN111965547A (en) * | 2020-09-27 | 2020-11-20 | 哈尔滨工业大学（威海） | Battery system sensor fault diagnosis method based on parameter identification method |
| CN112557906A (en) * | 2020-12-14 | 2021-03-26 | 湖南大学 | SOC and capacity online joint estimation method in full life cycle of power battery |
| CN113960476A (en) * | 2021-09-23 | 2022-01-21 | 北京理工大学 | Power battery fault monitoring method and system based on information physical fusion technology |
| CN114418465A (en) * | 2022-03-29 | 2022-04-29 | 湖北工业大学 | Data-driven power battery use behavior quantitative evaluation method |
| CN115320387A (en) * | 2022-10-17 | 2022-11-11 | 北京理工大学 | Vehicle fuel cell membrane humidity prediction method and system based on multi-source information fusion |
Non-Patent Citations (2)
| Title |
|---|
| 张凡 等: "数据驱动的车载动力电池使用行为分析", 《汽车技术》 * |
| 李建林 等: "面向调频的锂电池储能建模及仿真分析", 《电力***保护与控制》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116879763A (en) * | 2023-09-07 | 2023-10-13 | 上海融和元储能源有限公司 | Battery fault early warning method based on Kalman filtering algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115494401B (en) | 2023-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107957562B (en) | Online prediction method for residual life of lithium ion battery | |
| JP5944291B2 (en) | Battery parameter estimation apparatus and method | |
| CN103907030B (en) | The estimation unit of battery charge rate and the method for estimation of battery charge rate | |
| JP5400732B2 (en) | Parameter estimation device | |
| CN111965559B (en) | On-line estimation method for SOH of lithium ion battery | |
| CN115494401B (en) | Power electricity Chi Yun end data cleaning method based on information fusion | |
| CN109975717B (en) | Online calculation method for internal resistance of power battery | |
| CN105115573A (en) | Correction method and device for flood flow forecasting | |
| Park et al. | Capacity estimation of Li-ion batteries using constant current charging voltage with multilayer perceptron | |
| CN113391211A (en) | Method for predicting remaining life of lithium battery under small sample condition | |
| CN113791351B (en) | Lithium battery life prediction method based on transfer learning and difference probability distribution | |
| Greenbank et al. | Piecewise-linear modelling with automated feature selection for Li-ion battery end-of-life prognosis | |
| CN111965548B (en) | Battery system sensor fault diagnosis method based on state estimation method | |
| CN111458643B (en) | Abnormal battery screening method and device, electronic equipment and readable storage medium | |
| CN117252112A (en) | Method for training driving data model and method for estimating remaining life of battery | |
| CN110113030B (en) | Particle filtering algorithm for subsampling | |
| CN113076681A (en) | Lithium battery core and surface temperature estimation method based on filtering | |
| CN114200328B (en) | Lithium ion battery SOC estimation method under non-Gaussian Levy noise | |
| CN114418465B (en) | Data-driven power battery use behavior quantitative evaluation method | |
| CN114325449B (en) | SOH prediction method for lithium ion battery | |
| WO2022029892A1 (en) | Storage battery internal state estimation device and storage battery internal state estimation method | |
| CN113376534A (en) | Phase plane diagnosis method for early fault of power battery and advanced early warning system | |
| Baba et al. | Simultaneous Estimation of the SOC and Parameters of Batteries for HEV/EV | |
| CN115840157B (en) | Lithium battery electrical performance index coordination analysis system based on EOF analysis | |
| CN116400231B (en) | Battery multi-fault detection method and device of energy storage system and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |