CN115267555A - Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement - Google Patents
Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement Download PDFInfo
- Publication number
- CN115267555A CN115267555A CN202210943364.8A CN202210943364A CN115267555A CN 115267555 A CN115267555 A CN 115267555A CN 202210943364 A CN202210943364 A CN 202210943364A CN 115267555 A CN115267555 A CN 115267555A
- Authority
- CN
- China
- Prior art keywords
- temperature
- data
- battery
- sequence
- charging
- 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.)
- Withdrawn
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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K11/00—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00
- G01K11/32—Measuring temperature based upon physical or chemical changes not covered by groups G01K3/00, G01K5/00, G01K7/00 or G01K9/00 using changes in transmittance, scattering or luminescence in optical fibres
-
- 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]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Secondary Cells (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention discloses an energy storage system battery SOH evaluation system based on battery multipoint temperature measurement, which aims at the problem of accurate and real-time attenuation monitoring of battery SOH in an energy storage system. The model is used for optimizing and maintaining the battery in the energy storage system in actual operation, and is favorable for timely replacing the battery with abnormal attenuation, so that the safety of the battery of the energy storage system is guaranteed.
Description
Technical Field
The invention belongs to the technical field of prediction and evaluation of SOH attenuation in the actual operation process of an energy storage system, and particularly relates to an energy storage system battery SOH evaluation system based on battery multipoint temperature measurement.
Background
Since the industrial revolution, new energy automobile industry is emerging with the continuous thinking and improvement of environmental destruction and resource exhaustion brought by people to modern industrial systems based on fossil energy. With the continuous progress of science and technology, the electric automobile industry and corresponding energy storage systems at home and abroad are in a vigorous development period, and technologies such as new energy automobiles and energy storage system power exchange stations thereof and the like have opened a fierce tide all over the world. From the global market, according to EVsales data, the global new energy automobile sales volume in 2014 is 31.54 thousands, the number has broken through 200 thousands by 2018, the new energy automobile sales volume is increased by 5.4 times, and the annual average growth rate is as high as 52.44%. In recent years, the new energy automobile industry in China is in a vigorous development period under the support and guidance of national policies and the great investment of scientific research funds. The research of the corresponding energy storage system field is also continuously and rapidly developed, and the research of the energy storage field relates to the optimization of an energy structure and the long-term development of sustainable energy utilization, so that the battery fault prediction and the timely detection, maintenance and optimization in the energy storage system become important research subjects.
The energy storage battery is used as a heart in the energy storage system and is one of three electricity supporting the energy storage system and the traditional electric automobile. Generally, the performance indexes of a storage battery in an energy storage system mainly include energy, power density, high-temperature performance, low-temperature performance, energy storage performance and the like. However, the problem of performance degradation of energy storage batteries during use has plagued further development in the electric vehicle industry. Generally, the capacity attenuation of the storage battery reaches below 80%, and the storage battery is not suitable for being used as an energy storage battery of an electric automobile. The service life of the energy storage battery before the energy storage battery is retired is determined by the attenuation speed of the capacity of the energy storage battery, so the evaluation and prediction of the residual life and the state of health (SOH) of the battery become important research subjects in the field of energy storage batteries.
In the existing research, many scholars establish a capacity attenuation model based on single factors or multiple factors, and mainly attribute the influence factors of the residual service life SOH of the energy storage battery to temperature, depth of discharge, discharge rate and the like. Among the influencing factors, compared with the discharge rate and the discharge depth, the temperature is the most main influencing parameter of the performance of the energy storage battery in the vehicle stage, the design capacity of the energy storage battery for the vehicle is often overlarge, the situation of discharge near the maximum discharge rate rarely occurs in the use process of the electric vehicle, and if the discharge rate of the battery does not exceed the maximum discharge rate allowed by the battery, the change of the discharge rate does not have additional influence on the capacity decline. When the residual electric quantity (SOC) of a new energy automobile user is in a range of 30% -60%, the energy storage battery can be charged, the use behavior avoids the phenomenon of over-charging and over-discharging of the energy storage battery, meanwhile, the correlation exists between the surface temperature of the battery and the SOH, and the guess is proved by using a support vector regression analysis method. There is also a study on the problem of correlation between the wavelength of the temperature sensitive sensor and the battery capacity by a linear kalman filter method. From the above studies, it can be found that the operating temperature has an inseparable relationship with the battery residual life (SOH). In addition, in the research of thermal runaway faults, electric heating characteristics, battery safety and residual life of the energy storage battery of the new energy automobile, aiming at the aspect of temperature conduction, the anisotropic heat conduction process of the lithium ion battery is subjected to experiment and modeling, the heat transfer characteristics of the battery are subjected to parameter estimation and experiment simulation, parameters for battery thermal characteristic modeling are determined according to a graphical method, various acupuncture thermal runaway experiments are designed aiming at the thermal runaway phenomenon of the ternary lithium ion energy storage battery, a single acupuncture model is built, and three-dimensional modeling is carried out by coupling three factors of thermal runaway secondary reaction, joule heat and heat transfer; experimental research is carried out aiming at thermal runaway and spreading characteristics, and laboratory data such as internal and external temperature difference of the battery, thermal runaway spreading time and the like are found when thermal runaway occurs.
The invention adopts the data based on the battery multipoint temperature measurement under the real working condition and adopts the machine learning method to evaluate and predict the battery SOH, the prediction index related to the SOH attenuation is excavated by acquiring the data of the external temperature of the battery of the real vehicle and exploring the data, and the actual SOH attenuation prediction problem is solved from the aspect of external temperature acquisition.
Disclosure of Invention
The present invention is directed to solving at least one of the problems of the prior art; therefore, the invention provides an energy storage system battery SOH evaluation system based on battery multipoint temperature measurement.
In order to achieve the above object, an embodiment according to a first aspect of the present invention provides an energy storage system battery SOH evaluation system based on battery multipoint temperature measurement, including a data acquisition terminal, a data preprocessing terminal, a feature engineering processing terminal, and a model building terminal;
the data preprocessing terminal internally comprises a data deduplication unit, a null value processing unit, an abnormal temperature data processing unit and a temperature data conversion unit;
the data acquisition terminal acquires temperature data of power station data, acquires data of 16 temperature measurement points corresponding to each battery, and uses data in an experiment as optical fiber temperature measurement data of the power conversion station;
the data preprocessing terminal is used for carrying out data duplication removal, null value processing and abnormal value processing on the temperature data acquired by the temperature measuring point and then carrying out temperature data conversion after the processing is finished;
the characteristic engineering processing terminal constructs difference sequence characteristics of the same temperature points, constructs difference sequence characteristics of different temperature points, constructs a temperature increment sequence, and mainly comprises two aspects of temperature characteristic sequence extraction and SOH sequence acquisition: in the aspect of temperature characteristic sequence extraction, selecting temperature change in the charging process of an actual vehicle, and recording external multipoint temperature data;
the model construction terminal selects an algorithm and constructs a model according to various constructed characteristics such as difference sequence characteristics of the same temperature point, and fits the extracted related characteristic sequence with a target SOH through an LSTM method of deep learning or an integrated learning XGboost algorithm.
Preferably, when the data acquisition terminal acquires data, an optical fiber temperature measuring point is arranged on a battery compartment for battery replacement: each battery compartment is provided with 16 optical fiber temperature measuring points which are respectively arranged at the center and two sides of the battery compartment, the central optical fiber temperature measuring point is 4, and the non-central optical fiber temperature measuring point is 12.
Preferably, the data deduplication unit performs deduplication processing on the acquired temperature data, and retains one of the duplicate data;
the null value processing unit is used for filling null values existing in the temperature data and filling the null values by using the temperature value at the previous moment or the temperature values of other temperature measuring points at the current moment;
the abnormal temperature numerical value processing unit deletes abnormal high-temperature or low-temperature data through a box diagram method and preset temperature data based on a large amount of data;
and the temperature data conversion unit converts the temperature collected on the surface of the optical fiber to the actual temperature of the battery.
Preferably, the specific conversion mode of the temperature data conversion unit is as follows:
selecting and processing batteries with different periods, and mounting a thermistor on the batteries as a temperature sensor;
manually controlling the environmental temperature in the battery compartment to be different temperature gradients which are [0-10], [10,20], [20,30], [30,40];
discharging these cells offline (uniformly to 5% soc), and then charging in the battery compartment (tests were carried out at each temperature gradient);
respectively counting temperature data of an optical fiber temperature measuring point and temperature data acquired by a temperature sensor in different SOC intervals and different charging time intervals;
establishing a relation table among the temperature before light, the SOC interval, the charging time interval, the environment temperature interval and the temperature collected by the temperature sensor;
establishing a regression model for the relation table, and obtaining the corresponding relation between each characteristic and the real temperature of the battery by using logistic regression;
the interval of charging time selection is: [0-10], [10,20], [60+ min ]
During the modeling process, the interval is directly converted into numerical values, starting with 1 in sequence, such as the ambient temperature [0-10]]，[10,20]，[20,30]，[30,40]Is converted to 1,2,3,4 and finally, the conversion formula is obtained: t = T m +0.173*T e -0.036 × s-0.021 × t +0.83 note: t represents the real temperature of the battery, T _ m represents the core temperature mean value of the optical fiber temperature measuring point, T _ e represents the environment temperature, s represents the SOC interval, and T represents the charging time interval;
in the real SOH result calculation process, an ampere-hour integral method is adopted for data calculation, and after temperature data with considerable quality is obtained, the following method is adopted to calculate the capacity of charging rushing in each time:
the formula:
the method comprises the steps that n represents the number of data of charging at this Time, I represents the current magnitude in the charging process, time represents data Time, and delta Q is calibrated, the delta Q is standardized by the actual capacity of different SOC intervals of various types of batteries obtained through experiments in advance, the relative capacity of the different SOC intervals is shown in a table 1, and the value of the relative capacity is the ratio of the capacity of each interval to one tenth of rated capacity;
the formula after normalization is:
where Δ Q is a charging capacity calculated by ampere-hour integration, n is a ten-digit part of SOC where charging is started (n =2 if charging is started at 25), and j is a one-digit part where charging is started (j =5 if charging is started at 25);
processing abnormal values of Q, namely deleting the abnormal data due to the fact that a small number of data exist and the integral deviation is large although the charging data of the power changing station is pure;
carrying out moving average on Q to obtain a smooth Q sequence;
the SOH sequence is obtained by converting SOH using each value of the Q sequence divided by the starting Q.
Preferably, the multipoint temperature data comprises the following three temperature characteristic sequences:
temperature extraction is carried out on a plurality of temperature measuring points of each monomer in the charging process, and mainly a difference sequence, a rising speed sequence, an entropy sequence and the like of the temperature are extracted;
in the charging process, extracting a difference sequence, a rising speed sequence, an entropy sequence and the like of average temperatures among different monomers;
in the charging process, temperature increment sequences of unit SOC are extracted at all temperature measuring points of the monomers and the average temperature of different monomer individuals respectively.
Preferably, the step of fitting the extracted relevant feature sequence with the target SOH by the model construction terminal is as follows:
selecting battery surface temperature gradient sequence data in each charging process as training data, wherein the temperature gradient is data after the influence of the environmental temperature is removed;
selecting a charging SOC interval: the charging data amount of each battery is different, sequence data with the same length is needed to be used, and temperature gradient data of a charging process with the SOC between 35 and 100 is selected each time; 35 is selected as the initial SOC value and 35 is selected as the initial SOC value.
Calculating the temperature gradient: in the 35SOC-100SOC interval, temperature data are recorded every other SOC from the initial temperature, the final temperature gradient data are 66, 16 temperature measuring points are provided in total, and the final temperature gradient data of one charging are 66 x 16 matrix data;
converting real temperature gradient data: adding the current environment temperature, the charging SOC and the charging time by using the average value of the temperature data of the 8 th and 9 th temperature monitoring points; carrying out temperature conversion to obtain real temperature gradient data;
and constructing a battery temperature gradient sequence through the sequence data of the battery temperature, feeding back the attenuation state of the target SOH from the gradient information and the temperature related characteristic sequence, mining the corresponding relation between the temperature and the SOH, and predicting the attenuation condition of the SOH through the temperature sequence characteristic in each charging process after algorithm training and verification are finished.
Preferably, the data set is temperature gradient data of 66 × 16 per charge and corresponding calculated SOH data of the battery; and constructing a model, wherein the model is temperature gradient data, the gradient is based on the change of a time sequence and is a multi-dimensional characteristic, so that an LSTM model is selected as a training model.
Compared with the prior art, the invention has the beneficial effects that: the method comprises the steps of collecting battery pack data actually running in an energy storage system in two years, measuring the temperature from the outside of a battery, obtaining temperature information to generate a temperature gradient curve, and estimating the temperature inside the battery through the generated temperature gradient curve to judge the health state of the battery. The SOH attenuation state of the battery is predicted by researching the temperature gradient change of the battery and an LSTM neural network method, and an SOH attenuation prediction model is constructed. The model can be used for optimizing and maintaining the battery in the energy storage system in actual operation, and is favorable for timely replacing the battery with abnormal attenuation, thereby providing guarantee for the battery safety of the energy storage system. Compared with the laboratory environment, the invention is more suitable for the condition of the energy storage system which actually runs, and has higher theoretical significance and application value.
Drawings
FIG. 1 is a schematic block diagram of the principles of the present invention;
FIG. 2 is a general flow diagram of the inventive arrangement;
FIG. 3 is a network architecture diagram of the LSTM of the present invention;
FIG. 4 is a source visualization of raw data according to the present invention;
FIG. 5 is a comparison graph of the difference between the predicted value and the actual value according to the present invention;
fig. 6 is a graph showing the significance of the characteristics of the 16 temperature measuring points of the present invention.
Detailed Description
The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments, and it should be understood 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 derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present application provides an energy storage system battery SOH evaluation system based on battery multipoint temperature measurement, including a data acquisition terminal, a data preprocessing terminal, a feature engineering processing terminal, and a model building terminal;
the output end of the data acquisition terminal is electrically connected with the input end of the data preprocessing terminal, the output end of the data preprocessing terminal is electrically connected with the input end of the characteristic engineering processing terminal, and the output end of the characteristic engineering processing terminal is electrically connected with the input end of the model building terminal;
the data preprocessing terminal internally comprises a data deduplication unit, a null value processing unit, an abnormal temperature data processing unit and a temperature data conversion unit;
the data acquisition terminal acquires temperature data of power station data, data acquisition is carried out on 16 temperature measurement points corresponding to each battery, and data used in an experiment are optical fiber temperature measurement data of the power conversion station;
when data acquisition is carried out, optical fiber temperature measuring points are arranged on a battery compartment for replacing the battery: each battery compartment is provided with 16 optical fiber temperature measuring points which are respectively arranged at the center and two sides of the battery compartment, the central optical fiber temperature point is 4, and the non-central optical fiber temperature point is 12 (boundaries with different degrees), so that the battery can obtain the multipoint temperature outside the battery in the whole charging process in the charging process of the battery compartment, and the temperature precision is higher (the temperature is measured to two decimal points);
in the stage, because the directly acquired temperature data cannot be directly used for training the model due to the problems of repetition, null value, abnormality and the like, the data preprocessing terminal performs the following data preliminary processing process on the actual temperature data: carrying out data duplicate removal, null value processing and abnormal value processing on temperature data acquired by a temperature measuring point, and carrying out temperature data conversion;
in the data deduplication unit, data uploading has certain instability, so that a small amount of duplicate data may be caused, only one of the duplicate data is reserved, deduplication processing is performed on the acquired temperature data, and one of the duplicate data is reserved;
and the null value processing unit is used for processing the data with the effective temperature value in half of all the data. But some nulls may be generated due to data upload or fiber-optic problems. For these null values, we prefer to fill in, i.e. to use the temperature value at the previous time or the temperature values at other temperature measurement points at this time. If the temperature loss rate is large, the charging data is generally discarded;
the abnormal temperature numerical value processing unit: based on the temperature condition of a large amount of data under the condition of a box plot method and a business background, abnormal high-temperature or low-temperature data is deleted.
The temperature data conversion unit: because the temperature data acquired by the optical fiber are all temperature data of different surfaces of the battery, the temperature data are not real battery temperature data, and the difference between the surface temperature and the actual temperature is different at different stages of charging. In contrast, the actual experiment performed offline obtains the conversion process from the optical fiber surface collection temperature to the actual battery temperature, and the specific conversion process is as follows:
as shown in fig. 2, the experiment was performed by selecting batteries using different cycles, on which thermistors were mounted as temperature sensors;
artificially controlling the environmental temperature in the battery compartment to be different temperature gradients which are 0-10, 20,30 and 30,40 at present, and carrying out the following experiments on the basis;
discharging these cells offline (uniformly to 5% soc), and then charging in the battery compartment (tests were carried out at each temperature gradient);
respectively counting temperature data of an optical fiber temperature measuring point and temperature data acquired by a temperature sensor in different SOC intervals and different charging time intervals;
establishing a relation table among the pre-light temperature, the SOC interval, the charging time interval, the environment temperature interval and the temperature collected by the temperature sensor;
establishing a regression model for the relation table, and obtaining a corresponding relation between each characteristic and the real temperature of the battery by using logistic regression; note: the SOC selected interval is: [0,10], [10,20. ]., [90,100];
the interval selected for the charging time is: [0-10], [10, 20... ] and [60+ minute ]
In the modeling process, the interval is directly converted into numerical values, starting from 1 in sequence, such as the ambient temperature [0-10]]，[10,20]，[20,30]，[30,40]Is converted to 1,2,3,4 and finally, the conversion formula is obtained: t = T m +0.173*T e -0.036 s-0.021 t +0.83 note: t represents the real temperature of the battery, T _ m represents the core temperature mean value (the 8 th and the 9 th temperature mean values) of the optical fiber temperature measuring point, T _ e represents the ambient temperature, s represents an SOC interval, and T represents a charging time interval;
in the real SOH result calculation process, an ampere-hour integration method is also needed for data calculation, and after temperature data with considerable quality is obtained, the following method is adopted to calculate the capacity of each charging rush-in:
the formula:
the method comprises the steps of calibrating delta Q, wherein n represents the number of data of charging at this Time, I represents the current magnitude in the charging process, time represents the data Time, the delta Q is standardized by the actual capacity of different SOC intervals of each type of battery obtained through experiments in advance, the relative capacity of the different SOC intervals is shown in a table 1, and the value of the relative capacity is the ratio of the capacity of each interval to one tenth of rated capacity.
TABLE 1 relative capacities in different SOC intervals
The formula after normalization is:
where Δ Q is a charging capacity calculated by ampere-hour integration, n is a ten-digit part of SOC where charging is started (n =2 if charging is started at 25), and j is a one-digit part where charging is started (j =5 if charging is started at 25);
processing abnormal values of Q, namely deleting the abnormal data due to the fact that a small number of data exist and the integral deviation is large although the charging data of the power changing station is pure;
performing sliding average on Q to obtain a smoother Q sequence;
the SOH sequence is obtained by converting SOH and dividing each value of the Q sequence by the starting Q.
The characteristic engineering processing terminal constructs difference sequence characteristics of the same temperature points, constructs difference sequence characteristics of different temperature points, constructs a temperature increment sequence, and mainly comprises two aspects of temperature characteristic sequence extraction and SOH sequence acquisition: in the aspect of temperature characteristic sequence extraction, the method comprises the following steps of selecting temperature change in the charging process of an actual vehicle, recording external multipoint temperature data, and mainly comprising the following three temperature characteristic sequences:
1. and (3) extracting the temperature of a plurality of temperature measuring points of each monomer in the charging process, wherein the temperature is mainly extracted from a temperature difference sequence, a rising speed sequence, an entropy sequence and the like.
2. In the charging process, a difference sequence, a rising speed sequence, an entropy sequence and the like of average temperatures among different monomers are extracted.
3. In the charging process, temperature increment sequences of unit SOC are extracted at all temperature measuring points of the monomers and the average temperature of different monomers respectively.
In the aspect of obtaining the battery health SOH sequence, firstly, selecting a traditional ampere-hour integration method for measuring the capacity of each charge (selecting a fixed voltage interval); removing abnormal values according to the copper beam charged each time; finally, processing the capacity data by adopting a moving average method to obtain an SOH sequence; taking a part of the SOH sequence as a model training set, and fitting the temperature-related characteristic sequence with a target SOH sequence; the other part of the SOH sequence is used as a verification set of the fitting method, so that the robustness, the accuracy and the like of the model are verified;
after the characteristics are extracted, multipoint temperature measurement is carried out on the surface of each battery cell in a data acquisition stage; and the constructed data is evaluated and corrected based on the time unit: screening data with errors, and subsequently correcting the error data, wherein the steps comprise storing the average value of the missing value and the abnormal value and performing subsequent calculation on the average value; for the value with wrong time period, the time period is definitely obtained, and the data is adjusted and re-run; for values where the specification is wrong to calculate, the caliber is explicitly adjusted and the data sample quality is re-improved.
And the model construction terminal selects an algorithm and constructs a model according to various constructed characteristics such as the difference sequence characteristics of the same temperature point and the like. In the aspect of selection of a learning algorithm, the method fits the extracted related feature sequence with a target SOH through an LSTM method of deep learning or an integrated learning XGboost algorithm.
The method for selecting the long-short term memory neural network (LSTM) has excellent prediction accuracy on one hand, and can solve the problem of RNN gradient disappearance on the other hand, an innovative internal gate structure can be used, useful historical information is reserved, useless historical information is deleted, and accuracy is greatly improved. Meanwhile, the method mainly adopts a mechanism of controlling the gate and consists of a memory cell, an input gate, an output gate and a forgetting gate.
The method for predicting the SOH of the battery by adopting the LSTM neural network comprises the steps of performing serialization processing on vehicle-mounted battery charging data of a new energy lithium battery automobile which is actually acquired in two years, extracting input characteristics of the LSTM, using partial-score data in the battery data as a verification set, using indexes (such as maximum available capacity and the like) which embody SOH attenuation characteristics as output characteristics of the LSTM, and estimating the existing SOH condition of the battery by using a neural network method after training verification; therefore, a battery SOH attenuation prediction model which accords with actual vehicle conditions is constructed, the method can be suitable for estimating the SOH condition of the battery of the new energy vehicle in an actual scene, the key problem that the estimation method is difficult to adapt to complex working environments in a laboratory scene can be solved, and the method has important theoretical significance and application value.
And selecting the battery surface temperature gradient sequence data in each charging process as training data, wherein the temperature gradient is data after the influence of the environmental temperature is removed.
Selecting a charging SOC interval: the charging data amount of each battery is different, but sequence data with the same length is required to be used, so that the temperature gradient data of the charging process with the SOC between 35 and 100 is selected each time; 35 was chosen herein as the initial SOC value because, through statistics of the distribution of the initial SOC of each battery charge by the charging station, it was found that the percentage of batteries charged before 35SOC reached 95% or more, while the percentage charged before 30SOC reached 87%; therefore, selecting 35 as the initial SOC value can ensure both coverage of most charging conditions and a sufficiently large charging SOC interval. The end SOC =100 is because the battery of the charging station is charged to SOC =100 every time.
Calculating the temperature gradient: in the 35SOC-100SOC interval, temperature data is recorded every other SOC from the initial temperature, so that the final temperature gradient data is 66. There are 16 temperature measuring points, so that the temperature gradient data of the final charge is 66 × 16 matrix data
Converting real temperature gradient data: the average of the temperature data at the 8 th and 9 th temperature monitoring points is used, plus the current ambient temperature, charge SOC, charge time. Temperature transformation was performed by equation 1 to obtain true temperature gradient data (66 x 1 sequence data)
Data set: the data set is temperature gradient data for each charge 66 x 16 and corresponding calculated battery SOH data.
Constructing a model: because the data is temperature gradient data, the gradient is based on the change of time series and is a multi-dimensional characteristic, an LSTM model is selected as a training model.
And constructing a battery temperature gradient sequence through the sequence data of the battery temperature, feeding back the attenuation state of the target SOH from the gradient information and the temperature correlation characteristic sequence, and mining the corresponding relation between the temperature and the SOH. After the algorithm training and verification are finished, the SOH attenuation condition can be predicted through the temperature sequence characteristics in each charging process. The general scheme flow of the modeling of the patent is shown in fig. 3.
And model testing and online steps, namely performing testing on a test set by using the optimal model, counting test effects, and performing model online according to the test effects. Experiments made by the patent totally calculate 82216 charging data of 5600 batteries, respectively according to a training set: and (4) verification set: test set =7:2:1, the algorithm model obtains very good effects on a training set and a verification set, the effects on the test set are slightly lower than those of the training set and the verification set, but the difference is not large, and the effect basis is good. But even if the effect on the test set is good, the final effect verification needs to be carried out by using a real vehicle. The charging data of 5600 batteries and the corresponding optical fiber acquisition temperature data, wherein the battery type is mainly lithium iron phosphate. The temperature data collected by the fiber is shown in fig. 4.
The real data of 6 real vehicles are truly verified, and the statistical table is shown in table 2. For 6 vehicles, no. 1,2.. 5,6, detailed statistics were made on the charging time span and the number of times of charging of the battery. According to the statistical real data of 6 real vehicles, the offline actual capacity and SOH of the 6 vehicles are predicted, the absolute error of the rated capacity is analyzed, and the data verification result is shown in the table 3, so that the reliability of the model is further verified. To verify the validity of the SOH prediction model established herein, we selected 3 new vehicles for real vehicle testing, and the data distribution is shown in table 4. By simulating the charge and discharge behaviors of actual users, temperature data of the battery is recorded, and SOH prediction of the battery is carried out by using the model. And finally, comparing the actually measured SOC value of the battery with the SOH value predicted by the algorithm to obtain a relative error. The results of the experiment are shown in table 5, note: the rated capacity of the test cell in table 5 was 70.5AH. The error represented on the three latest battery data of the algorithm model is an average error larger than the effect on the test set, and after all, the error is a new battery which may have various dimensional differences with the previous battery, so that the effect predicted by the algorithm can be truly represented by the effect tested on the latest battery. From the prediction effect of the three batteries at present, the errors are all less than 3%, and the errors of two batteries are all less than 1%, which is enough to say that the current algorithm model has better generalization capability and higher prediction accuracy.
Table 2 basic conditions table of verification data
Table 3 data verification result distribution table
Table 4 data set distribution table
TABLE 5 real vehicle charging and discharging data table
As the actual training data volume is large and the average test index has been written in the above text, the comparison data of part of the training set and the test set is shown, as shown in fig. 5, the abscissa represents 20 sets of experimental data randomly selected in the test set, and the ordinate represents the comparison effect between the actual value and the predicted value of the SOH;
as a plurality of external temperature measuring points are distributed on the surface of the battery, the temperature measuring points at the center of the battery are numbered as 8 and 9, the number of the temperature measuring points extending from the center point to two sides to the marginal area is 1-16, the ordinate in the figure 6 represents the number of the temperature measuring points, the ordinate represents the importance sequence of the characteristics of each temperature measuring point, the numerical value of the importance degree of the abscissa calculates the information gain value through information entropy modeling, and the quantitative value of the importance degree of the characteristics is extracted. The closer the temperature measuring points at the center of the single battery are, the closer the measured values are to the actual temperature at the center of the battery, so that the overall characteristic ranking of the temperature measuring points is in a step-like distribution. Based on the data from the real vehicle testing (table 3), we conclude that: (1) The difference between the actual battery capacity (SOH) of the three vehicles and the SOH value predicted by the algorithm is small, and the relative errors are within 3 percent (the prediction error of the vehicle No. 01 is even less than 0.3 percent). (2) As the battery capacity (SOH) decreases, the prediction error of the algorithm increases gradually-from 0.27% to 2.5%;
in conclusion, the actual vehicle charging and discharging test proves that the model can achieve high prediction accuracy, the error does not exceed 3%, and the effectiveness of the SOH prediction model based on the multipoint temperature established in the text is fully verified. In addition, we also find that the prediction accuracy of the model gradually decreases as the SOH of the battery decreases. But due to the limitation of experimental conditions, we do not perform algorithm verification at low SOH values (which is also the key point of our follow-up work);
part of data in the formula is obtained by removing dimension and taking the value to calculate, and the formula is obtained by simulating a large amount of collected data through software and is closest to a real situation; the preset parameters and the preset threshold values in the formula are set by those skilled in the art according to actual conditions or obtained through simulation of a large amount of data.
Although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted for elements thereof without departing from the spirit and scope of the present invention.
Claims (7)
1. The system for evaluating the SOH of the battery of the energy storage system based on the multipoint temperature measurement of the battery is characterized by comprising a data acquisition terminal, a data preprocessing terminal, a characteristic engineering processing terminal and a model construction terminal;
the data preprocessing terminal internally comprises a data deduplication unit, a null value processing unit, an abnormal temperature data processing unit and a temperature data conversion unit;
the data acquisition terminal acquires temperature data of power station data, acquires data of 16 temperature measurement points corresponding to each battery, and uses data in an experiment as optical fiber temperature measurement data of the power conversion station;
the data preprocessing terminal is used for carrying out data duplication removal, null value processing and abnormal value processing on the temperature data acquired by the temperature measuring point and then carrying out temperature data conversion after the processing is finished;
the characteristic engineering processing terminal constructs difference sequence characteristics of the same temperature points, constructs difference sequence characteristics of different temperature points, constructs a temperature increment sequence, and mainly comprises two aspects of temperature characteristic sequence extraction and SOH sequence acquisition: in the aspect of temperature characteristic sequence extraction, selecting temperature change in the charging process of an actual vehicle, and recording external multipoint temperature data;
the model construction terminal selects an algorithm and constructs a model according to various constructed characteristics such as difference sequence characteristics of the same temperature point, and fits the extracted related characteristic sequence with a target SOH through an LSTM method of deep learning or an integrated learning XGboost algorithm.
2. The system for estimating SOH of the battery in the energy storage system based on the multipoint temperature measurement of the battery as claimed in claim 1, wherein when the data acquisition terminal acquires data, the data acquisition terminal arranges optical fiber temperature measurement points on a battery chamber for battery replacement by: each battery compartment is provided with 16 optical fiber temperature measuring points which are respectively arranged at the center and two sides of the battery compartment, the central optical fiber temperature measuring points are 4, and the non-central optical fiber temperature measuring points are 12.
3. The system for estimating battery SOH of an energy storage system based on battery multipoint temperature measurement according to claim 1, wherein the data deduplication unit performs deduplication processing on the acquired temperature data, and retains one of duplicate data;
the null value processing unit is used for filling null values existing in the temperature data and filling the null values by using the temperature value at the previous moment or the temperature values of other temperature measuring points at the current moment;
the abnormal temperature numerical value processing unit deletes abnormal high-temperature or low-temperature data through a box diagram method and preset temperature data based on a large amount of data;
and the temperature data conversion unit converts the temperature collected on the surface of the optical fiber to the actual temperature of the battery.
4. The system for estimating SOH of the energy storage system battery based on the multipoint temperature measurement of the battery as claimed in claim 3, wherein the specific conversion mode of the temperature data conversion unit is as follows:
selecting and processing batteries with different periods, and mounting a thermistor on the batteries as a temperature sensor;
manually controlling the environmental temperature in the battery compartment to be different temperature gradients which are [0-10], [10,20], [20,30], [30,40];
these cells were discharged offline (uniformly discharged to 5% soc) and then charged in the battery compartment (tests were performed at each temperature gradient);
respectively counting temperature data of an optical fiber temperature measuring point and temperature data acquired by a temperature sensor in different SOC intervals and different charging time intervals;
establishing a relation table among the temperature before light, the SOC interval, the charging time interval, the environment temperature interval and the temperature collected by the temperature sensor;
establishing a regression model for the relation table, and obtaining a corresponding relation between each characteristic and the real temperature of the battery by using logistic regression;
the interval selected for the charging time is: [0-10], [10, 20... ] and [60+ minute ]
In the modeling process, the interval is directly converted into numerical values, starting from 1 in sequence, such as the ambient temperature [0-10]]，[10,20]，[20,30]，[30,40]Is turned toTo 1,2,3,4, the resulting conversion equation is: t = T m +0.173*T e -0.036 s-0.021 t +0.83 note: t represents the real temperature of the battery, T _ m represents the mean value of the core temperature of the optical fiber temperature measuring point, T _ e represents the ambient temperature, s represents an SOC interval, and T represents a charging time interval;
in the real SOH result calculation process, an ampere-hour integral method is adopted for data calculation, and after temperature data with considerable quality is obtained, the following method is adopted to calculate the capacity of charging rushing in each time:
the formula:
the method comprises the steps that n represents the number of data of charging at this Time, I represents the current magnitude in the charging process, time represents data Time, and delta Q is calibrated, the delta Q is standardized by the actual capacity of different SOC intervals of various types of batteries obtained through experiments in advance, the relative capacity of the different SOC intervals is shown in a table 1, and the value of the relative capacity is the ratio of the capacity of each interval to one tenth of rated capacity;
the formula after normalization is:
where Δ Q is a charging capacity calculated by ampere-hour integration, n is a ten-digit part of SOC where charging is started (n =2 if charging is started at 25), and j is a one-digit part where charging is started (j =5 if charging is started at 25);
in the abnormal value processing of Q, although charging data of the power swapping station is relatively pure, a few data exist and the overall deviation is large, so that the abnormal data are deleted;
carrying out moving average on Q to obtain a smooth Q sequence;
the SOH sequence is obtained by converting SOH using each value of the Q sequence divided by the starting Q.
5. The energy storage system battery SOH estimation system based on battery multipoint temperature measurement according to claim 4, characterized in that the multipoint temperature data comprises a sequence of three temperature signatures including:
temperature extraction is carried out on a plurality of temperature measuring points of each monomer in the charging process, and mainly a difference sequence, a rising speed sequence, an entropy sequence and the like of the temperature are extracted;
in the charging process, extracting a difference sequence, a rising speed sequence, an entropy sequence and the like of average temperatures among different monomers;
in the charging process, temperature increment sequences of unit SOC are extracted at all temperature measuring points of the monomers and the average temperature of different monomers respectively.
6. The system for estimating battery SOH of the energy storage system based on battery multipoint temperature measurement as claimed in claim 5, wherein the concrete steps of the model building terminal fitting the extracted relevant feature sequence and the target SOH are as follows:
selecting battery surface temperature gradient sequence data in each charging process as training data, wherein the temperature gradient is data after the influence of the environmental temperature is removed;
selecting a charging SOC interval: the charging data amount of the battery at each time is different, sequence data with the same length is required to be used, and temperature gradient data of the charging process with the SOC between 35 and 100 is selected at each time; selecting 35 as an initial SOC value, and selecting 35 as an initial SOC value;
calculating the temperature gradient: in the 35SOC-100SOC interval, temperature data are recorded every other SOC from the initial temperature, the final temperature gradient data are 66, 16 temperature measuring points are provided in total, and the final temperature gradient data of one charging are 66 x 16 matrix data;
converting real temperature gradient data: adding the current ambient temperature, the charging SOC and the charging time to the average value of the temperature data of the 8 th and 9 th temperature monitoring points; carrying out temperature conversion to obtain real temperature gradient data;
and constructing a battery temperature gradient sequence through the sequence data of the battery temperature, feeding back the attenuation state of the target SOH from the gradient information and the temperature related characteristic sequence, mining the corresponding relation between the temperature and the SOH, and predicting the attenuation condition of the SOH through the temperature sequence characteristic in each charging process after algorithm training and verification are finished.
7. The system for estimating battery SOH of an energy storage system based on multiple point temperature measurements of batteries according to claim 6, wherein said data set is 66 x 16 per charge temperature gradient data and corresponding calculated battery SOH data; and constructing a model, wherein the model is temperature gradient data, the gradient is based on the change of a time sequence and is a multi-dimensional characteristic, so that an LSTM model is selected as a training model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210943364.8A CN115267555A (en) | 2022-08-08 | 2022-08-08 | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210943364.8A CN115267555A (en) | 2022-08-08 | 2022-08-08 | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115267555A true CN115267555A (en) | 2022-11-01 |
Family
ID=83749521
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210943364.8A Withdrawn CN115267555A (en) | 2022-08-08 | 2022-08-08 | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115267555A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115993683A (en) * | 2022-11-17 | 2023-04-21 | 华中科技大学 | Communication angle measurement integrated space laser communication device |
| CN116683061A (en) * | 2023-08-03 | 2023-09-01 | 太原科技大学 | Power battery thermal runaway prediction and suppression integrated system, method and storage medium |
| CN116826219A (en) * | 2023-08-28 | 2023-09-29 | 宁德时代新能源科技股份有限公司 | Battery, power utilization device and method for assembling battery |
| CN117057236A (en) * | 2023-08-14 | 2023-11-14 | 上海频准激光科技有限公司 | Optical fiber temperature measuring point evaluation system based on laser |
| CN117289168A (en) * | 2023-09-19 | 2023-12-26 | 北京翼新数智科技有限公司 | Product quality monitoring system based on sensor data and intelligent algorithm |
-
2022
- 2022-08-08 CN CN202210943364.8A patent/CN115267555A/en not_active Withdrawn
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115993683A (en) * | 2022-11-17 | 2023-04-21 | 华中科技大学 | Communication angle measurement integrated space laser communication device |
| CN116683061A (en) * | 2023-08-03 | 2023-09-01 | 太原科技大学 | Power battery thermal runaway prediction and suppression integrated system, method and storage medium |
| CN116683061B (en) * | 2023-08-03 | 2023-09-29 | 太原科技大学 | Power battery thermal runaway prediction and suppression integrated system, method and storage medium |
| CN117057236A (en) * | 2023-08-14 | 2023-11-14 | 上海频准激光科技有限公司 | Optical fiber temperature measuring point evaluation system based on laser |
| CN117057236B (en) * | 2023-08-14 | 2024-04-12 | 上海频准激光科技有限公司 | Optical fiber temperature measuring point evaluation system based on laser |
| CN116826219A (en) * | 2023-08-28 | 2023-09-29 | 宁德时代新能源科技股份有限公司 | Battery, power utilization device and method for assembling battery |
| CN117289168A (en) * | 2023-09-19 | 2023-12-26 | 北京翼新数智科技有限公司 | Product quality monitoring system based on sensor data and intelligent algorithm |
| CN117289168B (en) * | 2023-09-19 | 2024-05-28 | 北京翼新数智科技有限公司 | Product quality monitoring system based on sensor data and intelligent algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115267555A (en) | Battery SOH (State of health) evaluation system of energy storage system based on battery multipoint temperature measurement | |
| CN110031770B (en) | Method for rapidly obtaining capacity of all single batteries in battery pack | |
| CN106291372B (en) | Novel method for predicting residual life of lithium ion power battery | |
| CN112034356B (en) | GP-UKF-based online SOH estimation method for electric vehicle power battery | |
| Zheng et al. | Fault identification and quantitative diagnosis method for series-connected lithium-ion battery packs based on capacity estimation | |
| CN111239630A (en) | Energy storage battery service life prediction method and management system | |
| CN112986831A (en) | Lithium ion battery life prediction method based on correlation coefficient particle filtering | |
| CN112327167A (en) | Battery capacity diving risk assessment method and system | |
| CN116401585B (en) | Energy storage battery failure risk assessment method based on big data | |
| CN112433170B (en) | Method for identifying parameter difference of single batteries of series battery pack | |
| CN111064253A (en) | Battery health degree rapid evaluation method based on average discrete Frechet distance | |
| Zheng et al. | Real-time estimation of battery state of charge with metabolic grey model and LabVIEW platform | |
| CN112485693A (en) | Battery health state rapid evaluation method based on temperature probability density function | |
| CN112765794A (en) | Reliability analysis method of lithium iron phosphate battery based on accelerated life test | |
| CN115291131A (en) | Method and system for predicting cycle life and service temperature of lithium ion battery | |
| CN114646888A (en) | Assessment method and system for capacity attenuation of power battery | |
| CN114545275A (en) | Indirect prediction method for remaining service life of lithium ion battery | |
| CN117007975A (en) | Method for performing reinforcement learning on battery capacity attenuation assessment by collecting multi-point temperatures of battery cells of energy storage battery | |
| CN106599333A (en) | Method for estimating SOH of power supply | |
| CN115986229A (en) | Method for judging health degree of battery based on charging segment data | |
| CN116699445A (en) | Capacity prediction method and system for battery energy storage system | |
| CN112649735A (en) | Method for rapidly obtaining total capacity of battery pack | |
| CN116679208A (en) | Lithium battery residual life estimation method | |
| CN115800433A (en) | Battery pack consistency evaluation and grade evaluation method and device | |
| Wang et al. | Thermal fault diagnosis method of Lithium battery based on LSTM and battery physical model |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20221101 |
|
| WW01 | Invention patent application withdrawn after publication |