Flow battery performance analysis method based on current density distribution measurement

Abstract

The invention discloses a flow battery performance analysis method based on current density distribution measurement, and relates to the technical field of batteries. Compared with a previous battery performance measurement technology, the problem that the data acquisition dimension is relatively narrow is solved; the sampling frequency lacks dynamic adaptability; different operation conditions and charge states are not considered; the problem of lack of comprehensive performance evaluation and prediction under multi-parameter fusion is solved; in combination with real-time environment parameters, block data and the like, data acquisition dimensions are rich and comprehensive; the key performance indexes of the battery are analyzed in real time, the sampling frequency is further adjusted according to the dynamic change conditions of the indexes in the charging and discharging process, the optimal operation conditions and the like of the flow battery are analyzed, and the subsequent optimization capability on the overall performance of the battery is improved; various technical means are fused, the performance of the battery is comprehensively evaluated from different levels, and the optimal operation mode and potential problems of the battery under different operation conditions can be mined based on the analysis and prediction result.

Description

Technical Field

[0001] The present invention relates to the technical field of batteries, and in particular to a method for analyzing performance of a liquid flow battery based on current density distribution measurement. Background Art

[0002] As an advanced electrochemical energy storage technology, flow batteries have great application potential in renewable energy storage (such as solar energy, wind energy) and grid peak regulation. Its unique structure and working principle enable it to achieve large-capacity and long-term energy storage. However, flow batteries involve complex electrochemical processes, including the diffusion of ions in the electrolyte, charge transfer on the electrode surface, and the flow of electrolyte in the battery system. These processes affect each other and change dynamically during the battery's charge and discharge process, making accurate evaluation and tracking of flow battery performance a challenging task.

[0003] Traditional battery performance testing methods often focus on measuring macroscopic performance indicators, such as simply recording charge and discharge time, voltage, and current to calculate energy and power related parameters. These methods are of limited help in understanding the microscopic electrochemical processes inside flow batteries. Moreover, in actual operation, the performance of flow batteries will be dynamically affected by a variety of factors (such as temperature changes, electrolyte concentration fluctuations, current density changes, etc.), and traditional measurement indicators cannot capture the impact of these changes on battery performance in real time.

[0004] The existing Chinese patent with publication number CN116148681A discloses a method for predicting the performance of an iron-chromium liquid flow battery, including: obtaining the target iron-chromium liquid flow battery energy storage stack charge and discharge experimental data, wherein the charge and discharge experimental data are obtained under the same system complete charge and discharge cycle state; sorting the charge and discharge experimental data and converting the charge and discharge experimental data into independent variables; according to the liquid flow battery mechanism, using the independent variables as characteristic input variables, constructing a multivariate linear regression model, and predicting the battery performance indicators through the multivariate linear regression model. The invention obtains the target iron-chromium liquid flow battery energy storage stack charge and discharge experimental data, wherein the charge and discharge experimental data are obtained under the same system complete charge and discharge cycle state; sorting the charge and discharge experimental data and converting the charge and discharge experimental data into independent variables; using the independent variables as characteristic input variables, constructing a multivariate linear regression model, and predicting the battery performance indicators through the multivariate linear regression model. It can be seen that the present invention can predict the performance of iron-chromium liquid flow batteries, especially the coulombic efficiency, voltage efficiency and energy efficiency of iron-chromium liquid flow batteries. By establishing a multivariate linear regression model to predict battery performance, it can reduce the experimental costs of the R&D team, shorten the R&D time, and improve economic benefits.

[0005] Although the above method can meet most scenarios, research and practical application of the above method and existing technology have found that the above method and existing technology have at least the following defects:

[0006] 1. Only the macroscopic overall parameters of the battery during operation are measured, and the data dimension is relatively single, which makes it difficult to fully reflect the actual performance of the battery during the entire charge and discharge cycle and the comprehensive status under different working conditions.

[0007] 2. During the data collection process, the sampling frequency cannot be flexibly adjusted according to the actual battery conditions, resulting in the inability to accurately capture data under certain rapidly changing battery conditions, or collecting too much data when it is unnecessary, wasting resources.

[0008] 3. The diversity of battery performance under different operating conditions and charge states is not fully considered, and the accuracy and applicability of the prediction method under different working conditions will be limited.

[0009] 4. It mainly relies on simulation and corresponding flow velocity distribution analysis. The technical means are relatively single, and it is difficult to predict the potential fault hazards in battery operation and the impact on the overall performance of the battery. There are deficiencies in the comprehensive control of battery performance.

[0010] In order to solve the above problems, the present invention proposes a liquid flow battery performance analysis method based on current density distribution measurement. Summary of the invention

[0011] The purpose of the present invention is to propose a flow battery performance analysis method based on current density distribution measurement to solve the problems raised in the background technology:

[0012] The data collection dimension is narrow; the sampling frequency lacks dynamic adaptability; different operating conditions and charge states are not taken into account; and there is a lack of comprehensive performance evaluation and prediction under multi-parameter fusion.

[0013] In order to achieve the above object, the present invention adopts the following technical solutions:

[0014] The performance analysis method of a flow battery based on current density distribution measurement includes the following steps:

[0015] S1: Import real-time environmental parameter data related to battery operation, preset the initial sampling frequency and perform data collection, and set the sampling frequency adjustment rules;

[0016] S2: Run the preset charge and discharge program, measure and store data based on the printed board designed with block grids, adjust the operating conditions, and record the battery data and each block data under different conditions;

[0017] S3: Calculate the charge and discharge capacity and energy under different operating conditions, calculate the pump work in combination with the electrolyte flow rate, calculate the coulomb efficiency, voltage efficiency, energy efficiency and system efficiency under the different working conditions, draw the charge and discharge curve and efficiency curve, and analyze the optimal operating condition of the battery;

[0018] S4: filtering the block data recorded under the above conditions, and drawing the current density and voltage curve of each block as the battery state of charge changes in combination with the area of ​​each block; generating a current density distribution map according to the position of each block and the measurement principle;

[0019] S5: Calculate the current density distribution uniformity coefficient under different operating conditions and different battery charge states, draw a curve of the current density distribution uniformity coefficient as the battery charge state changes, and calculate the average value;

[0020] S6: Select seconds as the time characteristic period, extract the time characteristics of current density, voltage curve and current density distribution uniformity coefficient; use the minimum-maximum normalization method to normalize the current density and voltage data to obtain a data set, and divide the data set into a training set, a validation set and a test set;

[0021] S7: construct a first machine learning model based on the long short-term memory network, use mean square error, mean absolute error and root mean square error to evaluate the prediction performance of the first machine learning model, and predict the current density of each block, the voltage change trend and the change trend of the overall current density distribution uniformity coefficient;

[0022] S8: construct an anomaly monitoring model through an unsupervised learning algorithm, perform anomaly monitoring on the prediction result of the first machine learning model, and if the monitoring does not show anomaly, continue to perform anomaly monitoring on the prediction result of the first machine learning model, and if the monitoring shows anomaly, classify the detected abnormal data based on the anomaly classification model built by the decision tree to obtain the anomaly type;

[0023] S9: In the current density distribution diagram under different operating conditions, the current density distribution cloud diagram is screened; the current density distribution data is extracted according to the screened current density distribution cloud diagram, the current density distribution data is cleaned and normalized, the flow dead zone in the current density distribution cloud diagram is marked, and the training set, the validation set and the test set are divided according to the current density distribution cloud diagram;

[0024] S10: construct a second machine learning model based on a convolutional neural network, select cross entropy loss and squared absolute error loss as loss functions, use Adam as the optimizer, use the training set to train the second machine learning model, use the validation set to adjust the hyperparameters of the second machine learning model, and use the test set to verify the accuracy of the second machine learning model;

[0025] S11: Use the second machine learning model obtained by training optimization to process the current density distribution data under different operating conditions, automatically mark the flow dead zone position in the electrode frame, and summarize it;

[0026] S12: All processed and analyzed data are classified and saved according to time sequence and data type, and the stored data is visualized based on a visualization interface.

[0027] Preferably, the environmental parameter data in S1 includes temperature, humidity and pressure;

[0028] The sampling frequency adjustment rules described in S1 are as follows:

[0029] When the key performance indicator is within the preset threshold range, the sampling frequency is gradually reduced according to the preset time interval until it is reduced to the preset minimum sampling frequency;

[0030] When the key performance indicator exceeds the preset threshold range, the sampling frequency is increased until the performance stabilizes again or reaches the preset high-frequency sampling frequency;

[0031] In the sampling frequency adjustment rule, the minimum sampling frequency and the high-frequency sampling frequency are optimized and adjusted based on a machine learning algorithm.

[0032] Preferably, in S2, during the data collection process, the key performance indicators of the battery are analyzed in real time, and corresponding sampling frequency adjustment rules are executed according to the key performance indicators; the key performance indicators include current density change rate, voltage fluctuation amplitude and power change rate.

[0033] Preferably, a plurality of block grids are designed on the surface of the printed plate in S2, and the interval between each block grid is not less than 0.5 mm and not more than 2 mm. The groove of each block grid is filled with epoxy resin. After it solidifies, the back of the bipolar plate is cut until the solidified resin is exposed to obtain a printed plate designed based on the block grid.

[0034] Preferably, the voltage of each block in S5 is the voltage collected by the data acquisition card, the current of each block is the voltage of each block divided by the resistance of the resistor connected to each block, and the current density is the current of each block divided by the area corresponding to each block.

[0035] Preferably, the current density distribution uniformity in S6 is calculated as follows:

[0036]

[0037] Where U is the current density distribution uniformity coefficient, i a is the average current density of each block, and i is the current density of each block.

[0038] Preferably, the data set in S8 is also expanded by generating an adversarial network before being divided.

[0039] Preferably, the unsupervised learning algorithm in S10 is implemented using an autoencoder, the state data and environmental parameter data are input into the autoencoder, and the reconstruction error is calculated; when the reconstruction error exceeds a preset abnormal threshold, the data is determined to be abnormal.

[0040] Preferably, the training method of the first machine learning model includes:

[0041] K groups of experimental training data are collected in advance, and the experimental training data include current density, voltage data, the current density and voltage change trends of each block, and the change trend of the overall current density distribution uniformity coefficient;

[0042] Each group of experimental training data is used as the input of the first machine learning model. The first machine learning model uses the current density, voltage change trend and the change trend of the overall current density distribution uniformity coefficient of each block corresponding to each group of current density and voltage data as output, and uses the current density, voltage change trend and the change trend of the overall current density distribution uniformity coefficient of each actual block corresponding to each group of current density and voltage data as the prediction target; minimizes the sum of the prediction errors of the current density, voltage change trend and the change trend of the overall current density distribution uniformity coefficient of all blocks as the training target; trains the first machine learning model until the sum of the prediction errors converges, and stops the training;

[0043] The training method of the second machine learning model includes:

[0044] Collecting G group of model training data in advance, the model training data includes current density distribution data and flow dead zones corresponding to the current density distribution data;

[0045] Each set of model training data is used as the input of the second machine learning model. The second machine learning model takes the flow dead zone corresponding to each set of current density distribution data as output, and the actual flow dead zone corresponding to each set of current density distribution data as the prediction target; minimizing the sum of prediction errors of all flow dead zones is used as the training goal; the second machine learning model is trained until the sum of prediction errors reaches convergence and the training is stopped.

[0046] Compared with the prior art, the present invention provides a flow battery performance analysis method based on current density distribution measurement, which has the following beneficial effects:

[0047] The present invention combines the real-time environmental parameters related to the battery operation, and collects block data under different operating conditions, etc. The data collection dimensions are rich and comprehensive, and the overall operating status of the battery can be characterized from multiple angles; the key performance indicators of the battery are also analyzed in real time during the data collection process, and the sampling frequency is further adjusted, the optimal operating conditions of the battery are analyzed, etc. according to the dynamic changes of these indicators during the charging and discharging process, so as to improve the subsequent optimization capability of the overall performance of the battery; through dynamic adaptive sampling frequency adjustment, when the battery performance changes rapidly, sufficient data can be obtained in time to analyze the details of the performance change; various parameters under different operating conditions and different battery charge states, such as the current density distribution uniformity coefficient, etc., are calculated in detail, and their influence on the battery performance is analyzed, so that the characteristics of the battery under various working conditions can be more accurately grasped; based on the printed board designed with block grids for data measurement and collection, combined with the generated current density distribution diagram, problems such as uneven current distribution and flow dead zone inside the battery can be intuitively found; a variety of technical means are integrated to comprehensively evaluate the performance of the battery from different levels, and the optimal operating mode and potential problems of the battery under different operating conditions can be explored based on the analysis and prediction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flow chart of the method mentioned in Example 1 of the present invention;

[0049] Figure 2 Schematic diagram of the coulombic efficiency, voltage efficiency and energy efficiency of the battery mentioned in Example 1 of the present invention;

[0050] Figure 3 This is a schematic diagram of the charge and discharge curve mentioned in Example 1 of the present invention;

[0051] Figure 4 This is a schematic diagram of the current density curve mentioned in Example 1 of the present invention;

[0052] Figure 5 This is a schematic diagram of the voltage curve mentioned in Example 1 of the present invention;

[0053] Figure 6 Schematic diagram of the current density distribution uniformity coefficient curve mentioned in Example 1 of the present invention. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0055] The present invention combines the real-time environmental parameters related to battery operation, and collects block data under different operating conditions, etc. The data collection dimensions are rich and comprehensive, and the overall operating state of the battery can be portrayed from multiple angles; the key performance indicators of the battery are also analyzed in real time during the data collection process, and the sampling frequency is further adjusted according to the dynamic changes of these indicators during the charging and discharging process, and the optimal operating conditions of the battery are analyzed, so as to improve the subsequent optimization ability of the overall performance of the battery; through dynamic adaptive sampling frequency adjustment, when the battery performance changes rapidly, sufficient data is obtained in time to analyze the details of the performance change; various parameters under different operating conditions and different battery charge states, such as the current density distribution uniformity coefficient, etc., are calculated in detail, and their influence on battery performance is analyzed, so that the characteristics of the battery under various working conditions can be more accurately grasped; based on the printed board designed with block grids for data measurement and collection, combined with the generated current density distribution map, problems such as uneven current distribution and flow dead zone inside the battery can be intuitively found; a variety of technical means are integrated to comprehensively evaluate the performance of the battery from different levels, and the optimal operating mode and potential problems of the battery under different operating conditions can also be mined based on the analysis and prediction results. Specifically include the following contents.

[0056] Embodiment 1:

[0057] See also Figure 1 The method for analyzing the performance of a flow battery based on current density distribution measurement of the present invention comprises the following steps:

[0058] S1: Import real-time environmental parameter data related to battery operation, preset the initial sampling frequency and perform data collection, and set the sampling frequency adjustment rules; the details are as follows:

[0059] By changing the path read by the software to the address of the Excel file where the data is stored, single or batch data can be imported; specifically, real-time environmental parameter data related to battery operation (such as temperature, humidity, pressure, etc.) will be helpful for subsequent comprehensive analysis. Filtering algorithms (such as Kalman filtering, etc.) can be used to reduce noise on the imported data to remove noise interference during the measurement process and ensure the accuracy of the imported data.

[0060] According to the battery type and the estimated performance changes, an initial sampling frequency is set to start data collection. During the data collection process, the changes in key performance indicators of the battery are analyzed in real time, including the current density change rate, voltage fluctuation amplitude and power change rate.

[0061] When the battery performance is in a relatively stable state and the change rate of the key performance indicator is within the preset stable threshold range, the sampling frequency is gradually reduced at preset time intervals (such as every several sampling cycles), but the minimum sampling frequency limit is maintained to ensure that important information is not missed.

[0062] Once it is detected that the rate of change of a key performance indicator exceeds the stability threshold, such as a sudden increase in the rate of change of current density or a large fluctuation in voltage, the sampling frequency is immediately increased to quickly collect more data points until the performance stabilizes again or the preset high-frequency sampling frequency is reached.

[0063] Among them, the minimum sampling frequency and high-frequency sampling frequency are optimized and adjusted based on the machine learning algorithm. The sampling frequency adjustment history can also be reviewed regularly, and the threshold and adjustment step of the sampling frequency can be optimized and adjusted according to the long-term performance change trend of the battery to adapt to the characteristics of different batteries at different use stages.

[0064] S2: Running the charge and discharge program, measuring and storing data based on the printed board designed with block grids, adjusting the operating conditions such as load current density and electrolyte flow rate, and recording the battery data and each block data under different working conditions; the details are as follows:

[0065] Start a pre-set charge and discharge program, which performs charge and discharge operations on the battery according to a specific current control strategy to obtain detailed electrochemical information of the battery during the charge and discharge process. Then use a printed board with a block grid design, and each block can measure the electrical parameters of the corresponding area respectively. Through the data acquisition card connected to the printed board, the voltage data on each block is collected, so that detailed information of different local areas of the battery can be obtained. Among them, the processing process of the printed board with a block grid design includes: processing grooves between the grids on the upper surface of the bipolar plate, and the grooves cannot be completely penetrated. Fill the grooves of the grid with epoxy resin, wait for it to solidify, cut the back of the bipolar plate until the solidified resin is exposed, and then a bipolar plate with a completely block design can be processed, which also ensures electrical insulation between the grids. The same block operation is performed on the side of the printed circuit board that contacts the bipolar plate, and the block size of the printed circuit board can be slightly smaller than the block size of the bipolar plate. In order to reduce the contact resistance between the two, the blocks of the printed circuit board need to be gold-plated. Each block on the printed circuit board is connected to a resistor with a precise resistance value; a pair of pins are used to connect the two ends of the resistor to draw out the electric potential at both ends and arrange them on both sides of the printed circuit board. On the back of the printed circuit board, a large current collecting area is designed to ensure that all currents flowing out of / into the block converge here and then flow out of / into the current collecting plate. In order to achieve the measurement of the current density of each grid, all designs of the printed circuit board are to ensure that the internal circuit is: the block on the front of the printed circuit board - the resistor connected to the block - the current collecting area on the back of the printed circuit board. The pins on the printed circuit board are connected to the data acquisition card through wires, and the acquisition card is connected to the computer through a data cable. During the operation of the battery, the data acquisition card records and saves the voltage signal at both ends of each resistor.

[0066] The printed board designed based on the above-mentioned block grid is used for data measurement and collection. During the collection process, the key performance indicators of the battery are analyzed in real time, and the corresponding sampling frequency adjustment rules are executed according to the key performance indicators. During the charging and discharging process, the load current density can be changed manually or through an automated control system to observe the battery's response at different current densities; at the same time, the electrolyte flow rate is adjusted because the electrolyte flow rate affects the material transport and other processes inside the battery. For each different combination of load current density and electrolyte flow rate (i.e., different operating conditions), the data of each block measured from the block grid printed board is recorded in detail, and these data will be used for various analyses later.

[0067] S3: Calculate the charge and discharge capacity and energy under different operating conditions, calculate the pump work in combination with the electrolyte flow rate, calculate the coulomb efficiency, voltage efficiency, energy efficiency and system efficiency under the different operating conditions, draw the charge and discharge curve and efficiency curve, and analyze the optimal operating condition of the battery; the details are as follows:

[0068] By integrating the current and time data collected during the charge and discharge process, the charge and discharge capacity of the battery (usually in ampere-hours, Ah) can be obtained; then combined with the voltage data, the energy calculation formula (energy = voltage × capacity) is used for integration to calculate the energy value during the charge and discharge process. The above calculations are performed for different operating conditions (such as different load current density, electrolyte flow rate, etc.) to understand the impact of these conditions on battery capacity and energy.

[0069] According to the electrolyte flow rate and related physical parameters (such as electrolyte density, pipeline resistance coefficient, etc.), using the power calculation formula in fluid mechanics (pump work = flow rate × pressure difference and other related forms), combined with the flow of electrolyte in the battery system, the pump work required to maintain a specific electrolyte flow rate is calculated.

[0070] Calculate the coulombic efficiency, voltage efficiency and energy efficiency of the battery and draw the corresponding change diagram. Please refer to Figure 2 , the coulombic efficiency, voltage efficiency and energy efficiency of the battery can be intuitively obtained:

[0071] The coulombic efficiency is calculated as follows:

[0072]

[0073] Among them, η C is the Coulomb efficiency; Q fd Q is the amount of charge passing through the battery during discharge; cd is the amount of charge passing through the battery during the charging process; T fd is the discharge time; T cd is the charging time; I fd is the average current during the discharge process; Icd is the average current during the charging process;

[0074] Coulombic efficiency is a key indicator for measuring the degree of charge utilization during the battery charging and discharging process. By comparing the discharge output charge with the charging input charge, it can reflect the reversibility of the electrochemical reaction and the health of the battery. With the cycle use and aging of the battery, the coulombic efficiency may gradually decrease. This may be due to the decrease in the activity of the electrode material, the decomposition of the electrolyte, or the change in the internal structure of the battery. For example, when the coulombic efficiency of the battery drops from the initial 98% to below 90%, it indicates that the battery may have a more serious aging problem, such as the shedding of active substances on the electrode surface or the generation of a large number of irreversible reaction products in the electrolyte. This provides an important basis for battery fault diagnosis and life prediction.

[0075] The voltage efficiency is calculated as follows:

[0076]

[0077] Among them, η U is the voltage efficiency; U fd is the average voltage during the discharge process; U cd is the average voltage during the charging process;

[0078] Voltage efficiency reflects the voltage loss of the battery during the charging and discharging process. It is calculated by comparing the average discharge voltage with the average charge voltage. Voltage loss is mainly caused by polarization phenomena inside the battery (including ohmic polarization, concentration polarization, and activation polarization). For example, during the charging process, due to polarization, the charging voltage of the battery will be higher than its equilibrium voltage; while during the discharge process, the discharge voltage will be lower than the equilibrium voltage. By calculating the voltage efficiency, the degree of this voltage loss can be quantified, and then the magnitude of the energy loss inside the battery can be evaluated.

[0079] The energy efficiency is calculated as follows:

[0080]

[0081] Among them, η E is energy efficiency; E cd E is charging energy; fd is the discharge energy.

[0082] Energy efficiency is the product of coulombic efficiency and voltage efficiency. It comprehensively considers the charge utilization and voltage loss of the battery during the charging and discharging process, and can more comprehensively evaluate the performance of the battery.

[0083] Reference Figure 3, extract the charge and discharge capacity and charge and discharge voltage, and draw the charge and discharge curve. The charge and discharge curve (voltage-capacity curve) can intuitively show the changes in voltage and capacity of the battery during the charge and discharge process. From the shape of the curve, we can see some basic performance characteristics of the battery, such as the open circuit voltage, platform voltage, cut-off voltage and capacity of the battery. Various efficiency curves can be drawn, such as coulomb efficiency, energy efficiency and other curves that change with operating conditions. By observing and analyzing these curves, find out the combination of operating conditions that can make the battery perform best in terms of energy conversion, performance stability, etc., such as determining the most suitable parameters such as load current density and electrolyte flow rate.

[0084] S4: Filter the block data recorded under the above conditions, and draw the current density and voltage curve of each block as the battery state of charge changes in combination with the area of ​​each block; generate a current density distribution map based on the position of each block and the measurement principle; the details are as follows:

[0085] The block data previously recorded under different operating conditions are processed using a filtering algorithm to remove noise interference, making the data smoother and facilitating subsequent accurate analysis.

[0086] The voltage of each block is the voltage collected by the data acquisition card, the current of each block is the voltage divided by the resistance of the resistor connected to the block, and the current density is the current of each voltage divided by the area corresponding to each block. According to the area information of each block and the collected current and voltage data, the current density (current divided by area) corresponding to each block is calculated. Using a certain state parameter of the electrolyte (such as electrolyte concentration, temperature, etc.) as the horizontal axis, and the current density and voltage as the vertical axis, the curve of each block changing with the battery charge state is drawn. Please refer to Figure 4 , Figure 5 , intuitively presenting the changes in the electrical characteristics of each block under different battery charge states. Based on the position information of each block in the battery, combined with the measured value of current density and related measurement principles (such as the distribution law of current in different areas, etc.), the current density values ​​of each block are graphically displayed using drawing software to generate a current density distribution map, which can clearly show the spatial distribution of current density inside the battery.

[0087] S5: Calculate the current density distribution uniformity coefficient under different operating conditions and different battery charge states, draw a curve of the current density distribution uniformity coefficient as the battery charge state changes, and calculate the average value thereof; the details are as follows:

[0088] The current density distribution uniformity is calculated as follows:

[0089]

[0090] Where U is the current density distribution uniformity coefficient, ia is the average current density of each block, and i is the current density of each block.

[0091] Take the battery state of charge parameter as the horizontal axis and the calculated current density distribution uniformity coefficient as the vertical axis, and use the drawing tool to draw a curve of its change with the battery state of charge. For details, please refer to Figure 6 Through this curve, we can observe the influence trend of battery state of charge change on current density distribution uniformity. The average value of all current density distribution uniformity coefficients obtained under different operating conditions and battery state of charge is calculated. This average value can be used as an overall reference indicator to compare the approximate level of current density distribution uniformity under different working conditions or different experimental batches.

[0092] S6: Select seconds as the time characteristic period, extract the time characteristics of current density, voltage curve and current density distribution uniformity coefficient; use the minimum-maximum normalization method to normalize the current density and voltage data to obtain a data set, and divide the data set into a training set, a validation set and a test set; the details are as follows:

[0093] From the previously drawn current density curve, voltage curve and calculated current density distribution uniformity coefficient data, analyze their changes over time, extract time-related characteristic information such as periodic change characteristics (if any), change trends within a specific time period (increase, decrease, etc.), and explicitly select seconds as the basic time period feature to facilitate the unification of subsequent data processing and model analysis on a time scale.

[0094] Using the minimum-maximum normalization method, for the current density and voltage data, first find their respective minimum and maximum values, and then map the data to the [0,1] interval through the formula (normalized value = (original value - minimum value) / (maximum value - minimum value)) (it can also be further adjusted to [-1,1] and other intervals according to subsequent model requirements). This can eliminate the impact of different data magnitudes on subsequent machine learning model training and other operations, making the data more comparable and universal.

[0095] The normalized data is divided into training set, validation set and test set according to a certain ratio (commonly used ratios are 7:2:1 or 8:1:1, etc.). The training set is used to train the machine learning model so that the model can learn the patterns in the data; the validation set is used to adjust the model's hyperparameters (such as the number of layers and nodes of the neural network) during the model training process to prevent overfitting; the test set is used to finally evaluate the performance of the model on unseen data.

[0096] S7: Construct a first machine learning model based on the long short-term memory network, use mean square error, mean absolute error and root mean square error to evaluate the prediction performance of the first machine learning model, and predict the current density of each block, the voltage change trend and the change trend of the overall current density distribution uniformity coefficient; the details are as follows:

[0097] Determine the structure of the LSTM network, including the number of input layer nodes (determined according to the characteristic dimensions of the input data, such as the number of features such as current density and voltage), the number of layers and nodes in the hidden layer (which can be adjusted and optimized through experiments and experience), and the number of output layer nodes (corresponding to the output target number such as the current density, voltage change trend, and uniformity coefficient change trend to be predicted). Use a suitable deep learning framework to build the LSTM network model architecture and perform initialization settings.

[0098] The constructed LSTM model is trained using the divided training set, and the parameters of the model are adjusted through multiple iterations so that the model can learn the time series characteristics and internal laws in the data. At the same time, the mean square error (MSE, the average of the square of the difference between the predicted value and the true value), the mean absolute error (MAE, the average of the absolute value of the difference between the predicted value and the true value) and the root mean square error (RMSE, the square root of MSE) are selected as indicators for evaluating the prediction performance of the model. These indicators can reflect the degree of deviation between the model prediction results and the real data from different angles.

[0099] Use the trained model to predict data in a short period of time, and output the future trend of current density and voltage in each block, as well as the trend of the overall current density distribution uniformity coefficient. When analyzing the prediction results, focus on the time points and corresponding areas where there are obvious errors between the predicted values ​​and the true values. By further studying the causes of these differences (such as data anomalies, special rules that the model has not learned, etc.), the model can be improved or a deeper understanding of the battery operation can be obtained.

[0100] S8: Construct an anomaly monitoring model through an unsupervised learning algorithm, perform anomaly monitoring on the prediction results of the first machine learning model, and if there is no anomaly in the monitoring, continue to perform anomaly monitoring on the prediction results of the first machine learning model. If there is an anomaly in the monitoring, classify the detected abnormal data based on the anomaly classification model built by the decision tree to obtain the anomaly type; the details are as follows:

[0101] The battery data features under normal operating conditions are learned through unsupervised learning algorithms (such as autoencoders) to build a data anomaly detection model. The real-time collected data is input into the autoencoder to calculate the reconstruction error. When the reconstruction error exceeds the preset threshold, the data is judged to be abnormal, that is, the battery may fail. For example, if local corrosion occurs on the electrode surface, the difference between the data features and the normal state will increase, thereby triggering an anomaly detection alarm. At the same time, machine learning algorithms (such as random forest classifiers) are used to classify abnormal data and determine possible fault types (such as electrode failure, electrolyte failure, etc.) so that timely measures can be taken to repair or adjust them to ensure the safe and efficient operation of the battery system.

[0102] S9: In the current density distribution diagram under different operating conditions, the current density distribution cloud diagram is screened; the current density distribution data is extracted according to the screened current density distribution cloud diagram, the current density distribution data is cleaned and normalized, the flow dead zone in the current density distribution cloud diagram is marked, and the training set, the validation set and the test set are divided according to this; the details are as follows:

[0103] From the numerous current density distribution maps generated previously under different operating conditions, representative cloud maps that can clearly show the current density distribution characteristics are selected based on research objectives, analysis needs and other factors. These cloud maps present the spatial distribution of current density inside the battery in the form of intuitive colors, shapes, etc., which is convenient for subsequent observation and processing.

[0104] The selected current density distribution data is cleaned to remove possible erroneous data (such as outliers caused by measurement errors, etc.), duplicate data, etc. Then, a suitable normalization method is used (such as the minimum-maximum normalization similar to the previous one, by adjusting the parameters to map it to the [-1,1] interval) to ensure that the data of each block falls within the specified interval, which is convenient for subsequent model training and unified data scale.

[0105] Professionals identify and manually mark the dead zones (i.e. areas where the electrolyte flows slowly or almost not, which may affect battery performance) reflected in the current density distribution cloud map based on the features presented therein. Then, the processed data is divided into training set, validation set and test set according to a certain ratio (similar to the principle of dividing the data set above), in preparation for subsequent operations such as model training based on convolutional neural networks.

[0106] S10: Build a second machine learning model based on a convolutional neural network, select cross entropy loss and square absolute error loss as loss functions, use Adam as the optimizer, use the training set to train the second machine learning model, use the validation set to adjust the hyperparameters of the second machine learning model, and use the test set to verify the accuracy of the second machine learning model; the details are as follows:

[0107] Determine the network architecture of CNN, including parameters such as the number of convolutional layers, convolution kernel size, step size, the setting of pooling layers (such as maximum pooling, average pooling, etc. and corresponding parameters) and the structure of the fully connected layer, and build a CNN model suitable for processing current density distribution cloud map data, so that it can automatically extract feature information in the image, such as the texture, shape and other features of the current density distribution and the correlation features of the flow dead zone.

[0108] The cross entropy loss function is selected for classification to distinguish the flow dead zone from the normal flow area. The squared absolute error loss function is selected for the anchor box to mark the flow dead zone position. The model training is guided by calculating the difference between the predicted result and the true label. The Adam optimizer is used, which can adaptively adjust the model's learning rate and other parameters to speed up the model convergence and improve the training effect. The constructed CNN model is trained multiple times using the divided training set, so that the model can continuously learn the image features and classification rules in the data.

[0109] During the training process, according to the performance of the model on the validation set (such as changes in loss value and accuracy), the model's hyperparameters (such as the number of channels in the convolutional layer and the number of nodes in the fully connected layer) are adjusted to optimize the model performance and prevent overfitting. Finally, the test set is used to verify the final accuracy of the model, and the accuracy (the proportion of correctly predicted samples to the total number of samples) and recall (the proportion of samples that are actually positive and predicted to be positive to the actual number of positive samples) are used to comprehensively evaluate the performance of the model in identifying flow dead zones.

[0110] S11: Use the second machine learning model obtained by training optimization to process the current density distribution data under different operating conditions, judge the area within the electrode frame in each current density distribution map, automatically mark the flow dead zone position within the electrode frame, and summarize it; the details are as follows:

[0111] The previously collected large-scale current density distribution cloud map and other image data are input into the trained machine learning model based on the convolutional neural network. The model will judge the area within the electrode frame in each image based on the learned features and rules, and automatically mark the flow dead zone location. Then the flow dead zone location information marked in all images is summarized and sorted, such as the frequency and area size of flow dead zones in different areas, so as to further analyze the electrolyte flow characteristics inside the battery and its impact on battery performance.

[0112] S12: All processed and analyzed data are classified and stored according to time sequence and data type, and the stored data is dynamically visualized based on the visualization interface. The details are as follows:

[0113] All processed and analyzed data, including raw data, pre-processed data, performance calculation results at different time scales, current density distribution data, sampling frequency adjustment records, performance trend prediction results, and abnormal detection information, are classified and stored according to time sequence and data type. An efficient data storage format (such as HDF5 format) is used to facilitate subsequent query, analysis, and data mining.

[0114] Develop an intuitive visualization interface that can simultaneously display the evolution of battery performance at multiple time scales. For example, display the charge transfer rate change at the microscopic time scale and the energy efficiency change trend at the macroscopic time scale in the same chart, using different colors and line styles to distinguish different parameters and time scales.

[0115] The dynamic tracking process of battery performance over time is displayed in the form of animation or dynamic charts, including adaptive changes in sampling frequency, change curves of performance indicators at different time scales, etc. The abnormal detection results are displayed intuitively. When an abnormality is detected, the abnormal data points and corresponding fault type information are highlighted on the visual interface, which is convenient for operators to detect abnormalities in time and make corresponding decisions.

[0116] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for analyzing flow battery performance based on current density distribution measurement, characterized in that: The steps include: S1: Import real-time environmental parameter data related to battery operation, preset the initial sampling frequency and perform data collection, and set the sampling frequency adjustment rules; S2: Run the preset charge and discharge program, measure and store data based on the printed board designed with block grids, adjust the operating conditions, and record the battery data and each block data under different conditions; S3: Calculate the charge and discharge capacity and energy under different operating conditions, calculate the pump work in combination with the electrolyte flow rate, calculate the coulomb efficiency, voltage efficiency, energy efficiency and system efficiency under the different working conditions, draw the charge and discharge curve and efficiency curve, and analyze the optimal operating condition of the battery; S4: filtering the block data recorded under the above conditions, and drawing the current density and voltage curve of each block as the battery state of charge changes in combination with the area of ​​each block; generating a current density distribution map according to the position of each block and the measurement principle; S5: Calculate the current density distribution uniformity coefficient under different operating conditions and different battery charge states, draw a curve of the current density distribution uniformity coefficient as the battery charge state changes, and calculate the average value; S6: Select seconds as the time characteristic period and extract the time characteristics of current density, voltage curve and current density distribution uniformity coefficient; The current density and voltage data are normalized using the minimum-maximum normalization method to obtain a data set, and the data set is divided into a training set, a validation set, and a test set; S7: construct a first machine learning model based on the long short-term memory network, use mean square error, mean absolute error and root mean square error to evaluate the prediction performance of the first machine learning model, and predict the current density of each block, the voltage change trend and the change trend of the overall current density distribution uniformity coefficient; S8: construct an anomaly monitoring model through an unsupervised learning algorithm, perform anomaly monitoring on the prediction result of the first machine learning model, and if the monitoring does not show anomaly, continue to perform anomaly monitoring on the prediction result of the first machine learning model, and if the monitoring shows anomaly, classify the detected abnormal data based on the anomaly classification model built by the decision tree to obtain the anomaly type; S9: In the current density distribution diagram under different operating conditions, the current density distribution cloud diagram is screened; the current density distribution data is extracted according to the screened current density distribution cloud diagram, the current density distribution data is cleaned and normalized, the flow dead zone in the current density distribution cloud diagram is marked, and the training set, the validation set and the test set are divided according to the current density distribution cloud diagram; S10: construct a second machine learning model based on a convolutional neural network, select cross entropy loss and squared absolute error loss as loss functions, use Adam as the optimizer, use the training set to train the second machine learning model, use the validation set to adjust the hyperparameters of the second machine learning model, and use the test set to verify the accuracy of the second machine learning model; S11: Use the second machine learning model obtained by training optimization to process the current density distribution data under different operating conditions, automatically mark the flow dead zone position in the electrode frame, and summarize it; S12: All processed and analyzed data are classified and saved according to time sequence and data type, and the stored data is visualized based on a visualization interface.

2. The method for analyzing the performance of a flow battery based on current density distribution measurement according to claim 1, characterized in that: The environmental parameter data in S1 include temperature, humidity and pressure; The sampling frequency adjustment rules described in S1 are as follows: When the key performance indicator is within the preset threshold range, the sampling frequency is gradually reduced according to the preset time interval until it is reduced to the preset minimum sampling frequency; When the key performance indicator exceeds the preset threshold range, the sampling frequency is increased until the performance stabilizes again or reaches the preset high-frequency sampling frequency; In the sampling frequency adjustment rule, the minimum sampling frequency and the high-frequency sampling frequency are optimized and adjusted based on a machine learning algorithm.

3. The method for analyzing the performance of a flow battery based on current density distribution measurement according to claim 2, characterized in that: In S2, during the data collection process, the key performance indicators of the battery are analyzed in real time, and corresponding sampling frequency adjustment rules are executed according to the key performance indicators; the key performance indicators include current density change rate, voltage fluctuation amplitude and power change rate.

4. The method for analyzing the performance of a flow battery based on current density distribution measurement according to claim 1, characterized in that: The printed board described in S2 is designed with a plurality of block grids on its surface, and the interval between each block grid is not less than 0.5 mm and not more than 2 mm. The groove of each block grid is filled with epoxy resin. After it solidifies, the back of the bipolar plate is cut until the solidified resin is exposed to obtain a printed board designed based on the block grid.

5. The method for analyzing the performance of a flow battery based on current density distribution measurement according to claim 1, characterized in that: The voltage of each block of S5 is the voltage collected by the data acquisition card, the current of each block is the voltage of each block divided by the resistance of the resistor connected to each block, and the current density is the current of each block divided by the area corresponding to each block.

6. The method for analyzing performance of a flow battery based on current density distribution measurement according to claim 1, characterized in that: The current density distribution uniformity in S6 is calculated as follows: Where U is the current density distribution uniformity coefficient, i a is the average current density of each block, and i is the current density of each block.

7. The method for analyzing performance of a flow battery based on current density distribution measurement according to claim 1, characterized in that: The data set in S8 is also expanded by generating an adversarial network before division.

8. The method for analyzing performance of a flow battery based on current density distribution measurement according to claim 8, characterized in that: The unsupervised learning algorithm in S10 is implemented using an autoencoder, and the state data and environmental parameter data are input into the autoencoder to calculate the reconstruction error; when the reconstruction error exceeds a preset abnormal threshold, the data is determined to be abnormal.

9. The method for analyzing performance of a flow battery based on current density distribution measurement according to claim 1, characterized in that: The training method of the first machine learning model includes: K groups of experimental training data are collected in advance, and the experimental training data include current density, voltage data, the current density and voltage change trends of each block, and the change trend of the overall current density distribution uniformity coefficient; Each group of experimental training data is used as the input of the first machine learning model. The first machine learning model uses the current density, voltage change trend and the change trend of the overall current density distribution uniformity coefficient of each block corresponding to each group of current density and voltage data as output, and uses the current density, voltage change trend and the change trend of the overall current density distribution uniformity coefficient of each actual block corresponding to each group of current density and voltage data as the prediction target; minimizes the sum of the prediction errors of the current density, voltage change trend and the change trend of the overall current density distribution uniformity coefficient of all blocks as the training target; trains the first machine learning model until the sum of the prediction errors converges, and stops the training; The training method of the second machine learning model includes: Collecting G group of model training data in advance, the model training data includes current density distribution data and flow dead zones corresponding to the current density distribution data; Each set of model training data is used as the input of the second machine learning model. The second machine learning model takes the flow dead zone corresponding to each set of current density distribution data as output, and the actual flow dead zone corresponding to each set of current density distribution data as the prediction target; minimizing the sum of prediction errors of all flow dead zones is used as the training goal; the second machine learning model is trained until the sum of prediction errors reaches convergence and the training is stopped.

Images