Data processing system and method based on lithium battery dry method electrode preparation process

Abstract

The invention discloses a data processing system and method based on a lithium battery dry method electrode preparation process, and relates to the technical field of data processing.The method comprises the following steps that mixed material resistance data and environment and raw material information are collected and preprocessed; extracting features from the preprocessed resistance data, calculating a related change rate in combination with environment and raw material information, and establishing a regression analysis model; key factors influencing the resistance are screened out, and a data set containing a key factor set and corresponding powder resistance values is formed; the method comprises the following steps: adjusting controllable factors in a safety and process range, obtaining multiple groups of data about key factors and resistance values, establishing a database, monitoring the key factors in a production process in real time, calculating a difference degree with the data groups in the database, setting a tolerance threshold value, triggering an abnormal alarm when the difference degree exceeds the threshold value, and otherwise, determining a current resistance value. According to the invention, the resistance data can be processed in the preparation process of the dry electrode of the lithium battery.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a data processing system and method based on a dry-process electrode preparation process for a lithium battery. Background Art

[0002] With the vigorous development of the lithium battery industry, dry-process electrode preparation technology for lithium batteries has been widely used in the industry due to its unique advantages. In the pursuit of lithium battery products with higher energy density and better safety, precise control and optimization of the dry-process electrode preparation process has become the key. As an important link in dry-process electrode preparation, the stability of powder resistance during the mixing process has a vital impact on electrode performance and even the quality of the entire lithium battery. Therefore, the processing and analysis of data related to the powder resistance of the mixture during the dry-process electrode preparation of lithium batteries is of great significance.

[0003] However, the traditional method of processing the resistance data of mixed powders in the dry-process preparation of lithium battery electrodes has exposed many problems when facing the complex needs of modern lithium battery production: First, the resistance data is often greatly interfered by environmental noise, and the traditional method lacks effective data filtering and noise reduction processing, which affects the measurement accuracy; second, the resistance change in the mixing process is affected by multiple process parameters. The traditional single variable analysis method is difficult to fully reveal the key factors affecting the resistance change and cannot accurately quantify the effect of each factor on the resistance. In addition, the lack of efficient regression analysis and feature extraction methods makes it difficult to accurately adjust key parameters in the process optimization process; then, the real-time monitoring and anomaly detection capabilities are limited. Traditional process monitoring usually relies on manual experience or simple threshold setting, which makes it difficult to perform real-time analysis and anomaly detection of resistance data in complex process environments. Summary of the Invention

[0004] The purpose of the present invention is to provide a data processing system and method based on the dry process of preparing lithium battery electrodes to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a data processing method based on the dry-process electrode preparation process of a lithium battery, the method comprising the following steps:

[0006] Collect mixed material resistance data, environment and raw material information for preprocessing;

[0007] Extract features from the pre-processed resistance data, calculate the relevant change rate based on the environment and raw material information, and establish a regression analysis model;

[0008] Screen out the key factors that affect the resistance and form a data set containing a set of key factors and corresponding powder resistance values;

[0009] Adjust controllable factors within the safety and process range to obtain multiple sets of data on key factors and resistance values. Establish a database, monitor key factors in the production process in real time, calculate the difference with the data group in the database, and set a tolerance threshold. When the difference exceeds the threshold, trigger an abnormal alarm, otherwise determine the current resistance value.

[0010] The resistance measurement equipment is used to collect the powder resistance data of the mixed material at set time intervals to obtain the resistance measurement value sequence {R1, R2, ..., R n}, where n represents the number of collection times. Sensors are used to collect environmental data during the mixing process, including the temperature, humidity, and stirring speed inside the mixing equipment. Raw material information, including raw material type, batch, and amount, is also collected.

[0011] The Gaussian filter algorithm is applied to the collected powder resistance data to remove noise interference and obtain the preprocessed resistance data {R'1, R'2,…, R' n}, standardize environmental data and raw material information, and unify data of different magnitudes and units to the same scale.

[0012] The resistance characteristics are extracted from the pre-processed powder resistance data, including the average value, maximum value, minimum value, resistance change trend and fluctuation frequency. Combined with the environmental data and raw material information, the environmental factor change rate, stirring speed change rate and raw material dosage change rate are calculated. The extracted characteristic parameters are recorded as the characteristic vector X = [x1, x2, ..., x k ], where X represents the characteristic parameter set, k represents the total number of characteristic parameters extracted from the preprocessed powder resistance data, x1, x2, ..., x k The 1st, 2nd, ..., kth characteristic parameters respectively, wherein the environmental factor change rate includes the temperature change rate and humidity change rate inside the mixing equipment;

[0013] The calculation formula for the resistance change trend is as follows:

[0014]

[0015] Among them, ΔR' represents the change trend of resistance, R' i , R' i+1 They represent the resistance values ​​corresponding to the i-th and i+1-th data points in the preprocessed resistance data, respectively, and Δt represents the time interval of data collection;

[0016] The fluctuation frequency uses a spectrum analysis method to convert the resistance data from the time domain to the frequency domain to obtain the fluctuation frequency of the resistance;

[0017] A multiple linear regression model was used to construct a regression model with the extracted resistance characteristics, environmental factor change rate, stirring speed change rate, and raw material dosage change rate as independent variables and powder resistance value as the dependent variable. The regression model definition is as follows: R pred =β0+β1*x1+β2*x2+...+β k *x k +ε; where x1, x2, ..., x k Indicates the first, second, ..., k features obtained, β0, β1, β2, ..., β k Indicates the influence coefficient corresponding to the acquired feature, and ε indicates the error value.

[0018] According to actual production experience and historical data, the influence coefficient threshold θ is set, and the influence coefficient corresponding to the i-th feature extracted above |β i | is compared with the set influence coefficient threshold, when |β i When |>θ, the i-th feature is determined to have an impact on the powder resistance and is retained as a key factor. i When |≤θ, the feature is judged to have little effect on the resistance and is excluded, thereby obtaining the final set of key factors affecting the powder resistance {C(1,1),C(1,2),...,C(a,b)}, where a represents the number of key factors and b represents the number of features in each key factor that have an impact on the resistance.

[0019] The key factor set and the corresponding powder resistance value R are combined to form a data set related to the key factor set and the powder resistance value.

[0020] Under the premise of ensuring the mixing process and product quality requirements, the key factors in the mixing process are changed to obtain a total of z groups of data on the key factor set and the powder resistance value, thereby establishing a database on the powder resistance value and the key factor set;

[0021] In the real-time monitoring of the mixing production process, the difference V between the key factors in the current production environment and the rth group of key factors in the database is calculated in real time. r , the calculation formula is as follows:

[0022]

[0023] Among them, M b represents the characteristic value of the dth key factor monitored in real time, C r,d represents the dth eigenvalue of the rth group of key factors in the database, d total Indicates the total number of key features, and total indicates the data subscript;

[0024] Substitute r=1,2,...,z into the difference calculation formula to obtain the difference set {V1,V2,...,V r ,...,V z}, the minimum difference value obtained by comparison is V min , set the maximum tolerance of difference V max ;

[0025] When the minimum difference value is greater than the set maximum tolerance value, the powder resistance abnormality alarm is triggered, indicating that there is a production abnormality and the control of key factors needs to be checked;

[0026] When the minimum difference is less than or equal to the maximum tolerance of the difference, judge V min The powder resistance values ​​in the data group are the powder resistance values ​​under the current production environment.

[0027] A data processing system based on the dry-process electrode preparation process of lithium batteries includes a data acquisition module, a feature extraction module, a model construction module and a monitoring module. The data acquisition module is used to use a resistance measuring device to collect powder resistance data, environmental parameters and raw material information during the mixing process at set time intervals; the feature extraction module is used to extract resistance features from the pre-processed powder resistance data, and calculate the relevant change rate in combination with the environmental data and raw material information; the model construction module is used to construct a multivariate linear regression model, complete the screening of key factors, and store the processed data; the monitoring module is used to monitor the key factors in the production process in real time, calculate the difference with the data group in the database, and set a tolerance threshold. When the difference exceeds the threshold, an abnormal alarm is triggered; otherwise, the current resistance value is determined.

[0028] The data acquisition module includes a resistance data acquisition unit, an environmental parameter acquisition unit, a raw material information acquisition unit and a preprocessing unit. The resistance data acquisition unit is used to collect powder resistance data at set time intervals through a resistance measuring device to form a resistance measurement sequence; the environmental parameter acquisition unit is used to monitor environmental data such as temperature, humidity and stirring speed during the mixing process through sensors; the raw material information acquisition unit is used to record raw material type, batch and dosage information; the preprocessing unit is used to clean and standardize the collected resistance data, environmental parameters and raw material information; the output ends of the resistance data acquisition unit, environmental parameter acquisition unit and raw material information acquisition unit are connected to the input end of the preprocessing unit, and the output end of the preprocessing unit is connected to the input end of the feature extraction module.

[0029] The feature extraction module includes a resistance feature extraction unit and a change rate calculation unit. The resistance feature extraction unit is used to extract resistance features from the preprocessed powder resistance data, including the average value, maximum value, minimum value, resistance change trend and fluctuation frequency of the resistance; the change rate calculation unit is used to combine environmental data and raw material information to calculate the change rate of environmental factors, the change rate of stirring speed and the change rate of raw material usage; the output ends of the resistance feature extraction unit and the change rate calculation unit are connected to the input end of the model construction module.

[0030] The model construction module includes a regression model construction unit, a key factor screening unit and a data group construction unit. The regression model construction unit is used to use a multivariate linear regression model, take the extracted resistance characteristics, environmental factor change rate, stirring speed change rate and raw material usage change rate as independent variables, and powder resistance value as the dependent variable to construct a regression model; the key factor screening unit sets the influence coefficient threshold based on actual production experience and historical data, and compares the influence coefficient corresponding to the extracted characteristics with the set influence coefficient threshold to screen out key factors; the data group construction unit is used to combine the key factor set with the corresponding powder resistance value to form a data group about the key factor set and the powder resistance value. The output end of the regression model construction unit is connected to the input end of the key factor screening unit, and the output end of the key factor screening unit is connected to the input end of the data group construction unit.

[0031] The monitoring module includes a difference calculation unit, an optimal matching unit and an alarm judgment unit. The difference calculation unit is used to calculate the difference between the key factors in the current production environment and the key factor set in the database in real-time monitoring of the mixing production process; the optimal matching unit is used to calculate the minimum difference and determine the closest historical data group; the alarm judgment unit is used to compare the calculated minimum difference value with the set maximum tolerance of the difference. When the minimum difference value is greater than the set maximum tolerance of the difference, the powder resistance abnormality alarm is triggered, indicating that there is a production abnormality and the control of the key factors needs to be checked. When the minimum difference value is less than or equal to the set maximum tolerance of the difference, the powder resistance value in the data group where the minimum difference value is located is judged to be the powder resistance value under the current production environment. The output end of the difference calculation unit is connected to the input end of the optimal matching unit, and the output end of the alarm judgment unit is connected to the input end of the alarm judgment unit.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] 1. Different from traditional single-variable analysis methods, this method extracts multiple features from pre-processed resistance data, calculates the relevant change rate based on environmental and raw material information, and conducts comprehensive analysis using a multivariate linear regression model. This analysis method reveals the key factors affecting resistance changes and quantifies the influence of each factor on powder resistance.

[0034] 2. The present invention realizes real-time monitoring of resistance data during the mixing process. By calculating the difference between key factors in the current production environment and the data group in the database, and setting a tolerance threshold for comparison, anomalies can be detected;

[0035] 3. The present invention uses a Gaussian filtering algorithm to process the collected mixed powder resistance data, remove environmental noise interference, and improve the measurement accuracy of the resistance data. The filtered data reflects the actual changes in powder resistance during the mixing process, providing a data basis for subsequent analysis and decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Schematic diagram of the flow of the data processing method in the dry-process preparation process of lithium battery electrodes according to the present invention;

[0037] Figure 2 This is a schematic structural diagram of the data processing system in the dry-process lithium battery electrode preparation process according to the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] In an embodiment, Figure 1-Figure 2 As shown, the present invention provides a technical solution based on a data processing method in a dry-process lithium battery electrode preparation process, the method comprising the following steps:

[0040] Collect mixed material resistance data, environment and raw material information for preprocessing;

[0041] Extract features from the pre-processed resistance data, calculate the relevant change rate based on the environment and raw material information, and establish a regression analysis model;

[0042] Screen out the key factors that affect the resistance and form a data set containing a set of key factors and corresponding powder resistance values;

[0043] Adjust controllable factors within the safety and process range to obtain multiple sets of data on key factors and resistance values. Establish a database, monitor key factors in the production process in real time, calculate the difference with the data group in the database, and set a tolerance threshold. When the difference exceeds the threshold, trigger an abnormal alarm, otherwise determine the current resistance value.

[0044] The resistance measurement equipment is used to collect the powder resistance data of the mixed material at set time intervals to obtain the resistance measurement value sequence {R1, R2, ..., R n}, where n represents the number of collection times. Sensors are used to collect environmental data during the mixing process, including the temperature, humidity, and stirring speed inside the mixing equipment. Raw material information, including raw material type, batch, and amount, is also collected.

[0045] The Gaussian filter algorithm is applied to the collected powder resistance data to remove noise interference and obtain the preprocessed resistance data {R'1, R'2,…, R' n}, standardize environmental data and raw material information, and unify data of different magnitudes and units to the same scale.

[0046] Specifically, the resistance measurement equipment is used to collect the powder resistance data of the mixed material every 1 minute, for a total of 60 times, to obtain the resistance measurement value sequence {R1, R2, ..., R 60}, through the temperature sensor, humidity sensor and stirring speed sensor, synchronously collect the environmental data inside the mixing equipment. During these 60 minutes, the initial temperature value is 25℃ and gradually rises to 30℃; the humidity is maintained at 40% RH; the stirring speed is gradually increased from 100 rpm to 120 rpm. The raw materials used in this mixing are A and B. The batch of raw material A is 20240101, and the amount is 5kg; the batch of raw material B is 20240102, and the amount is 3kg;

[0047] The Gaussian filtering algorithm is applied to the collected resistance data to remove noise interference. After Gaussian filtering, the pre-processed resistance data {R'1, R'2, ..., R' 60}, the environmental data and raw material information were standardized, the temperature was converted from Celsius to Kelvin, the humidity was normalized, and the stirring speed was normalized according to the maximum speed of the equipment. After standardization, the temperature, humidity and stirring speed were unified within the scale range of [0,1].

[0048] The resistance characteristics are extracted from the pre-processed powder resistance data, including the average value, maximum value, minimum value, resistance change trend and fluctuation frequency. Combined with the environmental data and raw material information, the environmental factor change rate, stirring speed change rate and raw material dosage change rate are calculated. The extracted characteristic parameters are recorded as the characteristic vector X = [x1, x2, ..., xk ], where X represents the characteristic parameter set, k represents the total number of characteristic parameters extracted from the preprocessed powder resistance data, x1, x2, ..., x k The 1st, 2nd, ..., kth characteristic parameters respectively, wherein the environmental factor change rate includes the temperature change rate and humidity change rate inside the mixing equipment;

[0049] The calculation formula for the resistance change trend is as follows:

[0050]

[0051] Among them, ΔR' represents the change trend of resistance, R' i , R' i+1 They represent the resistance values ​​corresponding to the i-th and i+1-th data points in the preprocessed resistance data, respectively, and Δt represents the time interval of data collection;

[0052] The fluctuation frequency uses a spectrum analysis method to convert the resistance data from the time domain to the frequency domain to obtain the fluctuation frequency of the resistance;

[0053] A multiple linear regression model was used to construct a regression model with the extracted resistance characteristics, environmental factor change rate, stirring speed change rate, and raw material dosage change rate as independent variables and powder resistance value as the dependent variable. The regression model definition is as follows: R pred =β0+β1*x1+β2*x2+...+β k *x k +ε; where x1, x2, ..., x k Indicates the first, second, ..., k features obtained, β0, β1, β2, ..., β k Indicates the influence coefficient corresponding to the acquired feature, and ε indicates the error value.

[0054] According to actual production experience and historical data, the influence coefficient threshold θ is set, and the influence coefficient corresponding to the i-th feature extracted above |β i | is compared with the set influence coefficient threshold, when |β i When |>θ, the i-th feature is determined to have an impact on the powder resistance and is retained as a key factor. i When |≤θ, the feature is judged to have little effect on the resistance and is excluded, thereby obtaining the final set of key factors affecting the powder resistance {C(1,1),C(1,2),...,C(a,b)}, where a represents the number of key factors and b represents the number of features in each key factor that have an impact on the resistance.

[0055] The key factor set and the corresponding powder resistance value R are combined to form a data set related to the key factor set and the powder resistance value.

[0056] Specifically, features were extracted from the preprocessed resistance data, and the average resistance value was calculated to be 51.5Ω, the maximum value was 53Ω, and the minimum value was 50Ω. The resistance change trend was calculated as 0.8Ω / min using the resistance change trend calculation formula. The resistance data was converted from the time domain to the frequency domain using the spectrum analysis method, and the resistance fluctuation frequency was obtained as 0.5Hz. The change rate was calculated by combining the environmental data and raw material information. The temperature change rate was 0.083℃ / min, the humidity change rate was 0, and the stirring speed change rate was 0.33 rpm. 2 , the rate of change of the amount of raw material A is 0, the rate of change of the amount of raw material B is 0, and these characteristic parameters are recorded as the characteristic vector X=[x1,x2...x n ];

[0057] A multiple linear regression model was constructed using the extracted resistance characteristics, environmental factor change rate, stirring speed change rate, and raw material dosage change rate as independent variables and powder resistance value as the dependent variable.

[0058] Based on actual production experience and historical data, the influence coefficient threshold θ=1.5 is set. According to the influence coefficient of the resistance change trend |β1|=2>θ, the resistance change trend is determined to have an impact on the powder resistance and is retained as a key factor. The influence coefficient of the humidity change rate is relatively small, and each factor is judged one by one to obtain the set of key factors that affect the powder resistance {C(1,1),C(1,2),...,C(a,b)}.

[0059] Under the premise of ensuring the mixing process and product quality requirements, the key factors in the mixing process are changed to obtain a total of z groups of data on the key factor set and the powder resistance value, thereby establishing a database on the powder resistance value and the key factor set;

[0060] In the real-time monitoring of the mixing production process, the difference V between the key factors in the current production environment and the rth group of key factors in the database is calculated in real time. r , the calculation formula is as follows:

[0061]

[0062] Among them, M b represents the characteristic value of the dth key factor monitored in real time, C r,d represents the dth eigenvalue of the rth group of key factors in the database, d total Indicates the total number of key features, and total indicates the data subscript;

[0063] Substitute r=1,2,...,z into the difference calculation formula to obtain the difference set {V1,V2,...,V r ,...,V z}, the minimum difference value obtained by comparison is V min , set the maximum tolerance of difference V max ;

[0064] When the minimum difference value is greater than the set maximum tolerance value, the powder resistance abnormality alarm is triggered, indicating that there is a production abnormality and the control of key factors needs to be checked;

[0065] When the minimum difference is less than or equal to the maximum tolerance of the difference, judge V min The powder resistance values ​​in the data group are the powder resistance values ​​under the current production environment.

[0066] Specifically, in the subsequent real-time production monitoring, the difference V between the key factors in the current production environment and the rth group of key factors in the database is calculated in real time. r According to the key factors currently monitored in real time, the resistance change trend is 0.7Ω / min, the temperature change rate is 0.07℃ / min, and the stirring speed change rate is 0.3 rpm. 2 , compared with the third group of data in the database (resistance change trend is 0.9Ω / min, temperature change rate is 0.09℃ / min, stirring speed change rate is 0.35 rpm 2 ) calculates the difference to be 0.25, substitutes r from 1 to 10 into the difference calculation formula, and obtains the difference set {V1, V2, ..., V 10}, the minimum difference value is 0.25, the maximum tolerance of the difference is set to 0.3, and the calculated minimum difference value is compared with the set maximum tolerance of the difference to determine V min The powder resistance values ​​in the data group are the powder resistance values ​​under the current production environment.

[0067] A data processing system based on the dry-process electrode preparation process of lithium batteries includes a data acquisition module, a feature extraction module, a model construction module and a monitoring module. The data acquisition module is used to use a resistance measuring device to collect powder resistance data, environmental parameters and raw material information during the mixing process at set time intervals; the feature extraction module is used to extract resistance features from the pre-processed powder resistance data, and calculate the relevant change rate in combination with the environmental data and raw material information; the model construction module is used to construct a multivariate linear regression model, complete the screening of key factors, and store the processed data; the monitoring module is used to monitor the key factors in the production process in real time, calculate the difference with the data group in the database, and set a tolerance threshold. When the difference exceeds the threshold, an abnormal alarm is triggered; otherwise, the current resistance value is determined.

[0068] The data acquisition module includes a resistance data acquisition unit, an environmental parameter acquisition unit, a raw material information acquisition unit and a preprocessing unit. The resistance data acquisition unit is used to collect powder resistance data at set time intervals through a resistance measuring device to form a resistance measurement sequence; the environmental parameter acquisition unit is used to monitor environmental data such as temperature, humidity and stirring speed during the mixing process through sensors; the raw material information acquisition unit is used to record raw material type, batch and dosage information; the preprocessing unit is used to clean and standardize the collected resistance data, environmental parameters and raw material information; the output ends of the resistance data acquisition unit, environmental parameter acquisition unit and raw material information acquisition unit are connected to the input end of the preprocessing unit, and the output end of the preprocessing unit is connected to the input end of the feature extraction module.

[0069] The feature extraction module includes a resistance feature extraction unit and a change rate calculation unit. The resistance feature extraction unit is used to extract resistance features from the preprocessed powder resistance data, including the average value, maximum value, minimum value, resistance change trend and fluctuation frequency of the resistance; the change rate calculation unit is used to combine environmental data and raw material information to calculate the change rate of environmental factors, the change rate of stirring speed and the change rate of raw material usage; the output ends of the resistance feature extraction unit and the change rate calculation unit are connected to the input end of the model construction module.

[0070] The model construction module includes a regression model construction unit, a key factor screening unit and a data group construction unit. The regression model construction unit is used to use a multivariate linear regression model, take the extracted resistance characteristics, environmental factor change rate, stirring speed change rate and raw material usage change rate as independent variables, and powder resistance value as the dependent variable to construct a regression model; the key factor screening unit sets the influence coefficient threshold based on actual production experience and historical data, and compares the influence coefficient corresponding to the extracted characteristics with the set influence coefficient threshold to screen out key factors; the data group construction unit is used to combine the key factor set with the corresponding powder resistance value to form a data group about the key factor set and the powder resistance value. The output end of the regression model construction unit is connected to the input end of the key factor screening unit, and the output end of the key factor screening unit is connected to the input end of the data group construction unit.

[0071] The monitoring module includes a difference calculation unit, an optimal matching unit and an alarm judgment unit. The difference calculation unit is used to calculate the difference between the key factors in the current production environment and the key factor set in the database in real-time monitoring of the mixing production process; the optimal matching unit is used to calculate the minimum difference and determine the closest historical data group; the alarm judgment unit is used to compare the calculated minimum difference value with the set maximum tolerance of the difference. When the minimum difference value is greater than the set maximum tolerance of the difference, the powder resistance abnormality alarm is triggered, indicating that there is a production abnormality and the control of the key factors needs to be checked. When the minimum difference value is less than or equal to the set maximum tolerance of the difference, the powder resistance value in the data group where the minimum difference value is located is judged to be the powder resistance value under the current production environment. The output end of the difference calculation unit is connected to the input end of the optimal matching unit, and the output end of the alarm judgment unit is connected to the input end of the alarm judgment unit.

[0072] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A data processing method based on the dry-process electrode preparation process of lithium batteries, characterized by: The method comprises the following steps: Collect mixed material resistance data, environment and raw material information for preprocessing; Extract features from the pre-processed resistance data, calculate the relevant change rate based on the environment and raw material information, and establish a regression analysis model; Screen out the key factors that affect the resistance and form a data set containing a set of key factors and corresponding powder resistance values; Adjust controllable factors within the safety and process range to obtain multiple sets of data on key factors and resistance values. Establish a database, monitor key factors in the production process in real time, calculate the difference with the data group in the database, and set a tolerance threshold. When the difference exceeds the threshold, trigger an abnormal alarm, otherwise determine the current resistance value.

2. The data processing method according to claim 1, wherein: The resistance measurement equipment is used to collect the powder resistance data of the mixed material at set time intervals to obtain the resistance measurement value sequence {R1, R2, ..., R n }, where n represents the number of collection times. Sensors are used to collect environmental data during the mixing process, including the temperature, humidity, and stirring speed inside the mixing equipment. Raw material information, including raw material type, batch, and amount, is also collected. The Gaussian filter algorithm is applied to the collected powder resistance data to remove noise interference and obtain the preprocessed resistance data {R'1, R'2,…, R' n }, standardize environmental data and raw material information, and unify data of different magnitudes and units to the same scale.

3. The data processing method according to claim 2, wherein: The resistance characteristics are extracted from the pre-processed powder resistance data, including the average value, maximum value, minimum value, resistance change trend and fluctuation frequency. Combined with the environmental data and raw material information, the environmental factor change rate, stirring speed change rate and raw material dosage change rate are calculated. The extracted characteristic parameters are recorded as the characteristic vector X = [x1, x2, ..., x k ], where X represents the characteristic parameter set, k represents the total number of characteristic parameters extracted from the preprocessed powder resistance data, x1, x2, ..., x k The 1st, 2nd, ..., kth characteristic parameters respectively, wherein the environmental factor change rate includes the temperature change rate and humidity change rate inside the mixing equipment; A multiple linear regression model was used to construct a regression model with the extracted resistance characteristics, environmental factor change rate, stirring speed change rate, and raw material dosage change rate as independent variables and powder resistance value as the dependent variable. The regression model definition is as follows: R pred =β0+β1*x1+β2*x2+...+β k *x k +ε; where x1, x2, ..., x k Indicates the first, second, ..., k features obtained, β0, β1, β2, ..., β k Indicates the influence coefficient corresponding to the acquired feature, and ε indicates the error value.

4. The data processing method according to claim 3, wherein: According to actual production experience and historical data, the influence coefficient threshold θ is set, and the influence coefficient corresponding to the i-th feature extracted above |β i | is compared with the set influence coefficient threshold, when |β i When |>θ, the i-th feature is determined to have an impact on the powder resistance and is retained as a key factor. i When |≤θ, the feature is judged to have little effect on the resistance and is excluded, thereby obtaining the final set of key factors affecting the powder resistance {C(1,1),C(1,2),...,C(a,b)}, where a represents the number of key factors and b represents the number of features in each key factor that have an impact on the resistance. The key factor set and the corresponding powder resistance value R are combined to form a data set related to the key factor set and the powder resistance value.

5. The data processing method according to claim 4, characterized in that: Under the premise of ensuring the mixing process and product quality requirements, the key factors in the mixing process are changed to obtain a total of z groups of data on the key factor set and the powder resistance value, thereby establishing a database on the powder resistance value and the key factor set; In the real-time monitoring of the mixing production process, the difference V between the key factors in the current production environment and the rth group of key factors in the database is calculated in real time. r , the calculation formula is as follows: Among them, M b represents the characteristic value of the dth key factor monitored in real time, C r,d represents the dth eigenvalue of the rth group of key factors in the database, d total Indicates the total number of key features, and total indicates the data subscript; Substitute r=1,2,...,z into the difference calculation formula to obtain the difference set {V1,V2,...,V r ,...,V z }, the minimum difference value obtained by comparison is V min , set the maximum tolerance of difference V max ; When the minimum difference value is greater than the set maximum tolerance value, the powder resistance abnormality alarm is triggered, indicating that there is a production abnormality and the control of key factors needs to be checked; When the minimum difference is less than or equal to the maximum tolerance of the difference, judge V min The powder resistance values ​​in the data group are the powder resistance values ​​under the current production environment.

6. A data processing system based on the dry-process electrode preparation process for lithium batteries, characterized by: The system includes a data acquisition module, a feature extraction module, a model construction module and a monitoring module. The data acquisition module is used to use a resistance measuring device to collect powder resistance data, environmental parameters and raw material information during the mixing process at set time intervals; the feature extraction module is used to extract resistance features from the preprocessed powder resistance data, and calculate the relevant change rate in combination with the environmental data and raw material information; the model construction module is used to construct a multivariate linear regression model to complete the screening of key factors and store the processed data; the monitoring module is used to monitor the key factors in the production process in real time, calculate the difference with the data group in the database, and set a tolerance threshold. When the difference exceeds the threshold, an abnormal alarm is triggered; otherwise, the current resistance value is determined.

7. The data processing system based on the dry process of preparing lithium battery electrodes according to claim 6, characterized in that: The data acquisition module includes a resistance data acquisition unit, an environmental parameter acquisition unit, a raw material information acquisition unit, and a preprocessing unit. The resistance data acquisition unit is used to collect powder resistance data at set time intervals using a resistance measuring device to form a resistance measurement sequence; the environmental parameter acquisition unit is used to monitor environmental data such as temperature, humidity, and stirring speed during the mixing process using sensors; the raw material information acquisition unit is used to record raw material type, batch, and usage information; and the preprocessing unit is used to clean and standardize the collected resistance data, environmental parameters, and raw material information. The output ends of the resistance data acquisition unit, the environmental parameter acquisition unit and the raw material information acquisition unit are connected to the input end of the preprocessing unit, and the output end of the preprocessing unit is connected to the input end of the feature extraction module.

8. The data processing system based on the dry process of preparing lithium battery electrodes according to claim 7, characterized in that: The feature extraction module includes a resistance feature extraction unit and a change rate calculation unit. The resistance feature extraction unit is used to extract resistance features from the preprocessed powder resistance data, including the average value, maximum value, minimum value, resistance change trend and fluctuation frequency of the resistance; the change rate calculation unit is used to combine environmental data and raw material information to calculate the change rate of environmental factors, the change rate of stirring speed and the change rate of raw material usage; the output ends of the resistance feature extraction unit and the change rate calculation unit are connected to the input end of the model construction module.

9. The data processing system based on the dry process of preparing lithium battery electrodes according to claim 8, characterized in that: The model construction module includes a regression model construction unit, a key factor screening unit and a data group construction unit. The regression model construction unit is used to use a multivariate linear regression model, take the extracted resistance characteristics, environmental factor change rate, stirring speed change rate and raw material usage change rate as independent variables, and powder resistance value as the dependent variable to construct a regression model; the key factor screening unit sets the influence coefficient threshold based on actual production experience and historical data, and compares the influence coefficient corresponding to the extracted characteristics with the set influence coefficient threshold to screen out key factors; the data group construction unit is used to combine the key factor set with the corresponding powder resistance value to form a data group about the key factor set and the powder resistance value. The output end of the regression model construction unit is connected to the input end of the key factor screening unit, and the output end of the key factor screening unit is connected to the input end of the data group construction unit.

10. The data processing system based on the dry-process lithium battery electrode preparation process according to claim 9, characterized in that: The monitoring module includes a difference calculation unit, an optimal matching unit and an alarm judgment unit. The difference calculation unit is used to calculate the difference between the key factors in the current production environment and the key factor set in the database in real-time monitoring of the mixing production process; the optimal matching unit is used to calculate the minimum difference and determine the closest historical data group; the alarm judgment unit is used to compare the calculated minimum difference value with the set maximum tolerance of the difference. When the minimum difference value is greater than the set maximum tolerance of the difference, the powder resistance abnormality alarm is triggered, indicating that there is a production abnormality and the control of the key factors needs to be checked. When the minimum difference value is less than or equal to the set maximum tolerance of the difference, the powder resistance value in the data group where the minimum difference value is located is judged to be the powder resistance value under the current production environment. The output end of the difference calculation unit is connected to the input end of the optimal matching unit, and the output end of the alarm judgment unit is connected to the input end of the alarm judgment unit.

Images