A multi-stage pressure heat sealing system and method for pouch cells

Abstract

The application discloses a kind of multistage pressure heat sealing systems and methods of soft package battery, it is related to battery manufacturing technical field;Method includes first according to preset first pressure value and first temperature value to soft package battery is pre-sealed, obtains first soft package battery;Subsequently, first soft package battery is extracted vacuum and obtains second soft package battery, and it is divided into first partition and second partition;Subsequently, by first, second pressure heat sealing model, the second, third pressure values corresponding to two partitions are calculated respectively;Finally, in combination with preset second, third temperature value, heat sealing operation is carried out to two partitions, and the heat sealing of soft package battery is completed.The method first pre-seals for subsequent heat sealing to lay foundation, then in combination with partition design and accurate pressure calculation, different heat sealing control is realized for the characteristic difference of different regions of soft package battery, both avoid the limitations of traditional single parameter heat sealing, and also solve the problem of uneven heat sealing caused by lack of scientific basis of partition, to improve the sealing of soft package battery package.

Description

Technical Field

[0001] This invention belongs to the field of battery manufacturing technology, specifically relating to a multi-stage pressure heat sealing system and method for pouch batteries. Background Technology

[0002] With the rapid development of new energy vehicles and energy storage, the manufacturing of pouch batteries has received increasing attention, with heat sealing directly affecting battery quality. However, as pouch batteries evolve towards higher capacity and thinner profiles, traditional heat sealing processes are gradually revealing their bottlenecks. Traditional heat sealing uses a single parameter to process the battery, failing to consider the characteristic differences between the tab area and the non-tab area, easily leading to problems such as poor sealing. Existing partitioned heat sealing technologies either rely on manual or geometric methods to divide the area, lacking scientific quantitative basis; or the pressure model does not incorporate real-time deformation data or ignores the nonlinear relationship between thickness and pressure, resulting in poor heat sealing uniformity. Moreover, traditional clustering algorithms have low accuracy in dividing feature clusters, making it difficult to adapt to the needs of partitioning. Therefore, developing a multi-level pressure heat sealing method with precise partitioning and dynamic pressure regulation has become key to solving the current pain points.

[0003] Patent CN115799599A discloses a method for preparing an in-situ cured soft-pack battery and the in-situ cured soft-pack battery itself. The method includes stacking a positive electrode, a negative electrode, and a separator, followed by tab welding and aluminum-plastic film encapsulation to obtain a battery cell; injecting a polymeric electrolyte into the battery cell and performing a first vacuuming; placing the product from the previous step in a pressure vessel and, under a first constant temperature and pressure, allowing the polymeric electrolyte to wet the battery cell; removing the wetted battery cell from the pressure vessel, allowing it to return to room temperature, and then allowing it to stand; performing a second vacuuming on the stood battery cell; and, under a second constant temperature and varying pressure, achieving in-situ curing of the polymeric electrolyte to obtain the in-situ cured soft-pack battery. This invention can effectively improve the uniformity of polymer electrolyte distribution in the electrodes and has good prospects for industrial application. However, this invention does not address problems in the heat-sealing process of soft-pack batteries, such as differences in characteristics of different regions, lack of scientific quantitative basis for zoning, and poor heat-sealing uniformity, thus affecting the sealing performance of the soft-pack battery. Summary of the Invention

[0004] The purpose of this invention is to solve the problem of poor sealing performance caused by using only the same pressure to heat seal pouch batteries in the prior art, and to propose a multi-stage pressure heat sealing system and method for pouch batteries.

[0005] In a first aspect of this invention, a multi-stage pressure heat sealing method for a pouch cell is first proposed, the method comprising:

[0006] The first soft-pack battery is obtained by pre-sealing the soft-pack battery by setting a first pressure value and a first temperature value.

[0007] A vacuuming operation is performed on the first pouch cell to obtain a second pouch cell;

[0008] The second pouch cell is divided into regions to obtain a first region and a second region; the first region is a region susceptible to temperature effects; the second region is a region unaffected by temperature effects.

[0009] The second pressure value is obtained by calculating the pressure of the first partition using the first pressure heat sealing model;

[0010] The third pressure value is obtained by calculating the pressure of the second partition using the second pressure heat sealing model;

[0011] The first section of the second soft-pack battery is heat-sealed according to the second pressure value and the preset second temperature value.

[0012] The second section of the second soft-pack battery is heat-sealed according to the third pressure value and the preset third temperature value.

[0013] Optionally, pressure data and deformation data can be obtained by collecting pressure deformation data point by point on the second packaged soft-pack battery through the heat sealing head;

[0014] A feature parameter set is obtained by extracting features from the pressure data and the deformation data;

[0015] The feature parameter dataset is obtained by statistically analyzing the feature parameters of all collected locations.

[0016] A novel clustering algorithm is used to cluster and group the feature parameter dataset to obtain a first feature cluster and a second feature cluster; the first feature cluster consists of feature parameter data from regions affected by temperature; the second feature cluster consists of feature parameter data from regions unaffected by temperature.

[0017] The second packaged soft-pack battery is partitioned into a first partition and a second partition based on the first feature cluster and the second feature cluster.

[0018] Optionally, an analysis of variance is performed on the data corresponding to each feature parameter type in the feature parameter dataset to obtain an initial feature parameter weight set;

[0019] Local feature parameter weight sets are obtained by performing local variance analysis on the feature parameter dataset using a sliding window.

[0020] The initial parameter weight set and the local feature parameter weight set are fused to obtain a comprehensive feature parameter weight set;

[0021] Based on the comprehensive feature parameter weight set, the distance between every two feature parameter data in the feature parameter dataset is calculated to obtain the comprehensive distance set;

[0022] The local outlier factor value is calculated on the feature parameter dataset to obtain the local outlier value set of the feature parameters;

[0023] The first cluster center and the second cluster center are obtained by taking the two feature parameter data where the local outlier value of the feature parameter is greater than the preset density value and the comprehensive distance in the comprehensive distance set is greater than the preset distance as the cluster center;

[0024] The clustering distance value set is obtained by performing clustering calculations on all remaining feature parameter data in the feature parameter dataset with the first cluster center and the second cluster center;

[0025] Based on the clustering distance values ​​in the clustering distance value set, the feature parameter data are divided into corresponding clusters to obtain the first feature cluster and the second feature cluster.

[0026] Optionally, the real-time deformation rate can be obtained by comparing the real-time deformation of the first partition with the vacuuming time.

[0027] The real-time deformation rate is compared with a preset deformation rate threshold; if the real-time deformation rate is greater than the preset deformation rate threshold, then the first function P1=P_base×(1-α×) is used. ) Calculate the second pressure value; if the real-time deformation rate is less than or equal to the preset deformation rate threshold, then use the second function P1=P_base×(1+β× ) Calculate the second pressure value; where P_base is the base pressure value based on the preset partition material and thickness, v is the real-time deformation rate, v0 is the deformation rate threshold, and α and β are adjustment coefficients greater than zero.

[0028] Optionally, a second pressure heat sealing model can be obtained by constructing a nonlinear regression model for the second partition based on the mapping relationship between the thickness distribution of the second partition and the heat sealing pressure.

[0029] The expression for the second pressure heat sealing model is:

[0030] P2=P0×(a× +c× ), where P0 is the reference pressure value, The average thickness is the reference, T0 is the reference temperature value, P2 is the third pressure value, T is the preset third temperature value, δ is the average thickness of the second zone, and a, b and c are regression coefficients with a+c=1.

[0031] In a second aspect of this invention, a multi-stage pressure heat sealing system for a pouch cell is provided, comprising:

[0032] The first soft-pack battery pre-sealing module is used to pre-seal the soft-pack battery by pre-setting a first pressure value and a first temperature value to obtain the first soft-pack battery.

[0033] A vacuum module is used to perform a vacuuming operation on the first soft-pack battery to obtain a second soft-pack battery.

[0034] The partitioning module is used to divide the second pouch battery into regions to obtain a first partition and a second partition; the first partition is a region susceptible to temperature effects; the second partition is a region unaffected by temperature effects.

[0035] The second pressure value calculation module is used to calculate the pressure of the first partition using the first pressure heat sealing model to obtain the second pressure value;

[0036] The third pressure value calculation module is used to calculate the pressure of the second partition using the second pressure heat sealing model to obtain the third pressure value;

[0037] The first partition heat sealing module is used to heat seal the first partition of the second soft-pack battery according to the second pressure value and the preset second temperature value.

[0038] The second partition heat sealing module is used to heat seal the second partition of the second soft-pack battery according to the third pressure value and the preset third temperature value.

[0039] Optionally, the partitioning module includes:

[0040] The data acquisition module is used to collect pressure and deformation data point by point from the second packaged soft-pack battery through the heat sealing head.

[0041] The feature parameter extraction module is used to extract features from the pressure data and the deformation data to obtain a feature parameter set;

[0042] The data statistics module is used to statistically analyze the feature parameters of all collected locations to obtain a feature parameter dataset.

[0043] The clustering module is used to cluster the feature parameter dataset using a novel clustering algorithm to obtain a first feature cluster and a second feature cluster; the first feature cluster consists of feature parameter data from regions affected by temperature; the second feature cluster consists of feature parameter data from regions unaffected by temperature.

[0044] The partitioning module is used to partition the second packaged soft-pack battery according to the first feature cluster and the second feature cluster to obtain a first partition and a second partition.

[0045] Optionally, the clustering module includes:

[0046] The initial weight set generation module is used to perform variance analysis on the data corresponding to each feature parameter type in the feature parameter dataset to obtain the initial feature parameter weight set.

[0047] The local weight set generation module is used to perform local variance analysis on the feature parameter dataset through a sliding window to obtain a local feature parameter weight set.

[0048] The comprehensive weight set generation module is used to fuse the initial parameter weight set and the local feature parameter weight set to obtain a comprehensive feature parameter weight set;

[0049] The comprehensive distance set calculation module is used to calculate the distance between every two feature parameter data in the feature parameter dataset according to the comprehensive feature parameter weight set to obtain the comprehensive distance set.

[0050] The local outlier set calculation module is used to calculate the local outlier factor value of the feature parameter dataset to obtain the local outlier set of feature parameters;

[0051] The cluster center generation module is used to take two feature parameter data in the feature parameter local outlier value set where the local outlier value of the feature parameter is greater than a preset density value and the comprehensive distance in the comprehensive distance set is greater than a preset distance as cluster centers to obtain the first cluster center and the second cluster center;

[0052] The clustering distance value calculation module is used to perform clustering calculations with the first cluster center and the second cluster center in the feature parameter dataset to obtain a clustering distance value set;

[0053] The feature cluster generation module is used to divide the feature parameter data into corresponding clusters based on the cluster distance values ​​in the cluster distance value set to obtain the first feature cluster and the second feature cluster.

[0054] Optionally, the second pressure value calculation module includes:

[0055] The real-time deformation rate module is used to obtain the real-time deformation rate by measuring the ratio of the real-time deformation of the first partition to the vacuuming time.

[0056] The second calculation module is used to obtain the ratio of the real-time deformation of the first partition to the vacuuming time to obtain the real-time deformation rate.

[0057] The real-time deformation rate is compared with a preset deformation rate threshold; if the real-time deformation rate is greater than the preset deformation rate threshold, then the first function P1=P_base×(1-α×) is used. ) Calculate the second pressure value; if the real-time deformation rate is less than or equal to the preset deformation rate threshold, then use the second function P1=P_base×(1+β× ) Calculate the second pressure value; where P_base is the base pressure value based on the preset partition material and thickness, v is the real-time deformation rate, v0 is the deformation rate threshold, and α and β are adjustment coefficients greater than zero.

[0058] Optionally, the third pressure value calculation module includes:

[0059] The model building module is used to construct a nonlinear regression model for the second partition based on the mapping relationship between the thickness distribution of the second partition and the heat sealing pressure to obtain the second pressure heat sealing model;

[0060] The expression for the second pressure heat sealing model is:

[0061] P2=P0×(a× +c× ), where P0 is the reference pressure value, The average thickness is the reference, T0 is the reference temperature value, P2 is the third pressure value, T is the preset third temperature value, δ is the average thickness of the second zone, and a, b and c are regression coefficients with a+c=1.

[0062] The beneficial effects of this invention are as follows: This invention proposes a method that involves pre-sealing a pouch battery according to preset first pressure and first temperature values ​​to obtain a first pouch battery; then, evacuating the first pouch battery to obtain a second pouch battery, which is then divided into a first zone and a second zone; subsequently, using first and second pressure heat-sealing models, the second and third pressure values ​​corresponding to the two zones are calculated respectively; finally, combined with preset second and third temperature values, heat-sealing is performed on the two zones to complete the heat sealing of the pouch battery. This method first lays a stable foundation for subsequent heat sealing through pre-sealing, and then, by combining zone design and precise pressure calculation, achieves differentiated heat sealing control based on the characteristic differences of different areas of the pouch battery. This avoids the limitations of traditional single-parameter heat sealing and solves the problem of uneven heat sealing caused by the lack of scientific basis for zone division, thereby improving the sealing performance of pouch battery packaging. Attached Figure Description

[0063] The invention will now be further described with reference to the accompanying drawings.

[0064] Figure 1 A flowchart illustrating a multi-stage pressure heat sealing method for a pouch battery provided in an embodiment of the present invention;

[0065] Figure 2 This is a framework diagram of a multi-stage pressure heat sealing system for a pouch battery provided in an embodiment of the present invention. Detailed Implementation

[0066] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0067] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] This invention provides a multi-stage pressure heat sealing method for pouch batteries. See also... Figure 1 , Figure 1 A flowchart illustrating a multi-stage pressure heat sealing method for a pouch cell provided in an embodiment of the present invention. The method includes the following steps:

[0069] S101, the first soft-pack battery is obtained by pre-sealing the soft-pack battery by setting a first pressure value and a first temperature value;

[0070] S102, perform a vacuuming operation on the first pouch battery to obtain the second pouch battery;

[0071] S103, the second soft-pack battery is divided into regions to obtain a first partition and a second partition;

[0072] S104, the second pressure value is obtained by calculating the pressure of the first zone using the first pressure heat sealing model;

[0073] S105, the third pressure value is obtained by calculating the pressure of the second zone using the second pressure heat sealing model;

[0074] S106, heat seal the first section of the second soft-pack battery according to the second pressure value and the preset second temperature value;

[0075] S107, heat seal the second section of the second soft pack battery according to the third pressure value and the preset third temperature value;

[0076] The first zone is a region susceptible to temperature changes; the second zone is a region unaffected by temperature changes.

[0077] This invention provides a multi-stage pressure heat-sealing method for pouch batteries. The method involves first pre-sealing the pouch battery according to preset first pressure and first temperature values ​​to obtain a first pouch battery. Then, the first pouch battery is evacuated to obtain a second pouch battery, which is then divided into a first zone and a second zone. Subsequently, using first and second pressure heat-sealing models, second and third pressure values ​​corresponding to the two zones are calculated respectively. Finally, combined with preset second and third temperature values, heat-sealing is performed on the two zones to complete the heat sealing of the pouch battery. This method first lays a stable foundation for subsequent heat sealing through pre-sealing, and then, by combining zone design and precise pressure calculation, achieves differentiated heat sealing control based on the characteristic differences of different regions of the pouch battery. This avoids the limitations of traditional single-parameter heat sealing and solves the problem of uneven heat sealing caused by a lack of scientific basis for zone division, thereby improving the sealing performance of the pouch battery packaging.

[0078] In one implementation, the preset first pressure value, preset first temperature value, preset second temperature value, and preset third temperature value are determined by a technician.

[0079] In one implementation, firstly, based on the characteristics of the inner heat-sealing layer of the aluminum-plastic film used in the pouch battery, the overall size of the battery, and the distribution of the electrolyte after injection, a suitable first temperature value and a first pressure value are preset. The first temperature value needs to be precisely controlled between 130-160℃ and the heat resistance limit of the aluminum foil to ensure that the heat-sealing layer can fully melt without damaging the outer structure. The preset first pressure value needs to ensure that the molten heat-sealing layer is tightly adhered while avoiding compression of the battery's internal electrode core, which could lead to deformation or electrolyte leakage. Subsequently, the pouch battery with completed electrolyte injection is placed in the pressing mold of the pre-sealing equipment. The mold is heated to the target temperature according to the preset first temperature value, and the first pressure value is applied to press the edge area of ​​the aluminum-plastic film, so that the inner heat-sealing layer melts and bonds under high temperature and high pressure, forming a preliminary sealed cavity structure. Finally, a first pouch battery with pre-fixed edges and internal connectivity to external vacuum equipment is obtained. This process avoids the aluminum-plastic film from collapsing and wrinkling due to the lack of a fixed structure during subsequent vacuuming, and also prevents electrolyte leakage during the vacuuming process.

[0080] In one implementation, a first pouch cell is transferred to a vacuum pump. The device connects to a pre-reserved vacuum channel in the battery via a sealed interface, ensuring a seamless seal between the interface and the aluminum-plastic film to prevent external air ingress. The device is then activated to evacuate the battery's internal cavity using negative pressure. The evacuation rate must be controlled to avoid excessive local negative pressure that could cause core misalignment or electrolyte adsorption at the channel opening. This ensures the removal of air and electrolyte volatiles between the core layers and between the aluminum-plastic film and the core without damaging the internal structure. When the device detects that the vacuum level has reached a preset standard, vacuuming stops. At this point, the battery's interior is nearly vacuum-like and briefly maintained by the pre-sealed structure, ultimately resulting in a second pouch cell with internal gas removal and preliminary sealing. This process reduces the risk of battery bulging during cycling, improves electrolyte and core wetting, and ensures battery capacity and cycle life.

[0081] In one implementation, the battery is first pre-sealed at a preset pressure and temperature, then vacuumed. Next, the battery is divided into different regions. For each region, the pressure is calculated using a corresponding pressure heat-sealing model. Finally, heat-sealing is performed on each region at a preset temperature. This tiered pressure heat-sealing method allows for precise pressure application based on the characteristics of different regions, ensuring each region is heat-sealed under appropriate pressure conditions. This significantly improves the accuracy and reliability of the heat-sealing process, effectively avoiding sealing defects caused by uneven pressure, and strongly guaranteeing the sealing quality of the pouch battery, thereby optimizing battery performance and lifespan.

[0082] In one embodiment, pressure data and deformation data are obtained by collecting pressure deformation data point by point on the second packaged soft-pack battery through a heat sealing head.

[0083] A feature parameter set is obtained by extracting features from pressure data and deformation data;

[0084] The feature parameter dataset is obtained by statistically analyzing the feature parameters of all collected locations.

[0085] A novel clustering algorithm is used to cluster and group the feature parameter dataset to obtain a first feature cluster and a second feature cluster; the first feature cluster consists of feature parameter data from regions affected by temperature; the second feature cluster consists of feature parameter data from regions unaffected by temperature.

[0086] The second packaged soft-pack battery is divided into a first partition and a second partition based on the first feature cluster and the second feature cluster.

[0087] In one implementation, when collecting pressure deformation data point by point on the second packaged soft-pack battery through the heat sealing head, the heat sealing head applies specific pressure to each point on the battery surface according to the preset collection path and interval, and records the pressure magnitude (pressure data) and deformation data generated under the pressure at each point in real time, thereby comprehensively obtaining the response of different positions of the battery under pressure.

[0088] In one implementation, when extracting features based on pressure and deformation data, each set of collected pressure and deformation data is analyzed to extract key indicators that reflect the mechanical properties and deformation patterns at that location: local stiffness coefficient, deformation relaxation rate, deformation uniformity index, and area under the pressure-deformation curve. These indicators are then integrated to form a feature parameter set, simplifying the original data and highlighting key information. When statistically analyzing the feature parameter data for all collected locations, the indicators in the feature parameter set corresponding to each collection point are summarized and arranged according to a unified format and order, forming a set containing feature information from all points, i.e., a feature parameter dataset. This provides a complete and systematic data foundation for subsequent cluster analysis.

[0089] In one implementation, when dividing the second packaged soft-pack battery into regions based on the first feature cluster and the second feature cluster, the physical locations of all the sampling points belonging to the first feature cluster on the battery are connected and integrated to form a continuous region, which is the first region; similarly, the physical locations of all the sampling points belonging to the second feature cluster are integrated to form the second region, thereby completing the region division of the battery.

[0090] In one implementation, pressure deformation data is collected point by point through the heat sealing head. After feature extraction and statistical analysis, a dataset is formed. Then, a novel clustering algorithm is used to cluster and group the data to divide the battery into zones. The advantage is that it can accurately and meticulously capture the pressure and deformation characteristics of different locations on the battery. Based on these characteristics, the zones are scientifically and rationally divided, so that when appropriate heat sealing parameters are used for different zones, they can better fit the actual conditions of each area. This improves the uniformity and reliability of heat sealing, effectively avoids sealing defects caused by mismatched heat sealing parameters, ensures the sealing quality of the soft-pack battery, and thus optimizes battery performance and service life.

[0091] In one embodiment, variance analysis is performed on the data corresponding to each feature parameter type in the feature parameter dataset to obtain an initial feature parameter weight set;

[0092] Local feature parameter weights are obtained by performing local variance analysis on the feature parameter dataset using a sliding window.

[0093] The initial parameter weight set and the local feature parameter weight set are fused to obtain the comprehensive feature parameter weight set;

[0094] The comprehensive distance set is obtained by calculating the distance between every two feature parameter data in the feature parameter dataset based on the comprehensive feature parameter weight set;

[0095] The local outlier factor is calculated on the feature parameter dataset to obtain the local outlier value set of the feature parameters;

[0096] The first cluster center and the second cluster center are obtained by taking the two feature parameter data where the local outlier value of the feature parameter is greater than the preset density value and the comprehensive distance in the comprehensive distance set is greater than the preset distance.

[0097] The cluster distance value set is obtained by performing clustering calculations on all remaining feature parameter data in the feature parameter dataset with the first cluster center and the second cluster center;

[0098] Based on the cluster distance values ​​in the cluster distance value set, the feature parameter data are divided into corresponding clusters to obtain the first feature cluster and the second feature cluster.

[0099] In one implementation, when performing ANOVA on the data corresponding to each feature parameter type in the feature parameter dataset, the dispersion of each feature parameter across all data points is calculated. The contribution of the feature parameter to the data differences is evaluated by the variance magnitude; a larger variance indicates a more significant role for the feature parameter in distinguishing data. Based on this, the contribution of each feature parameter is quantified into a weight value, forming an initial feature parameter weight set. When performing local ANOVA on the feature parameter dataset using a sliding window, a fixed-size sliding window moves segment by segment in the dataset, calculating the local variance of each feature parameter within each window. This reflects the dispersion characteristics of the feature parameter in different local data regions. The weight of each feature parameter in the corresponding local region is then determined based on the magnitude of the local variance, and the summaries yield a local feature parameter weight set. When merging the initial parameter weight set and the local feature parameter weight set, a preset fusion strategy, such as weighted average or rule combination, is used to integrate the weight values ​​of the corresponding feature parameters in the two weight sets, balancing the influence of global statistical characteristics and local regional characteristics on feature importance. Finally, a comprehensive feature parameter weight set that comprehensively reflects the overall and local importance of the feature parameters is obtained.

[0100] In one implementation, when calculating the distance between any two feature parameters in the feature parameter dataset based on the comprehensive feature parameter weight set, the contribution of different feature parameters in the distance calculation is weighted according to the comprehensive weight. Methods such as weighted Euclidean distance and weighted Manhattan distance are used to calculate the weighted distance between any two data points. All calculation results are then organized into a comprehensive distance set to more accurately reflect the actual differences between data points.

[0101] In one implementation, when calculating the local outlier factor value of the feature parameter dataset, the density relationship between each data point and its surrounding neighboring data points is analyzed to assess the degree to which the data point deviates from the local dense region. The higher the degree of outlier, the larger the local outlier factor value. The local outlier factor values ​​of all data points are summarized to form a feature parameter local outlier value set, which is used to identify possible outliers or potential cluster centers.

[0102] In one implementation, when using two feature parameter data points where the local outlier value in the feature parameter set is greater than the preset density value and the comprehensive distance in the feature distance set is greater than the preset distance as cluster centers, candidate points with a high degree of outlierness are first screened out, and then two points that are far apart from each other are selected from these candidate points to ensure that they can serve as initial cluster centers with obvious differences, namely the first cluster center and the second cluster center.

[0103] In one implementation, when clustering all remaining feature parameter data in the feature parameter dataset with the first and second cluster centers, the distance between each remaining data point and the two cluster centers is calculated separately, resulting in a distance value from each data point to the two centers. All these distance values ​​together constitute a cluster distance value set, providing a basis for subsequent data attribution determination. When dividing the feature parameter data into corresponding clusters based on the cluster distance values ​​in the cluster distance value set, the distance from each data point to the first and second cluster centers is compared, and the data point is assigned to the cluster containing the closer cluster center, ultimately forming two relatively concentrated datasets, namely the first feature cluster and the second feature cluster, completing the clustering and grouping process.

[0104] In one implementation, a comprehensive weight is determined by combining analysis of variance with sliding window local analysis, which can comprehensively consider the overall and local importance of feature parameters. The distance between feature parameters is calculated based on the comprehensive weight, and cluster centers are selected by combining local outlier factors, which can accurately capture the core features of data distribution and the influence of outliers. Finally, data is grouped according to the cluster distance, which makes the obtained feature clusters more in line with the inherent distribution law of the data, and the division results are more scientific and accurate. This provides a reliable basis for subsequent cluster-based partitioning and improves the accuracy and rationality of the overall process.

[0105] In one embodiment, the real-time deformation rate is obtained by obtaining the ratio of the real-time deformation of the first partition to the vacuuming time.

[0106] The real-time deformation rate is compared with a preset deformation rate threshold; if the real-time deformation rate is greater than the preset deformation rate threshold, then the first function P1=P_base×(1-α×) is applied. ) Calculate the second pressure value; if the real-time deformation rate is less than or equal to the preset deformation rate threshold, then use the second function P1=P_base×(1+β× ) Calculate the second pressure value; where P_base is the base pressure value based on the preset partition material and thickness, v is the real-time deformation rate, v0 is the deformation rate threshold, and α and β are adjustment coefficients greater than zero.

[0107] In one implementation, the values ​​of α and β are fitted based on historical experience.

[0108] In one implementation, when obtaining the ratio of the real-time deformation of the first partition to the vacuuming time to obtain the real-time deformation rate, the deformation change of the first partition during the heat sealing process is monitored in real time, and the corresponding vacuuming time is recorded. The real-time deformation at the same moment is divided by the vacuuming time at that moment to obtain the real-time deformation rate that reflects the deformation speed of the partition, thereby quantifying the speed of deformation.

[0109] In one implementation, if the real-time deformation rate is greater than a preset deformation rate threshold, i.e., v>v0, then the first function P1=P_base×(1-α×) is used. The second pressure value is calculated, where P_base is a pre-set base pressure value based on the material properties and thickness of the partition, and α is a positive zero adjustment coefficient. At this point, an excessively fast deformation rate is suppressed by adding a pressure value proportional to the excess deformation rate to the base pressure. If the real-time deformation rate is less than or equal to the preset deformation rate threshold, i.e., v ≤ v0, the second function P1 = P_base × (1 + β × ... Calculate the second pressure value, where β is a positive adjustment coefficient. At this point, by subtracting the pressure value proportional to the insufficient deformation rate from the base pressure, deformation is appropriately promoted to ensure the stability of the heat sealing process.

[0110] In one implementation, by monitoring the deformation rate of the first partition in real time and comparing it with a preset threshold, the deformation state during the heat sealing process can be dynamically sensed. Then, according to different rate conditions, the pressure value is calculated using the corresponding function, which can achieve precise adaptive adjustment of the pressure. When the deformation rate is too fast, the pressure is increased to suppress excessive deformation, and when it is too slow, the pressure is appropriately reduced to avoid excessive pressure. This ensures that the heat sealing pressure always matches the real-time deformation state of the partition, effectively guaranteeing the stability and consistency of the heat sealing process and improving the reliability of the packaging quality.

[0111] In one embodiment, a nonlinear regression model is constructed for the second partition based on the mapping relationship between the thickness distribution of the second partition and the heat sealing pressure to obtain the second pressure heat sealing model;

[0112] The expression for the second pressure heat sealing model is:

[0113] P2=P0×(a× +c× ), where P0 is the reference pressure value, The average thickness is the reference, T0 is the reference temperature value, P2 is the third pressure value, T is the preset third temperature value, δ is the average thickness of the second zone, and a, b and c are regression coefficients with a+c=1.

[0114] In one implementation, P0, The values ​​of T0 are set by the technicians. a, b, and c are model parameters obtained through nonlinear regression analysis, used to fit the effects of thickness and temperature on pressure. The values ​​of a, b, and c are obtained by fitting historical data.

[0115] In one implementation, when constructing a nonlinear regression model for the second partition based on the mapping relationship between the thickness distribution and the heat sealing pressure to obtain the second pressure heat sealing model, the thickness data at different locations of the second partition are first collected to form thickness distribution information. At the same time, the pressure values ​​applied during actual heat sealing under these thickness distributions are recorded to establish the correspondence between the thickness distribution and the heat sealing pressure. Then, based on these mapping relationships, a suitable nonlinear regression algorithm is selected, with thickness distribution-related parameters and preset temperature as independent variables and heat sealing pressure as the dependent variable. The algorithm is used to fit and train the data to construct a model that can describe the nonlinear relationship between them, thus obtaining the second pressure heat sealing model.

[0116] In one implementation, this method of constructing a second pressure heat-sealing model for the second zone directly correlates the thickness distribution of the non-tab area with the heat-sealing pressure, and accurately quantifies the relationship between the two through a nonlinear regression model. The model expression incorporates preset temperature and average thickness parameters, and combined with regression coefficients, dynamically calculates the appropriate second pressure value. This ensures that the pressure adjustment closely matches the physical characteristics of the non-tab area, avoiding deviations from empirical pressure settings, improving the matching degree between the heat-sealing pressure and the actual conditions of the area, thereby guaranteeing the heat-sealing quality of the non-tab area and enhancing the scientific rigor and precision of the overall packaging process.

[0117] Based on the same inventive concept, embodiments of the present invention also provide a multi-stage pressure heat sealing system for a pouch battery. See also Figure 2 , Figure 2 A framework diagram of a multi-stage pressure heat sealing system for a pouch battery provided in an embodiment of the present invention includes:

[0118] The first soft-pack battery pre-sealing module is used to pre-seal the soft-pack battery by pre-setting a first pressure value and a first temperature value to obtain the first soft-pack battery.

[0119] A vacuum module is used to perform a vacuuming operation on the first pouch cell to obtain a second pouch cell.

[0120] The partitioning module is used to divide the second pouch battery into regions to obtain a first partition and a second partition; the first partition is a region susceptible to temperature effects; the second partition is a region unaffected by temperature effects.

[0121] The second pressure value calculation module is used to calculate the pressure of the first zone using the first pressure heat sealing model to obtain the second pressure value;

[0122] The third pressure value calculation module is used to calculate the pressure of the second zone using the second pressure heat sealing model to obtain the third pressure value;

[0123] The first partition heat sealing module is used to heat seal the first partition of the second soft pack battery according to the second pressure value and the preset second temperature value.

[0124] The second partition heat sealing module is used to heat seal the second partition of the second soft-pack battery according to the third pressure value and the preset third temperature value.

[0125] This invention provides a multi-stage pressure heat-sealing system for pouch batteries. The system first pre-seales the pouch battery according to preset first pressure and first temperature values ​​to obtain a first pouch battery. Then, it evacuates the first pouch battery to obtain a second pouch battery, which is then divided into a first zone and a second zone. Subsequently, using first and second pressure heat-sealing models, second and third pressure values ​​corresponding to the two zones are calculated respectively. Finally, combined with preset second and third temperature values, heat-sealing is performed on the two zones to complete the heat sealing of the pouch battery. This method first lays a stable foundation for subsequent heat sealing through pre-sealing, and then, combined with zone design and precise pressure calculation, achieves differentiated heat sealing control based on the characteristic differences of different areas of the pouch battery. This avoids the limitations of traditional single-parameter heat sealing and solves the problem of uneven heat sealing caused by a lack of scientific basis for zoning, thereby improving the sealing performance of the pouch battery packaging.

[0126] In one embodiment, the acquisition module is used to collect pressure and deformation data point by point from the second packaged soft-pack battery through the heat sealing head.

[0127] The feature parameter extraction module is used to extract features from pressure data and deformation data to obtain a set of feature parameters.

[0128] The data statistics module is used to statistically analyze the feature parameters of all collected locations to obtain a feature parameter dataset.

[0129] The clustering module is used to cluster the feature parameter dataset using a novel clustering algorithm to obtain a first feature cluster and a second feature cluster; the first feature cluster consists of feature parameter data from regions affected by temperature; the second feature cluster consists of feature parameter data from regions unaffected by temperature.

[0130] The partitioning module is used to partition the second packaged soft-pack battery according to the first feature cluster and the second feature cluster to obtain the first partition and the second partition.

[0131] In one embodiment, the clustering module includes:

[0132] The initial weight set generation module is used to perform variance analysis on the data corresponding to each feature parameter type in the feature parameter dataset to obtain the initial feature parameter weight set.

[0133] The local weight set generation module is used to obtain the local feature parameter weight set by performing local variance analysis on the feature parameter dataset through a sliding window.

[0134] The comprehensive weight set generation module is used to fuse the initial parameter weight set and the local feature parameter weight set to obtain the comprehensive feature parameter weight set;

[0135] The comprehensive distance set calculation module is used to calculate the distance between every two feature parameter data in the feature parameter dataset based on the comprehensive feature parameter weight set to obtain the comprehensive distance set.

[0136] The local outlier set calculation module is used to calculate the local outlier factor values ​​of the feature parameter dataset to obtain the local outlier set of feature parameters;

[0137] The cluster center generation module is used to select two feature parameter data points, where the local outlier value of the feature parameter is greater than a preset density value and the comprehensive distance in the comprehensive distance set is greater than a preset distance, as cluster centers to obtain the first cluster center and the second cluster center.

[0138] The clustering distance calculation module is used to perform clustering calculations between all other feature parameter data in the feature parameter dataset and the first and second cluster centers to obtain a clustering distance value set;

[0139] The feature cluster generation module is used to divide the feature parameter data into corresponding clusters based on the cluster distance values ​​in the cluster distance value set to obtain the first feature cluster and the second feature cluster.

[0140] In one embodiment, the second pressure value calculation module includes:

[0141] The real-time deformation rate module is used to obtain the real-time deformation rate by measuring the ratio of the real-time deformation of the first partition to the vacuuming time.

[0142] The second calculation module compares the real-time deformation rate with a preset deformation rate threshold; if the real-time deformation rate is greater than the preset deformation rate threshold, then the first function P1=P_base×(1-α×) is used. ) Calculate the second pressure value; if the real-time deformation rate is less than or equal to the preset deformation rate threshold, then use the second function P1=P_base×(1+β× ) Calculate the second pressure value; where P_base is the base pressure value based on the preset partition material and thickness, v is the real-time deformation rate, v0 is the deformation rate threshold, and α and β are adjustment coefficients greater than zero.

[0143] In one embodiment, the third pressure value calculation module includes:

[0144] The model building module is used to construct a nonlinear regression model for the second partition based on the mapping relationship between the thickness distribution of the second partition and the heat sealing pressure, so as to obtain the second pressure heat sealing model.

[0145] The expression for the second pressure heat sealing model is:

[0146] P2=P0×(a× +c× ), where P0 is the reference pressure value, The average thickness is the reference, T0 is the reference temperature value, P2 is the third pressure value, T is the preset third temperature value, δ is the average thickness of the second zone, and a, b and c are regression coefficients with a+c=1.

[0147] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A multi-stage pressure heat sealing method for a pouch cell, characterized in that, The method includes: The first soft-pack battery is obtained by pre-sealing the soft-pack battery by setting a first pressure value and a first temperature value. A vacuuming operation is performed on the first pouch cell to obtain a second pouch cell; The second pouch cell is divided into regions to obtain a first region and a second region; the first region is a region susceptible to temperature effects; the second region is a region unaffected by temperature effects. The second pressure value is obtained by calculating the pressure of the first partition using the first pressure heat sealing model; The third pressure value is obtained by calculating the pressure of the second partition using the second pressure heat sealing model; The first section of the second soft-pack battery is heat-sealed according to the second pressure value and the preset second temperature value. The second compartment of the second soft-pack battery is heat-sealed according to the third pressure value and the preset third temperature value. The second pouch battery is divided into regions to obtain a first partition and a second partition, including: Pressure data and deformation data are obtained by collecting pressure deformation data point by point through the heat sealing head on the second soft pack battery. A feature parameter set is obtained by extracting features from the pressure data and the deformation data; The feature parameter dataset is obtained by statistically analyzing the feature parameters of all collected locations. A novel clustering algorithm is used to cluster and group the feature parameter dataset to obtain a first feature cluster and a second feature cluster; the first feature cluster consists of feature parameter data from regions affected by temperature; the second feature cluster consists of feature parameter data from regions unaffected by temperature. The second soft-pack battery is partitioned according to the first feature cluster and the second feature cluster to obtain a first partition and a second partition; The first and second feature clusters are obtained by clustering the feature parameter dataset using a novel clustering algorithm. Perform variance analysis on the data corresponding to each feature parameter type in the feature parameter dataset to obtain the initial feature parameter weight set; Local feature parameter weight sets are obtained by performing local variance analysis on the feature parameter dataset using a sliding window. The initial feature parameter weight set and the local feature parameter weight set are fused to obtain a comprehensive feature parameter weight set; Based on the comprehensive feature parameter weight set, the distance between every two feature parameter data in the feature parameter dataset is calculated to obtain the comprehensive distance set; The local outlier factor value is calculated on the feature parameter dataset to obtain the local outlier value set of the feature parameters; The first cluster center and the second cluster center are obtained by taking the two feature parameter data where the local outlier value of the feature parameter is greater than the preset density value and the comprehensive distance in the comprehensive distance set is greater than the preset distance as the cluster center; The clustering distance value set is obtained by performing clustering calculations on all remaining feature parameter data in the feature parameter dataset with the first cluster center and the second cluster center; Based on the clustering distance values ​​in the clustering distance value set, the feature parameter data are divided into corresponding clusters to obtain the first feature cluster and the second feature cluster.

2. The multi-stage pressure heat sealing method for a soft-pack battery according to claim 1, characterized in that, The second pressure value is obtained by calculating the pressure of the first partition using the first pressure heat sealing model, including: The real-time deformation rate is obtained by comparing the real-time deformation of the first partition with the vacuuming time. The real-time deformation rate is compared with a preset deformation rate threshold; if the real-time deformation rate is greater than the preset deformation rate threshold, then the first function is used. Calculate the second pressure value; if the real-time deformation rate is less than or equal to a preset deformation rate threshold, then use the second function... Calculate the second pressure value; where, The base pressure value is based on the pre-set material and thickness of the partition. For real-time deformation rate, The deformation rate threshold, and The adjustment coefficient is greater than zero.

3. The multi-stage pressure heat sealing method for a soft-pack battery according to claim 1, characterized in that, The third pressure value is obtained by calculating the pressure of the second partition using the second pressure heat sealing model, including: Based on the mapping relationship between the thickness distribution of the second partition and the heat sealing pressure, a nonlinear regression model is constructed for the second partition to obtain the second pressure heat sealing model; The expression for the second pressure heat sealing model is: Where P0 is the reference pressure value. As the baseline average thickness, The reference temperature value This is the third pressure value. To preset the third temperature value, The average thickness of the second partition. , and For regression coefficients and .

4. A multi-stage pressure heat sealing system for a pouch battery, characterized in that, The system includes: The first soft-pack battery pre-sealing module is used to pre-seal the soft-pack battery by pre-setting a first pressure value and a first temperature value to obtain the first soft-pack battery. A vacuum module is used to perform a vacuuming operation on the first soft-pack battery to obtain a second soft-pack battery. The partitioning module is used to divide the second pouch battery into regions to obtain a first partition and a second partition; the first partition is a region susceptible to temperature effects; the second partition is a region unaffected by temperature effects. The second pressure value calculation module is used to calculate the pressure of the first partition using the first pressure heat sealing model to obtain the second pressure value; The third pressure value calculation module is used to calculate the pressure of the second partition using the second pressure heat sealing model to obtain the third pressure value; The first partition heat sealing module is used to heat seal the first partition of the second soft-pack battery according to the second pressure value and the preset second temperature value. The second partition heat sealing module is used to heat seal the second partition of the second soft pack battery according to the third pressure value and the preset third temperature value; The partitioning module includes: The data acquisition module is used to collect pressure and deformation data of the second soft-pack battery point by point through the heat sealing head. The feature parameter extraction module is used to extract features from the pressure data and the deformation data to obtain a feature parameter set; The data statistics module is used to statistically analyze the feature parameters of all collected locations to obtain a feature parameter dataset. The clustering module is used to cluster the feature parameter dataset using a novel clustering algorithm to obtain a first feature cluster and a second feature cluster; the first feature cluster consists of feature parameter data from regions affected by temperature; the second feature cluster consists of feature parameter data from regions unaffected by temperature. The partitioning module is used to partition the second soft-pack battery according to the first feature cluster and the second feature cluster to obtain a first partition and a second partition. The clustering module includes: The initial weight set generation module is used to perform variance analysis on the data corresponding to each feature parameter type in the feature parameter dataset to obtain the initial feature parameter weight set. The local weight set generation module is used to perform local variance analysis on the feature parameter dataset through a sliding window to obtain a local feature parameter weight set. The comprehensive weight set generation module is used to fuse the initial feature parameter weight set and the local feature parameter weight set to obtain a comprehensive feature parameter weight set; The comprehensive distance set calculation module is used to calculate the distance between every two feature parameter data in the feature parameter dataset according to the comprehensive feature parameter weight set to obtain the comprehensive distance set. The local outlier set calculation module is used to calculate the local outlier factor value of the feature parameter dataset to obtain the local outlier set of feature parameters; The cluster center generation module is used to take two feature parameter data in the feature parameter local outlier value set where the local outlier value of the feature parameter is greater than a preset density value and the comprehensive distance in the comprehensive distance set is greater than a preset distance as cluster centers to obtain the first cluster center and the second cluster center; The clustering distance value calculation module is used to perform clustering calculations with the first cluster center and the second cluster center in the feature parameter dataset to obtain a clustering distance value set; The feature cluster generation module is used to divide the feature parameter data into corresponding clusters based on the cluster distance values ​​in the cluster distance value set to obtain the first feature cluster and the second feature cluster.

5. The multi-stage pressure heat sealing system for a soft-pack battery according to claim 4, characterized in that, The second pressure value calculation module includes: The real-time deformation rate module is used to obtain the real-time deformation rate by measuring the ratio of the real-time deformation of the first partition to the vacuuming time. The second calculation module is used to compare the real-time deformation rate with a preset deformation rate threshold; if the real-time deformation rate is greater than the preset deformation rate threshold, then the first function is used. Calculate the second pressure value; if the real-time deformation rate is less than or equal to a preset deformation rate threshold, then use the second function... Calculate the second pressure value; where, The base pressure value is based on the pre-set material and thickness of the partition. For real-time deformation rate, The deformation rate threshold, and The adjustment coefficient is greater than zero.

6. The multi-stage pressure heat sealing system for a soft-pack battery according to claim 4, characterized in that, The third pressure value calculation module includes: The model building module is used to construct a nonlinear regression model for the second partition based on the mapping relationship between the thickness distribution of the second partition and the heat sealing pressure to obtain the second pressure heat sealing model; The expression for the second pressure heat sealing model is: Where P0 is the reference pressure value. As the baseline average thickness, The reference temperature value This is the third pressure value. To preset the third temperature value, The average thickness of the second partition. , and For regression coefficients and .

Images