Cell connection assembly assembly method and related assembly system

Abstract

The application provides an electric core connection assembly assembling method and a related assembling system, and belongs to the technical field of electrical element assembly manufacturing. The method comprises the following steps: obtaining an initial feature data set of each contact point in a to-be-crimped electric core connection assembly, generating differentiated crimping parameters for each contact point based on the initial feature data set and a mapping model; performing crimping on each contact point according to the differentiated crimping parameters, and collecting the dynamic contact resistance of the contact point in real time, calculating the relative deviation of the dynamic contact resistance and a target contact resistance, and solving and correcting the crimping force and the crimping stroke in real time based on the relative deviation to obtain the crimped electric core connection assembly; and checking the contact characteristics of the contact points in the crimped electric core connection assembly, and outputting the crimped and assembled qualified electric core connection assembly. The method can optimize the electric core connection assembly assembling process, and improve the interface bonding stability and long-term use reliability of the contact points of the contact electric core connection assembly.

Description

Technical Field

[0001] This application belongs to the field of manufacturing technology for electrical components and assemblies, and particularly relates to a method for assembling a battery cell connection assembly and a related assembly system. Background Technology

[0002] The Cells Contact System (CCS) is the core component for achieving electrical connection and signal acquisition between power batteries and energy storage battery modules.

[0003] Currently, the field of cell connection component contact point crimping assembly generally adopts the open-loop crimping method with globally unified process parameters. After crimping, the contact resistance has a large dispersion, which can easily lead to poor consistency of cell voltage sampling accuracy and abnormal heating of local contact points, thereby affecting the service reliability and service life of the battery module. Summary of the Invention

[0004] In view of this, this application provides a method and system for assembling a battery cell connection component, which combines the characteristics of the contact points to complete the configuration of differentiated process parameters, and coordinates with real-time dynamic parameter correction during the crimping process to optimize the assembly process of the battery cell connection component and improve the stability of the contact interface and the reliability of long-term use.

[0005] In a first aspect, this application provides a method for assembling a battery cell connection assembly, the method comprising the following steps: Obtain the initial feature dataset of each contact point in the battery cell connection assembly to be crimped, and generate differentiated crimping parameters for each contact point based on the initial feature dataset and the mapping model; Each contact point is crimped according to the differentiated crimping parameters, and the dynamic contact resistance of the contact point is collected in real time. The relative deviation between the dynamic contact resistance and the target contact resistance is calculated. Based on the relative deviation, the crimping force and crimping stroke are calculated and corrected in real time to obtain the crimped cell connection assembly. The contact characteristics of the contact points in the crimped cell connection assembly are verified, and a qualified crimped cell connection assembly is output.

[0006] Optionally, generating differentiated crimping parameters for each contact point based on the initial feature dataset and mapping model includes: A mapping model of contact resistance, crimping force, and crimping stroke is constructed. Based on the initial feature dataset and the mapping model, the target contact resistance corresponding to each contact point is determined. Using the target contact resistance as the optimization objective, the crimping force, crimping stroke, and holding time are optimized within the preset process constraints to obtain differentiated crimping parameters for each contact point.

[0007] Optionally, the construction of the mapping model of contact resistance, crimping force, and crimping stroke includes: Based on Holm's contact resistance theory, the contact resistance at the contact point is decomposed into shrinkage resistance and film resistance; Based on Hertzian contact theory, a nonlinear relationship between the shrinkage resistance and the pressing force and pressing stroke is established. Based on the initial feature dataset, a relationship between the film resistance and the surface morphology of the contact point and the coating characteristics is established. By combining the nonlinear relationship between the shrinkage resistance and the pressing force and pressing stroke, and the relationship between the film resistance and the surface morphology of the contact point and the coating characteristics, a nonlinear mapping mechanism model of contact resistance, pressing force and pressing stroke is obtained. Based on the measured initial resistance of the contact point, the nonlinear mapping mechanism model is fitted and calibrated to obtain the mapping model.

[0008] Optionally, the fitting and calibration of the nonlinear mapping mechanism model based on the measured initial resistance at the contact point includes: Using the measured initial resistance of all contact points in the same batch as the calibration benchmark, the least squares method is adopted, and the minimum mean square error between the calculated value and the measured value is used as the optimization objective to iteratively optimize the fitting parameters of the nonlinear mapping mechanism model. When the mean square error meets the preset threshold and the model fit meets the preset requirements, the calibration of the nonlinear mapping mechanism model is completed.

[0009] Optionally, the iterative optimization of the fitting parameters of the nonlinear mapping mechanism model includes: Based on the initial feature dataset of the contact points, the contact coefficient of the membrane layer is differentiated and corrected. Based on the elastic-plastic properties of the contact point material, the plastic deformation parameters corresponding to the pressing stroke are adapted and corrected to achieve iterative optimization of the fitting parameters of the nonlinear mapping mechanism model. During the iteration process, the goodness of fit of the model is constrained to be no less than the preset lower limit.

[0010] Optionally, optimizing the crimping force, crimping stroke, and holding time within the preset process constraints includes: For each contact point, an optimization objective function is constructed based on the initial feature dataset and mapping model; An optimization algorithm is used to iteratively optimize within the preset process constraints, and outputs the crimping force, crimping stroke and holding time that meet the optimization objective function requirements for each contact point, as the corresponding differentiated crimping parameters.

[0011] Optionally, constructing the optimization objective function includes: The optimization objective is to minimize the deviation between the contact resistance after crimping and the target contact resistance, and to minimize the dispersion of the contact resistance of all contact points in the same batch. A corresponding weighting coefficient is assigned to each optimization objective, and a penalty term for exceeding the process boundary constraints is added to form the optimization objective function.

[0012] Optionally, the step of calculating the relative deviation between the dynamic contact resistance and the target contact resistance, and calculating and correcting the crimping force and crimping stroke in real time based on the relative deviation, includes: The dynamic contact resistance acquired in real time is subjected to sliding window filtering, and the relative deviation between the filtered dynamic contact resistance and the target contact resistance is calculated. Based on the relative deviation, an incremental control algorithm is used to calculate the crimping force correction and crimping stroke correction in real time, and the current crimping parameters are corrected in real time according to the crimping force correction and crimping stroke correction.

[0013] Optionally, the step of verifying the contact characteristics of the contact points in the crimped cell connection assembly includes: Electrochemical impedance spectroscopy was used to detect the contact characteristics of the contact points in the crimped battery cell connection assembly, and the detection data was obtained. The detection data is fitted based on an equivalent circuit model to obtain the interfacial charge transfer resistance at the contact point. The cross-validation was performed based on the interface charge transfer resistance and the measured contact resistance after crimping, and the pass / fail status was determined by combining the batch contact resistance dispersion.

[0014] Secondly, this application provides a cell connection assembly system, the cell connection assembly system including a main control device and multiple functional components respectively communicatively connected to the main control device, the main control device including: The parameter generation module is used to obtain the initial feature dataset of each contact point in the battery cell connection assembly to be crimped, and generate differentiated crimping parameters for each contact point based on the initial feature dataset and the mapping model. The crimping control module is used to perform crimping on each contact point according to the differentiated crimping parameters, and to collect the dynamic contact resistance of the contact point in real time, calculate the relative deviation between the dynamic contact resistance and the target contact resistance, and calculate and correct the crimping force and crimping stroke in real time based on the relative deviation to obtain the crimped cell connection assembly. The verification output module is used to verify the contact characteristics of the contact points in the crimped cell connection assembly and output a qualified crimped cell connection assembly.

[0015] This application first obtains the initial feature dataset of each contact point in the cell connection assembly to be crimped. Based on the initial feature dataset and mapping model, differentiated crimping parameters are generated for each contact point. Then, crimping is performed on each contact point according to the differentiated crimping parameters, and the dynamic contact resistance of the contact point is collected in real time. The relative deviation between the dynamic contact resistance and the target contact resistance is calculated. Based on the relative deviation, the crimping force and crimping stroke are calculated and corrected in real time to obtain the crimped cell connection assembly. Finally, the contact characteristics of the contact points in the crimped cell connection assembly are verified, and a qualified crimped cell connection assembly is output. Because this application adopts a process control method of single-contact differentiated parameter adaptation, real-time correction during crimping, and full interface verification after crimping, it can reduce the dispersion of contact resistance and improve crimping consistency. At the same time, it can improve voltage acquisition accuracy and long-term service reliability, effectively reduce the risk of contact degradation, overheating failure, and thermal runaway, and also improve mass production yield, meeting the high precision and high reliability requirements of mass production. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the assembly steps of the battery cell connection assembly method provided in the embodiments of this application; Figure 2 This is a schematic diagram illustrating the steps for generating differentiated crimping parameters provided in an embodiment of this application; Figure 3 This is a schematic diagram of the process for constructing a mapping model provided in an embodiment of this application; Figure 4 This is a schematic diagram of the process for finding the objective function based on an embodiment of this application; Figure 5 This is a schematic diagram of the real-time correction process for crimping parameters provided in the embodiments of this application; Figure 6 This is a schematic diagram of the contact characteristic verification process provided in an embodiment of this application; Figure 7 This is a schematic diagram of the main control equipment of the battery cell connection assembly system provided in the embodiments of this application; Figure 8 This is a schematic diagram of the hardware architecture of the battery cell connection component assembly system provided in the embodiments of this application; Figure 9 This is a schematic diagram of the hardware architecture of a cell connection component assembly system provided in another embodiment of this application; Figure 10 This is a schematic diagram of the structure of the cell connection assembly provided in the embodiments of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. 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.

[0019] The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0020] In this document, references to "embodiment" or "implementation" mean that a particular feature, structure, or characteristic described in connection with an embodiment or implementation may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0021] Before introducing the technical solution of this application, let's go over the technical issues in related technologies in detail.

[0022] In existing cell connection system (CCS) components, the crimping of contact points and busbars in flexible printed circuit boards (FPCs) generally uses uniform crimping parameters for the entire batch. This cannot adapt to individual differences such as multiple contact points within a single component, contact point plating loss, surface flatness, and roughness between different components in the same batch. At the same time, the contact resistance dispersion of the contact points after crimping is too large, and the process capability index does not meet the automotive-grade mass production standard.

[0023] Furthermore, traditional crimping is an open-loop operation mode, which only completes the crimping action according to preset parameters. It lacks a dynamic contact resistance real-time monitoring and parameter closed-loop correction mechanism, making it difficult to offset the contact resistance fluctuations caused by material elastic-plastic deformation and interface micro-slippage during the crimping stage, further widening the differences in electrical performance of each contact point within a batch.

[0024] Furthermore, only static resistance sampling tests were conducted after crimping, without comprehensive verification of the interface impedance characteristics of all contact points. This fails to effectively identify micro-defects and latent contact problems at the contact point interfaces. During long-term use, under temperature cycling and mechanical vibration, the contact point interfaces are highly susceptible to oxidation and deterioration of contact performance. This can lead to signal drift in the Battery Management System (BMS) cell acquisition system, abnormal heating at the contact points, and in severe cases, even thermal runaway of the battery pack, posing a significant safety hazard.

[0025] Please see Figure 1 and Figure 10 This application provides a method for assembling a battery cell connection assembly, the method comprising the following steps: S10. The cell connection assembly system obtains the initial feature dataset of each contact point in the cell connection assembly to be crimped, and generates differentiated crimping parameters for each contact point based on the initial feature dataset and the mapping model.

[0026] Among them, the CellsContactSystem (CCS) is the core structure in the power battery module that realizes the electrical connection of cells and the signal acquisition of the Battery Management System (BMS). The contact points laid on it are the key docking points for the busbar and the flexible printed circuit board (FPC) acquisition end to be crimped and matched. Each contact point has independent physical characteristic parameters due to differences in production and processing, and there is no uniform solidification property.

[0027] The initial feature dataset is a set of basic parameters of each contact point collected in advance through actual measurement, including but not limited to quantitative data such as contact point coating thickness, surface roughness, flatness error, substrate material properties and initial native contact resistance, which can characterize the individual processing differences and basic interface state of a single contact point.

[0028] The mapping model is a nonlinear correlation model that is pre-constructed and fitted and calibrated based on the contact resistance-related contact theory. It can establish a quantitative correspondence between the initial characteristics of the contact point, the crimping process parameters and the contact resistance, and can match the adaptive crimping process benchmark based on the unique initial characteristic data of each contact point.

[0029] In one possible embodiment, relying on the complete initial feature dataset of each contact point, the feature data of each single point is entered into the mapping model one by one to complete the association calculation. Based on the interface characteristics and processing differences of each contact point, a unique matching differential pressing parameter is generated, thereby avoiding the situation where batch uniform parameters cannot adapt to the individual differences of single points.

[0030] S20. The cell connection assembly system performs crimping on each contact point according to the differentiated crimping parameters, and collects the dynamic contact resistance of the contact point in real time. It calculates the relative deviation between the dynamic contact resistance and the target contact resistance, and calculates and corrects the crimping force and crimping stroke in real time based on the relative deviation to obtain the crimped cell connection assembly.

[0031] The differentiated crimping parameters are the combination parameters of crimping force, crimping stroke and holding time corresponding to each contact point, which are adapted to the interface physical characteristics of a single contact point.

[0032] The dynamic contact resistance is a real-time value of the contact resistance at the interface of the contact point, which is continuously sampled throughout the entire crimping assembly process. It can provide feedback on the real-time impact of dynamic changes such as the elastic-plastic deformation of the contact point material and the micro-slippage of the interface on the contact conductivity during the crimping process.

[0033] The relative deviation is the quantitative difference ratio between the real-time dynamic contact resistance and the target contact resistance required for mass production compliance. It is used to determine whether the current crimping status meets the standard and whether process parameters need to be adjusted, and to quantify the fluctuation range of contact performance during the crimping process.

[0034] In one possible embodiment, the initial crimping operation is completed according to the differentiated crimping parameters corresponding to each contact point. Dynamic contact resistance data is continuously collected throughout the crimping process, and the relative deviation from the target contact resistance is calculated in real time. Based on the deviation value, the correction amount of crimping force and crimping stroke is calculated in real time, and the dynamic closed-loop adjustment of crimping process parameters is completed simultaneously, thereby offsetting the contact resistance fluctuations caused by various interference factors during the crimping process.

[0035] S30. The cell connection assembly system verifies the contact characteristics of the contact points in the crimped cell connection assembly and outputs a qualified crimped cell connection assembly.

[0036] The contact characteristic verification can be understood as the detection of the conductivity, contact stability and latent defects of the contact point interface after crimping.

[0037] In one possible embodiment, the assembly qualification criteria include the compliance of the measured contact resistance of the contact points, the batch contact resistance dispersion, and the interface contact stability, so that the output cell connection assembly meets the requirements of high-precision mass production process and ensures the reliability of the cell connection assembly under long-term service conditions.

[0038] In one possible embodiment, the contact characteristics of each contact point of the press-formed cell connection assembly are checked one by one. Based on the check data, the pass / fail determination is completed, and unqualified products with substandard contact characteristics or hidden defects are eliminated. Only the qualified cell connection assemblies are output for subsequent battery module assembly processes.

[0039] In summary, compared to the existing cell connection assembly crimping process which uses uniform parameters, resulting in high contact resistance dispersion, poor consistency, and insufficient long-term reliability, this application first obtains the initial feature dataset of each contact point in the cell connection assembly to be crimped. Based on the initial feature dataset and mapping model, differentiated crimping parameters are generated for each contact point. Then, crimping is performed on each contact point according to the differentiated crimping parameters, and the dynamic contact resistance of the contact point is collected in real time. The relative deviation between the dynamic contact resistance and the target contact resistance is calculated. Based on the relative deviation, the crimping force and crimping stroke are calculated and corrected in real time to obtain the crimped cell connection assembly. Finally, the contact characteristics of the contact points in the crimped cell connection assembly are verified, and a qualified crimped cell connection assembly is output. Because this application adopts a process control method of single-contact point differential parameter adaptation, real-time correction during crimping, and full interface verification after crimping, it can reduce contact resistance dispersion and improve crimping consistency, while improving voltage acquisition accuracy and long-term service reliability. It can effectively reduce the risk of contact degradation, overheating failure and thermal runaway, and also improve mass production yield, meeting the high precision and high reliability requirements of mass production.

[0040] In one specific embodiment of this application, the initial feature dataset of each contact point in the battery cell connection assembly to be pressed is obtained, including the acquisition and preprocessing of the multi-dimensional initial feature dataset of the contact points to be pressed.

[0041] Specifically, the system acquires the unique ID of each contact point of the CCS component to be crimped, the total number of contact points N, the preset basic parameters of the contact points (contact point material, rated plating thickness, and design target contact resistance), and the raw sampling data output by the multi-dimensional initial feature acquisition subsystem.

[0042] Furthermore, for each contact point i to be pressed (i=1,2,...,N), four original feature data are simultaneously collected: initial contact resistance, coating thickness, surface roughness, and flatness error, forming a single contact point original feature vector. The 3σ criterion is used to remove outliers from the sampling sequence of each feature; the removal judgment formula is as follows: Where, x ij The feature value of the j-th sample at the i-th contact point; This is the arithmetic mean of the sampling sequence corresponding to this feature; This represents the standard deviation of the sampling sequence corresponding to this feature.

[0043] Furthermore, the feature data after outlier removal is subjected to min-max normalization to eliminate the dimensional differences between different features. The normalization formula is: in, This is the normalized eigenvalue of the j-th contact point; This is the minimum value of this feature among all contact points in the same batch; This is the maximum value of this feature across all contact points in the same batch.

[0044] Furthermore, a normalized initial feature dataset for single contact points is obtained. Feature matrix of all contacts in the same batch .

[0045] Please see Figure 2 The step of generating differentiated crimping parameters for each contact point based on the initial feature dataset and mapping model includes the following steps: S11. Construct a mapping model of contact resistance, crimping force and crimping stroke. Based on the initial feature dataset and the mapping model, determine the target contact resistance corresponding to each contact point.

[0046] The mapping model is a nonlinear correlation mechanism model established based on the fundamental theory of contact interface conduction. It can reproduce the influence of changes in the magnitude of the pressing force and the displacement of the pressing stroke on the overall contact resistance of the contact point under actual pressing conditions.

[0047] The contact resistance is an indicator of the electrical conductivity of the contact point, determining the accuracy and stability of the battery management system's acquisition of cell voltage and temperature signals. The crimping force is the clamping force applied to the contact interface during the crimping process, representing the external force condition that changes the contact interface's adhesion state. The crimping stroke is the actual downward displacement of the crimping actuator, reflecting the degree of contact interface deformation and the actual effective contact area.

[0048] In one possible embodiment, after the mapping model is constructed, the initial feature datasets of each contact point obtained from previous measurements are entered into the model one by one. Relying on the correlation calculation logic built into the mapping model, and combining the differentiated inherent characteristics of different contact points such as their surface morphology, plating state, substrate material, and surface cleanliness, the exclusive target contact resistance corresponding to each independent contact point is calculated and determined. The target contact resistance is not a fixed resistance value standard uniformly set for the entire batch, but rather a conduction resistance value matched to the physical state of the contact point itself. This satisfies both the electrical conduction requirements and the structural tolerance of the contact point, avoiding the problem of a uniform resistance standard not matching the actual operating conditions of the contact point.

[0049] S12. Taking the target contact resistance as the optimization target, optimize the crimping force, crimping stroke and holding time within the preset process constraints to obtain the differentiated crimping parameters for each contact point.

[0050] Specifically, the preset process constraint range limits the maximum value of the crimping force, the effective displacement range of the crimping stroke, and the reasonable duration of the holding time, so as to avoid the process parameters obtained by optimization solution from exceeding the actual production execution range, and to avoid various process abnormalities such as crimping overload damaging the contact point structure, insufficient crimping force and loose fit.

[0051] In one possible embodiment, within the preset process constraints, with the target contact resistance as the optimization objective, a combined optimization calculation is performed on the crimping force, crimping stroke, and holding time. By continuously adjusting the combination of these three types of parameters, and repeatedly deduce the actual contact resistance forming effect corresponding to different parameter combinations through a mapping model, the parameter combination that can conform to the target contact resistance while taking into account the interface bonding strength and long-term use stability is gradually selected. Finally, this set of parameter combinations is determined as the differentiated crimping parameters for matching the corresponding contact point.

[0052] As can be seen, this embodiment first establishes a multivariate correlation mapping model, then matches the target contact resistance based on the initial characteristics of the contact points, and finally optimizes multi-dimensional process parameters within the process boundary. By combining theoretical models with measured feature data to achieve intelligent parameter matching, the crimping process parameters configured for each contact point can conform to its actual physical characteristics, reducing the crimping quality gap between different contact points, effectively reducing the contact resistance dispersion of batch products, improving the crimping consistency of products in the same batch, and fully adapting to the high-precision, high-stability, and high-consistency mass production requirements of battery cell connection components.

[0053] Please see Figure 3 The construction of the mapping model for contact resistance, crimping force, and crimping stroke includes the following steps: S111. Based on Holm's contact resistance theory, the contact resistance at the contact point is decomposed into shrinkage resistance and film resistance.

[0054] Specifically, based on the principles of Holm's contact resistance theory, the magnitude of contact resistance is mainly determined by two parts: shrinkage resistance and film resistance. The formation mechanisms and influencing factors of shrinkage resistance and film resistance are fundamentally different. If the contact resistance is directly modeled as a whole, the model will fail to accurately reflect the impact of different factors on the contact resistance, thus affecting the accuracy of subsequent parameter matching. Therefore, in the initial stage of mapping model construction, the contact resistance at the contact point is first decomposed to distinguish the differences between shrinkage resistance and film resistance, thereby effectively improving the accuracy of the mapping model.

[0055] S112. Based on Hertzian contact theory, establish the nonlinear relationship between the shrinkage resistance and the pressing force and pressing stroke, and based on the initial feature dataset, establish the relationship between the film resistance and the surface morphology and coating characteristics of the contact point.

[0056] The shrinkage resistance is the resistance generated when current passes through the contact point due to the contraction of the current lines. Its magnitude is mainly related to the actual contact area, pressing pressure, and material resistivity; the larger the contact area and the tighter the pressing, the smaller the shrinkage resistance. The film resistance is the resistance formed by the oxide layer, contaminants, and coatings on the contact point surface. Its magnitude is related to the surface morphology of the contact point, coating thickness, coating material, and the degree of environmental oxidation. By decomposing the contact resistance, the influence of different factors on the contact resistance can be analyzed in a targeted manner, ensuring that the mapping model can accurately capture the changing patterns of the contact resistance.

[0057] Specifically, for the shrinkage resistance, a nonlinear relationship between it and the crimping force and crimping stroke is constructed based on Hertzian contact theory. Hertzian contact theory is mainly used to describe the pressure distribution and deformation law during the contact of elastomers, and can reflect the nonlinear relationship between crimping force, crimping stroke and contact area.

[0058] S113. Combine the nonlinear relationship between the shrinkage resistance and the pressing force and the pressing stroke, and the relationship between the film resistance and the surface morphology of the contact point and the coating characteristics to obtain a nonlinear mapping mechanism model of contact resistance, pressing force and pressing stroke.

[0059] The nonlinear mapping mechanism model reflects the interaction relationships between various parameters. For example, changes in crimping force not only affect the shrinkage resistance but also indirectly influence the stability of the film resistance by altering the adhesion of the contact interface. Differences in surface roughness at the contact points further affect the matching relationship between the total contact resistance and the crimping parameters through changes in film resistance. By combining these two types of relationships, the nonlinear mapping mechanism model effectively covers the factors influencing contact resistance.

[0060] S114. Based on the measured initial resistance of the contact point, the nonlinear mapping mechanism model is fitted and calibrated to obtain the mapping model.

[0061] In one possible embodiment, measured initial resistance data from different contact points within the same batch are first collected to reflect the initial state of the contact points. Using this measured initial resistance data as a calibration benchmark, it is substituted into a nonlinear mapping mechanism model. Through iterative fitting, relevant parameters in the model are adjusted to ensure that the deviation between the model-calculated contact resistance value and the measured initial resistance value is controlled within a preset range, thus ensuring that the model's calculation results are highly consistent with actual operating conditions.

[0062] In the fitting and calibration process, the fitting error is controlled through multiple iterations to ensure that the model accurately reflects the changes in actual contact resistance, crimping force, and crimping stroke, thus ensuring that the model's prediction accuracy meets the requirements of automotive-grade mass production. Simultaneously, the model is further calibrated using initial characteristic data of the contact points, enabling it to adapt to contact points with different coatings and surface morphologies, thereby improving the model's versatility and adaptability.

[0063] As can be seen, the mapping model constructed in this embodiment combines the core logic of Holm contact resistance theory and Hertz contact theory, while also incorporating measured data and individual characteristics of contact points. This solves the problems of traditional mapping models lacking theoretical support, having insufficient fitting accuracy, and being unable to adapt to individual differences in contact points. The mapping model provides a theoretical computational basis for determining the target contact resistance and optimizing differentiated pressing parameters, ensuring that the pressing parameters of each contact point are highly adapted to its own characteristics, improving the precision of the pressing process, and thus meeting the actual needs of large-scale, high-precision mass production of CCS components.

[0064] In one specific embodiment of this application, the construction of the mapping model of contact resistance, crimping force, and crimping stroke specifically includes obtaining a normalized initial feature dataset X for a single contact point. i Physical parameters of the contact point material (volume resistivity ρ, elastic modulus E, Poisson's ratio ν, resistivity of the coating film ρf), and contact point design parameters (equivalent radius of curvature R).

[0065] Furthermore, based on Holm's contact resistance theory, a fundamental mechanism model of single-contact-point contact resistance is constructed, with contact resistance R... c Due to the shrinkage resistance R s and film resistance R f It consists of two parts, and the formula is: Furthermore, construct a shrinkage resistor R sThe nonlinear model shows that the shrinkage resistance is determined by the actual contact area at the contact point, and it varies nonlinearly with the pressing force and pressing stroke. The formula is: in, ρ is the volume resistivity of the contact material, in Ω·m; a is the actual contact radius of the contact point, in meters.

[0066] Furthermore, based on Hertzian contact theory, a nonlinear correlation model is constructed between the actual contact radius *a* and the pressing force *F*, with the following formula: Where F is the crimping force applied to the contact point, in N; R is the equivalent radius of curvature of the contact surface, in m; ρ is the Poisson's ratio of the contact material, dimensionless; E is the elastic modulus of the contact material, in Pa.

[0067] Furthermore, a correlation model is constructed between the pressing stroke S and the actual contact radius a, considering the corrections for material elastic-plastic deformation and initial morphological characteristics. The formula is: in, The plastic deformation of the material at the contact point, in meters (m), is given by the initial feature dataset X. i The coating thickness and surface roughness are fitted and corrected.

[0068] Furthermore, construct the film resistance R f The nonlinear model, considering the influence of coating thickness and surface morphology on the contact interface, is given by the following formula: in, The resistivity of the film coating on the surface of the contact point is expressed in Ω·m. The membrane contact coefficient is dimensionless and is derived from the initial feature dataset X. i The fitted values ​​range from 0 to 1, with α being closer to 1 the better the surface quality of the contact point.

[0069] Furthermore, by combining the above formulas, a complete nonlinear mapping model of contact resistance-pressing force-pressing stroke is constructed, in matrix form as follows: Furthermore, the least squares method is used to fit and calibrate the model. The calibration objective is to minimize the mean square error between the calculated model value and the measured initial contact resistance. The optimization formula is: Wherein, MSE is the mean square error, which is used to characterize the overall deviation level between the model's calculated value and the measured value. The smaller the MSE value, the higher the model's prediction accuracy; N is the total number of contact points to be calibrated in the same batch of battery cell connection components; i is the contact point number. Calculate the contact resistance for the model of the i-th contact point; R 0i Let be the measured initial contact resistance at the i-th contact point; min is the optimization objective, representing the solution of model fitting parameters to minimize the mean square error (MSE). After calibration, the model fit goodness of fit must satisfy... .

[0070] Furthermore, a calibrated nonlinear mapping model of single-contact point contact resistance-crimping force-crimping stroke is obtained. Model fitting parameter set.

[0071] Optionally, the fitting and calibration of the nonlinear mapping mechanism model based on the measured initial resistance at the contact point includes: Using the measured initial resistance of all contact points in the same batch as the calibration benchmark, the least squares method is adopted, and the minimum mean square error between the calculated value and the measured value is used as the optimization objective to iteratively optimize the fitting parameters of the nonlinear mapping mechanism model. When the mean square error meets the preset threshold and the model fit meets the preset requirements, the calibration of the nonlinear mapping mechanism model is completed.

[0072] Among them, the measured initial resistance of all contact points in the same batch is the real resistance data obtained by testing each contact point one by one with special testing equipment. It covers the individual differences in plating status, surface roughness and flatness of different contact points, and can reflect the actual electrical characteristics and physical state of contact points in the same batch, avoiding the impact of abnormal data of a single contact point on the calibration accuracy.

[0073] In one possible embodiment, the measured initial resistance data of all contact points in the same batch are matched with the calculated data of the nonlinear mapping mechanism model to construct an error function. The fitting parameters in the model are then gradually adjusted with the minimum mean square error as the optimization objective. These fitting parameters are used in the nonlinear mapping mechanism model to correlate contact resistance, crimping force, and crimping stroke, determining the degree to which the model's calculation results match the actual working conditions.

[0074] In one possible embodiment, during the iterative optimization process, after each adjustment of the fitting parameters, the mean square error between the model's calculated value and the measured initial resistance value is calculated, and it is determined whether this error meets a preset threshold. The mean square error is an indicator of the model's calculation accuracy; the smaller the value, the smaller the deviation between the model's calculated result and the measured data, and the higher the model's accuracy. The preset threshold is determined by considering product accuracy requirements, the operating accuracy of the production equipment, and the reasonable fluctuation range of contact resistance, ensuring that the fitted model can meet actual production needs and avoiding deviations in subsequent crimping parameter settings due to excessive errors.

[0075] The model fit score is used to reflect the degree to which the model fits the measured data. The higher the fit score, the more accurately the model can reflect the relationship between contact resistance, crimping force and crimping stroke, and the better it can adapt to the individual differences of different contact points.

[0076] Optionally, the preset requirements for the model fit are formulated by combining the principles of Holm contact resistance theory and Hertz contact theory with the crimping process requirements in actual production, and mainly include multiple dimensions such as model fitting accuracy, parameter adaptability and data correlation.

[0077] As can be seen, the model fitting and calibration method provided in this embodiment achieves accurate calibration of the nonlinear mapping mechanism model through a process of determining the calibration benchmark, selecting the fitting algorithm, setting the optimization target, iteratively optimizing parameters, and achieving the calibration. The calibrated model can accurately reflect the nonlinear relationship between contact resistance, crimping force, and crimping stroke, while also taking into account the individual differences of contact points in the same batch. This provides a reliable theoretical basis for subsequently determining the target contact resistance based on the initial feature dataset and optimizing crimping parameters, ensuring that the subsequently generated differentiated crimping parameters have high accuracy and adaptability, and effectively improving the consistency and reliability of the crimping quality of batch CCS components.

[0078] Optionally, the iterative optimization of the fitting parameters of the nonlinear mapping mechanism model includes: Based on the initial feature dataset of the contact points, the contact coefficient of the membrane layer is differentiated and corrected. Based on the elastic-plastic properties of the contact point material, the plastic deformation parameters corresponding to the pressing stroke are adapted and corrected to achieve iterative optimization of the fitting parameters of the nonlinear mapping mechanism model. During the iteration process, the goodness of fit of the model is constrained to be no less than the preset lower limit.

[0079] The coating thickness and surface oxidation degree vary at different contact points. If a uniform film contact coefficient is used, the contact resistance calculated by the model will deviate significantly from the actual situation, thus affecting the rationality of the crimping parameters. Therefore, in this embodiment, the film contact coefficient is adjusted individually for the initial characteristics of each contact point.

[0080] In one possible embodiment, for contact points with thicker coatings and lighter surface oxidation, the film contact coefficient is reduced to ensure the accuracy of contact resistance calculation; for contact points with thinner coatings and more severe surface oxidation, the film contact coefficient is increased to compensate for the resistance loss caused by the oxide layer, thereby achieving a precise match between the film contact coefficient and the individual characteristics of the contact point.

[0081] Specifically, the differential correction of the film contact coefficient involves combining the coating material, thickness, and surface contamination of the contact point, referring to Holm contact resistance theory, clarifying the correlation between film resistance and film contact coefficient, and using linear fitting to independently correct the film contact coefficient of each contact point, thereby ensuring the accurate calculation of film resistance at each contact point.

[0082] During the pressing process, the contact material undergoes elastoplastic deformation. Since the substrate material and thickness vary at different contact points, their elastoplastic deformation patterns also differ. Using a uniform plastic deformation parameter could lead to some contact points being pressed too tightly or too loosely, causing material damage or poor contact. Therefore, in this embodiment, based on the elastoplastic test data of the contact material, a correlation between plastic deformation and pressing stroke is established. The plastic deformation amount corresponding to the pressing stroke is then adjusted according to the material characteristics of different contact points.

[0083] In one possible embodiment, for contact points with thicker and harder substrates, where plastic deformation is more difficult, the plastic deformation parameter is appropriately increased to ensure an effective contact area is formed during pressing. For contact points with thinner and lower substrates, the amount of plastic deformation is controlled to avoid excessive deformation that could damage the contact points. Simultaneously, based on actual deformation feedback during the pressing process, the plastic deformation parameter is dynamically adjusted to ensure it matches the pressing stroke and the elastic-plastic properties of the material, thereby achieving stable control of the contact resistance during pressing.

[0084] In one possible embodiment, with minimizing contact resistance as the optimization objective, the corrected film contact coefficient and plastic deformation parameters are substituted into the nonlinear mapping mechanism model. The deviation between the model's fitted value and the actual contact resistance is calculated, and the fitting parameters are adjusted according to the magnitude of the deviation. This process is repeated until the deviation between the model-calculated contact resistance and the actual measured resistance is within an acceptable range, and the goodness of fit meets the preset requirements, thus completing one iteration of optimization. To improve optimization efficiency, a step-by-step iterative approach can be adopted, adjusting a set of parameters each time and comparing the fitting effects before and after optimization to gradually approach the optimal parameter combination.

[0085] Optionally, after each parameter correction, the model fit and contact resistance deviation are recalculated. If the preset requirements are not met, the parameters are adjusted until the fitting parameters of all contact points meet the adaptation requirements and the model fit is stable above the preset lower limit.

[0086] As can be seen, this embodiment can achieve precise optimization of the fitting parameters of the nonlinear mapping mechanism model by differentiating the contact coefficient of the film layer, adapting and adjusting the amount of plastic deformation, and constraining the goodness of fit, thus ensuring a high degree of matching between the fitting parameters and the individual characteristics of the contact point, thereby taking into account both production efficiency and product quality.

[0087] Please see Figure 4 The optimization of the pressing force, pressing stroke, and holding time within the preset process constraints includes the following steps: S121. For each contact point, construct an optimization objective function based on the initial feature dataset and mapping model.

[0088] The optimization objective function is a quantitative mathematical function used to evaluate the quality of the combination of crimping process parameters. This objective function is constructed based on the actual working conditions of the contact points represented by the initial feature dataset, combined with the performance change patterns derived from the mapping model. The optimization objective function can uniformly quantify and integrate multiple objectives, such as contact resistance approaching a set target value, crimping deformation within a reasonable material tolerance range, and stable interface bonding.

[0089] S122. An optimization algorithm is used to iteratively optimize within the preset process constraints, and the crimping force, crimping stroke and holding time that meet the optimization objective function requirements are output for each contact point as the corresponding differentiated crimping parameters.

[0090] The preset process constraint range is a range of legal values ​​for process parameters that are comprehensively defined by combining actual industrial production conditions, equipment operating limits, and product structural safety baselines. It is determined by the rated output pressure range of the crimping equipment, the mechanical stroke limit of the equipment, the elastic deformation limit of the substrate at the contact point of the CCS component, the anti-extrusion damage threshold of the coating structure, and mature mass production process specifications in the industry.

[0091] The preset process constraint range is divided into hard boundary conditions such as the maximum allowable value and minimum working value of the crimping force, the effective displacement range of the crimping stroke, the shortest stable holding time and the longest working time, etc., to avoid the optimized process parameters exceeding the actual execution capacity of the equipment and causing equipment malfunctions; and to avoid excessive crimping force and excessive crimping stroke causing structural damage such as extrusion cracking of the contact point coating, while preventing the process parameters from being too small, which may cause quality problems such as loose crimping, insufficient interface contact and easy loosening after long-term use.

[0092] The optimization algorithm is an intelligent optimization algorithm with the ability to solve multiple variables simultaneously and compare it iteratively. It can simultaneously control and calculate three types of variables: pressing force, pressing stroke, and holding time.

[0093] In one possible embodiment, under the premise that all process parameters are limited to a preset constraint range, the optimization algorithm continuously adjusts the combination of the three types of process parameters according to the performance evaluation criteria set by the optimization objective function specific to the corresponding contact point. Each time a new set of parameters is generated, it is simultaneously imported into the nonlinear mapping mechanism model to deduce and calculate the actual contact resistance, interface deformation state, and interface bonding stability of the contact point after forming under the action of that set of parameters. The calculation results are then compared in real time with the expected values ​​set within the optimization objective function.

[0094] As can be seen, this embodiment constructs an optimization objective function for each contact point, completes quantitative modeling based on the mapping model and initial feature data, and then iteratively optimizes and matches process parameters within the process constraint range through optimization algorithms. This can fully accommodate individual differences such as surface morphology and coating state of different contact points, and match specific crimping force, crimping stroke, and holding time. At the same time, based on process boundary constraints, it avoids contact point structural damage and crimping failure caused by parameter exceeding limits, reduces the contact resistance difference of products in the same batch, and improves the overall consistency of crimping operations, thereby effectively improving the assembly quality and mass production stability of CCS components.

[0095] Optionally, the construction of the optimization objective function includes: taking the minimum deviation between the contact resistance after crimping and the target contact resistance and the minimum dispersion of the contact resistance of all contact points in the same batch as the optimization objective, configuring a corresponding weight coefficient for each optimization objective, and adding a penalty term for exceeding the process boundary constraints to form the optimization objective function.

[0096] The contact resistance after crimping at the contact point deviates from the target contact resistance to the minimum, ensuring that the actual contact resistance of each contact point matches the preset target contact resistance after crimping. The target contact resistance is determined based on the initial feature dataset of the contact points and the calibrated mapping model, adapting to the surface morphology, coating characteristics, and substrate properties of each individual contact point.

[0097] Among them, the contact resistance dispersion of all contact points in the same batch is the smallest, which is used to reduce the difference in contact resistance values ​​between different contact points in the same batch.

[0098] The weighting coefficients range from 0 to 1, and the sum of the weighting coefficients for the two optimization objectives is 1. In one possible embodiment, the weighting coefficients can be flexibly adjusted according to actual production needs: if the focus is on the conductivity accuracy of a single contact point, such as for high-rate power batteries with CCS modules, then a higher weighting coefficient is configured for the minimum single-point resistance deviation; if the focus is on the consistency of batch production, such as for the mass production of large-scale energy storage battery modules, then a higher weighting coefficient is configured for the minimum batch dispersion.

[0099] The penalty term is a quantitative penalty factor set for process parameters exceeding preset process boundary constraints. It constrains non-compliant parameter combinations that occur during the optimization process, preventing the optimization algorithm from solving for process parameters that exceed the equipment's operating limits and the material's tolerance range. The value of the penalty term is positively correlated with the degree to which the parameter exceeds the constraint range: the more the parameter exceeds the constraint range, the larger the penalty term value, and the worse the overall function value; if the parameter is completely within the preset process constraint range, the penalty term value is 0, and it does not affect the normal solution of the optimization objective.

[0100] As can be seen, the optimization objective function constructed in this embodiment not only ensures that the contact resistance of a single contact point conforms to the preset standard, but also effectively reduces the dispersion difference of the contact resistance of the entire batch, thus balancing single-point conductivity and batch assembly consistency. By configuring weight coefficients, the optimization focus can be flexibly adjusted according to different application scenarios to adapt to diversified mass production needs. Furthermore, the addition of a process boundary penalty term can automatically avoid exceeding process parameters during parameter optimization, effectively preventing problems such as excessive crimping damaging the contact point structure or insufficient crimping causing poor contact, further improving the crimping quality of CCS components.

[0101] In one specific embodiment of this application, the optimization of the crimping force, crimping stroke, and holding time within the preset process constraints specifically includes the following process.

[0102] First, obtain the nonlinear mapping model. Preset the boundary conditions of the crimping process (crimping force F) min and F max , crimping stroke S min and S max Holding time T min and T max ), target contact resistance The total number of contact points in the same batch, N.

[0103] Furthermore, a multi-objective optimization objective function is constructed. The core optimization objectives are to minimize the deviation between the contact resistance after crimping and the target value, minimize the contact resistance dispersion within a batch, and ensure that the crimping parameters are within the process boundaries. The objective function formula is as follows: ; in, The overall objective function value is optimized for multiple objectives, where F represents the pressing force, S represents the pressing stroke, and T represents the holding time, which are the differentiated pressing parameters to be optimized; w1, w2, and w3 are weighting coefficients, which are dimensionless and satisfy w1+w2+w3=1, and are preset according to the mass production process requirements. This is the standard deviation of the contact resistance after all contact points in the same batch are crimped, used to control batch dispersion. This is a boundary penalty term; it is applied when the parameter exceeds the preset process boundary. When the parameter is within the boundary, P=0.

[0104] Furthermore, the prior distribution of Bayesian optimization is set to a Gaussian process, and the kernel function is the Matern5 / 2 kernel adapted to nonlinear characteristics. The kernel function formula is as follows: in, The kernel function output value represents the two sets of compression parameter vectors. and The degree of correlation between the two sets of parameters; the larger the value, the closer the crimping effect of the two sets of parameters. , These are two sets of crimping parameter vectors to be optimized; , represents the Euclidean distance between two sets of parameter vectors; l is the length scale parameter, which is dimensionless; Let V be the signal variance, which is dimensionless.

[0105] Furthermore, the acquisition function is set as the desired improvement (EI) function, balancing the exploration and utilization of the optimization process. The formula for the acquisition function is: Among them, f min f(x) represents the minimum value of the objective function among the currently sampled parameters; f(x) represents the predicted value of the objective function for the parameter x to be sampled.

[0106] Furthermore, Bayesian iterative optimization is performed, with a preset maximum number of iterations of 50. In each iteration, the optimal sampling point is selected based on the EI function, the objective function value is calculated by substituting it into the nonlinear mapping model, and the Gaussian process model is updated until the iteration converges or the maximum number of iterations is reached.

[0107] Furthermore, after the iteration converges, the parameter combination that minimizes the objective function value is output as the differentiated crimping parameter for the i-th contact point.

[0108] Furthermore, obtain the vector of differentiated crimping parameters for a single contact point. Parameter matrix of all contacts crimped in the same batch .

[0109] Please see Figure 5 The calculation of the relative deviation between the dynamic contact resistance and the target contact resistance, and the real-time calculation and correction of the crimping force and crimping stroke based on the relative deviation, includes the following steps: S21. Perform sliding window filtering on the real-time acquired dynamic contact resistance and calculate the relative deviation between the filtered dynamic contact resistance and the target contact resistance.

[0110] Due to various factors such as equipment vibration during the crimping process, environmental electromagnetic interference, and minor errors in the testing instruments, the real-time acquired dynamic contact resistance data may contain abnormal fluctuations and random interference signals. Directly using the raw acquired data for parameter adjustment can easily lead to deviations in the parameter correction direction, affecting the conductivity stability and structural safety of the contact point. This embodiment employs a sliding window filter to process the real-time acquired dynamic contact resistance data. It calculates a weighted average or median of multiple sets of resistance data within the window, thereby filtering out instantaneous outliers and random interference while preserving the true trend of contact resistance changes.

[0111] In one possible embodiment, the sliding window moves synchronously as data acquisition progresses. Each time a new set of dynamic contact resistance data is acquired, it is included in the current window, while the earliest acquired set of data in the window is removed. The average or median of all data in the window is calculated by scrolling to obtain the filtered contact resistance value.

[0112] S22. Based on the relative deviation, an incremental control algorithm is used to calculate the crimping force correction and crimping stroke correction in real time, and the current crimping parameters are corrected in real time according to the crimping force correction and crimping stroke correction.

[0113] The incremental control algorithm is an intelligent optimization algorithm with multi-variable collaborative solution capability. It can simultaneously handle multiple variables such as pressing force, pressing stroke and holding time, and continuously adjust the parameter combination through iterative loops.

[0114] In one possible embodiment, the incremental control algorithm uses filtered dynamic contact resistance data as a basis, combined with a preset target contact resistance value in the mapping model, to calculate the deviation between the current crimping parameters and the optimal state. Based on this, it dynamically adjusts the crimping force, the displacement of the crimping stroke, and the holding time. After each parameter adjustment, the mapping model simulates and predicts the changes in contact resistance, structural deformation, and long-term stability after the adjustment, determining whether the adjusted parameters can make the contact resistance closer to the target value and whether a better conductivity can be achieved without damaging the coating and substrate.

[0115] As can be seen, this embodiment uses a sliding window filter to process the real-time acquired dynamic contact resistance, which can effectively filter out data noise caused by on-site vibration and electromagnetic interference, and eliminate abnormal fluctuation data. Based on the filtered resistance value, the relative deviation is accurately calculated, and combined with an incremental control algorithm, the crimping force and crimping stroke correction are quickly calculated, enabling real-time control of the crimping process. This allows for timely compensation for resistance fluctuations caused by material elastic-plastic deformation and interface micro-slippage, avoiding performance deviations caused by the solidification of crimping parameters. This effectively stabilizes the contact interface bonding state, further reduces the contact resistance fluctuation range, improves the uniformity of crimping, ensures long-term stability of BMS signal acquisition, and improves the assembly yield and long-term operational reliability of the cell connection assembly.

[0116] In one specific embodiment of this application, the specific process of real-time acquisition and closed-loop parameter correction of dynamic contact resistance during the crimping process includes first obtaining the single-contact differentiated crimping parameter P. i The dynamic contact resistance sequence R acquired at high frequency during the crimping process ct Sampling frequency f s Target contact resistance .

[0117] Furthermore, before the crimping process is performed, parameter P is... i The command is sent to the crimping execution unit, which performs the crimping action according to the preset crimping speed, and simultaneously at the sampling frequency. Collect dynamic contact resistance sequence during the crimping process , Sampling time, .

[0118] Furthermore, a sliding window filtering algorithm is used to filter the dynamic contact resistance sequence in real time to eliminate sampling noise and electromagnetic interference.

[0119] Furthermore, a real-time deviation feedback model is constructed to calculate the relative deviation between the current filtered contact resistance and the target contact resistance. .

[0120] Furthermore, based on relative deviation An incremental (Incremental-Integral-Derivative) control algorithm is adopted to calculate the correction increment of the pressing force and pressing stroke in real time, thereby realizing closed-loop control of the pressing process. The correction formula is as follows: ; ; in, This represents the correction increment of the clamping force at time t, in N; K represents the correction increment for the crimping stroke at time t, in meters (m). p K i K d These are the proportional, integral, and derivative coefficients for the pressure control, respectively, and are dimensionless. These are the proportional, integral, and derivative coefficients for the crimping stroke control, respectively, and are dimensionless. These represent the relative deviations of the dynamic contact resistance from the target contact resistance at time t, time t-1, and time t-2, respectively.

[0121] Furthermore, after the crimping stroke reaches the corrected target stroke, the pressure holding stage begins, with a holding time of the corrected T. i During the pressure holding process, the dynamic contact resistance is continuously collected. If the relative deviation is... If the pressure exceeds the ±5% threshold, the holding pressure will be adjusted in real time, and the pressing operation will be stopped after the holding pressure is completed.

[0122] Furthermore, the full-time dynamic resistance dataset of the crimping process and the corrected final crimping execution parameters are obtained. Measured static contact resistance after crimping .

[0123] Please see Figure 6 The contact characteristic verification of the contact points in the crimped cell connection assembly includes the following steps: S31. Electrochemical impedance spectroscopy is used to detect the contact characteristics of the contact points in the crimped battery cell connection assembly to obtain detection data.

[0124] Among them, electrochemical impedance spectroscopy is used to detect the contact characteristics of the contact points in the crimped battery cell connection assembly. By applying alternating excitation signals of different frequencies, the impedance response characteristics of the contact interface under different operating conditions are collected.

[0125] Optionally, the detection data obtained by electrochemical impedance spectroscopy includes multi-dimensional raw information such as impedance modulus, phase angle, and real and imaginary parts of impedance at different frequencies, which can reflect multiple physical states such as the conductivity of the substrate inside the contact point, the contact surface adhesion state, the distribution of interfacial thin-layer material, and the size of micro gaps.

[0126] S32. Fit the detection data based on the equivalent circuit model to obtain the interface charge transfer resistance of the contact point.

[0127] The equivalent circuit model is a standardized analytical model built based on the actual interface structure and charge conduction mechanism after the contact point is pressed and formed. It can abstract and decompose the electrical behavior of the physical contact interface into a combination of basic electrical components such as resistors and capacitors, and match the conductivity characteristics of the contact point substrate, the characteristics of the interface film layer and the charge conduction path of the contact surface.

[0128] In one possible embodiment, electrochemical impedance spectroscopy (EIS) data is imported into a pre-prepared equivalent circuit model. A data fitting algorithm is used to match and calibrate the measured data with the model's theoretical curve, eliminating irrelevant interference components such as substrate impedance and lead impedance, and thus determining the interface charge transfer resistance specific to the contact point. This interface charge transfer resistance reflects the effectiveness of surface oxide layer removal during the crimping process, the number of micro-contact points formed, and the tightness of the adhesion between the two contact surfaces. It can effectively predict the performance degradation trend of the contact point under complex automotive operating conditions such as long-term vibration and high / low temperature cycling.

[0129] S33. Cross-verify the interface charge transfer resistance and the measured contact resistance after crimping, and determine the pass / fail status by combining the batch contact resistance dispersion.

[0130] The measured contact resistance after crimping is a direct and intuitive conductivity value obtained using conventional electrical testing methods, which can quickly determine whether the immediate conductivity of the contact point meets basic usage standards. By cross-validating the interface charge transfer resistance with the measured contact resistance after crimping, it is possible to judge the immediate performance of the product based on the measured contact resistance, and to identify quality defects such as bonding gaps, oxidation residues, and incomplete crimping within the interface based on the interface charge transfer resistance.

[0131] In one possible embodiment, various parameter qualification thresholds are pre-set based on industry automotive-grade component inspection standards. First, it is determined whether the measured contact resistance and interface charge transfer resistance of a single contact point are both within the preset qualification range, completing the single-point contact quality verification. Then, the overall contact resistance dispersion of the same batch of products is statistically calculated to confirm that the dispersion value is lower than the mass production control limit. When both single-point dual parameter compliance and overall batch dispersion compliance are met simultaneously, the cell connection assembly is determined to be a qualified assembly product. If a single-point parameter exceeds the standard, or the overall batch resistance dispersion exceeds the allowable range, it is uniformly determined to be a non-qualified product.

[0132] As can be seen, this embodiment can identify quality problems such as micro-defects at the interface and poor contact by cross-verifying the interface charge transfer resistance and the measured contact resistance. At the same time, it can make a comprehensive judgment by combining the batch contact resistance dispersion, realize single-point performance verification and batch consistency control, reduce the risk of failure such as BMS signal acquisition drift and contact point overheating, thereby effectively preventing the hidden danger of battery pack thermal runaway and improving product quality and long-term operational safety.

[0133] In one specific embodiment of this application, a detailed operational procedure is provided for verifying the consistency and passability of the impedance spectrum of contact points after crimping. First, the static contact resistance after crimping is obtained. The original impedance spectrum data output by the EIS test unit and the preset pass / fail threshold.

[0134] Furthermore, for each crimped contact point i, full-band EIS test data were collected to obtain the real part of the impedance at different frequencies. virtual part Construct a Nyquist curve dataset.

[0135] Furthermore, the EIS data is analyzed using an equivalent circuit fitting method. The equivalent circuit adopts the industry-known R... s -(R ct -Q) model, where R s R is the ohmic resistance of the system. ct Q is the interface charge transfer resistance (corresponding to the actual contact resistance of the interface at the contact point), and Q is a constant phase angle element (corresponding to the double-layer capacitance characteristic of the interface).

[0136] Furthermore, the interfacial charge transfer resistance at each contact point was obtained through fitting. ,and Cross-validation is performed, and the validation decision formula is as follows: The verification is successful if the calculation result is less than or equal to the preset qualified threshold; otherwise, it is determined that there is a minor defect in the contact point interface and it is directly marked as unqualified.

[0137] Furthermore, calculate the number of qualified contact points in the same batch. arithmetic mean with standard deviation The batch contact resistance dispersion is obtained according to the following formula: .

[0138] Furthermore, a multi-dimensional qualification criterion is constructed. A single contact point must simultaneously meet all of the following conditions to be considered qualified; for components in the same batch, all contact points must be qualified for the batch to be considered qualified: the static contact resistance after crimping meets the following requirements. EIS cross-validation deviation ≤10%; contact resistance dispersion of contact points in the same batch meets CV. Rct ≤2.

[0139] Furthermore, we obtained the single contact point conformity assessment results, the overall conformity assessment results of the same batch of CCS components, and the contact point interface characteristic evaluation report.

[0140] In another specific embodiment of this application, the entire lifecycle traceability management of crimping data and the continuous iterative update of the nonlinear mapping model are realized. Specifically, the full-process crimping operation data output from the above assembly method, the finished product contact characteristic qualification judgment results, and the long-term service status tracking data after the product is off the production line are used as a unified input data source.

[0141] Furthermore, firstly, the initial characteristic data, customized crimping parameters, dynamic monitoring data of the crimping process, finished product performance verification data, and quality judgment results corresponding to each contact point are bound and archived with the unique identification code of each contact point and uniformly stored in the industrial database. This enables full lifecycle traceability and management of the entire crimping process data. Secondly, complete production data of qualified batches are selected to form a new training sample set. Based on this sample set, incremental training is carried out on the nonlinear mapping mechanism model built above, and the fitting parameters within the model are updated synchronously to continuously optimize the model's calculation logic and steadily improve the model's prediction accuracy of contact resistance under different operating conditions. At the same time, combined with feedback data from long-term service tracking of the product, the weight coefficients within the optimization objective function used in the process optimization are dynamically adjusted to further optimize the adaptability of the crimping process and improve the reliability of the cell connection assembly under long-term service conditions.

[0142] Please see Figure 7 This application also provides a cell connection assembly system 100, which includes a main control device 200 and multiple functional components respectively communicatively connected to the main control device 200. The main control device 200 includes: The parameter generation module 201 is used to obtain the initial feature dataset of each contact point in the battery cell connection assembly to be crimped, and generate differentiated crimping parameters for each contact point based on the initial feature dataset and the mapping model. The crimping control module 202 is used to perform crimping on each contact point according to the differentiated crimping parameters, and to collect the dynamic contact resistance of the contact point in real time, calculate the relative deviation between the dynamic contact resistance and the target contact resistance, and calculate and correct the crimping force and crimping stroke in real time based on the relative deviation to obtain the crimped cell connection assembly. The verification output module 203 is used to verify the contact characteristics of the contact points in the crimped cell connection assembly and output a qualified crimped cell connection assembly.

[0143] The main control device 200 of the battery cell connection assembly system is capable of executing the battery cell connection assembly method provided in any embodiment of this application.

[0144] It should be noted that this embodiment is only a preferred architecture of the battery cell connection component assembly system 100. In actual industrial applications, the battery cell connection component assembly system 100 has various implementation forms and layout combinations, and is not limited to this fixed structure. At the hardware deployment level, each functional module can be set as an integrated device according to the production line layout requirements, suitable for compact production scenarios with small workstations; alternatively, each module can be separated into independent single-unit devices, relying on conveyor lines to complete material flow, adapting to large-scale mass production scenarios on long production lines.

[0145] Furthermore, in terms of functional configuration, the battery cell connection assembly system 100 can add or remove auxiliary functional units according to product production specifications and precision requirements. For small-batch customized production, some data calculation units can be simplified, while for mass production of high-precision automotive products, expansion units such as multi-dimensional environmental monitoring and multi-channel synchronous detection can be added. In terms of control implementation, it can complete the entire assembly operation using a local stand-alone autonomous control mode, or it can be connected to a cloud management platform to achieve remote parameter distribution, production data aggregation, and unified batch process control.

[0146] The specific workflow, data operation logic, and job judgment method executed by the parameter generation module 201, the crimping control module 202, and the verification output module 203 are all matched one by one with the corresponding technical steps in the assembly method described above, and will not be repeated here.

[0147] Please see Figure 8 and Figure 9 In one possible embodiment of this application, the hardware system architecture of the battery cell connection assembly system 100 includes a main control device 200 and multiple functional components that are communicatively connected to the main control device 200. The multiple functional components include a workpiece positioning and clamping subsystem 300, a multi-dimensional initial feature acquisition subsystem 400 for contact points, a servo pressing execution and dynamic acquisition subsystem 500, and a full consistency verification subsystem 600 after pressing. All hardware units are connected through an industrial bus.

[0148] The main control device 200 controls the positioning and clamping subsystem 300 of the workpiece to be crimped to fix and visually position the battery cell connection assembly to be crimped, so that each contact point in the battery cell connection assembly to be crimped moves to the preset crimping station, and ensures that the coaxiality error between the contact point and the crimping punch is within the allowable range.

[0149] In one possible embodiment, the workpiece positioning and clamping subsystem 300 includes a servo positioning slide. The workpiece positioning and clamping subsystem 300 may further include a vacuum adsorption fixture, an industrial charge-coupled device (CCD) vision alignment unit, a busbar, or a flexible printed circuit (FPC) limiting fixture.

[0150] The main control device 200 controls the multi-dimensional initial feature acquisition subsystem 400 of the contact point to collect the initial contact resistance, coating thickness, surface roughness and flatness error of each contact point to be pressed, forming an initial feature dataset.

[0151] In one possible embodiment, the multi-dimensional initial feature acquisition subsystem 400 for contact points includes a micro-resistance scanning unit. The multi-dimensional initial feature acquisition subsystem 400 may further include a laser confocal thickness measurement unit, a three-dimensional topography scanning unit, and a linear servo scanning motion module.

[0152] The main control device 200 controls the servo crimping execution and dynamic acquisition subsystem 500 to perform crimping actions according to the generated differentiated crimping parameters, and simultaneously acquires the dynamic contact resistance of the contact points in real time during the crimping process. At the same time, it receives the crimping force correction amount and crimping stroke correction amount calculated in real time by the main control device based on the relative deviation between the dynamic contact resistance and the target contact resistance, and completes the closed-loop correction of the crimping parameters to obtain the crimped cell connection assembly.

[0153] In one possible embodiment, the servo crimping execution and dynamic acquisition subsystem 500 includes a servo crimping module. The servo crimping execution and dynamic acquisition subsystem 500 may further include a crimping punch, a resistance acquisition unit, a pressure holding control unit, and a shielded signal transmission cable.

[0154] The main control device 200 controls the full consistency verification subsystem 600 after crimping to perform electrochemical impedance spectroscopy detection on each contact point in the crimped cell connection assembly, obtain contact characteristic data, and output a sorting signal according to the preset qualification judgment criteria to separate the crimped and assembled qualified components.

[0155] In one possible embodiment, the full consistency verification subsystem 600 after crimping consists of an electrochemical impedance spectroscopy testing unit and a defective product sorting execution unit.

[0156] As can be seen, the cell connection assembly system 100 provided in this embodiment, compared with the existing cell connection assembly crimping process which uses uniform parameters and results in high contact resistance dispersion, poor consistency, and insufficient long-term reliability, firstly obtains the initial feature dataset of each contact point in the cell connection assembly to be crimped through the parameter generation module 201, and generates differentiated crimping parameters for each contact point based on the initial feature dataset and mapping model; then, through the crimping control module 202, crimps each contact point according to the differentiated crimping parameters, and collects the dynamic contact resistance of the contact point in real time, calculates the relative deviation between the dynamic contact resistance and the target contact resistance, and calculates and corrects the crimping force and crimping stroke in real time based on the relative deviation to obtain the crimped cell connection assembly; finally, through the verification output module 203, the contact characteristics of the contact points in the crimped cell connection assembly are verified, and a qualified crimped cell connection assembly is output. Because this application adopts a process control method of single-contact differential parameter adaptation, real-time correction during crimping, and full interface verification after crimping, it can reduce contact resistance dispersion and improve crimping consistency. At the same time, it can improve voltage acquisition accuracy and long-term service reliability, effectively reduce the risk of contact degradation, overheating failure and thermal runaway, and also improve mass production yield, meeting the high precision and high reliability requirements of mass production.

[0157] In this application, the terms "embodiment" and "implementation" mean that a specific feature, structure, or characteristic described in connection with an embodiment can be included in at least one embodiment of this application. The appearance of these phrases in various locations throughout the specification does not necessarily refer to the same embodiment, nor are they independent or alternative embodiments mutually exclusive with other embodiments. Those skilled in the art will understand, explicitly and implicitly, that the embodiments described in this application can be combined with other embodiments. Furthermore, it should be understood that the features, structures, or characteristics described in the various embodiments of this application can be arbitrarily combined to form another embodiment that does not depart from the spirit and scope of the technical solution of this application, provided there is no contradiction between them.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail with reference to the above preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions to the technical solutions of this application should not depart from the spirit and scope of the technical solutions of this application.

Claims

1. A method for assembling a battery cell connection assembly, characterized in that, The method includes the following steps: An initial feature dataset is obtained for each contact point in the battery cell connection assembly to be crimped. Based on the initial feature dataset and the mapping model, differentiated crimping parameters are generated for each contact point. The mapping model is constructed based on Holm contact resistance theory and Hertz contact theory. The differentiated crimping parameters include crimping force, crimping stroke, and holding time. Generating differentiated crimping parameters for each contact point based on the initial feature dataset and the mapping model includes: constructing a mapping model of contact resistance, crimping force, and crimping stroke; determining the target contact resistance for each contact point based on the initial feature dataset and the mapping model; and optimizing the crimping force, crimping stroke, and holding time within a preset process constraint range, using the target contact resistance as the optimization objective, to obtain the results for each contact point. Differentiated crimping parameters; the construction of a mapping model for contact resistance, crimping force, and crimping stroke includes: based on Holm contact resistance theory, decomposing the contact resistance of the contact point into shrinkage resistance and film resistance; based on Hertz contact theory, establishing a nonlinear correlation between the shrinkage resistance and crimping force and crimping stroke, and based on an initial feature dataset, establishing a correlation between the film resistance and the surface morphology and coating characteristics of the contact point; combining the nonlinear correlation between the shrinkage resistance and crimping force and crimping stroke and the correlation between the film resistance and the surface morphology and coating characteristics of the contact point to obtain a nonlinear mapping mechanism model for contact resistance, crimping force, and crimping stroke; and fitting and calibrating the nonlinear mapping mechanism model based on the measured initial resistance of the contact point to obtain the mapping model. Crimping is performed on each contact point according to the differentiated crimping parameters, and the dynamic contact resistance of the contact point is collected in real time. The relative deviation between the dynamic contact resistance and the target contact resistance is calculated. Based on the relative deviation, the crimping force and crimping stroke are calculated and corrected in real time to obtain the crimped cell connection assembly. The calculation of the relative deviation between the dynamic contact resistance and the target contact resistance, and the calculation and correction of the crimping force and crimping stroke based on the relative deviation, include: performing sliding window filtering on the real-time collected dynamic contact resistance, calculating the relative deviation between the filtered dynamic contact resistance and the target contact resistance; based on the relative deviation, using an incremental control algorithm to calculate the crimping force correction amount and the crimping stroke correction amount in real time, and correcting the current crimping parameters in real time according to the crimping force correction amount and the crimping stroke correction amount. The contact characteristics of the contact points in the crimped cell connection assembly are verified, and a qualified crimped cell connection assembly is output. The verification of the contact characteristics of the contact points in the crimped cell connection assembly includes: using electrochemical impedance spectroscopy to detect the contact characteristics of the contact points in the crimped cell connection assembly and obtaining detection data; fitting the detection data based on an equivalent circuit model to obtain the interfacial charge transfer resistance of the contact points; cross-validating the interfacial charge transfer resistance with the measured contact resistance after crimping, and determining the pass / fail status based on the batch contact resistance dispersion.

2. The method as described in claim 1, characterized in that, The fitting and calibration of the nonlinear mapping mechanism model based on the measured initial resistance at the contact point includes: Using the measured initial resistance of all contact points in the same batch as the calibration benchmark, the least squares method is adopted, and the minimum mean square error between the calculated value and the measured value is used as the optimization objective to iteratively optimize the fitting parameters of the nonlinear mapping mechanism model. When the mean square error meets the preset threshold and the model fit meets the preset requirements, the calibration of the nonlinear mapping mechanism model is completed.

3. The method as described in claim 2, characterized in that, The iterative optimization of the fitting parameters of the nonlinear mapping mechanism model includes: Based on the initial feature dataset of the contact points, the contact coefficient of the membrane layer is differentiated and corrected. Based on the elastic-plastic properties of the contact point material, the plastic deformation parameters corresponding to the pressing stroke are adapted and corrected to achieve iterative optimization of the fitting parameters of the nonlinear mapping mechanism model. During the iteration process, the goodness of fit of the model is constrained to be no less than the preset lower limit.

4. The method as described in claim 1, characterized in that, The optimization of the crimping force, crimping stroke, and holding time within the preset process constraints includes: For each contact point, an optimization objective function is constructed based on the initial feature dataset and mapping model; An optimization algorithm is used to iteratively optimize within the preset process constraints, and outputs the crimping force, crimping stroke and holding time that meet the optimization objective function requirements for each contact point, as the corresponding differentiated crimping parameters.

5. The method as described in claim 4, characterized in that, The construction of the optimization objective function includes: The optimization objective is to minimize the deviation between the contact resistance after crimping and the target contact resistance, and to minimize the dispersion of the contact resistance of all contact points in the same batch. A corresponding weighting coefficient is assigned to each optimization objective, and a penalty term for exceeding the process boundary constraints is added to form the optimization objective function.

6. A cell connection assembly system, characterized in that, The cell connection assembly system includes a main control device and multiple functional components that are communicatively connected to the main control device. The main control device includes: The parameter generation module is used to acquire the initial feature dataset of each contact point in the battery cell connection assembly to be crimped. Based on the initial feature dataset and the mapping model, it generates differentiated crimping parameters for each contact point. The mapping model is constructed based on Holm contact resistance theory and Hertz contact theory. The differentiated crimping parameters include crimping force, crimping stroke, and holding time. Generating differentiated crimping parameters for each contact point based on the initial feature dataset and the mapping model includes: constructing a mapping model of contact resistance, crimping force, and crimping stroke; determining the target contact resistance for each contact point based on the initial feature dataset and the mapping model; and optimizing the crimping force, crimping stroke, and holding time within a preset process constraint range, using the target contact resistance as the optimization objective, to obtain the parameters for each contact point. Differentiated crimping parameters at the contact point; the construction of a mapping model for contact resistance, crimping force, and crimping stroke includes: based on Holm contact resistance theory, decomposing the contact resistance of the contact point into shrinkage resistance and film resistance; based on Hertz contact theory, establishing a nonlinear correlation between the shrinkage resistance and crimping force and crimping stroke, and based on an initial feature dataset, establishing a correlation between the film resistance and the surface morphology and coating characteristics of the contact point; combining the nonlinear correlation between the shrinkage resistance and crimping force and crimping stroke and the correlation between the film resistance and the surface morphology and coating characteristics of the contact point to obtain a nonlinear mapping mechanism model for contact resistance, crimping force, and crimping stroke; and fitting and calibrating the nonlinear mapping mechanism model based on the measured initial resistance of the contact point to obtain the mapping model. The crimping control module is used to perform crimping on each contact point according to the differentiated crimping parameters, and to collect the dynamic contact resistance of the contact point in real time, calculate the relative deviation between the dynamic contact resistance and the target contact resistance, and calculate and correct the crimping force and crimping stroke in real time based on the relative deviation to obtain the crimped cell connection assembly; the calculation of the relative deviation between the dynamic contact resistance and the target contact resistance, and the calculation and correction of the crimping force and crimping stroke in real time based on the relative deviation, includes: performing sliding window filtering on the real-time collected dynamic contact resistance, calculating the relative deviation between the filtered dynamic contact resistance and the target contact resistance; based on the relative deviation, using an incremental control algorithm to calculate the crimping force correction amount and the crimping stroke correction amount in real time, and correcting the current crimping parameters in real time according to the crimping force correction amount and the crimping stroke correction amount; The verification output module is used to verify the contact characteristics of the contact points in the crimped cell connection assembly and output a qualified crimped cell connection assembly. The verification of the contact characteristics of the contact points in the crimped cell connection assembly includes: using electrochemical impedance spectroscopy to detect the contact characteristics of the contact points in the crimped cell connection assembly and obtaining detection data; fitting the detection data based on an equivalent circuit model to obtain the interfacial charge transfer resistance of the contact points; cross-validating the interfacial charge transfer resistance with the measured contact resistance after crimping, and determining the pass / fail status based on the batch contact resistance dispersion.

Images