Process arrangement and method for assembling a battery system

A self-learning process optimizes thermal paste application in battery assembly for electric vehicles, reducing material waste and press-in forces while ensuring efficient heat dissipation.

DE102023112765A1Pending Publication Date: 2024-11-21AUDI AG
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
DE102023112765
Authority / Receiving Office
DE · DE
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-15
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing battery assembly processes for electric vehicles require excessive use of thermal paste to ensure heat dissipation, leading to high material consumption, increased component weight, and high press-in forces.

Method used

A self-learning process arrangement with integrated sensors and control loops adjusts the application of thermal paste based on actual layer thickness and surface profiles, optimizing the quantity and distribution to minimize air gaps and reduce material usage.

Benefits of technology

Reduces thermal paste consumption and press-in forces while maintaining effective heat dissipation by optimizing the thermal paste application process through self-learning and adaptive control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

The invention relates to a process arrangement for assembling a battery system consisting of battery modules (7) inserted into a battery housing (1), each of which can be inserted into the battery housing (1) in a module insertion process, in which the respective battery module (7) is inserted into the battery housing (1) in an insertion step, and an application step is preceding the insertion step, in which an application unit (19) applies at least one bead of thermal paste (17) to the housing base (11) to fill a gap between the battery module (7) and the housing base (11), wherein the application unit (19) is integrated into a control circuit (R) in which an electronic control unit (21) adjusts the flow rate of thermal paste applied by the application unit (19) as a function of a setpoint (zS(x,y)), thereby forming a thermal paste layer (13) on the housing base. (11).According to the invention, the process arrangement is designed as a self-learning system.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a process arrangement for assembling a battery system for an electrically powered vehicle according to the preamble of claim 1 and to a method for assembling such a battery system according to the preamble of claim 10.

[0002] The high-voltage battery of an electric vehicle comprises battery modules arranged in a battery housing. Battery assembly of this type takes place in a production facility where, with the help of an industrial robot in a fully automated process chain, an empty battery housing is first prepared. The housing interior is divided by partition walls into chambers, each of which contains a battery module.

[0003] Subsequently, a module setting process is carried out in which the battery modules are inserted into a battery housing in a setting direction with a retraction movement in one setting step until an assembly position is reached in which the respective battery module rests correctly on a housing-side contact contour.

[0004] Before the insertion step, an application process step is performed in which a highly viscous thermal paste is applied to the housing base to prevent a thermally insulating air gap between the battery module and the housing base during assembly. After the thermal paste has been applied, the battery module is pressed against the thermal paste with a contact force during the module insertion process.

[0005] To ensure proper heat dissipation, it is important that the gap between the battery module and the housing base is almost completely filled with thermal paste, so that air pockets in the gap can be prevented as far as possible. For this reason, in a conventional application step, the thermal paste is applied to the housing base in an excess quantity that is significantly larger than the actually required fill quantity. This leads to excessive material consumption of the costly thermal paste and an excessively high component weight of the high-voltage battery. It also requires high press-in forces to insert the battery module into the thermal paste in its assembly position.

[0006] From DE 10 2021 105 355 A1 a detection device and a method for applying a thermally conductive compound to a cooling base of a battery housing are known.

[0007] The object of the invention is to provide a process arrangement and a method for assembling a battery system for an electrically powered vehicle, in which the use of thermal paste can be reduced compared to the prior art while otherwise maintaining the same heat dissipation performance.

[0008] The object is achieved by the features of claim 1 or 10. Preferred developments of the invention are disclosed in the subclaims.

[0009] The invention is based on a process arrangement for assembling a battery system for an electrically powered vehicle. The battery system consists of battery modules inserted into a battery housing. Each battery module is inserted into the battery housing in a module insertion process. The module insertion process includes a insertion step in which the respective battery module is inserted into the battery housing until it reaches an assembly position. This insertion step is preceded by an application step in which an application unit applies at least one bead of thermally conductive paste to the housing base to fill a gap between the battery module and the housing base.

[0010] The application unit is integrated into a control circuit in which an electronic control unit adjusts a thermal paste flow rate dispensed by the application unit depending on a target specification, whereby a thermal paste layer forms on the housing base. According to the characterizing part of claim 1, the process arrangement is a self-learning system with which at least one teach-in module insertion process can be carried out based on a teach-in target specification. For this purpose, the process arrangement can have a correction module with which the teach-in target specification can be corrected into a corrective target specification during the teach-in module insertion process. A subsequent module insertion process taking place after the teach-in module insertion process can therefore be carried out based on the corrective target specification.In the subsequent module installation process following the teach-in module installation process, the control unit no longer adjusts the thermal paste flow rate based on the original teach-in target value, but rather based on the correction target value generated in the previous teach-in module installation process. The correction is performed based on an actual value of the thermal paste layer, in particular an actual layer thickness, which is detected by sensors at the end of the application step and / or after completion of the installation process.

[0011] In a specific implementation, the correction module is in signal communication with at least one first test unit. After completion of the application step, the first test unit can perform a test process in which a sensor detects an actual value, in particular the actual layer thickness, of the thermal paste layer. The test unit also has a comparator module that determines a difference value from a comparison of the actual value and the stored target value. Based on this difference value, the correction module generates the target correction specification, which can be used as the basis for the subsequent module installation process following the module installation process.

[0012] The determined target correction value can be read and stored in a database that is part of the control loop. This makes the target correction value available for the application step of the subsequent module setting process.

[0013] It is preferred if the testing process of the first testing unit can be carried out before the setting step, i.e. when the thermal paste layer is not yet pressed.

[0014] Alternatively and / or in addition to the first test unit, the correction module can be in signal communication with at least one second test unit. After completion of the setting step, this can perform a test process in which a sensor detects an actual value, in particular the actual layer thickness, of the thermal paste layer. The second test unit can also have a comparator module in which a difference value can be determined from a comparison of the actual value and a stored target value. Based on this difference value, the correction module can generate the target correction specification. It is preferred if both the first test unit and the second test unit are available. In this case, the target correction specification is generated based on the difference value determined by the first and second test units.The difference values ​​generated by the first test unit and the second test unit reflect different information contents: In the first test unit, the actual value is recorded when the thermal paste layer is not yet pressed, while in the second test unit, the actual value is recorded when the thermal paste layer is pressed.

[0015] In a preferred embodiment, a sensor for detecting a thermal paste fill level or a related parameter, such as the actual profile height of the thermal paste bead applied by the application unit, can be integrated into the control loop. The electronic control unit of the control loop can have a comparator module that determines a difference value from the comparison of the actual profile height and the target learning value / correction target value. Based on this difference value, the electronic control unit controls the application unit.

[0016] The control / regulation circuit can also be assigned a compensation unit with a sensor that detects unevenness in the housing base. The compensation unit can have a comparator module that determines a difference between the actual surface profile and a stored reference value. Based on this difference, the teach-in target value / correction target value can be adjusted depending on the detected surface profile of the housing base.

[0017] In one technical implementation, the sensor of the first inspection unit can be a linear scanner that optically scans the surface of the unpressed thermal paste layer. The sensor of the second inspection unit, on the other hand, can be an X-ray device that performs an X-ray inspection of the contact surfaces of the thermal paste layer with the battery module and the housing base. The sensor for detecting the thermal paste fill level can be a photo-optical sensor, which can, for example, detect an outlet cross-section of the application unit through which the thermal paste bead emerges. Alternatively and / or additionally, the sensor for detecting the actual surface profile of the housing base surface can also be a linear scanner that can optically scan the housing base surface.

[0018] An embodiment of the invention is described below with reference to the attached figures.

[0019] They show: Fig. 1 to 7 are different representations illustrating the process arrangement for assembling a battery system.

[0020] In the Fig. 1 shows an assembled high-voltage battery system for an electrically powered vehicle in a partial sectional view. Accordingly, the battery system comprises a battery housing 1 with a housing cover 2. The housing interior is divided by partition walls 3 into chambers 5, in each of which a battery module 7 is arranged. Of the chambers 5, Fig. 1 only shows a chamber 5 with a battery module 7 positioned therein. The battery module 7 is according to the Fig. 1 is supported on a housing-side contact contour 9. To reduce heat-insulating air inclusions 15 ( Fig. 4) between the battery module and the housing base 11, the gap between the battery module 7 and the housing base 11 is largely completely filled with a thermal paste layer 13.

[0021] The thermal paste layer 13 consists according to the Fig. 4 consists of adjacently arranged thermal paste beads 17, which have an oval cross-section. This results in the Fig. 4, between the contact surface 12 of the battery module base or the housing base 11 and the thermal paste layer 13. The aim is for the air inclusions 15 to cover only the smallest possible surface area of ​​the contact surfaces 12. For example, a 90% surface connection of the thermal paste layer 13 to the contact surfaces 6 can be desired, so that the contact surface area covered by the air inclusions 15 would be 10%.

[0022] During battery assembly, each of the battery modules 7 is positioned in the associated chamber 5 of the battery housing 1 in a module placement process. The module placement process comprises an application step and a placement step. In the placement step, the respective battery module 7 is inserted into the chamber 5 until it reaches the contact contour 9. The placement step is preceded by the application step, in which an application unit 19 ( Fig. 5) the thermal paste beads 17 are applied to the housing base 11 to fill the gap between the battery module 7 and the housing base 11.

[0023] During the module installation process, it is important to apply the thermal paste filler quantity without excess material and in a targeted manner to the housing base 11. This saves thermal paste material; moreover, the installation pressure required during the installation step, with which the battery module 7 is pressed onto the thermal paste layer 13, can be reduced.

[0024] For this purpose, according to the Fig. 2 a process arrangement with a control circuit R, a compensation unit 29 and a first and second test unit 35, 41 is provided, with which the module setting process in a Fig. 2 indicated process sequence t1 to t4 is feasible. For reasons of clarity, Fig. 2 does not show the entire battery housing 1 to be equipped with the battery modules 7, but only the housing base 11 of a single chamber 5 of the battery housing 1. The housing base 11 is in the Fig. 2 at different process times t1 to t4 of the process sequence.

[0025] A core of the invention is that the battery assembly is carried out in at least one or more learning module setting processes ( Fig. 2) as well as subsequent module setting processes ( Fig. 7). In the learning module setting process ( Fig. 2) the control loop R is extended by the two test units 35, 41, the correction module 27 and the target specification unit 10 in order to optimize a training target specification zS(x,y) of the control loop R to a correction target specification zSK(x,y). The subsequent module setting process ( Fig. 7) no longer takes place using the original training target specification zS (x,y), but using the already optimized correction target specification zSK(x,y), so that the above components (i.e. test units 35, 41, correction module 27, target specification unit 10) in the Fig. 7 are omitted.

[0026] The control circuit R consists of the learning module setting process according to the Fig. 2 consists of an electronic control unit 21, the application unit 19, a sensor 23 and the target specification unit 10, which provides the learning target specification zS(x,y) for the control loop R. The correction module 27 and a database 26 are interposed in the signal path from the target specification unit 10 to the control unit 21. In the learning module setting process ( Fig. 2) the correction module 27 corrects a learning target value zS(x,y) read from the target value unit 10 to the correction target value zSK(x,y). This is read into the database 26 and stored there. In the subsequent module setting process ( Fig. 7), a correction target value zSK(x,y) is read from the database 26—that is, no longer from the target value unit 10—and sent to the control unit 21. Based on the correction target value zSK(x,y), the control unit 21 sets an optimized thermal paste flow rate during the application step, which is applied by the application unit 19 as a thermal paste bead 17 onto the housing base 11. The sensor 23 detects the applied thermal paste quantity or a parameter correlating therewith, such as an actual bead height z1(x,y). The control unit 21 determines a difference value Δz1(x,y) from a comparison of the actual bead height z1(x,y) with the correction target value zSK(x,y). Based on the difference value Δz1 (x,y), the electronic control unit 21 controls the application unit 19.

[0027] In the process arrangement according to the invention, both the learning module setting process and the subsequent module setting process are carried out taking into account any existing unevenness in the floor t ( Fig. 3) in the housing base 11. For this purpose, the compensation unit 29 has a sensor 31 which measures an actual surface profile t(x,y) ( Fig. 3) of the housing base 11. In addition, the compensation unit 29 has a comparator module 33, which determines a difference value Δt(x,y) from the actual surface profile t(x,y) and a stored reference value tref(x,y), which is added in a summing module of the correction target specification zSK(x,y).

[0028] The (in the Fig. The teach-in module placement process (outlined in Figure 2) takes place, for example, at the beginning of a battery module assembly or after a tool change. Accordingly, at process time t1, the still unpopulated housing base 11 is prepared. Using the compensation unit 29, the actual surface profile t(x,y) is recorded, and the difference value Δt(x,y) is calculated from this. Further in the process, at process time t2, the application step takes place, in which the application unit 19 applies the thermal paste beads 17 to the housing base 11.

[0029] After completion of the application step, the still unpressed thermal paste layer 13 is formed on the housing base 11 at process time t3. At process time t3, the first test unit 35 starts a test procedure in which a sensor 37 detects an actual layer thickness z2(x,y) of the thermal paste layer 13. A comparator module 39 of the first test unit 35 determines a difference value Δz2(x,y) from a comparison of the actual layer thickness z2(x,y) and a stored target value zS2(x,y). The difference value Δz2(x,y) is passed to the correction module 27.

[0030] Subsequently, at process time t4, the setting step takes place, during which the thermal paste layer 14 is pressed between the battery module 7 and the housing base 11. After completion of the setting step, the second test unit 41 starts another test process. In this test process, a sensor 43 detects the size and / or position of the air inclusions 15 or a correlating parameter, such as an actual layer thickness z3(x,y) of the pressed thermal paste layer 13. A comparator module 45 determines a difference value Δz3(x,y) from a comparison of the actual layer thickness z3(x,y) and a stored target value zS3(x,y). This difference is also passed to the correction module 27. In correction block 27, the learning target value zS(x,y) is corrected into the correction target value zSK(x,y) based on the two difference values ​​Δz2(x,y) and Δz3(x,y).

[0031] With the help of the two test units 35, 41, the target specification unit 10 and the correction module 27, a self-learning system is provided which operates as follows: The learning module setting process is carried out on the basis of the learning target specification zS(x,y) stored in the target specification unit 10.

[0032] During the learning module setting process, the correction module 27 automatically corrects the learning target value zS(x,y) to the correction target value zSK(x,y), which is stored in the database 26. The correction target value zSK(x,y) is then available for the subsequent module setting process.

[0033] In at least one learning module setting process, the correction target value zSK(y,x) is determined from the learning target value zS(x,y), which forms a starting point, using the correction module 27. This is read into the database 26 and stored therein. The subsequent module setting process ( Fig. 7) can be carried out on the basis of the correction target specification zSK(y,x) stored in the database 26.

[0034] The application unit 19 is according to the Fig. 5 shows a conveyor device whose conveyor housing has a screw that conveys the thermal paste. The screw is mounted in the screw housing for longitudinal displacement to adjust the flow rate, allowing an outlet nozzle cross-section between an outlet nozzle and the screw and thus an exiting thermal paste flow rate to be adjusted. The thermal paste flow rate or a parameter correlating therewith can be detected by sensor 23. For example, sensor 23 can be a photo-optical sensor that detects the longitudinal displacement of the conveyor screw.

[0035] The sensor 43 used in the second test unit 41 is preferably a radiometric device as described in Fig. 6 is indicated. This has an X-ray or gamma radiation generating X-ray or gamma radiation source 42 and an imaging detector 44, wherein the X-ray or gamma radiation is generated as reflected X-rays in the Fig. 6 is directed towards the bottom of the housing, so that the size and / or position of the air inclusions 15 ( Fig. 4) are detectable at the contact surfaces 12 of the thermal paste layer 13 to the battery module 7 and to the housing base 11.

[0036] The target value zS3(x,y) stored in the second test unit 41 can be identical to the gap width between the battery module 7 and the housing base 11. The target value zS2(x,y) stored in the first test unit 35, however, can be greater than the target value zS3(x,y) of the second test unit 41.

[0037] In an alternative embodiment, the target value zS3(x,y) stored in the second test unit 41 can be a maximum value that specifies a maximum permissible contact surface area that can be covered by the air inclusions 15. For example, the maximum value can be 10% of the contact surface 12 of the thermal paste layer 13 to the battery module 7, so that the permissible surface connection area of ​​the contact surface 12 is at least 90%. List of reference symbols 1 battery case 3 partition walls 5 chambers 7 Battery module 9 Plant contour 11 Case back 12 Contact surface between thermal paste layer and battery module / housing base 13 Thermal paste layer 15 air pockets 17 Thermal paste bead 19 Application unit 21 Control unit 23 Sensor 26 Database 27 Correction module 29 Balancing unit 31 Sensor 33 Comparator module 35 first test unit 37 Sensor 39 Comparator module 41 second test unit 42 X-ray or gamma radiation source 43 Sensor 44 imaging detector 45 Comparator module R control loop QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2021 105 355 A1

[0006]

Claims

[1] Process arrangement for assembling a battery system which is formed from battery modules (7) inserted into a battery housing (1), each battery module (7) of which can be inserted into the battery housing (1) in a module insertion process, in which in one setting step, the respective battery module (7) can be inserted into the battery housing (1), and in an application step preceding the setting step, an application unit (19) applies at least one thermal paste bead (17) to the housing base (11) in order to fill a gap between the battery module (7) and the housing base (11), wherein the application unit (19) is integrated into a control circuit (R) in which an electronic control unit (21) adjusts a quantity of thermal paste applied by the application unit (19) as a function of a target specification, specifically with the formation of a thermal paste layer (13) on the housing base (11), characterized bythat the process arrangement is a self-learning system with which at least one learning module setting process can be carried out on the basis of a learning target specification (zS(x,y)), and that the process arrangement has a correction module (27) with which the learning target specification (zS(x,y)) can be corrected into a correction target specification (zSK(x,y)) during the learning module setting process, so that a subsequent module setting process can be carried out on the basis of the correction target specification (zSK(x,y)). [2] Process arrangement according to claim 1, characterized bythat the correction module (27) is in signal connection with at least one first test unit (35) which, after completion of the application step, carries out a test process on the still unpressed thermal paste layer (13), and that on the basis of the test result the correction module (27) generates the desired correction specification (zSK(x,y)), and that in particular in the test process a sensor (37) detects an actual value (z2(x,y)), in particular actual layer thickness, of the thermal paste layer (13), and a comparator module (39) determines a difference value (Δz2(x,y)) from a comparison of the actual value (z2(x,y)) and a stored desired value (zs2(x,y)), and that the correction module (27) generates the desired correction specification (zSK(x,y)) on the basis of the difference value (Δz2(x,y)). [3] Process arrangement according to claim 1 or 2, characterized bythat the control circuit (R) has a database (26) in which the correction target specification (zSK(x,y)) can be read in and stored, and that in the subsequent module setting process the application step can be carried out on the basis of the correction target specification (zSK(x,y)). [4] Process arrangement according to one of claims 1, 2 or 3, characterized by that the testing process of the first testing unit (35) can be carried out before the setting step, i.e. when the thermal paste layer (13) is still unpressed. [5] Process arrangement according to one of the preceding claims, characterized bythat the correction module (27) is in signal connection with at least one, in particular second, test unit (41), which carries out a test process of the thermal paste layer (13) after completion of the setting step, and that on the basis of the test result, the correction module (27) generates the correction target specification (zSK(x,y)), and that in particular in the test process, a sensor (43) detects an actual value (z3(x,y)), in particular actual layer thickness, of the thermal paste layer (13), and a comparator module (45) determines a difference value (Δz3(x,y)) from a comparison of the actual value (z3(x,y)) and a stored target value (zS3(x,y)), and that in particular the correction module (27) generates the correction target specification (zSK(x,y)) on the basis of the difference value (Δz3(x,y)), and that in particular in the second test unit (41) in contrast to the first test unit (35) the actual value is recorded when the thermal paste layer (13) is pressed. [6] Process arrangement according to one of the preceding claims, characterized by that a sensor (23) for detecting the actual thermal paste quantity applied or a parameter correlating therewith, in particular e.g. actual profile height (z1 (x,y)) of the applied thermal paste bead (17), is integrated in the control circuit (R), and that the electronic control unit (21) determines a difference value (Δz1 (x,y)) from a comparison of the actual thermal paste quantity with the learning target specification / correction target specification, and that the electronic control unit (21) controls the application unit (19) on the basis of the difference value (Δz1 (x,y)). [7] Process arrangement according to one of the preceding claims, characterized bythat the control / regulating circuit (R) is assigned a compensation unit (29) for detecting floor unevenness (t), which has a sensor (31) that detects an actual surface profile (t(x,y)) of the housing floor surface, and a comparator module (33) that determines a difference value (Δt(x,y)) from the actual profile height (t(x,y)) and a stored reference value (tref), and that the learning target specification / correction target specification can be adapted on the basis of the difference value (Δt(x,y)). [8] Process arrangement according to one of claims 2 to 7, characterized by that the sensor (37) of the first testing unit (35) is a linear scanner which optically detects the surface of the still unpressed thermal paste layer (13), and / or that the sensor (43) of the second testing unit (41) is an X-ray device which carries out an X-ray test by means of reflected light X-rays of the contact surfaces of the thermal paste layer (13) to the battery module (7) and to the housing base (11). [9] Process arrangement according to one of claims 6 to 8, characterized by that the sensor (23) for detecting the actual quantity of thermal paste or a parameter correlating therewith (z1 (x,y)) is a photo-optical sensor by means of which an outlet nozzle cross-section of the application unit (19) through which the thermal paste bead (17) emerges can be detected, and / or that the sensor (31) for detecting the actual profile height (t(x,y)) of the housing base surface is a linear scanner which optically detects the housing base surface. [10] Method for assembling a battery system for an electrically powered vehicle according to one of the preceding claims.

Citation Information

Patent Citations

  • Method for determining the quality status of a component used during a setting process and control device for monitoring a setting process.

    DE102021104146A1

  • Determination device and method for applying a thermally conductive compound to a cooling base of a battery housing

    DE102021105355A1

  • Method and apparatus for planning and carrying out a process for applying a thermal paste using machine learning methods

    DE102022200945A1