Method for assembling a plurality of cells into a battery module

Abstract

A method for assembling a multitude of cells into a battery module uses a controller with a processor and tangible, non-volatile memory. The method includes positioning the multitude of cells in a test rig for pre-assembly testing. The test rig is designed to accommodate a selected group of cells at a time. The method includes performing a pulse power test on the selected group via the test rig. The method includes receiving corresponding voltage waveform data, partly based on the pulse power test. A corresponding internal resistance parameter of the multitude of cells is monitored by the controller, partly based on the corresponding voltage waveform data.The process involves arranging the multitude of cells into sorted groups, partly based on the respective internal resistance parameter, and assembling the sorted groups in a predefined pattern within the battery module.

Description

[0001] The present disclosure generally relates to the management of energy storage in a vehicle. More specifically, the disclosure relates to a system and method for automatically managing a dual-design battery pack. The use of mobile platforms that utilize a rechargeable energy source, both as an exclusive power source and as a non-exclusive power source, has increased significantly in recent years. Energy storage with battery packs can store and release electrochemical energy as needed in a specific operating mode. Electrochemical energy can be used for propulsion, heating or cooling the passenger compartment, powering vehicle accessories, and other purposes. The different cells in the battery packs can be characterized by different states of charge, power, and capacity rates.To obtain a complete picture of cell quality, time-consuming measurements at a specific state of charge are usually required.

[0002] DE 10 2011 101 793 A1 discloses a method for selecting electrochemical cells in the production of a battery having a number of electrochemical cells, comprising the following steps: acquiring parameter data of an individual electrochemical cell to be examined, transmitting the acquired parameter data to a control unit, assigning the electrochemical cell to the acquired parameter data and determining by means of the control unit whether a predetermined relationship of the parameter data with respect to predetermined parameter values ​​exists for the electrochemical cell assigned to the parameter data.The method may further comprise the following steps: supplying the electrochemical cell assigned to the parameter data to a first production line for producing a first type of battery if the existence of the predetermined relationship is determined in the determining step, or supplying the electrochemical cell assigned to the parameter data to a second production line for producing a second type of battery if the non-existence of the predetermined relationship is determined in the determining step.

[0003] CN 1 15 902 628A discloses a method for calculating the internal resistance of a battery. Under preset fixed operating conditions, a first voltage deviation between a simulated voltage, which is simulated according to a battery model determined by the current stable internal resistance, and an actual voltage obtained by testing is calculated in real time. The initial value of the current stable internal resistance is the initial stable resistance obtained by prior testing of the battery. Furthermore, the current stable internal resistance of the battery is corrected according to the first voltage deviation calculated in real time, and the corrected current stable internal resistance is used as the real-time aging resistance.

[0004] DE 10 2011 017 113 A1 discloses a method for determining state variables of a rechargeable battery using a triple extended Kalman filter, in which a first Kalman filter is used to calculate the state of charge and the fast and slow overvoltages generated by the current, a second Kalman filter is used to calculate the internal resistance, and a third Kalman filter is used to calculate the cell capacity.

[0005] Disclosed herein is a method of assembling a plurality of cells into a battery module using a controller having a processor and tangible, non-volatile memory. The method includes positioning the plurality of cells in a test fixture for pre-assembly testing. The test fixture is configured to receive a selected group of the plurality of cells at a time. A pulse power test is performed on the selected group via the test fixture. The method includes receiving, via the controller, corresponding voltage waveform data based in part on the pulse power test. A corresponding internal resistance parameter of the plurality of cells is tracked via the controller based in part on the corresponding voltage waveform data.The method includes arranging the plurality of cells into sorted groups based in part on the respective internal resistance parameter, and assembling the sorted groups in a predefined pattern in the battery module.

[0006] The method includes integrating a probe, a current source, and a voltage measurement device into the test device, the probe having a first and a second end. Performing the pulse power test includes connecting a respective positive end and a respective negative end of the selected group to the first and second ends of the probe, injecting a test pulse into the selected group via the current source, and acquiring the respective voltage waveform data of the selected group via the voltage measurement device.

[0007] The method may include using a first robotic arm to transport the plurality of cells into a plurality of storage containers, each of which is assigned to a predefined range of the respective internal resistance parameter. The method may include using a second robotic arm capable of removing the plurality of cells from the plurality of storage containers.

[0008] The battery module may contain a plurality of parallel cell groups. In one embodiment, the plurality of cells on the battery module are arranged in a ranking order via the second robot arm such that the plurality of parallel cell groups are arranged in ascending order of their respective internal resistance parameters. The battery module may consist of an inner and an outer part. In one embodiment, the plurality of cells are arranged on the battery module via the second robot arm such that the majority of cells with a relatively high internal resistance parameter are located in the inner section and the majority of cells with a relatively low internal resistance parameter are located in the outer section.

[0009] The method may include integrating a Kalman filter into the observer module. The method may include determining standard deviation values ​​of the respective internal resistance parameter among the plurality of cells and developing a binary threshold for accepting and rejecting the plurality of cells based in part on the standard deviation values.

[0010] In one embodiment, the method comprises performing a statistical analysis to determine whether the respective internal resistance parameter for one of the plurality of cells is an outlier compared to the respective internal resistance parameter for a remainder of the plurality of cells. The method may include incorporating a principal component analysis into the statistical analysis, including determining the respective coefficients of a first principal mode for each of the plurality of cells. The assembled battery module may be installed in a vehicle.

[0011] Disclosed herein is a system for assembling a plurality of cells into a battery module. The system includes a tester configured to accept a selected group of the plurality of cells at a given time for conducting a pulse power test, wherein the plurality of cells are positioned in the tester for pre-assembly testing. The system includes a controller in communication with the tester, the controller including a processor and a tangible, non-volatile memory having instructions recorded therein.

[0012] The controller is configured to receive corresponding voltage waveform data based in part on the pulse power test, the controller being configured to track a corresponding internal resistance parameter of the plurality of cells based in part on the corresponding voltage waveform data and an observer module.

[0013] The majority of cells are divided into sorted groups based in part on their respective internal resistance parameters. The sorted groups are arranged in a predefined pattern within the battery module, with the predefined pattern based on a value of the respective internal resistance parameter. Fig. 1 is a schematic diagram of a system for assembling a plurality of cells into a battery module, the system having a controller; Fig. 2 is a schematic flow diagram of a method for assembling a plurality of cells into a battery module; Fig. 3A, Fig. 3B, Fig. 3C are example schematic diagrams each showing the load current, voltage and an internal resistance tracking parameter used in a system of Fig. 1 impulse power test; and Fig. 4A and Fig. 4B are schematic example diagrams, each illustrating statistical analysis data for an example set of battery cells.

[0014] With reference to the drawings, in which like reference numerals refer to like components, Fig. 1 is a schematic diagram illustrating a system 10 for assembling a plurality of cells 12 into a battery module 14. The plurality of cells 12 may have different chemistries, including, but not limited to, lithium-ion, lithium-iron, nickel-metal hydride, and lead-acid batteries. It should be understood that the number of battery cells in each battery module 14 may vary depending on the application.

[0015] As in Fig. 1, the battery module 14 may be incorporated into a rechargeable energy storage device in a vehicle V that may be partially or fully electric. It should be understood that the battery module 14 may take many different forms and have additional components. The electric vehicle V may be a mobile platform, such as a passenger car, a sport utility vehicle, a light truck, a heavy-duty vehicle, an ATV, a minivan, a bus, a transit vehicle, a bicycle, a mobile robot, an agricultural implement (e.g., a tractor), a sporting equipment (e.g., a golf cart), a boat, an airplane, and a train.

[0016] Cell qualification procedures for determining battery cell suitability for assembly generally rely on battery cell measurements compared to a nominal range. To obtain a complete picture of cell quality, the DC internal resistance must typically be measured at approximately 50% of the state of charge. The state of charge of individual cells within a large batch (e.g., from a supplier) varies, so directly measuring the DC internal resistance is both complex and time-consuming (e.g., it requires charging and discharging).

[0017] As described below, the system 10 provides a method 100 for qualifying and sorting the plurality of cells 12 for module assembly based on the estimation of a DC internal resistance of each cell in the plurality of cells 12, regardless of their individual state of charge. Referring to Fig. 1, the system 10 comprises a controller C having at least one processor P and at least one memory M (or a non-transitory, tangible, computer-readable storage medium) on which instructions for executing a portion of a method 100 for assembling the plurality of cells 12 into the battery module 14 can be recorded. The memory M can store executable instruction sets, and the processor P can execute the instruction sets stored in the memory M.

[0018] The system 10 is capable of estimating the DC cell internal resistance using Kalman filtering to enable rapid cell sorting for assembly of the battery module 14. The plurality of cells 12 are sorted into sorted groups based in part on the respective internal resistance parameter tracked by the controller C. Prior to assembly, each of the plurality of cells 12 is pulsed power tested in a tester 16. The controller C is in communication with the tester 16 and receives the respective voltage waveform data of the plurality of cells 12 based in part on the pulsed power test. The method 100 includes adding tracking parameters to the Kalman filter observer and tuning them to isolate and track a respective internal resistance parameter of the plurality of cells 12.Complementing cell qualification testing with an estimation of DC cell internal resistance minimizes changes to the assembly line.

[0019] With reference to Fig. 2 shows a flowchart of the method 100 implemented in the controller C of Fig. 1 is stored and can be executed by it. The method 100 can be executed in the form of computer-readable code or instructions that can be executed on the controller C of Fig. 1 and partially executable by it. The method 100 does not have to be performed in the order specified here. It is also understood that some steps may be omitted. The method 100 can be executed dynamically.

[0020] According to Block 102 of Fig. 2, the method 100 includes positioning the plurality of cells 12 in the test fixture 16 for pre-assembly testing. The test fixture 16 is configured to accommodate a selected group 18 (e.g., cell B1 and cell B2) from the plurality of cells 12 for a pulse power test. The number of cells in the selected group 18 can be varied depending on the application. In one embodiment, the selected group 18 comprises six cells from the plurality of cells 12.

[0021] The controller C is capable of performing a pulse power test on the selected group 18 via the test device 16. The test device 16 includes a probe 20 having a first end 22 and a second end 24, each connected to a positive and a negative end of the selected group 18. With reference to Fig. 1, the test device 16 includes a current source 26 for injecting a test pulse into the cells of the selected group 18, which together form an electrical circuit. The same test pulse is applied to each cell of the selected group 18. Fig. 3A-C show example graphs for the load current, the cell voltage curve, and the internal resistance parameter versus time, respectively, for an example pulse power test.

[0022] With reference to Fig. Figure 3A shows an example graph of the load current 210 in a pulse power test, with amplitude (in amperes) on the vertical axis Y1 and time on the horizontal axis T. With reference to Fig. 3A is the amplitude of the load current between time zero and time t1, zero, negative current between time t1 and time t2, zero between time t2 and time t3, positive current between time t3 and time t4, and zero after time t4. The load current can be specified by an HPPC (Hybrid Pulse Power Cycle) time curve, which is determined by monitoring the change in the internal voltage of the battery cell while simultaneously changing the charge and discharge current within the battery cell's operating range.

[0023] With reference to Fig. 1, a voltage measuring device 28 is designed to record the respective voltage-time curve of each cell in the selected group 18 that is subjected to the pulse power test. The pulse power test is repeated for each of the plurality of cells 12. In the Fig. In the embodiment shown in Figure 1, the battery cells B1, B2 are held in position between the current probes by a first robot arm 30 with a peripheral device or end effector 32. The end effector 32 enables it to perform its assigned tasks and may include a gripper, a sensor, or another tool.

[0024] Continuing to block 104, the controller C is configured to receive the respective voltage curve of the plurality of cells 12 via the voltage measuring device 28. With reference to Fig. 3B shows exemplary cell voltage curves for a plurality of cells 12, which correspond to the load current in Fig. 3A. In Fig. In Figure 3B, the vertical axis Y2 shows the amplitude (in volts) and the horizontal axis T shows time.

[0025] Continuing with block 106, controller C is configured to track an internal resistance parameter of each of the plurality of cells based in part on the voltage waveform data, the pulse power test settings / parameters, and an observer module. The tracked internal resistance parameter may be proportional to the absolute value of the cell's DC internal resistance. With reference to Fig. 3C shows an example of an internal resistance parameter that corresponds to the load current in Fig. 3A and the cell voltage curve in Fig. 3B. The vertical axis Y3 in Fig. 3C shows the amplitude and the horizontal axis T the time. The traces of the tracking parameters 225 in Fig. 3C correspond to the voltage curves 220 in Fig. 3B. The track of the tracking parameter 235 in Fig. 3C corresponds to the voltage curve 230 in Fig. 3B.

[0026] The tracking parameter of the internal resistance for each of the plurality of cells 12 is determined from the voltage curve using an observer module. The observer module can be embedded in the controller C. The observer module can be based on a Kalman filter module. The Kalman filter module can rely on a cell model developed from various tests that were carried out beforehand. As one skilled in the art knows, the Kalman filter module works recursively and runs in real time using the current state, the previously calculated state, and its uncertainty matrix. The Kalman filter module can work in a two-step process. For example, in a first step, the Kalman filter module creates estimates of the current state variables (here, the DC internal resistance) along with their uncertainties. Once the result of the next measurement (with a discrete error amount such asrandom noise) is observed, these estimates are updated using a weighted average, with estimates with higher confidence receiving more weight.

[0027] Proceeding to block 108, the method 100 may include removing outlier cells from the plurality of cells 12. The controller C may be configured to determine the standard deviation values ​​of the respective internal resistance parameter among the plurality of cells 12. The controller C may develop a binary threshold for accepting and rejecting the plurality of cells 12 based in part on the standard deviation values. In other words, the binary threshold may "accept" the cells with a standard deviation below a predefined value and "reject" the cells with a standard deviation above the predefined value.

[0028] Controller C may perform a statistical analysis to determine whether the respective internal resistance parameter for one of the plurality of cells 12 is an outlier compared to the respective internal resistance parameter for the remaining cells 12. In one embodiment, the statistical analysis includes a principal component analysis that transforms a set of measurement curves into a set of weighted sums of linearly uncorrelated (orthonormal) measurement curves, referred to as principal modes.

[0029] An example of performing principal component analysis is described below. Controller C is configured to subtract an average of the individual internal resistance tracking parameter curves from each of the individual internal resistance tracking parameter curves to obtain a difference of each of the individual internal resistance tracking parameter curves from the average. The deviation of each individual internal resistance tracking parameter voltage curve from the average can then be described by a series of principal modes, e.g., a first principal mode, a second principal mode, a third principal mode, etc. The coefficients (i.e., weighting) of the first principal mode can be graphically represented to illustrate the status of each of the plurality of cells 12.

[0030] With reference to Fig. 4A shows the coefficients of the first main mode for a first group of 310 healthy cells. As in Fig. As shown in Figure 4A, the coefficients of the first principal mode each surround zero on the x-axis if there are no outlier cells. With reference to Fig. 4B, however, if there is an outlier cell, the coefficients of the first main mode for the outlier cell 320 shift the first group 310 to one side and the coefficients for the outlier cell 320 to the other side of a zero value in the diagram. In this way, the outlier cell 320 is identified, which can be removed from the first group 310 (consisting of healthy cells), as shown in Fig. 4B. This occurs because the sum of the coefficients corresponding to each of the multiple cells 12 being tested is zero.

[0031] Continuing with block 110, the method 100 includes classifying the plurality of cells into sorted groups based in part on the respective internal resistance parameter. Referring to Fig. 1, the first robotic arm 30 may be configured to transport the plurality of cells 12 into a plurality of storage bins 40. The storage bins 40 are pre-sorted according to internal resistance values. In other words, the storage bins 40 may each be assigned to the plurality of cells with a respective internal resistance parameter falling within a predefined range (e.g., milliohms). For example, the first bin 42 may be reserved for the cells that fall within a first resistance range (e.g., 31.1-31.5 milliohms), the second bin 44 may be reserved for the cells that fall within a second resistance range (e.g., 31.5-32.0 milliohms), the third bin 46 may be reserved for the cells that fall within a third resistance range (e.g., 32.0-32.5 milliohms), a fourth bin (not shown), etc.After the DC cell internal resistance of the plurality of cells 12 has been measured, ordered and sorted, the battery module 14 can be built in various arrangements.

[0032] Continuing with block 112, the method 100 includes assembling the sorted groups into the battery module 14 in a predefined pattern. The predefined pattern is based on the magnitude of the respective internal resistance parameter. The controller C may be configured to synchronize the position of the cell with the associated power to present it to a second robot arm 50. Referring to Fig. 1, the second robot arm 50 is adapted to retrieve the plurality of cells 12 from the plurality of storage bins 40 using a peripheral device or end effector 52. The first robot arm 30 and / or the second robot arm 50 may be mounted on a linear axis or a mobile robot to reach the various storage bins 40. The first robot arm 30 and / or the second robot arm 50 may be part of a pick-and-place robot, a gantry robot, or another robotic system.

[0033] As in Fig. 1, the battery module 14 includes a plurality of parallel cell groups 60. The parallel cell groups 60 refer to at least two groups of n cells arranged in parallel. Each of the parallel cell groups 60 within the battery module 14 is populated with cells that have similar values ​​of DC internal cell resistance, i.e., come from the same sorted group. The technical advantage is that minimizing the cell-to-cell variations in DC internal cell resistance in parallel cell groups 60 within the battery module 14 results in each cell in the group making an equal contribution to battery charging / discharging, resulting in better battery utilization. The majority of the cells 12 can be arranged in a combination of series and parallel connections in an array (PXS). For example, a 4P6S array has four parallel cell groups with six cells (connected in series) in each parallel cell group 60.

[0034] In one embodiment, the second robotic arm 50 is configured to arrange the plurality of cells 12 in the battery module 14 in a ranking pattern such that the parallel cell groups 60 are arranged in ascending order of the internal resistance parameter (ie, the first parallel cell group has the smallest internal resistance and the last parallel cell group has the largest internal resistance).

[0035] Referring to Fig. 1, the battery module 14 has a central or inner section 62 surrounded by one or more outer sections 64, 66. In another embodiment, the predefined pattern is temperature-ordered such that the parallel cell groups 60 in the battery module 14 have cells arranged such that the highest DC internal resistance is in the inner section 62, where the operating temperature is highest. In other words, the arrangement is such that the majority of cells 12 with a relatively high internal resistance parameter are arranged in the inner section 62 of the battery module 14. The majority of cells 12 with a relatively low internal resistance parameter are arranged in the outer sections 64, 66 of the battery module 14.This approach has the technical advantage that as the cell with the highest internal resistance parameter heats up, its internal resistance parameter drops more sharply than the cell near the edge (or outer sections 64, 66) with lower temperature, which then causes the overall module internal resistance parameter map to flatten during operation.

[0036] Continuing to block 114, the method 100 includes inserting the assembled battery module 14 into the vehicle V.

[0037] In summary, the system 10 estimates the cell DC internal resistance of a plurality of cells 12 (e.g., small-format lithium-ion cells) used in the construction of a battery module 14 to enable cell sorting during the module assembly process. Each battery cell undergoes a pulse power test prior to assembly, with a current source 26 applying a load current. A voltmeter 28 records the cell's voltage-time curve. The rapid estimation of the cell DC internal resistance parameter enables rapid cell sorting and reduces the module assembly / production line cycle time. Furthermore, cell sorting increases the discharge energy delivered by the battery module 14.

[0038] As used herein, the terms "dynamic" and "dynamic" describe steps or methods that execute in real time and are characterized by monitoring or otherwise determining the state of parameters and regularly or periodically updating the state of the parameters during the execution of a routine or between iterations of the execution of the routine.

[0039] The control C of Fig. 1 may be an integral part or a separate module that is functionally connected to other controls of the vehicle V. The control C of Fig. 1 includes a computer-readable medium (also called a processor-readable medium), including a non-transitory (e.g., tangible) medium that is involved in providing data (e.g., instructions) that can be read by a computer (e.g., by a processor of a computer).

Claims

[1] Method (100) for assembling a plurality of cells (12) into a battery module (14) with a controller (C) having a processor (P) and a tangible, non-volatile memory (M), the method (100) comprising: Positioning (102) the plurality of cells (12) in a test device (16) for the pre-assembly test, wherein the test device (16) is designed to receive a selected group (18) of the plurality of cells (12) at a time; Performing an impulse power test on the selected group (18) via the test device (16); Receiving (104) corresponding voltage waveform data based in part on the pulse power test via the controller (C); Tracking (106) a respective internal resistance parameter of the plurality of cells (12), based in part on the respective voltage waveform data and an observer module, via the controller (C); Arranging (110) the plurality of cells (12) in sorted groups, partly based on the respective internal resistance parameter; and assembling (112) the sorted groups in a predefined pattern in the battery module (14), wherein the predefined pattern is based on a size of the respective internal resistance parameter; Installing a probe (20), a current source (26) and a voltage measuring device (28) into the test device (16), the probe (20) having a first end (22) and a second end (24), wherein performing the pulse power test comprises: Connecting a respective positive end and a respective negative end of the selected group (18) to the first end (22) and the second end (24) of the probe (20); and Feeding a test pulse into the selected group (18) via the current source (26) and obtaining the respective voltage waveform data of the selected group (18) via the voltage measuring device (28). [2] The method (100) of claim 1, further comprising: Using a first robot arm (30) to transport the plurality of cells (12) into a plurality of storage containers (40), wherein the plurality of storage containers (40) are each assigned to predefined ranges of the respective internal resistance parameter. [3] The method (100) of claim 2, further comprising: Inserting a second robot arm (50) which is suitable for removing the plurality of cells (12) from the plurality of storage containers (40), wherein the battery module (14) has a plurality of parallel cell groups (60). [4] The method (100) of claim 3, further comprising: Arranging the plurality of cells (12) on the battery module (14) in a ranking order via the second robot arm (50) such that the multiple parallel cell groups (60) are in an increasing order of the respective internal resistance parameter. [5] The method (100) of claim 3, further comprising: Arranging the plurality of cells (12) on the battery module (14) via the second robot arm (50), wherein the battery module (14) has an inner section (62) and an outer section (64, 66) such that the plurality of cells (12) with a relatively high internal resistance parameter are arranged in the inner section (62) and the plurality of cells (12) with a relatively low internal resistance parameter are arranged in the outer section (64, 66). [6] The method (100) of claim 1, further comprising: Integrating a Kalman filter into the observer module. [7] The method (100) of claim 6, further comprising: Determining standard deviation values ​​of the respective internal resistance parameter among the plurality of cells (12); and Develop a binary threshold for accepting and rejecting the majority of cells (12) based in part on the standard deviation values. [8] The method (100) of claim 6, further comprising: Performing a statistical analysis to determine whether the respective internal resistance parameter for one of the plurality of cells (12) is an outlier compared to the respective internal resistance parameter for a remainder of the plurality of cells (12). [9] The method (100) of claim 8, further comprising: Incorporating a principal component analysis into the statistical analysis, including determining the respective coefficients of a first principal mode for each of the plurality of cells (12).

Images