A pole piece slurry cell printing method based on flow field control

Abstract

The present application relates to the technical field of battery pole piece manufacturing, in particular to a kind of pole piece slurry battery cell printing method based on flow field control, comprising: by core layer pump, upper sheath layer pump and lower sheath layer pump, slurry is supplied to die head with independently adjustable gap section, and lower sheath layer, core layer and upper sheath layer are formed on continuous walking base material;Collect die head pressure drop, flow, core layer pump speed, walking speed and area density detection value;Based on the momentum conservation relationship of steady incompressible fluid, three-layer flow model of parallel flat narrow slit flow is established, and the thickness of upper and lower sheath layer is calculated;According to target area density and target sheath layer thickness, calculate error;Through fast, medium and slow adjustment loop, respectively adjust the tooth frequency compensation flow, sheath layer basic flow, core layer pump speed and die head lip gap which are opposite to the phase of tooth frequency flow fluctuation, so as to maintain the stability of printing process and the consistency of area density.

Description

Technical Field

[0001] This invention relates to the field of battery electrode manufacturing technology, specifically to a method for printing electrode slurry cells based on flow field control. Background Technology

[0002] Currently, electrode slurry printing typically employs single-layer or multi-layer coating methods, controlling coating thickness and areal density by adjusting slurry supply flow rate, die gap, and conveyor speed. Existing methods struggle to accurately detect the flow state between the core and sheath layers, sheath thickness, and interface integrity. They also struggle to coordinate and manage periodic pump flow fluctuations, changes in slurry solids content, and lateral areal density deviations, leading to problems such as uneven sheath layer distribution, areal density fluctuations, and reduced coating consistency during the printing process. Summary of the Invention

[0003] The purpose of this invention is to provide a method for printing electrode paste cells based on flow field control, which avoids problems such as uneven sheath layer distribution, surface density fluctuation and reduced coating consistency during the printing process, and makes it easier to control the flow state of the core layer and sheath layer, sheath layer thickness and interface integrity.

[0004] The objective of this invention can be achieved through the following technical solutions: The slurry is supplied to the die head with an independently adjustable lip gap section through the core layer pump, the upper sheath layer pump and the lower sheath layer pump, and the slurry is coated once on the continuous conveyor substrate to form the lower sheath layer, the core layer and the upper sheath layer stacked in sequence. During the coating process, the following data were collected: die head pressure drop, upper sheath flow rate, lower sheath flow rate, core pump speed, belt conveyor speed, and areal density. Based on the die head pressure drop, upper sheath flow rate, lower sheath flow rate, and core pump speed, combined with preset die head structural parameters and preset slurry viscosity, the preset die head structural parameters include die head width, die lip straight section length, and die head clearance. The preset slurry viscosity includes core layer viscosity, upper sheath viscosity, and lower sheath viscosity. Based on the steady-state incompressible fluid momentum conservation relationship, a three-layer flow model of parallel flat plate narrow slit flow is established, and the upper sheath thickness and lower sheath thickness are calculated. Set preset target areal density and preset target sheath thickness, and calculate areal density error and sheath thickness error respectively; set up fast adjustment loop, medium adjustment loop and slow adjustment loop. The fast adjustment loop adjusts the tooth frequency compensation flow of the upper and lower sheath pumps according to the tooth frequency flow fluctuation corresponding to the core pump speed. Tooth frequency flow fluctuation refers to the periodic flow fluctuation corresponding to the tooth frequency and its harmonic components in the core pump supply flow, which is equal to the product of the core pump speed and the preset core pump tooth number; tooth frequency compensation flow refers to the compensation flow output by the upper and lower sheath pumps that is opposite in phase to the tooth frequency flow fluctuation; the medium adjustment loop adjusts the basic flow of the upper and lower sheath pumps according to the sheath thickness error; the slow adjustment loop adjusts the core pump speed and the lip clearance of each section of the die head according to the areal density error. The above-mentioned acquisition, calculation and adjustment process is executed cyclically according to the sampling period. In each sampling period, the fast adjustment loop, the medium adjustment loop and the slow adjustment loop update their respective adjustment values ​​independently. The outputs of the three adjustment loops are superimposed in the order of fast adjustment loop, medium adjustment loop and slow adjustment loop and then output to the corresponding actuators to cyclically adjust the flow field state and surface density.

[0005] Preferably, the die head includes five independently adjustable lip gap sections arranged laterally, a manifold, and a die lip straight section; the pressure difference between the two locations is measured by two pressure sensors respectively installed at the manifold inlet and the die lip straight section inlet as the die head pressure drop; When calculating the thickness of the upper and lower sheath layers, the apparent viscosity is first calculated based on the measured die head pressure drop, die head gap, die head width, die lip straight section length, and total flow rate. Then, the core layer viscosity and the corresponding sheath layer viscosity are combined to obtain the ratio of the upper or lower sheath layer thickness to the die head gap. The above calculation is applicable to steady-state incompressible parallel plate narrow slit flow.

[0006] Preferably, five sets of four microelectrodes are uniformly embedded in the lower wall of the straight section of the die lip along the transverse direction; during the coating process, the reference pressure drop in the sheathless state when the upper sheath pump and the lower sheath pump are stopped, as well as the impedance amplitude of the five sets of four microelectrodes under 5kHz and 100kHz excitation are also collected. The sheath integrity index is determined based on the pressure drop of the mold head in the sheathed state and the reference pressure drop in the unsheathed state, and the ratio of the low-frequency impedance amplitude to the high-frequency impedance amplitude corresponding to each group of microelectrodes is calculated; the core solid content is determined based on the calibration relationship of the impedance ratio. When the calculated deviation of the core layer solid content is greater than the set solid content threshold, a sinusoidal temperature modulation with a temperature rise and fall range of 1.5K centered on the set temperature is applied to the lower wall of the straight section of the die lip. The core layer solid content detection result is corrected according to the differential response of the die head pressure drop with temperature change. When the calculated deviation of the core layer solid content is less than the set solid content threshold, the current core layer solid content detection result is maintained.

[0007] Preferably, an asymmetric three-layer co-extrusion process is used, wherein the core layer is a slurry with a solid content of 68wt% to 78wt%, the binder mass fraction of the lower sheath slurry forming the lower sheath layer is greater than the binder mass fraction of the upper sheath slurry forming the core layer to form a binder-rich layer with a wet film thickness of 8μm to 15μm, and the upper sheath slurry forms a porous layer with a wet film thickness of 3μm to 6μm.

[0008] Preferably, the areal density detection value is measured by a downstream areal density meter; the fast adjustment loop uses a comb notch filter with a center frequency determined by the core pump speed and the preset number of core pump teeth to remove the tooth frequency and harmonic components of the tooth frequency in the flow data to obtain transient flow data. Substituting transient flow data into the three-layer flow model, the instantaneous mass flow rate of the core layer, the mass solid content of the core layer, the instantaneous mass flow rate of the upper sheath layer, the mass solid content of the upper sheath layer, the instantaneous mass flow rate of the lower sheath layer, the mass solid content of the lower sheath layer, the instantaneous conveyor speed, and the die width are used to calculate the transient surface density of dry-based solids. Based on the distance along the belt travel direction from the die head outlet to the downstream surface density meter and the instantaneous belt travel speed, the surface density detection value is delayed and corrected. The difference between the maximum and minimum surface density values ​​at each transverse detection point within the same time section is calculated as the transverse surface density range.

[0009] Preferably, the closed-loop bandwidth of the fast adjustment loop is 200Hz to 1kHz; the closed-loop bandwidth of the medium adjustment loop is 1Hz to 5Hz, and the control target of the medium adjustment loop is to keep the sheath integrity index within the range of 0.60±0.05; the closed-loop bandwidth of the slow adjustment loop is 0.05Hz to 0.2Hz, and the slow adjustment loop adjusts the lip gap of five independently adjustable sections according to the transverse surface density difference.

[0010] Preferably, pressure, impedance, flow rate, and areal density data are synchronized using a unified timestamp; outliers in the head pressure drop signal are removed using a Hamper filtering method with a window length of 15 sampling points, a scale parameter σ estimated based on the absolute deviation of the median within the window, and a judgment threshold of 3σ. When any channel in any of the five groups of four microelectrodes has no effective voltage data for 200ms, linear interpolation is performed based on the impedance amplitude of the corresponding electrode in the same group or the horizontally adjacent group.

[0011] Preferably, when the viscosity ratio of the upper or lower sheath layer to the core layer is greater than or equal to 50, the ratio of the upper or lower sheath layer thickness to the die clearance is less than or equal to 1.5%, and a non-tooth-frequency periodic peak value from the die pressure drop is detected in the 30Hz to 80Hz frequency band, with an amplitude greater than or equal to a set signal-to-noise ratio threshold determined by the pressure spectrum under stable three-layer flow conditions, the flow state is determined to be abnormal and an abnormality flag is set; after determining the abnormality, the rotational speed of the upper sheath pump and the lower sheath pump is increased by 20%; If the non-tooth frequency periodic peak does not disappear within 3 seconds, stop the output of the fast regulation circuit and the medium regulation circuit, keep the core pump running, and enter the degraded operation state; If the peak value is less than or equal to the set signal-to-noise ratio threshold and remains so for 1 second, the abnormality flag is removed, and the rotational speeds of the upper and lower sheath pumps are restored to their pre-abnormal base rotational speeds by decreasing by 5% every 500ms. In the event of a power outage, a backup pressure-maintaining circuit consisting of a backup storage tank and a switching valve continuously supplies the upper and lower sheath slurries to the die head.

[0012] Preferably, the upper sheath pump and the lower sheath pump are plunger pumps with periodic flow compensation capability; the fast adjustment loop generates a tooth frequency flow fluctuation template based on the flow data of the core pump in multiple rotation cycles according to the encoder angle of the core pump, calculates the amplitude and phase of the tooth frequency flow fluctuation template, and controls the upper sheath pump and the lower sheath pump to generate compensation flow with opposite phase. Every 30 minutes of operation, reverse the shafts of the upper and lower sheath pumps and return the flow to the die inlet at a flow rate three times the rated flow rate for 1 second. Repeat this process three times. Then, with the upper and lower sheath pumps stopped and the core supply maintained, remeasure the reference pressure drop in the sheathless state.

[0013] Preferably, the effectiveness of the channels for the five sets of four microelectrodes and two pressure sensors is detected, and the detection criteria are whether the signal amplitude and impedance are within the set range. When the signal loss time of at least three sensor channels is detected to be greater than or equal to the preset time threshold, the average value of the sheath integrity index and sheath thickness within the last 30 seconds is used as the controller hold value, the output of the fast adjustment loop and the medium adjustment loop is stopped, and the conveyor speed is reduced to 60% of the rated speed; when there are fewer than three failed channels and the channel validity detection standard is met for 5 consecutive seconds, the multi-channel failure mark is removed, and the closed-loop input is re-established with three consecutive valid sampling cycles.

[0014] The beneficial effects of this invention are: This invention utilizes a three-layer co-extrusion feeding system consisting of a core layer, an upper sheath layer, and a lower sheath layer. By combining the results of die pressure drop, sheath layer flow rate, pump speed, microelectrode impedance, and areal density detection, it collaboratively detects and classifies the sheath layer thickness, sheath layer integrity, core layer solid content, and transverse areal density. This reduces the wall shearing effect on the core layer slurry and improves the coating consistency reduction caused by uneven sheath layer distribution, periodic pump flow fluctuations, solid content variations, and transverse areal density deviations. As a result, it achieves stable one-time forming printing of high-solids-content electrode slurry, and has the advantages of accurate sheath layer state detection, good transverse consistency, and high operational reliability. Attached Figure Description

[0015] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic flowchart of the electrode paste cell printing method based on flow field control according to the present invention. Detailed Implementation

[0016] 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 some embodiments of the present invention, and not all embodiments. 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.

[0017] Please see Figure 1 A method for printing electrode paste cells based on flow field control includes the following steps: Sp1. Slurry is supplied to the die head with an independently adjustable lip gap section through the core layer pump, upper sheath layer pump and lower sheath layer pump, and coated on the continuous conveyor belt substrate in one step to form the lower sheath layer, core layer and upper sheath layer stacked in sequence; during the coating process, the die head pressure drop, upper sheath layer flow rate, lower sheath layer flow rate, core layer pump speed, conveyor belt speed and areal density are measured. Sp2. Based on the die head pressure drop, upper sheath flow rate, lower sheath flow rate, and core pump speed, combined with preset die head structural parameters and preset slurry viscosity, the preset die head structural parameters include die head width, die lip straight section length, and die head clearance, and the preset slurry viscosity includes core layer viscosity, upper sheath viscosity, and lower sheath viscosity. Based on the steady-state incompressible fluid momentum conservation relationship, a three-layer flow model of parallel flat plate narrow slit flow is established, and the upper sheath thickness and lower sheath thickness are calculated. Sp3, set the preset target surface density and preset target sheath thickness, and calculate the surface density error and sheath thickness error respectively; Sp4. Set up fast, medium, and slow adjustment loops: The fast adjustment loop adjusts the tooth frequency compensation flow of the upper and lower sheath pumps according to the tooth frequency flow fluctuation corresponding to the core pump speed. The tooth frequency flow fluctuation refers to the periodic flow fluctuation corresponding to the tooth frequency and its harmonic components in the core pump supply flow, where the frequency is equal to the product of the core pump speed and the preset core pump tooth number. The tooth frequency compensation flow refers to the compensation flow output by the upper and lower sheath pumps, which is opposite in phase to the tooth frequency flow fluctuation. The medium adjustment loop adjusts the basic flow of the upper and lower sheath pumps according to the sheath thickness error. The slow adjustment loop adjusts the core pump speed and the lip gap of the die head with an independently adjustable lip gap section according to the areal density error. Sp5. Execute steps Sp1 to Sp4 in a cyclical manner according to the sampling cycle. In each sampling cycle, the fast adjustment loop, the medium adjustment loop, and the slow adjustment loop update their respective adjustment values ​​independently. The outputs of the three adjustment loops are superimposed in the order of fast adjustment loop, medium adjustment loop, and slow adjustment loop and then output to the corresponding actuators to cyclically adjust the flow field state and surface density. In one embodiment, in Sp1, the die head includes five independently adjustable lip gap sections arranged laterally, a manifold, and a die lip straight section; the pressure difference between the two locations is measured as the die head pressure drop by two pressure sensors respectively located at the inlet of the manifold and the die lip straight section. In Sp2, the thickness of the upper sheath and the thickness of the lower sheath are calculated using the following formulas: ; ; In the formula, For apparent viscosity, For the measured pressure drop of the die head, For die head clearance, The width of the die head. The average density of the slurry, The length of the straight section of the mold lip. For total mass flow rate, The thickness of the upper sheath or the lower sheath. For core layer viscosity, This corresponds to the viscosity of the upper sheath or the viscosity of the lower sheath. Five sets of four microelectrodes were uniformly embedded laterally on the lower wall of the straight section of the die lip; in Sp1, the reference pressure drop in the sheathless state when the upper and lower sheath pumps were stopped was also collected. and five sets of four microelectrodes in and Impedance amplitude under excitation; In Sp2, the sheath integrity index is determined based on the pressure drop of the die head in the sheathed state and the reference pressure drop in the unsheathed state. And calculate the impedance ratio corresponding to each group of microelectrodes. ; The core layer solid content is determined based on the calibration relationship of the impedance ratio R. When the calculated deviation of the core layer solid content is greater than or equal to the set solid content threshold determined based on the percentile value of the absolute deviation between the solid content back-calculated value at each calibration point and the known value, a sinusoidal temperature modulation with a temperature rise and fall range of 1.5K centered on the current stable wall temperature is applied to the lower wall surface of the straight section of the die lip. The core layer solid content detection result is corrected based on the differential response of the die head pressure drop with temperature change. When the calculated deviation is less than the set solid content threshold, the current core layer solid content detection result is maintained.

[0018] This embodiment is applicable to the one-time forming printing of high solids content electrode slurry on a continuous conveyor substrate; the printing device includes a core layer pump, an upper sheath layer pump, a lower sheath layer pump, a three-layer co-extrusion die, a conveyor mechanism, a pressure sensor, a four-microelectrode array, a surface density meter, and a controller connected to each actuator. The controller is implemented by combining an industrial control computer with a mold head side control unit. The mold head side control unit is responsible for collecting pressure, impedance, flow rate and pump speed signals, while the industrial control computer is responsible for calculating the upper sheath thickness, lower sheath thickness, surface density error and various adjustment values, and sending the results to the core pump, upper sheath pump, lower sheath pump and mold head gap adjustment mechanism. In this embodiment, the core layer refers to the main slurry layer that undertakes the main function of transporting active substances; the upper sheath layer and the lower sheath layer refer to the auxiliary slurry layers located on both sides of the core layer, which are used to change the wall shear state and interface composition. The pressure drop at the die head refers to the pressure difference generated during the process of the slurry flowing from the manifold inlet to the inlet of the straight section of the die lip; the target sheath thickness refers to the pre-defined upper and lower sheath thickness settings. Areal density error is the difference between the detected areal density value and the target areal density; sheath thickness error is the difference between the calculated sheath thickness and the target sheath thickness; Before printing begins, the controller reads the die head structure parameters and the slurry viscosity parameters. The die head structure parameters include the die head width, the length of the straight section of the die lip, and the die head gap. The slurry viscosity parameters include the core layer viscosity, the upper sheath layer viscosity, and the lower sheath layer viscosity. None of the three viscosities are obtained by using the single-point viscosity under static conditions. Instead, they are obtained by linear interpolation based on the slurry rheology calibration table, with the current temperature and the operating shear rate determined by the total flow rate and the die head clearance as inputs. The operating shear rate is determined by dividing the total flow rate by the average shear rate obtained by dividing the die width by the square of the die gap. The controller first calculates the total flow rate based on the current core flow rate, upper sheath flow rate, and lower sheath flow rate, and then substitutes it into this rule to obtain the operating shear rate. If the total flow rate is updated to a new value within the same sampling period, the corresponding viscosity will be refreshed once at the end of the sampling period to avoid using different viscosity values ​​in a single sampling period. The rheological calibration table should cover at least 3K above and below the expected operating temperature and 0.5 to 1.5 times the expected shear rate. Linear interpolation should be used between adjacent calibration points. When the calibration range is exceeded, the thickness closed-loop update should be stopped and the effective viscosity value of the most recent 10 seconds should be maintained. At the same time, an out-of-range status signal should be output. The core layer pump, upper sheath layer pump and lower sheath layer pump simultaneously supply material to the die head. The three slurries form a layered flow of lower sheath layer, core layer and upper sheath layer inside the die head and are coated onto the continuous conveyor substrate from the die head outlet. In Sp1, the controller continuously collects the die head pressure drop, upper sheath flow rate, lower sheath flow rate, core pump speed, conveyor speed, and areal density detection value. In order to obtain the die head pressure drop, this embodiment sets a pressure sensor at the manifold inlet and the straight section inlet of the die lip, respectively. The output signals of the two pressure sensors are sampled synchronously and then subtracted. The resulting differential pressure value is used as the real-time die head pressure drop. The flow rates of the upper and lower sheath layers are output by the flow detection components on their respective branches, the core pump speed is output by the core pump encoder or drive feedback signal, the conveyor speed is output by the speed detection component of the conveyor mechanism, and the areal density detection value is output by the areal density meter downstream of the die head. Since the surface density meter corresponds to the coating state after it has exited the die, the controller uses the detected value as a slow correction value, while the pressure, flow rate and speed signals are used as real-time inputs that directly reflect the flow state inside the die. When obtaining the reference pressure drop in the sheathless state, within one reference acquisition cycle, the controller stops the upper and lower sheath pumps from feeding, and only keeps the core pump feeding, so that the core slurry passes through the die head alone; During the baseline acquisition, the core flow rate, conveyor speed, die gap, and slurry temperature were maintained within ±1%, ±1%, ±0.5%, and ±0.2K of the normal operating settings, respectively. The pressure drop data was continuously collected for 5 seconds, and the average value of the 3-second pressure drop data was used as the reference pressure drop in the sheathless state. If the range of pressure drop within 3 seconds exceeds 2% of the average value, the current baseline is invalid and a new data acquisition is required. The reference pressure drop reflects the reference flow resistance under the combined effect of the current die geometry, core slurry state, and operating conditions; after the reference acquisition is completed, the controller restarts the upper sheath pump and the lower sheath pump, and enters the three-layer co-extrusion printing state; In Sp2, the controller establishes a three-layer flow model of steady-state incompressible parallel flat plate narrow slit flow based on real-time die head pressure drop, upper sheath flow rate, lower sheath flow rate, core pump speed, and pre-input die head structure parameters and slurry viscosity parameters. The core pump speed is converted into core flow rate using the pump speed-flow rate calibration table, and then added to the upper sheath flow rate and lower sheath flow rate to obtain the total flow rate; The speed and flow rate calibration table is obtained at no fewer than five calibration points covering the normal speed range, according to the slurry used; after maintaining a stable feed at each calibration point, the average flow rate is recorded for 10 seconds, and linear interpolation is used between adjacent calibration points; the controller performs the conversion of the core layer flow rate in the following order: First, read the two adjacent calibration points where the current rotation speed is located; if the rotation speed falls exactly on a calibration point, then directly take the average flow rate corresponding to that calibration point; if it is between two calibration points, then perform linear interpolation according to the rotation speed ratio; if it is lower than the lowest calibration point or higher than the highest calibration point, then do not extrapolate, directly mark the core flow rate result of this cycle as invalid, and keep the average value of the effective core flow rate in the last 2 seconds to participate in subsequent judgment, but do not send it into the calculation of the new thickness value; The controller first calculates according to the recorded apparent viscosity formula. Then let each Take the viscosity of the upper sheath and the viscosity of the lower sheath, and obtain the initial values ​​of the upper sheath thickness and the lower sheath thickness according to the sheath thickness formula; To avoid uncertainty in the distribution of upper and lower sheath thickness due to using only a total pressure drop, the controller further uses the measured flow rates of the upper and lower sheaths for consistency verification. Specifically, a three-layer velocity distribution is established using two initial thickness values, and the velocity distribution function within each layer region is as follows: ; in, These are the coordinates in the thickness direction. For the corresponding layer viscosity, and To obtain the scalar coefficients obtained by simultaneously solving the velocity continuity and shear stress continuity equations at each layer interface and the no-slip boundary conditions at the die wall, the velocity distribution function is integrally integrated within its corresponding layer thickness interval to obtain the predicted flow rates of the upper and lower sheath layers. Compare the predicted flow rate with the actual flow rate of the corresponding branch, and adjust the thickness accordingly based on the ratio of the actual flow rate to the predicted flow rate; the adjustment is performed in the following fixed order: If the relative deviation between the predicted flow rate and the measured flow rate on one side is greater than 2%, then keep the thickness on the other side unchanged, and only multiply the thickness on that side by the ratio of the measured flow rate to the predicted flow rate; check whether the sum of the corrected thicknesses on both sides reaches 30% of the die head gap. If it reaches or exceeds this upper limit, then push the thickness to be updated back to the boundary value where the sum is equal to 30% of the die head gap by the same scaling ratio, and record this sampling period as the boundary-constrained period. Then the velocity distribution and predicted flow rate are recalculated, and the next verification is performed. After each correction, the velocity distribution and predicted flow rate are recalculated, with a maximum of five iterations. When the relative deviation between the predicted flow rate and the measured flow rate of the upper and lower sheaths is no greater than 2%, the thickness of this iteration is output. If any deviation is greater than 2% after five iterations, the result is not sent to the closed loop. Instead, the average effective thickness of the last 2 seconds is maintained, and the current sampling period is marked as model inconsistency. The head pressure drop is used to constrain the overall flow resistance, and the flow rates of the upper and lower sheath branches are used to determine the thickness distribution on both sides. The thickness calculation also includes an applicability check; when the total flow rate is not greater than 1 / 3 of the full scale of the corresponding flow sensor... The pressure drop of the die head should not exceed 1 / 3 of the full scale of the pressure sensor. If any viscosity data exceeds the rheological calibration range, or if the calculated thickness is less than zero, or the sum of the upper sheath thickness and the lower sheath thickness is greater than 30% of the die gap, the thickness result of this sampling cycle is deemed invalid; the adjustment loop is only restored after three consecutive valid sampling cycles to avoid boundary data directly causing sudden changes in pump volume; Simultaneously with Sp2, the controller utilizes the pressure drop of the die head in the sheath state. and reference pressure drop in the sheathless state Calculate the sheath integrity index ; If the sheath is continuous and the lubrication effect meets the set conditions, then the sheath pressure drop relative to the reference pressure drop will decrease. The pressure drop increases; however, if the sheath is too thin, unstable in distribution, or partially failed, the sheath pressure drop will rise again. decline; when Less than 5 of the full scale of the pressure sensor Or, the current core flow rate, die gap, and slurry temperature deviate from the data collection reference by more than 2... 1 At 0.5K, the controller pauses updates. The calculation will continue once the operating conditions are restored or the reference pressure drop is re-acquired. During the update pause, the controller will not generate new ones. The value is maintained for the most recent 10 valid sampling periods. The average value is used as the displayed value; if the update condition is not restored for more than 10 seconds, then... Relevant inputs are removed from the centering adjustment loop, and only the effect of thickness error on the centering adjustment loop is retained; In this embodiment, five sets of four microelectrodes are uniformly embedded in the lower wall of the straight section of the die lip along the transverse direction to determine the influence of changes in the core layer composition on the flow state; each set of four microelectrodes consists of a set of collinear electrodes and is used to measure the local impedance amplitude at different excitation frequencies. During the acquisition cycle, the controller applies 5kHz and 100kHz excitation to the five groups of four microelectrodes respectively, and reads the impedance amplitude of each group of electrodes at the two frequencies, calculating the impedance ratio. ; The calibration relationship between impedance ratio and core solid content was established under the same electrode spacing, excitation amplitude, and slurry formulation as during operation; Select at least six solid content calibration points that cover the expected core solid content range and are no more than 2wt% apart. At each calibration point, obtain no less than 30 sets of effective impedance ratios continuously at a set temperature. Use the average impedance ratio and the known solid content to form a calibration table that corresponds one-to-one. During operation, linear interpolation is used between adjacent calibration points; if the measured impedance ratio exceeds the calibration table range, extrapolation is not performed, and the channel is marked as out of range. When all five groups of electrodes are effective, the median of the solid content of the five groups is taken as the overall solid content of the core layer, and the results of each group are retained to judge the lateral local deviation, so as to avoid the single local high or low value directly changing the overall solid content result. When not all five groups of electrodes are effective, the controller processes them in the following order: when there are no fewer than three effective groups, they are first sorted from left to right according to their horizontal position, and the median of the solid content results of the effective groups is still taken as the overall solid content of the core layer. When the number of valid groups is less than three, the overall solid content of the core layer will not be updated in this sampling period. Instead, the median result of the last five valid sampling periods will be maintained, and this period will be marked as a low-confidence period for solid content. The solid content threshold is determined based on the calibration repeatability: the absolute value of the difference between the back-calculated solid content value and the known value at each calibration point is taken, and the 95th percentile of all absolute deviations is taken as the basic threshold. When the value is less than 0.5wt%, take 0.5wt%; when it is greater than 2wt%, take 2wt%. The calculation deviation of the core solid content during operation is the absolute value of the difference between the impedance ratio conversion result and the target solid content of the current formula. Temperature modulation is only activated if the deviation exceeds the set solid content threshold for five consecutive sampling cycles, in order to eliminate false triggering caused by a single impedance disturbance. Temperature modulation uses the current stable wall temperature as the center temperature, with a temperature rise and fall range of 1.5K, a modulation period of 10s, and is executed continuously for three complete cycles. The controller synchronously records the temperature and the pressure drop of the die head, and performs linear fitting on the heating and cooling segments of each cycle respectively. The fitting slope is used as the differential response of the pressure drop of the die head as a function of temperature. Cycles with a fitting determination coefficient lower than 0.95 are not included in the correction. The correction relationship is calibrated in advance by different slurries with known solid contents under the same flow rate, gap and temperature modulation conditions, forming a two-dimensional lookup table with impedance ratio and pressure drop temperature differential response as input variables and solid content as output variable; During operation, the initial solid content range is first determined based on the impedance ratio, and then bilinear interpolation is performed between adjacent calibration points based on the differential response to obtain the corrected core solid content. Bilinear interpolation is performed in the following order: First, within the two adjacent calibration intervals where the impedance ratio is located, take two solid content nodes adjacent to the current differential response to obtain four calibration nodes surrounding the current operating point; If all four nodes are present, first calculate two intermediate solid content values ​​at the same impedance ratio node along the differential response direction, then perform a linear interpolation on these two intermediate solid content values ​​along the impedance ratio direction, and output the corrected core solid content. If there is a lack of adjacent nodes in any direction, the bilinear interpolation is abandoned and the initial solid content obtained by separately converting the impedance ratio is maintained; if there are fewer than two effective periods in the three periods, the correction is abandoned and the solid content result before temperature modulation is maintained, and no correction value not supported by the calibration range is input to the control loop. In Sp3, the controller reads the pre-input target surface density, upper sheath target thickness, and lower sheath target thickness; the controller subtracts the target surface density from the surface density detection value to obtain the surface density error; the controller subtracts the corresponding target thickness from the calculated upper sheath thickness and lower sheath thickness respectively to obtain the upper sheath thickness error and lower sheath thickness error. The thickness errors of the upper and lower sheath layers are used as inputs for adjusting the basic flow rate of the upper and lower sheath layer pumps, respectively, while the areal density error is used as inputs for adjusting the core layer pump speed and die head gap. In Sp4, the input of the fast adjustment loop is the tooth frequency flow fluctuation corresponding to the core pump speed; the controller determines the tooth frequency based on the core pump speed and number of teeth, and obtains the amplitude and phase of the tooth frequency fluctuation based on the flow data of multiple consecutive pump rotation cycles; The compensation waveforms of the upper and lower sheath pumps are output in opposite phase relative to the core tooth frequency fluctuations. The initial compensation amplitude is allocated according to the ratio of the basic flow rates of the upper and lower sheaths to the total basic flow rates of the two sheaths. Every ten tooth frequency cycles, the controller compares the total flow tooth frequency amplitude before and after compensation; if the amplitude decreases after compensation, the compensation amplitude for the next cycle is increased by 5 times the current core tooth frequency fluctuation amplitude. When the amplitude increases after compensation, it decreases by 5. ; A single change shall not exceed 2% of the rated flow rate of the corresponding sheath pump. The peak compensation flow shall not exceed 20% of its base flow. This forms a convergent amplitude adjustment process and avoids compensation commands exceeding the pump's dynamic feeding range. The intermediate regulating loop adjusts the base flow rates of the two sheath pumps separately based on the thickness errors of the upper and lower sheath layers; the intermediate regulating loop adjusts the base flow rates of the two sheath pumps every 200... Updated once: The absolute value of the thickness error is no greater than 0.5. Maintain a baseline flow rate; greater than 0.5. and not greater than 2 Each time, it will be calculated at 0.5 times the current base flow rate. Adjustment; Greater than 2 and no more than 5 At that time, adjust 1 each time. Greater than 5 At that time, adjust 2 each time. When the thickness is lower than the target value, the base flow rate of the corresponding pump is increased; when the thickness is higher than the target value, the base flow rate of the corresponding pump is decreased. The base flow rate is limited to the effective range of the pump calibration. When the adjustment direction is reversed in two consecutive update cycles, the adjustment range of the next cycle is halved to suppress the back-and-forth fluctuations near the target value. The basic update cycle of the slow adjustment loop is set to 5 seconds, but when the calculated material transport time... When the time exceeds 5 seconds, the update cycle of the slow adjustment loop is automatically and dynamically extended to [a later value]. To prevent integral saturation of the control system; maintain the core pump speed when the absolute value of the areal density error is no greater than 0.5% of the target areal density; if the absolute value of the error is greater than 0.5%,... and not greater than 2 At that time, each time it is 0.5 times the current speed. Adjustment; Greater than 2 At that time, press 1 / 2 of the current speed each time. Adjustments: When the areal density detection value is lower than the target value, increase the core layer pump speed; when it is higher than the target value, decrease the core layer pump speed. After adjusting the core layer pump speed, maintain at least two slow adjustment cycles before allowing the die head section gap to participate in the correction, thereby distinguishing between deviations caused by insufficient total material supply and uneven lateral gap. When the areal density meter provides data for five lateral detection areas, each area is mapped to its corresponding mold section; if the areal density of a certain area is higher than the target value by 1 for two consecutive slow adjustment cycles, the areal density of that area will be higher than the target value by 1. At the above times, the lip gap in the corresponding section decreases by 1 each time. ; for two consecutive periods, the value was below the target value by 1. When the above is true, increase by 1 each time. The deviation does not exceed 1. Time remains unchanged; After each adjustment, wait at least one material transport time from the die outlet to the areal density meter before evaluating the effect. The cumulative adjustment amount of each section relative to the initial gap is limited to ±20. Within this range, the gap difference between adjacent sections is limited to 10. Within; if the lateral deviation cannot be eliminated even after reaching any limit, stop further widening of the gap difference and output the inspection signal for the corresponding section; In Sp5, the controller cyclically performs the acquisition, calculation, and adjustment process according to the sampling cycle; in each cycle, the pressure signal, flow signal, pump speed signal, impedance signal, and areal density signal are uniformly sent to the controller; The controller first checks the validity of the data and the applicability of the model, then updates the sheath thickness, sheath integrity index, core solid content and various errors, and outputs the control commands corresponding to the fast adjustment loop, medium adjustment loop and slow adjustment loop respectively. The outputs of the three loops are superimposed in the order of fast, medium and slow. The fast adjustment loop only changes the periodic compensation flow, the medium adjustment loop only changes the sheath basic flow, and the slow adjustment loop only changes the core pump average speed and the lip clearance of each section, so as to avoid the same execution quantity being repeatedly corrected by different loops at the same time. In this embodiment, the main high-solids slurry passes through the die head in the form of a core layer, while the sheath layers on both sides play a role in regulating the flow near the wall. The high shear force near the wall is concentrated in the sheath layer, reducing the damage to the internal structure of the core layer. The current collector side and the air side can form interface layers with different functions through the lower sheath layer and the upper sheath layer, respectively. Meanwhile, by using a calculation link consisting of die head pressure drop, measured flow rate of two sheath layers, impedance ratio, core layer pump speed and areal density detection value, the flow rate consistency of the thickness calculation results can be checked, and the feed amount and die head gap can be corrected according to a clear effective range, threshold and adjustment step size, thereby maintaining the stability of the printing process and the consistency of areal density.

[0019] In one embodiment, in Sp1, an asymmetric three-layer co-extrusion process is used. The core layer is a slurry with a solid content of 68wt% to 78wt%. The binder mass fraction of the lower sheath slurry forming the lower sheath layer is greater than the binder mass fraction of the upper sheath slurry forming the core layer. The lower sheath slurry forms a binder-rich layer with a wet film thickness of 8μm to 15μm, and the upper sheath slurry forms a porous layer with a wet film thickness of 3μm to 6μm. In Sp1, the areal density is measured by a downstream areal density meter; in Sp4, the fast-adjustment loop uses a comb-shaped notch filter with a center frequency determined by the core pump speed and the preset number of core pump teeth to remove the tooth frequency and its harmonic components from the flow data, thus obtaining transient flow data; the transient flow data is then substituted into the three-layer flow model to calculate the dry-based transient areal density of solids. ; In the formula, The transient surface density of a dry-based solid. This represents the instantaneous mass flow rate of the core layer. The solid content of the core layer. This represents the instantaneous mass flow rate of the upper sheath. The solid content of the upper sheath. This represents the instantaneous mass flow rate of the lower sheath. The mass solids content of the lower sheath. The instantaneous belt speed, The width of the die head; Based on the distance along the belt travel from the die outlet to the downstream surface density meter and instantaneous belt speed ,according to Delay correction is applied to the surface density detection value, and the difference between the maximum and minimum surface density values ​​at each transverse detection point within the same time section is calculated as the transverse surface density range. In Sp4, the closed-loop bandwidth of the fast-adjustment loop is 200Hz to 1kHz; The closed-loop bandwidth of the intermediate regulation loop is 1Hz to 5Hz, and the sheath integrity index S controlled by the intermediate regulation loop is maintained within the range of 0.60±0.05. The closed-loop bandwidth of the slow adjustment loop is 0.05Hz to 0.2Hz. The slow adjustment loop adjusts the lip gap of five independently adjustable lip gap sections according to the lateral surface density difference.

[0020] This embodiment is used to illustrate the material configuration, transient areal density calculation, and coordination of the three types of adjustment loops under asymmetric three-layer co-extrusion conditions; the main execution components in this embodiment are still the controller, core layer pump, upper sheath pump, lower sheath pump, die head gap adjustment mechanism, and downstream areal density meter; Compared with the aforementioned embodiments, this embodiment focuses on illustrating the correspondence between material-level design and control actions in different frequency bands; The asymmetric three-layer co-extrusion process in this embodiment refers to the core layer, lower sheath layer, and upper sheath layer having different compositions and thicknesses, and the lower sheath layer and upper sheath layer each performing different functions; The core layer uses a slurry with a solid content of 68 to 78 wt%. Slurries within this range have an active material loading capacity that exceeds the preset loading threshold. When directly subjected to shear force exceeding the set shear threshold on the die wall, it is easy to cause uneven distribution of the binder near the current collector interface. Therefore, in this embodiment, the binder mass fraction in the lower sheath slurry is set to be higher than that in the core layer, so that the lower sheath layer forms a wet film with a thickness of 8 to 15 on the substrate side. Adhesive-rich layer; The binder enrichment layer refers to the lower sheath layer having a higher binder ratio than the core layer in the wet film state. After subsequent drying, this layer is retained on the current collector side to improve the bonding state between the main slurry and the substrate. Unlike the lower sheath, the upper sheath mainly faces the air side, and its function is not to enhance the bonding with the substrate, but to form a thinner surface porous structure. Therefore, in this embodiment, the wet film thickness of the upper sheath slurry is 3-6 mm. The porous layer; the upper sheath layer is less than the set thickness threshold, and together with the core layer, it forms a three-layer flow, reducing the shearing effect of the core layer on the wall of the die head, and forming a surface layer with a thickness less than the blocking threshold after film formation. The asymmetric arrangement of the lower and upper sheaths in terms of thickness and composition allows the current collector side and the air side to obtain different interface conditions, respectively. During operation, a downstream surface density meter continuously measures the surface density value after coating. The surface density meter is located downstream of the die head, and the measured surface density value corresponds to the state after the coating has traveled a preset distance, not the immediate state at the die head exit. The controller measures the distance along the conveyor belt from the die head exit to the downstream surface density meter. and instantaneous belt speed Calculate transport delay ; The controller maps the sampled value output by the areal density meter forward to the mold outlet time one transport delay in advance, thereby obtaining areal density data consistent with the flow state at that time; The controller retrieves the current time from each surface density sampling point using its timestamp and subtracts the timestamp. The corresponding flow rate and conveyor speed are recorded; if the moment falls between two stored sampling points, the two data points are linearly interpolated according to the time ratio; if the reverse lookup time is earlier than the cache start point, the areal density sampling point will not participate in this closed-loop update. The controller must maintain a historical buffer length that is at least 1.5 times the maximum transport delay to ensure that the corresponding data can be found during delay correction. When the instantaneous conveyor speed changes by more than 5 within one sampling period At that time, the controller divides the sampling period into multiple sub-intervals with approximately constant speed, calculates the transport delay of each sub-interval, and performs a time lookup. When the instantaneous change in conveyor speed does not exceed 5 At that time, the transport delay is calculated using the velocity at the start of the sampling period to avoid the same detection value being mapped repeatedly; When the instantaneous conveyor speed is less than 10% of the normal operating set speed At that time, no new delay conversion and areal density closed-loop update are performed. The previous effective delay correction result is maintained, and the calculation is resumed only after the belt speed recovers to above the threshold. The controller determines the tooth frequency based on the core pump speed and the number of teeth of the core pump to extract transient flow data for use in the fast regulation loop; after the tooth frequency is determined, the controller calls the comb notch filter with the center frequency corresponding to that tooth frequency to process the flow data related to the core, upper sheath and lower sheath. The discrete-time transfer function of the comb notch filter is designed as follows: ; in, for Transform the complex variable, The delay order is obtained by rounding down the ratio of the industrial control computer's sampling frequency to the corresponding notch filter center frequency. The pole radius parameter set according to the 8% bandwidth requirement ( ); The filtered data serves as transient flow data without the tooth frequency and its harmonic periodic components, while the separated tooth frequency and its harmonic components serve as the basis for compensation in the fast adjustment loop. To ensure that the filtering process has defined boundaries, the controller uses the core pump tooth frequency as the first notch center frequency, and its integer multiples as the subsequent notch center frequencies, until the notch point is stopped from being added when the multiples are higher than half of the sampling frequency of the fast adjustment loop. The bandwidth of each notch point is taken as 8 times the corresponding center frequency. If the interval between a certain notch center frequency and an adjacent center frequency is less than twice the bandwidth, then the next notch point with the higher frequency is canceled to avoid overlap of adjacent notch regions. In this embodiment, the filtered flow rate data output by the comb notch filter is the sole input for calculating the transient surface density of dry-based solids; The controller inputs the filtered three flow data points into the dry-based solid transient surface density calculation process according to their timestamps. Used to characterize the slow-variable surface density state after removing the core pump tooth frequency and its harmonic pulsations. The filtered tooth frequency and its harmonic components are used only as the basis for compensation in the fast adjustment loop and are no longer added back. The calculation input is so that the tooth frequency component will not be repeatedly included or canceled in the calculation of transient surface density of dry-based solids; To evaluate the impact of tooth frequency disturbance on the total feed rate, the controller separately saves the difference between the original flow rate data and the filtered flow rate data as the tooth frequency disturbance monitoring quantity. This monitoring quantity is only used to evaluate the tooth frequency amplitude before and after compensation and does not change the tooth frequency amplitude. The calculation results; After obtaining the transient flow data, the controller calculates the instantaneous mass flow rate of the core layer. Instantaneous mass flow rate of the upper sheath Instantaneous mass flow rate of the lower sheath Core layer quality solid content upper sheath mass solids content lower sheath mass solids content Instantaneous belt speed and die head width Substitute into the following formula to calculate the transient surface density of the dry-based solid: ; In the above formula, the units for the instantaneous mass flow rates of the three layers are unified as follows: The solid content is a dimensionless mass fraction, and the product of the instantaneous conveyor speed and the die width is... Therefore, the calculated Units are ; The controller first multiplies the instantaneous mass flow rate of each of the three layers by its respective mass solid content to obtain the supply of solid components in the three layers at the corresponding time; then it adds the three together to obtain the total solid supply at that time. The instantaneous conveyor speed and die width represent the substrate area per unit time, and the transient surface density of dry-based solids per unit area is obtained by dividing them; due to the core layer mass solid content... It can be obtained by converting and correcting the aforementioned impedance ratio, so the calculation result can simultaneously reflect the impact of feed fluctuations and core solid content changes on surface density; when When the flow area is less than the preset minimum area threshold, the controller does not calculate new flow. While maintaining the most recent effective transient surface density value; the preset minimum area flow threshold is taken as 10 times the product of the normal operating speed and the die width. To avoid an abnormal amplification of transient surface density due to an excessively small denominator during low-speed start-stop phase; In terms of lateral uniformity control, this embodiment uses the delayed-corrected areal density detection value to calculate the lateral areal density range. The specific method is that the controller reads the surface density values ​​of each transverse detection point of the surface density meter within the same time section, finds the maximum and minimum values, and takes the difference between the two as the transverse surface density range. The transverse surface density range characterizes the degree of difference in the material output state at different positions in the transverse direction of the die head and the uniformity of transverse coating; the transverse surface density range is fed into the slow adjustment loop to determine the lip gap correction direction and correction range of the five independently adjustable sections. To ensure a definite correspondence between cross sections at the same time, the controller only calculates the range for lateral detection points that have completed the same delay correction batch; If a certain transverse detection point is missing in this batch, the transverse surface density range of this batch will not be updated, and the valid results of the previous batch will be retained. In this embodiment, the intermediate adjustment loop uses the sheath thickness error as the direct adjustment basis and the sheath integrity index as the adjustment basis. As a supervisory constraint on the sheath state, the controller first calculates the adjustment direction and adjustment amount of the basic flow of the two sheath pumps based on the thickness error of the upper and lower sheaths respectively. Before outputting this adjustment amount, check Is it at 0.60? Within the range; when When the thickness is below 0.55, prioritize increasing the base flow rate of the sheath pump on the side with insufficient thickness; if both the upper and lower sheath thickness errors are positive, determine the side to be adjusted based on the larger absolute value of the thickness errors on both sides; when When the value is higher than 0.65, the base flow rate of the sheath pump on the side with the corresponding excessive thickness should be reduced first. If the thickness on both sides exceeds the target value, the side with the larger absolute value of the thickness error on both sides shall be selected for priority adjustment; when When the value is between 0.55 and 0.65, the basic flow rate is adjusted only according to the sheath thickness error. Thus, the sheath thickness error determines the specific adjustment object and direction of the adjustment loop. The sheath integrity index only imposes priority constraints on the adjustment action when it exceeds the target range, and does not replace the calculation and adjustment of the sheath thickness error. In this embodiment, the fast adjustment circuit, medium adjustment circuit, and slow adjustment circuit are divided into different frequency ranges; the working bandwidth of the fast adjustment circuit is set to 200Hz to 1kHz, which is used to handle the tooth frequency and its related rapid fluctuations. The operating bandwidth of the intermediate regulating loop is set to 1 to 5 Hz, and the control objective of the intermediate regulating loop is to improve the sheath integrity index. Maintain within the range of 0.60±0.05; when When the flow rate is below the lower limit of the range, increase the base flow rate of the sheath pump on the side where insufficient thickness or decreased sheath integrity occurs; when When the flow rate exceeds the upper limit of the range, reduce the base flow rate of the sheath pump on the side where the thickness is too large; when When within the range, adjustment is made only based on the corresponding sheath thickness error; the operating bandwidth of the slow adjustment loop is set to 0.05–0.5 Hz, and the areal density detection value after delay correction and the lateral areal density range are received; To avoid control command conflicts caused by inconsistencies between update times and closed-loop bandwidth in different loops, the controller uses closed-loop bandwidth as an upper limit constraint for signal processing and command updates. The fast adjustment loop updates only the tooth frequency compensation component, the medium adjustment loop updates only the sheath basic flow rate, and the slow adjustment loop updates only the core pump average speed and the lip clearance of each section. The middle regulating circuit every 200 Perform a thickness error calculation and basic flow update, but after limiting and low-pass processing, the effective change frequency of the output does not exceed 5. ; When the sheath integrity index exceeds 0.55 to 0.65, the corresponding basic flow rate is adjusted according to the aforementioned priority constraints. After returning to this range, the adjustment is once again dominated by the thickness error. After issuing the gap correction command, the slow adjustment loop waits at least for the material transport time from the die outlet to the areal density meter, and evaluates the effect of the command after the delay correction is completed. During the waiting period, the areal density value that has not yet been matched with the material segment is not repeatedly corrected; the instructions of the three types of loops are written into independent compensation amount, base amount and average set amount respectively, and are combined into execution instructions by the controller in the same output cycle to avoid the same execution amount being repeatedly overwritten by different loops; Through the above material configuration and control, the core layer can maintain a high solids content feed, the lower sheath layer forms a thicker binder-rich layer on the substrate side, and the upper sheath layer forms a thinner porous layer on the air side. At the same time, the problems of rapid pulsation, sheath layer maintenance and lateral uniformity are handled by adjusting loops with different bandwidths respectively. Through the above configuration, the areal density stability, lateral consistency and interface layer function allocation are taken into account in the same device.

[0021] In one embodiment, in Sp1, pressure, impedance, flow rate, and areal density data are synchronized with a unified timestamp; outliers in the head pressure drop signal are removed using a Hamper filter method with a window length of 15 sampling points, a scale parameter σ estimated based on the absolute deviation of the median within the window, and a decision threshold of 3σ. When any channel in the five groups of four microelectrodes has no effective voltage data for 200ms, linear interpolation is performed based on the impedance amplitude of the corresponding electrode in the same group or the horizontally adjacent group; if neither the corresponding electrode in the same group nor the horizontally adjacent group can provide a valid detection value, the most recent effective voltage data of the channel is retained, and the impedance value of the channel is not updated in the current sampling period. When the viscosity ratio of the upper or lower sheath layer to the core layer is greater than or equal to 50, the ratio of the thickness of the upper or lower sheath layer to the die gap is less than or equal to 1.5%, and a non-tooth frequency periodic peak value from the die pressure drop is detected in the 30Hz to 80Hz frequency band with an amplitude greater than or equal to the set signal-to-noise ratio threshold determined by the pressure spectrum under the stable three-layer flow state, the flow state is determined to be abnormal and an abnormality mark is set. After an anomaly is detected, the speed of the upper and lower sheath pumps is increased by 20%. If the non-tooth frequency periodic peak does not disappear within 3 seconds, the output of the fast adjustment circuit and the medium adjustment circuit is stopped, the core pump is kept running, and the operation is degraded. If the peak value is less than or equal to the set signal-to-noise ratio threshold and is maintained for 1 second, the anomaly mark is removed, and the speed of the upper and lower sheath pumps is restored to the base speed before the anomaly by decreasing by 5% every 500ms. In the event of a power failure, the upper and lower sheath slurry is continuously supplied to the die head through the backup pressure-holding circuit consisting of the backup storage tank and the switching valve. The upper and lower sheath pumps are plunger pumps with periodic flow compensation capability. The fast adjustment loop generates a tooth frequency flow fluctuation template based on the flow data of the core pump over multiple rotation cycles according to the encoder angle of the core pump. It calculates the amplitude and phase of the tooth frequency flow fluctuation template and controls the upper and lower sheath pumps to generate compensation flow with opposite phase. Every 30 minutes of operation, reverse the shafts of the upper and lower sheath pumps and return the flow to the die inlet at a flow rate three times the rated flow rate for 1 second. Repeat this process three times. Then, with the upper and lower sheath pumps stopped and the core layer supply maintained, remeasure the reference pressure drop in the sheathless state. In SP5, the channel validity of five sets of four microelectrodes and two pressure sensors is detected by the standard that the signal amplitude and impedance are within the set range. When the signal loss time of at least three sensor channels is detected to be greater than or equal to the preset time threshold, the average value of the sheath integrity index and sheath thickness within the last 30 seconds is used as the controller hold value, the output of the fast adjustment loop and the medium adjustment loop is stopped, and the conveyor speed is reduced to 60% of the rated speed; when there are fewer than three failed channels and the channel validity detection standard is met for 5 consecutive seconds, the multi-channel failure mark is removed, and the closed-loop input is re-established with three consecutive valid sampling cycles.

[0022] This implementation method is used to illustrate data synchronization, pressure signal preprocessing, microelectrode failure compensation, abnormal flow judgment, tooth frequency compensation template generation, self-cleaning benchmark retest, and degraded operation mode when multiple channels fail. The execution entities in this embodiment include a mold head side control unit, an industrial computer, an upper sheath pump, a lower sheath pump, a core pump, a conveyor belt mechanism, a spare liquid storage tank, and a switching valve; the mold head side control unit is responsible for high-frequency signal acquisition and preliminary data processing, while the industrial computer is responsible for anomaly detection and adjustment command generation; In Sp1, the die head side control unit performs unified timestamp synchronization on pressure, impedance, flow rate and areal density data; unified timestamp synchronization means that under the same sampling reference, a unified timestamp is added to the pressure sensor output, four microelectrode output, upper sheath flow rate, lower sheath flow rate and areal density meter output, so that the controller can correspond the data within the same sampling period to the same material state in subsequent calculations. After time synchronization is completed, the controller will arrange various types of data into the buffer in chronological order, and then call the corresponding processing program for subsequent calculations. The controller uses the pressure sampling clock as the primary time base to ensure consistency between subsequent delay correction and multi-sensor alignment. Impedance, flow rate, and areal density data are all converted to this primary time base. If two adjacent original sampling points of a certain type of data span a main time base sampling time, a synchronization value is added to that time according to the time ratio; if the continuous missing time exceeds one sampling period of its own, the data of that type at that main time base time is marked as missing and is directly overwritten without the previous value. Since the pressure sensor output may contain burr signals, this embodiment uses the Hamper filtering method to remove burrs from the pressure data before calculating the pressure drop of the die head. In practice, the controller uses a sliding window consisting of 15 consecutive sampling points. Within each window, the median is first calculated, then the absolute deviation of each sampling point from the median is calculated, and the median of these absolute deviations is used to estimate the scale parameter. ; The controller calculates the absolute value of the difference between each sample point within the window and the median. Comparison; when the deviation of a certain sampling point exceeds When this happens, the controller identifies the sampling point as a glitch and replaces the original value with the middle value in the window. When the deviation does not exceed When the original value is retained, the above processing will prevent sudden spikes from directly entering the pressure drop calculation link, which can reduce the error in subsequent sheath thickness calculation and anomaly detection. For cases where the sliding window is located at the beginning or end of the data stream and cannot simultaneously capture symmetrical data before and after, the controller adopts the order of prioritizing data before the current point and then filling in the missing data by capturing data after the current point. If the number of valid points in the entire window is less than 9, the window will not be subject to Hampel determination, the original pressure value will be retained directly, and the window will be marked as a low confidence window. For the four-microelectrode array, this implementation specifies that when any channel in any of the five groups of four microelectrodes has no effective voltage data for 200ms, the compensation processing is initiated. Here, "no effective voltage data" means that during continuous monitoring, the output of this channel is missing or does not meet the acquisition conditions, and cannot be used to calculate the dual-frequency impedance amplitude. After this condition is triggered, the controller first searches for the detection values ​​of the adjacent electrodes in the same group. If the data of the adjacent electrodes in the same group is valid, the temporary substitute value of the missing channel is filled in by linear interpolation. If adjacent electrodes in the same group also cannot provide a valid reference, the detection value of the corresponding electrode in the adjacent horizontal group is read, and a substitute value is calculated according to the linear change relationship between adjacent groups. After the replacement value is completed, the controller uses the substitute value to continue to complete the impedance ratio calculation to prevent the entire group of horizontal data from losing continuity due to a short-term disconnection of a single channel. The controller selects reference values ​​and executes the replacement value rule in the following fixed order: prioritizes the left and right adjacent electrodes in the same group. When only one side of the left and right adjacent electrodes is valid, the valid value of that single side is directly taken as the substitute value; when there is no valid reference in the same group, the corresponding electrode with the same number in the horizontally adjacent group is selected; when the corresponding electrodes of the left and right adjacent groups are valid at the same time, the average value of the two is taken. When only one adjacent group is valid, the value of that side is taken; when none of the above reference values ​​exist, no value is added, and the channel is marked as unavailable in this sampling period. In the fast-adjustment loop, the upper sheath pump and the lower sheath pump in this embodiment are plunger pumps with periodic flow compensation capability. In order to make the compensation action correspond to the core pump pulsation, the controller reads the flow data of the core pump in multiple rotation cycles and averages it by the encoder angle. The average value of the flow rate data falling within the same angle interval is calculated by dividing the rotation cycle of one or more core pumps into several angle intervals, and then averaging the flow rate data to obtain the flow rate fluctuation curve corresponding to the rotation angle. The controller then extracts the tooth frequency flow fluctuation template based on the curve and calculates the amplitude and phase of the template. The phase in the template refers to the angular position of the peak tooth frequency fluctuation relative to the zero position of the encoder in the core pump encoder angular coordinate. It is used to unify the timing reference between the core pump pulsation and the compensation actions of the upper and lower sheath pumps. The phase-opposite compensation flow refers to the compensation waveforms output by the upper and lower sheath pumps being offset by 180° in the angular coordinates relative to the tooth frequency flow fluctuation template, thereby providing reverse compensation within the angular range of core pulsation increase. During the compensation process, the controller writes the original flow data, filtered flow data, template amplitude and phase difference into the same circular buffer. Each record in the buffer carries three status flags: valid, rebuilt, and degraded. Once a channel is linearly interpolated, it only participates in amplitude and phase calculations in the current sampling period. If no real data is found for three consecutive sampling periods, the channel switches to a degraded state and freezes the interpolation results, no longer participating in template reconstruction. The tooth frequency template is updated using a verification-then-replacement sequence. Only when the correlation coefficient of two consecutive complete rotation cycles is higher than the set threshold, the new template is used to cover the old template; otherwise, the most recent stable template is retained as the input of the fast adjustment loop to avoid transient glitches directly changing the compensation direction. The determination of abnormal flow state is performed after the thickness calculation, viscosity conversion and pressure spectrum analysis are completed; the controller calculates the viscosity ratio of the upper sheath layer to the core layer and the viscosity ratio of the lower sheath layer to the core layer, and calculates the ratio of the upper sheath layer thickness to the die head gap and the ratio of the lower sheath layer thickness to the die head gap, respectively. Pressure spectrum analysis uses the head pressure drop signal after Hampel filtering. The controller calculates the periodic peaks in the 30-80Hz frequency band within a continuous 1s data window, and removes the tooth frequency and its integer multiples obtained from the core pump speed and number of teeth from the candidate peaks; the frequency tolerance during removal is ±5% of the corresponding tooth frequency. The signal-to-noise ratio threshold is set by the pressure spectrum obtained by the current printing recipe under a stable three-layer flow state. The controller uses the root mean square value of the non-tooth frequency background amplitude in the 30-80Hz frequency band as the noise reference, and takes 6 times the noise reference as the set signal-to-noise ratio threshold. The controller determines that the flow state is abnormal only when the three conditions of viscosity ratio, thickness ratio and non-tooth frequency periodic peak are simultaneously met on the same side of the sheath and two consecutive spectral data windows are met. After detecting the anomaly, the controller simultaneously increases the speed of both the upper and lower sheath pumps by 20%, which is added to the base speed given by the middle regulating loop, and is maintained for 3 seconds; within these 3 seconds, the controller continues to detect non-tooth frequency periodic peaks according to the same spectrum rules. If the peak value is lower than the set signal-to-noise ratio threshold and remains so for 1 second, the abnormality flag is removed, and the speed of the upper sheath pump and the lower sheath pump is restored to the base speed before the abnormality by decreasing by 5% every 500ms. If the non-tooth frequency periodic peak does not disappear within 3 seconds, stop the output of the fast adjustment circuit and the medium adjustment circuit, keep the core layer pump running at the average speed before entering the degradation, and prohibit the lip gap of the die section from continuing to expand the lateral difference. During degraded operation, the controller only retains the monitoring functions of pressure, conveyor speed and areal density. It is only allowed to re-enter the normal closed loop after the pressure peak is lower than the set signal-to-noise ratio threshold for 10 consecutive seconds and the two sheath pumps re-establish stable flow. When power is lost, the backup pressure-holding circuit consisting of the backup liquid storage tank and the switching valve is automatically connected to the upper and lower sheath inlets of the die head; the backup liquid storage tank is pre-filled with sheath liquid compatible with the upper and lower sheaths, and the outlet of the liquid storage tank is connected to the upper sheath inlet and the lower sheath inlet respectively through a one-way valve and a flow-limiting orifice; When the power supply is normal, the switching valve keeps the main feed branch connected. When the power is cut off or the controller is de-energized, it is reset to the standby branch connected position by the spring, so that the standby liquid storage tank continuously supplies the upper sheath layer slurry and the lower sheath layer slurry to the die head at the set pressure. The pressure holding pressure is set to 80% of the average pressure of the upper and lower sheath inlet during normal operation. The backup supply time is not less than the time required to complete the shutdown conveyor belt and the pressure relief of the die head. This backup pressure holding circuit does not participate in the normal thickness closed loop. Its function is to maintain the coverage of the upper and lower sheath slurry near the die lip during the moment of power failure and shutdown, thereby reducing the direct retention of high solids core layer slurry on the wall surface. Every 30 minutes of operation, the controller performs a self-cleaning action on the upper and lower sheath branches. Before the self-cleaning starts, the controller freezes the output of the fast adjustment circuit and the medium adjustment circuit, keeps the core pump feeding material at the current average speed, and records the average die head pressure drop, upper sheath base flow rate and lower sheath base flow rate within 30 seconds before the self-cleaning. The controller reverses the shafts of the upper and lower sheath pumps, returning the flow to the die inlet at a flow rate three times the rated flow rate for 1 second, then stops for 1 second before reversing the flow again, repeating this process three times. After the three refluxes are completed, the upper sheath pump and the lower sheath pump resume forward feeding and stabilize for 5 seconds; if the flow rates of both sheaths return to within ±2% of the baseline flow rate before self-cleaning, the reference pressure drop retesting step is initiated; otherwise, the most recent effective reference pressure drop is maintained and a sheath branch check signal is output. During the baseline pressure drop retest, the controller stops the upper and lower sheath pumps, keeps the core supply, belt speed, die gap and slurry temperature near the normal operating settings, and collects pressure drop data in the sheathless state. The controller continuously collects data for 5 seconds and uses the average value of the 3-second pressure drop data as the new reference pressure drop for the sheathless state. If the range of the pressure drop within 3 seconds exceeds 2% of the average value, the retest is invalid and the original reference pressure drop is used again. After the new reference pressure drop takes effect, the controller restarts the upper and lower sheath pumps, and restores the fast and medium regulation loops after the thickness calculation is valid for three consecutive sampling cycles. In SP5, the controller performs channel validity detection on five sets of four microelectrodes and two pressure sensors. The detection targets for the four microelectrode channels include the excitation voltage amplitude, response voltage amplitude, and impedance value calculated from the dual-frequency impedance amplitude. The detection targets for the pressure sensor channels include the pressure signal amplitude, zero-point drift, and sampling continuity. The set range is determined jointly by the sensor range and the operating calibration value. The pressure signal amplitude should be between 1% and 95% of the sensor's full scale, and the zero drift should not exceed 0.5% of the full scale for 10 consecutive seconds; the impedance values ​​of the four microelectrodes should be between the upper and lower limits of the impedance range of the corresponding solid content calibration table, and the excitation voltage amplitude should be within ±5% of the set excitation amplitude; if any channel exceeds the above range or has no sampling value for a continuous period, the controller will mark that channel as a failed channel. The set time threshold is 500ms; when the signal loss time of at least three sensor channels exceeds 500ms, the controller enters multi-channel failure degradation processing. After entering this process, the controller calculates the average value of the effective sheath integrity index in the last 30 seconds, as well as the average value of the effective upper sheath thickness and lower sheath thickness in the last 30 seconds, and uses these average values ​​as the controller hold values. If the number of valid samples for a certain holding value is less than 50% of the total number of samples in the last 30 seconds, then the average value of the 10 valid sampling periods before the downgrade is used as the corresponding holding value. The controller stops the output of the fast and medium regulating loops, maintains the upper and lower sheath pumps at their baseline flow rates before degradation, and reduces the conveyor belt speed to 60% of its rated speed. In degraded state, the slow adjustment loop no longer increases the lip gap difference of the five sections, and is only allowed to maintain the current gap or execute the stop command after manual confirmation; The controller will remove the multi-channel failure flag and re-establish the closed-loop input after the failure channels are less than three and the channel validity detection criteria are met for 5 consecutive seconds.

Claims

1. A method for printing electrode paste cells based on flow field control, characterized in that, Includes the following steps: Sp1. Slurry is supplied to the die head with an independently adjustable lip gap section through the core layer pump, upper sheath layer pump and lower sheath layer pump, and coated on the continuous conveyor belt substrate in one step to form the lower sheath layer, core layer and upper sheath layer stacked in sequence; during the coating process, the die head pressure drop, upper sheath layer flow rate, lower sheath layer flow rate, core layer pump speed, conveyor belt speed and areal density are measured. Sp2. Based on the die head pressure drop, upper sheath flow rate, lower sheath flow rate, and core pump speed, combined with preset die head structural parameters and preset slurry viscosity, the preset die head structural parameters include die head width, die lip straight section length, and die head clearance. The preset slurry viscosity includes core layer viscosity, upper sheath viscosity, and lower sheath viscosity. A three-layer flow model of parallel flat plate narrow slit flow is established based on the steady-state incompressible fluid momentum conservation relationship. The upper sheath thickness and lower sheath thickness are calculated using the following formula: ； ； In the formula, For apparent viscosity, To measure the pressure drop of the die head, For die head clearance, The width of the die head. The average density of the slurry, The length of the straight section of the mold lip. For total mass flow rate, The thickness of the upper sheath or the lower sheath. For core layer viscosity, This corresponds to the viscosity of the upper sheath or the viscosity of the lower sheath. Sp3, set the preset target surface density and preset target sheath thickness, and calculate the surface density error and sheath thickness error respectively; Sp4. A fast adjustment loop, a medium adjustment loop, and a slow adjustment loop are configured: The fast adjustment loop adjusts the tooth frequency compensation flow of the upper and lower sheath pumps based on the tooth frequency flow fluctuation corresponding to the core pump speed; the tooth frequency flow fluctuation refers to the periodic flow fluctuation corresponding to the tooth frequency and its harmonic components in the core pump supply flow, where the frequency is equal to the product of the core pump speed and the preset core pump tooth number; the tooth frequency compensation flow refers to the compensation flow output by the upper and lower sheath pumps, which is opposite in phase to the tooth frequency flow fluctuation; the medium adjustment loop adjusts the basic flow of the upper and lower sheath pumps based on the sheath thickness error; the slow adjustment loop adjusts the core pump speed and the lip gap of the die head with an independently adjustable lip gap section based on the areal density error. Sp5. Execute steps Sp1 to Sp4 in a cyclical manner according to the sampling period. In each sampling period, the fast adjustment loop, the medium adjustment loop, and the slow adjustment loop independently update their respective adjustment values. The outputs of the three adjustment loops are superimposed in the order of fast adjustment loop, medium adjustment loop, and slow adjustment loop and then output to the corresponding actuators to cyclically adjust the flow field state and surface density.

2. The electrode paste cell printing method based on flow field control according to claim 1, characterized in that, In Sp1, the die head includes five independently adjustable lip gap sections arranged laterally, a manifold, and a die lip straight section; the pressure difference between the two positions is measured by two pressure sensors respectively set at the inlet of the manifold and the die lip straight section as the die head pressure drop.

3. The electrode paste cell printing method based on flow field control according to claim 2, characterized in that, Five sets of four microelectrodes are uniformly embedded in the transverse direction on the lower wall surface of the straight section of the mold lip; In Sp1, the reference pressure drop in the sheathless state when the upper sheath pump and the lower sheath pump are stopped is also collected. And the five sets of four microelectrodes in 5 and 100 Impedance amplitude under excitation; in Sp2, the sheath integrity index is determined based on the head voltage drop in the sheathed state and the reference voltage drop in the unsheathed state. ; And calculate the impedance ratio corresponding to each group of microelectrodes. According to the impedance ratio The calibration relationship determines the core solids content; When the calculated deviation of the solid content of the core layer is greater than or equal to the percentile value of the absolute deviation between the solid content back-calculated value and the known value at each calibration point, a sinusoidal temperature modulation with a temperature rise and fall range of 1.5K centered on the current stable wall temperature is applied to the lower wall surface of the straight section of the die lip, and the core layer solid content detection result is corrected according to the differential response of the die head pressure drop with temperature change; when the calculated deviation is less than the set solid content threshold, the current core layer solid content detection result is maintained.

4. The electrode paste cell printing method based on flow field control according to claim 1, characterized in that, In Sp1, an asymmetric three-layer co-extrusion process is used. The core layer is a slurry with a solid content of 68wt% to 78wt%. The binder mass fraction of the lower sheath slurry forming the lower sheath layer is greater than that of the upper sheath slurry forming the core layer. The lower sheath slurry forms a binder-rich layer with a wet film thickness of 8μm to 15μm, and the upper sheath slurry forms a porous layer with a wet film thickness of 3μm to 6μm.

5. The electrode paste cell printing method based on flow field control according to claim 3, characterized in that, In Sp1, the areal density is measured by a downstream areal density meter; in Sp4, the fast adjustment loop uses a comb-shaped notch filter with a center frequency determined by the core pump speed and the preset number of core pump teeth to remove the tooth frequency and its harmonic components from the flow data, thus obtaining transient flow data; the transient flow data is then substituted into the three-layer flow model to calculate the dry-based transient areal density of solids. ； In the formula, The transient surface density of a dry-based solid. This represents the instantaneous mass flow rate of the core layer. The solid content of the core layer. This represents the instantaneous mass flow rate of the upper sheath. The solid content of the upper sheath. This represents the instantaneous mass flow rate of the lower sheath. The mass solids content of the lower sheath. The instantaneous belt speed, The width of the die head; Based on the distance along the belt travel from the die outlet to the downstream surface density meter and instantaneous belt speed ,according to Delay correction is applied to the surface density detection values, and the difference between the maximum and minimum surface density values ​​at each transverse detection point within the same time section is calculated as the transverse surface density range.

6. The electrode paste cell printing method based on flow field control according to claim 5, characterized in that, In Sp4, the closed-loop bandwidth of the fast-adjustment loop is 200 Hz to 1 kHz; The closed-loop bandwidth of the intermediate regulating loop is 1Hz to 5Hz, and the intermediate regulating loop controls the sheath integrity index S to remain within the range of 0.60±0.05; The closed-loop bandwidth of the slow adjustment circuit is 0.05Hz to 0.2Hz. The slow adjustment circuit adjusts the lip-mouth gap of the five independently adjustable lip-mouth gap sections according to the lateral surface density difference.

7. The electrode paste cell printing method based on flow field control according to claim 3, characterized in that, In Sp1, pressure, impedance, flow rate, and areal density data are synchronized using a unified timestamp. An outlier in the head voltage drop signal was removed by using a Hamper filter method with a window length of 15 sampling points, an estimation of the scale parameter σ based on the absolute deviation of the median within the window, and a judgment threshold of 3σ. When any channel in the five groups of four microelectrodes is continuously 200 When there is no effective voltage data, linear interpolation is performed based on the impedance amplitude of the corresponding electrodes in the same group or in the transverse adjacent group.

8. The electrode paste cell printing method based on flow field control according to claim 3, characterized in that, When the viscosity ratio of the upper or lower sheath layer to the core layer is greater than or equal to 50, the ratio of the thickness of the upper or lower sheath layer to the die gap is less than or equal to 1.5%, and a non-tooth frequency periodic peak value with an amplitude greater than or equal to the set signal-to-noise ratio threshold determined by the pressure spectrum under the stable three-layer flow state is detected in the 30Hz to 80Hz frequency band from the die pressure drop, the flow state is determined to be abnormal and an abnormality mark is set. After detecting the anomaly, the rotational speeds of the upper sheath pump and the lower sheath pump are increased by 20%. If the non-tooth frequency periodic peak does not disappear within 3 seconds, stop the output of the fast adjustment circuit and the medium adjustment circuit, keep the core pump running, and enter the degraded operation state. If the peak value is less than or equal to the set signal-to-noise ratio threshold and remains so for 1 second, the abnormality flag is removed, and the rotational speeds of the upper sheath pump and the lower sheath pump are restored to their pre-abnormal base rotational speeds by decreasing by 5% every 500ms. In the event of a power outage, a backup pressure-maintaining circuit consisting of a backup storage tank and a switching valve continuously supplies the upper sheath slurry and the lower sheath slurry to the die head.

9. The electrode paste cell printing method based on flow field control according to claim 3, characterized in that, The upper sheath pump and the lower sheath pump are plunger pumps with periodic flow compensation capability; The fast adjustment loop generates a tooth frequency flow fluctuation template based on the flow data of the core pump over multiple rotation cycles, according to the encoder angle of the core pump, calculates the amplitude and phase of the tooth frequency flow fluctuation template, and controls the upper and lower sheath pumps to generate compensating flow with opposite phases. Every 30 minutes of operation, the shafts of the upper sheath pump and the lower sheath pump are reversed, and the flow is returned to the die inlet at a flow rate three times the rated flow rate for 1 second. This is repeated three times. Then, the reference pressure drop in the sheathless state is measured again while the upper and lower sheath pumps are stopped and the core layer supply is maintained.

10. The electrode paste cell printing method based on flow field control according to claim 5, characterized in that, In SP5, the channel validity of five sets of four microelectrodes and two pressure sensors is detected by the standard that the signal amplitude and impedance are within the set range. When the signal loss time of at least three sensor channels is detected to be greater than or equal to a preset time threshold, the controller holds the average value of the sheath integrity index and sheath thickness over the past 30 seconds as the hold value, stops the output of the fast adjustment loop and the medium adjustment loop, and reduces the belt conveyor speed to 60% of the rated speed. ; When fewer than three channels fail and the channel validity detection criteria are met for 5 consecutive seconds, the multi-channel failure flag is removed, and a closed-loop input is re-established with three consecutive valid sampling cycles.

Images