A detection system for new energy battery production

Abstract

The application discloses a kind of detection systems for new energy battery production, it is related to new energy battery detection technical field, including: sensing acquisition module adopts double-channel spectral confocal displacement sensor, for synchronously collecting the displacement data on the upper and lower surfaces of pole piece;Edge computing module is used for real-time processing thickness data and identifying abnormality;Self-calibration compensation module integrates miniature standard reflector plate and temperature and humidity sensor, for automatically calibrating zero point and eliminating environmental drift;The application integrates spectral confocal displacement sensing, temperature and humidity sensing and image recognition technology, constructs multi-source data acquisition network, carries out real-time fusion analysis to thickness data, environmental parameter and surface topography feature by preset algorithm, realizes thickness detection and defect diagnosis integration, solve the problem that traditional equipment function is single, easily disturbed by environment.

Description

Technical Field

[0001] This invention relates to the field of new energy battery testing technology, and more specifically to a testing system for new energy battery production. Background Technology

[0002] The manufacturing of current new energy batteries (such as lithium-ion batteries) requires extremely high consistency, and the uniformity of electrode coating directly affects battery capacity, cycle life, and safety. Therefore, inspection during the electrode coating process is a crucial step. Current technologies primarily employ contact thickness measurement or conventional laser thickness measurement. However, in practice, contact thickness measurement and conventional laser thickness measurement have several drawbacks: first, they are susceptible to environmental vibrations and temperature drift, resulting in significant accuracy fluctuations; second, they cannot achieve full-width continuous scanning, creating blind spots; third, data lag makes real-time feedback and adjustment of coating machine parameters difficult; and fourth, they lack intelligent diagnostic capabilities, relying on manual intervention for anomaly identification.

[0003] Therefore, how to propose a testing system for new energy battery production and overcome the shortcomings of existing technologies is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] In view of this, the present invention provides a testing system for new energy battery production, which realizes non-contact, high-precision, intelligent, and embeddable non-destructive testing in production line closed-loop control. To achieve the above objectives, the present invention adopts the following technical solution: A testing system for new energy battery production, comprising: The module includes a sensor acquisition module, an edge computing module, a self-calibration and compensation module, and a communication control module. The sensing and acquisition module uses a dual-channel spectral confocal displacement sensor to synchronously acquire displacement data of the upper and lower surfaces of the electrode. The edge computing module is used to calculate the electrode thickness data and identify anomalies based on the displacement data of the upper and lower surfaces of the electrode. The self-calibration compensation module is used to automatically calibrate the zero point and eliminate environmental drift; The communication control module has a built-in EtherCAT master controller, which is used to connect the sensor acquisition module, edge computing module, and self-calibration compensation module to the EtherCAT master controller for closed-loop adjustment.

[0005] Optionally, it also includes: a coating machine servo system and a coating machine doctor blade, wherein the coating machine servo system is signal-connected to the coating machine doctor blade, and the EtherCAT master station controller is signal-connected to the coating machine servo system.

[0006] Optionally, the dual-channel spectral confocal displacement sensor is arranged laterally at the exit of the coating machine's scraper.

[0007] Optionally, the edge computing module includes: using a CNN-LSTM hybrid neural network model to identify three types of abnormal patterns: abrupt changes in thickness, gradual slopes, and periodic fluctuations, and generating corresponding alarm signals.

[0008] Optionally, the self-calibration compensation module includes a miniature standard reflector and a temperature and humidity sensor, which automatically triggers zero-point calibration every 30 minutes, collects environmental data through the temperature and humidity sensor, and performs real-time temperature drift compensation on the collected results.

[0009] Optionally, the communication control module is used to send an adjustment command to the coating machine servo system within 100ms when a local thickness deviation of the electrode is detected to exceed ±2μm, so as to adjust the angle of the coating machine blade.

[0010] Optionally, the sensing acquisition module adopts a linear array multi-point layout, with a horizontal scanning range covering the full width of the electrode, acquiring no less than 1,000 data points per second to generate a thermal map of the electrode thickness.

[0011] Optionally, the edge computing module has a built-in data storage unit for storing no less than 72 hours of detection data and supports export via USB interface.

[0012] Optionally, it also includes: the communication control module uploads the test data to the cloud-based quality traceability platform via industrial Ethernet.

[0013] Optionally, it also includes an OTA remote upgrade function module for updating the edge computing module and the self-calibration compensation module via a wireless network.

[0014] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a testing system for new energy battery production, which has the following beneficial effects: This invention breaks through the limitations of traditional single-sensor detection by integrating spectral confocal displacement sensing, temperature and humidity sensing, and image recognition technologies to construct a multi-source data acquisition network. Through a preset algorithm, it performs real-time fusion analysis of thickness data, environmental parameters, and surface morphology features, not only improving thickness detection accuracy to ±0.3μm but also simultaneously identifying microscopic defects such as scratches and bulges on the electrode surface. This achieves integrated thickness detection and defect diagnosis, solving the problems of traditional equipment being functionally limited and susceptible to environmental interference. The design of this invention is based on a real-time feedback module of edge computing, with a built-in CNN-LSTM hybrid neural network model. It can dynamically match the production conditions according to the detection data. When the electrode thickness deviation is detected to exceed the threshold, the system automatically generates adjustment instructions and directly controls the angle of the coating machine's scraper and the material feeding speed through the EtherCAT bus, forming a closed-loop control of detection-analysis-adjustment. Compared with the traditional lag-based manual adjustment, the response time is shortened to less than 0.5 seconds, which improves the consistency of electrode thickness. This invention employs a sliding standard reflector and an intelligent calibration algorithm, automatically triggering zero-point calibration every 30 minutes to eliminate measurement errors caused by temperature drift and mechanical wear. Simultaneously, the equipment adopts a modular design, allowing the sensing unit, computing module, and communication interface to be independently disassembled and replaced, shortening maintenance time, adapting to the modification needs of different coating machine models, and reducing production line upgrade costs. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 This invention provides a structural framework diagram of a testing system for new energy battery production.

[0017] Figure 2 The structural framework diagram of the coating machine provided by the present invention.

[0018] Figure 3 This is a schematic diagram of the installation of the dual-channel spectral confocal displacement sensor provided by the present invention.

[0019] Among them, 1-coating machine servo system, 2-coating machine doctor blade, 3-dual-channel spectral confocal displacement sensor. Detailed Implementation

[0020] 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.

[0021] This invention discloses a testing system for new energy battery production, such as... Figure 1 As shown, it includes: The module includes a sensor acquisition module, an edge computing module, a self-calibration and compensation module, and a communication control module. The sensing and acquisition module uses a dual-channel spectral confocal displacement sensor to synchronously acquire displacement data of the upper and lower surfaces of the electrode. The edge computing module is used to calculate the electrode thickness data and identify anomalies based on the displacement data of the upper and lower surfaces of the electrode. The self-calibration compensation module is used to automatically calibrate the zero point and eliminate environmental drift; The communication control module has a built-in EtherCAT master controller, which is used to connect the sensor acquisition module, edge computing module, and self-calibration compensation module to the EtherCAT master controller for closed-loop adjustment.

[0022] Furthermore, such as Figure 2 As shown, it also includes: a coating machine, the coating machine including a coating machine servo system and a coating machine doctor blade, the coating machine servo system being signal-connected to the coating machine doctor blade, and the EtherCAT master station controller being signal-connected to the coating machine servo system.

[0023] Furthermore, the dual-channel spectral confocal displacement sensor is arranged laterally at the exit of the coating machine's doctor blade, achieving a measurement repeatability accuracy of ±0.3μm and a response frequency of not less than 5kHz, meeting the testing requirements of high-speed production lines with speeds exceeding 80m / min.

[0024] Furthermore, the edge computing module includes: using a CNN-LSTM hybrid neural network model to identify three types of abnormal patterns: abrupt changes in thickness, gradual slopes, and periodic fluctuations, and generating corresponding alarm signals.

[0025] Furthermore, the self-calibration compensation module includes a miniature standard reflector and a temperature and humidity sensor, which automatically triggers zero-point calibration every 30 minutes, collects environmental data through the temperature and humidity sensor, and performs real-time temperature drift compensation on the collected results.

[0026] Furthermore, the communication control module has a built-in EtherCAT master controller, which is used to send an adjustment command to the coating machine servo system within 100ms when the local thickness deviation of the electrode is detected to exceed ±2μm, so as to adjust the angle of the coating machine blade.

[0027] Furthermore, the sensor acquisition module adopts a linear array multi-point layout, with a horizontal scanning range covering the full width of the electrode sheet, acquiring no less than 1000 data points per second to generate a thermal map of the electrode sheet thickness.

[0028] Furthermore, the edge computing module has a built-in data storage unit for storing no less than 72 hours of detection data, and supports export via USB interface.

[0029] Furthermore, it also includes: support for integration with MES systems, with the communication control module uploading detection data to the cloud-based quality traceability platform via industrial Ethernet, enabling data visualization of the entire production process.

[0030] Furthermore, it also includes an OTA remote upgrade module, which is used to update the edge computing module and self-calibration compensation module via wireless network without the need for on-site disassembly and maintenance.

[0031] In a specific embodiment, a scratch defect identification technology approach can also be used: Leveraging the micron-level spot scanning capability of a spectral confocal displacement sensor, high-density point cloud data is collected from the electrode surface. Using a pre-stored wavelength-distance calibration mapping, a complete three-dimensional elevation profile of the surface is reconstructed. The system automatically compares recessed areas below the reference plane within the profile with a preset scratch depth threshold library. Combined with the stray light suppression characteristics of the confocal optical path, it accurately distinguishes between real scratches and ordinary reflections, capable of identifying minute scratches with depths down to the sub-micron level.

[0032] The technical approach for identifying bulge defects is as follows: In three-dimensional elevation data, the raised areas above the electrode reference surface are extracted. Combined with the surface texture features of the image recognition module acquired simultaneously, the area, height, and edge gradient of the raised areas are matched in multiple dimensions to eliminate the interference of normal coating thickness fluctuations of the electrode and accurately identify bulge-type bulge defects. At the same time, quantitative parameters such as the specific height and area occupied by the bulge can be output simultaneously.

[0033] Multi-source data collaborative verification mechanism: Real-time fusion of morphological elevation data from spectral confocal imaging, surface texture features from image recognition, and environmental compensation parameters from temperature and humidity sensors. Through a preset fusion algorithm, measurement errors caused by environmental temperature and humidity fluctuations are eliminated, avoiding misjudgment by a single sensor. This enables synchronous and stable identification of microscopic defects such as scratches and bulges, and forms a linkage verification with thickness detection data to ensure the reliability of the detection results.

[0034] In a specific embodiment, the design of this invention is based on a real-time feedback module using edge computing, with a built-in CNN-LSTM hybrid neural network model. This module can dynamically match production conditions based on detection data. When the detected electrode thickness deviation exceeds a threshold, the system automatically generates adjustment commands, directly controlling the coating machine's scraper angle and feed speed via the EtherCAT bus, forming a closed-loop control system of detection-analysis-adjustment. Specifically: 1. Establishment of a graded thickness deviation threshold system: Based on the nominal electrode thickness in the lithium battery coating industry, a three-level progressive deviation threshold is set: the first-level warning threshold is ±1.5μm, the second-level adjustment threshold is ±3μm, and the third-level emergency stop threshold is ±5μm. All thresholds are dynamically calibrated using a CNN-LSTM hybrid neural network model combined with real-time collected environmental temperature and humidity and slurry viscosity fluctuation data to avoid false triggering of fixed thresholds when operating conditions change.

[0035] 2. Deviation-Scraper Angle Precise Mapping Mechanism: When the thickness deviation triggers the first-level warning threshold, the scraper angle is adjusted upward by 1.2° for every positive deviation of +1μm (coating is too thick) and downward by 1.2° for every negative deviation of -1μm (coating is too thin), based on the original baseline of 30°, with a single adjustment step not exceeding ±2°; when the deviation enters the second-level adjustment range, the model automatically superimposes the scraper gap compensation parameter, and simultaneously corrects the gap value based on the angle adjustment, to offset the uneven coating problem caused by scraper deflection, and ensure that the coating thickness quickly returns to the nominal range.

[0036] 3. Deviation-Feed Speed ​​Dynamic Matching Rule: Based on the doctor blade angle adjustment, the foil feeding speed of the coating machine is synchronously linked: when the coating thickness exceeds +2μm, the feeding speed is increased by 8%-12% from the current benchmark value to accelerate the slurry spreading rate; when the coating thickness exceeds -2μm, the feeding speed is simultaneously reduced by 8%-12% to prolong the residence time of the slurry in the doctor blade gap and complete the coating thickness. A fluctuation limit of ±1m / min is set throughout the speed adjustment process to avoid sudden speed changes that could cause coating tension loss of control.

[0037] 4. EtherCAT bus closed-loop synchronous control logic: All threshold judgments, angle and speed adjustment commands are sent to the coating machine servo system via the EtherCAT bus with a 3ms synchronization cycle. The entire chain time for deviation data acquisition, model calculation, and command output is controlled within 0.5 seconds. At the same time, a 5-point moving average filtering algorithm is used to filter instantaneous data fluctuations and avoid frequent jitter of the actuator, ultimately achieving dynamic convergence of thickness deviation and controlling the lateral thickness deviation of the electrode sheet within ±3% of the nominal thickness.

[0038] In a specific implementation, a testing system for new energy battery production, such as... Figure 3 As shown, it includes: The module includes a sensor acquisition module, an edge computing module, a self-calibration and compensation module, and a communication control module. The sensing module employs a dual-channel spectral confocal displacement sensor to simultaneously acquire displacement data from the upper and lower surfaces of the electrode. The module uses a linear array multi-point layout, with a lateral scanning range covering the full width of the electrode, acquiring no less than 1000 data points per second to generate a thermal map of the electrode thickness. The dual-channel spectral confocal displacement sensor is laterally positioned at the coating machine exit, achieving a measurement repeatability accuracy of ±0.3μm and a response frequency of no less than 5kHz, meeting the testing requirements of high-speed production lines exceeding 80m / min.

[0039] The dual-channel spectral confocal displacement sensor 5, model SCI04020, is suitable for precision measurement in demanding and harsh environments, and performs well in scenarios requiring large working distances, such as image instrument inspection.

[0040] The edge computing module incorporates an AI chip for real-time processing of thickness data and anomaly identification: a CNN-LSTM hybrid neural network model automatically identifies three types of anomaly patterns—abrupt thickness changes, gradual slopes, and periodic fluctuations—and generates corresponding alarm signals. Simultaneously, the edge computing module has a built-in data storage unit capable of storing at least 72 hours of detection data, which can be exported via a USB interface.

[0041] The self-calibration compensation module integrates a miniature standard reflector and a temperature and humidity sensor to automatically calibrate the zero point and eliminate environmental drift. The self-calibration compensation module automatically triggers zero-point calibration every 30 minutes, collects environmental data through the temperature and humidity sensor, and performs real-time temperature drift compensation on the measurement results.

[0042] The communication control module supports EtherCAT bus output for interface with the EtherCAT master controller to achieve closed-loop regulation. The communication control module has a built-in EtherCAT master controller, which can send adjustment commands to the coating machine servo system within 100ms when a local thickness deviation of the electrode exceeds ±2μm, dynamically adjusting the doctor blade angle.

[0043] It supports integration with MES systems, uploading test data to a cloud-based quality traceability platform via industrial Ethernet to achieve full-process data visualization. The instrument also features OTA remote upgrade capabilities, enabling updates to AI models and calibration algorithms via wireless network, eliminating the need for on-site disassembly and maintenance.

[0044] Example 1: Special type for high-speed coating production line

[0045] This embodiment is applicable to high-speed electrode coating production lines with a speed of 80m / min or higher. The specific structure and workflow are as follows: 1. Hardware configuration: Three sets of dual-channel spectral confocal displacement sensors are arranged horizontally at the exit of the coating machine blade, with a sensor spacing of 20cm, covering the full width of the electrode sheet; the edge computing module adopts Huawei Ascend 310 chip, with a built-in lightweight defect recognition model; it is equipped with EtherCAT communication interface, which directly connects to the EtherCAT master controller of the production line.

[0046] 2. Workflow: When the electrode passes through the sensor detection area, the displacement data of the upper and lower surfaces are synchronously transmitted to the edge calculation module. The model calculates the thickness value in real time and compares it with the preset standard. When a local thickness deviation > ±2μm is detected, the system immediately sends an adjustment signal to the servo motor of the coating machine's scraper to finely adjust the scraper angle by 0.1°. If five consecutive detection points show abnormalities, an audible and visual alarm is automatically triggered and the defect location information is recorded.

[0047] Specifically, the formula for calculating the thickness of a dual-channel spectral confocal displacement sensor is as follows: ; in, This is the initial calibration spacing between the top and bottom sensors; This is the real-time distance from the upper surface of the electrode to the probe, measured by the upper confocal spectral sensor. This is the real-time distance from the lower surface of the electrode to the probe, measured by the lower spectral confocal sensor. The equivalent average refractive index is the electrode coating.

[0048] The model calculates the thickness value in real time and compares it with a preset standard. When a local thickness deviation > ±2μm is detected, the system immediately sends an adjustment signal to the servo motor of the coating machine's scraper to finely adjust the scraper angle by 0.1°. If five consecutive detection points show abnormalities, an audible and visual alarm is automatically triggered and the defect location information is recorded.

[0049] 3. Calibration Mechanism: Before the production line starts each day, the miniature standard reflector automatically slides to the sensor detection area, and the system collects the standard reflection signal to complete the zero-point calibration; during production, dynamic calibration is performed every 30 minutes, and environmental errors are compensated by temperature and humidity sensor data to ensure stable detection accuracy throughout the day.

[0050] Before the daily production line starts, the miniature standard reflector automatically slides to the sensor detection area, and the system collects the standard reflection signal to complete zero-point calibration. During production, dynamic calibration is performed every 30 minutes, and the environmental error compensation calculation formula is as follows: ; in, The raw thickness measurement value directly output by the sensor; Temperature drift compensation coefficient preset for the system (unit: μm / ℃); This is the difference between the current ambient temperature and the zero-point calibration reference temperature. This is the difference between the current ambient relative humidity and the zero-point calibration reference humidity.

[0051] By combining temperature and humidity sensor data to compensate for environmental errors, the detection accuracy remains stable throughout the day.

[0052] Example 2: Laboratory R&D Dedicated Type

[0053] This embodiment is designed for small-batch electrode development scenarios in the laboratory, optimizing flexibility and data visualization capabilities: 1. Hardware configuration: It adopts a single set of movable spectral confocal sensors, with a manually adjustable bracket to adapt to experimental electrode sheets of different sizes; the calculation module integrates a 10.1-inch touch screen to display the thickness curve and defect distribution heat map in real time; it supports USB data export and docking with laboratory data analysis software.

[0054] 2. Workflow: Place the electrode sheet on a precision worktable, set the detection parameters via the touchscreen, and the sensor scans the electrode surface at a uniform speed, adjustable between 5-20 mm / s. After inspection, the system automatically generates a report including the average thickness, standard deviation, defect type, and location. R&D personnel can adjust the coating process parameters based on the data.

[0055] 3. Calibration Mechanism: Before each experiment, the calibration program is manually started, and a standard thickness sample is placed in the detection area. The system automatically calibrates the sensor sensitivity. The built-in temperature compensation algorithm corrects the measurement data in real time according to the laboratory ambient temperature to meet the requirements of detection accuracy in R&D scenarios.

[0056] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0057] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A testing system for new energy battery production, characterized in that, include: The module includes a sensor acquisition module, an edge computing module, a self-calibration and compensation module, and a communication control module. The sensing and acquisition module uses a dual-channel spectral confocal displacement sensor to synchronously acquire displacement data of the upper and lower surfaces of the electrode. The edge computing module is used to calculate the electrode thickness data and identify anomalies based on the displacement data of the upper and lower surfaces of the electrode. The self-calibration compensation module is used to automatically calibrate the zero point and eliminate environmental drift; The communication control module has a built-in EtherCAT master controller, which is used to connect the sensor acquisition module, edge computing module, and self-calibration compensation module to the EtherCAT master controller for closed-loop adjustment.

2. The testing system for new energy battery production according to claim 1, characterized in that, Also includes: The coating machine servo system and the coating machine doctor blade are connected by signals, and the EtherCAT master station controller is connected by signals to the coating machine servo system.

3. The testing system for new energy battery production according to claim 2, characterized in that, The dual-channel spectral confocal displacement sensor is arranged laterally at the exit of the coating machine's scraper.

4. The testing system for new energy battery production according to claim 1, characterized in that, The edge computing module includes a processing module that identifies three types of abnormal patterns: sudden thickness changes, gradual slopes, and periodic fluctuations, and generates corresponding alarm signals.

5. The testing system for new energy battery production according to claim 1, characterized in that, The self-calibration compensation module includes a miniature standard reflector and a temperature and humidity sensor, which automatically triggers zero-point calibration every 30 minutes. It collects environmental data through the temperature and humidity sensor and performs real-time temperature drift compensation on the collected results.

6. The testing system for new energy battery production according to claim 1, characterized in that, The communication control module is used to send an adjustment command to the coating machine servo system within 100ms when a local thickness deviation of the electrode is detected to exceed ±2μm, so as to adjust the angle of the coating machine blade.

7. The testing system for new energy battery production according to claim 1, characterized in that, The sensor acquisition module adopts a linear array multi-point layout, with a horizontal scanning range covering the full width of the electrode sheet. It acquires no less than 1,000 data points per second and generates a thermal map of the electrode sheet thickness.

8. The testing system for new energy battery production according to claim 1, characterized in that, The edge computing module has a built-in data storage unit for storing no less than 72 hours of detection data and supports export via USB interface.

9. A testing system for new energy battery production according to claim 1, characterized in that, Also includes: The communication control module uploads the test data to the cloud-based quality traceability platform via industrial Ethernet.

10. A testing system for new energy battery production according to claim 1, characterized in that, It also includes an OTA remote upgrade module, used to update the edge computing module and self-calibration compensation module via wireless network.

Images