CN-122016879-A - Battery cell detection method, system, electronic equipment and storage medium

CN122016879ACN 122016879 ACN122016879 ACN 122016879ACN-122016879-A

Abstract

The application relates to the technical field of battery cell detection, and discloses a battery cell detection method, a system, electronic equipment and a storage medium, wherein the battery cell detection method comprises the steps of acquiring a preliminary image of a battery cell under an initial angle; the method comprises the steps of identifying a preliminary defect area in a preliminary image, obtaining defect data corresponding to the preliminary defect area, obtaining a target observation angle according to the defect data, obtaining a target image of a battery cell at the target observation angle, and judging whether the battery cell is qualified or not through the preliminary image and the target image. The embodiment of the application ensures that the quantity and the angles of the detection visual angles corresponding to the battery cells with different defect degrees are different, and the detection precision of the battery cells and the detection of the battery cells are considered.

Inventors

SHEN SENYI
GUAN XUEDAN
WU ZHONGMING

Assignees

远景能源有限公司

Dates

Publication Date: 20260512
Application Date: 20251229

Claims (11)

1. The battery cell detection method is characterized by comprising the following steps of: acquiring a preliminary image of the battery cell at an initial angle; Identifying a preliminary defect area in the preliminary image and acquiring defect data corresponding to the preliminary defect area; acquiring a target observation angle according to the defect data, and acquiring a target image of the battery cell at the target observation angle; judging whether the battery cell is qualified or not through the preliminary image and the target image.
2. The method of claim 1, wherein the obtaining the target observation angle according to the defect data comprises: Constructing a covariance matrix according to the defect data; performing eigenvalue decomposition on the covariance matrix to obtain a defect main direction vector; And obtaining a target observation angle according to the defect main direction vector.
3. The method of claim 2, wherein the obtaining the target observation angle according to the defect principal direction vector comprises: matching the defect main direction vector in a defect model library; If the defect main direction vector is matched with the target observation angle, searching a corresponding first observation angle in the defect model library, and taking the first observation angle as the target observation angle, or searching a corresponding first observation angle in the defect model library, calculating a second observation angle according to the defect main direction vector, and taking the first observation angle and the second observation angle as the target observation angle; and if the two main directions are not matched, calculating a second observation angle according to the main direction vector of the defect, and taking the second observation angle as the target observation angle.
4. The method for detecting a battery cell according to claim 3, characterized in that the method further comprises: when the covariance matrix is subjected to eigenvalue decomposition, a first eigenvalue and a second eigenvalue are also obtained; obtaining elongation according to the first characteristic value and the second characteristic value, and determining a defect type according to the elongation; the calculating to obtain the second observation angle according to the defect main direction vector includes: When the defect type is linear or long-strip defect, taking the normal direction of the main defect direction vector as a first additional observation angle, and acquiring the second observation angle according to the first additional observation angle and the initial angle; And under the condition that the defect type is a bulk defect, a block defect or a diffuse defect, calculating an external rectangle surrounding the minimum area of the preliminary defect image, determining a first direction of a long axis and a second direction of a short axis of the external rectangle, taking the vertical direction of the first direction and the vertical direction of the second direction as two second additional observation angles, and acquiring the two second observation angles according to the second additional observation angles and the initial angle.
5. The method of claim 1, wherein the identifying the preliminary defect region in the preliminary image comprises: Calculating the average gray value and gray standard deviation of any pixel point in the initial image in a neighborhood window; marking the pixel points meeting the preset condition as abnormal points, wherein the preset condition is I (p) -mu local ∣>k×σ local , I (p) is the gray value of the pixel points, mu_local is the average gray value of the pixel points, sigma_local is the gray standard deviation of the pixel points, and k is a sensitivity coefficient; and identifying abnormal point clusters which are continuous in space and the number of which exceeds a preset area threshold value as the preliminary defect area.
6. The method for detecting a battery cell according to claim 5, wherein the sensitivity coefficient is obtained by: Determining a first target coordinate point on a preset first ROC curve according to a target expected false alarm rate and/or an expected detection rate, and determining the sensitivity coefficient according to the first target coordinate point, wherein each coordinate point of the first ROC curve is bound with the corresponding sensitivity coefficient; The acquisition mode of the preset area threshold value is as follows: Determining a second target coordinate point on a preset second ROC curve according to the target expected false alarm rate and/or the expected detection rate, and determining the preset area threshold according to the second target coordinate point, wherein each coordinate point of the second ROC curve is bound with the corresponding preset area threshold.
7. The method of claim 1, wherein the identifying the preliminary defect region in the preliminary image comprises: acquiring an edge map of the initial image; comparing the edge map with a pre-stored standard cell edge structure, identifying an abnormal edge profile in the edge map, and identifying the abnormal edge profile as the preliminary defect region.
8. The method for detecting a battery cell according to claim 1, wherein before the preliminary image of the battery cell at the initial angle is acquired, further comprising: And determining a detection angle sequence according to the model of the battery cell, wherein the detection angle sequence comprises at least one initial angle.
9. The battery cell detection system is characterized by comprising an X-ray source, a battery cell carrier, a multi-axis precise movement mechanism, a flat panel detector and a main controller; the multi-axis precise movement mechanism is used for driving the X-ray source or driving the battery cell carrier, and the main controller is used for realizing the battery cell detection method according to any one of claims 1 to 8.
10. An electronic device comprising at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the cell detection method of any one of claims 1 to 8.
11. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the cell detection method according to any one of claims 1 to 8.

Description

Battery cell detection method, system, electronic equipment and storage medium Technical Field The present application relates to the field of battery cell detection technologies, and in particular, to a battery cell detection method, a system, an electronic device, and a storage medium. Background The conventional main flow scheme mainly adopts a fixed angle detection scheme aiming at the detection of the battery cell at the present stage, namely an X-ray source and a detector are installed at a fixed and optimized angle, and the fixed angle detection scheme is specially used for detecting Overhang of the battery cell, namely that a negative pole piece exceeds the size allowance of a positive pole piece in the length and width directions, but has the problems of single function and limited detection range, and has the problems of low detection rate of non-Overhang defects and missing killing. In order to solve the problems of the fixed angle detection scheme, a multi-station combined detection scheme, an off-line sampling detection scheme and a rotation detection scheme exist at the present stage, wherein the multi-station combined detection scheme is characterized in that a plurality of X-ray detection stations are arranged on a production line, each station adopts different fixed angles and is respectively responsible for detecting different types of defects, the off-line sampling detection scheme adopts manual sampling detection for some special defects in off-line high-precision CT equipment with a multi-angle adjustment function, but 100% on-line full detection cannot be realized, the 100% on-line full detection cannot be used as a supplementary analysis means, batch quality problems in the production process cannot be effectively controlled, the rotation detection scheme is mainly simple and improved in mechanical structure, the actual detection effect is poor, and the operation flexibility is limited by a part of schemes, so that the off-line sampling detection scheme is not suitable for large-scale or special-shaped battery cores. Therefore, the current cell detection scheme cannot give consideration to both detection accuracy and detection efficiency. Disclosure of Invention The embodiment of the application aims to provide a battery cell detection method, a system, electronic equipment and a storage medium, so that the battery cell detection precision and the detection efficiency are both considered. In order to solve the technical problems, the embodiment of the application provides a battery cell detection method, which comprises the steps of obtaining a preliminary image of a battery cell under an initial angle, identifying a preliminary defect area in the preliminary image and obtaining defect data corresponding to the preliminary defect area, obtaining a target observation angle according to the defect data, obtaining a target image of the battery cell at the target observation angle, and judging whether the battery cell is qualified or not through the preliminary image and the target image. The embodiment of the application also provides a battery cell detection system which comprises an X-ray source, a battery cell carrier, a multi-axis precise movement mechanism, a flat panel detector and a main controller, wherein the multi-axis precise movement mechanism is used for driving the X-ray source or the battery cell carrier, and the main controller is used for realizing the battery cell detection method. The embodiment of the application also provides electronic equipment, which comprises at least one processor and a memory in communication connection with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the cell detection method. The embodiment of the application also provides a computer readable storage medium which stores a computer program, and the computer program realizes the battery cell detection method when being executed by a processor. In some embodiments, the obtaining the target observation angle according to the defect data comprises constructing a covariance matrix according to the defect data, performing eigenvalue decomposition on the covariance matrix to obtain a defect main direction vector, and obtaining the target observation angle according to the defect main direction vector. In some embodiments, the obtaining the target observation angle according to the defect main direction vector includes matching the defect main direction vector in a defect model library, searching a corresponding first observation angle in the defect model library to serve as the target observation angle if the defect main direction vector is matched, or searching a corresponding first observation angle in the defect model library to serve as the target observation angle, calculating to obtain a second observation angle according