Search

CN-121988535-A - Classification method and system for gradient utilization of batteries

CN121988535ACN 121988535 ACN121988535 ACN 121988535ACN-121988535-A

Abstract

The embodiment of the invention provides a classification method and a classification system for gradient utilization of batteries, and belongs to the technical field of battery classification. The method comprises the steps of carrying out matrix arrangement on the series battery packs to obtain a first temperature matrix of the series battery packs, sharpening each element in the first temperature matrix according to a convolution kernel matrix to obtain a second temperature matrix, carrying out binarization operation on each element in the second temperature matrix to obtain a third temperature matrix, judging whether the elements in the third temperature matrix are equal to a preset threshold value, eliminating the elements from the first temperature matrix under the condition that the elements in the third temperature matrix are equal to the preset threshold value, and taking the single battery corresponding to the elements as an independent gradient. According to the invention, classification is realized through the temperature test of one-time charge and discharge operation, capacity measurement is not needed, a great amount of time is avoided, electric energy waste is avoided, the temperature difference among the single batteries is highlighted through sharpening convolution, and the influence of the positions of the single batteries in matrix arrangement is eliminated.

Inventors

  • ZENG GUOJIAN
  • JI XIANG
  • ZHUANSUN MINGMING
  • YANG YANHUI

Assignees

  • 安徽锐能科技有限公司

Dates

Publication Date
20260508
Application Date
20251205

Claims (10)

  1. 1. A method of classifying battery gradient utilization, the method comprising: Performing matrix arrangement on the series battery packs to obtain a first temperature matrix of the series battery packs; Sharpening each element in the first temperature matrix according to the convolution kernel matrix to obtain a second temperature matrix; performing binarization operation on each element in the second temperature matrix to obtain a third temperature matrix; judging whether the elements in the third temperature matrix are equal to a preset threshold value or not; under the condition that the elements in the third temperature matrix are equal to a preset threshold value, removing the elements from the first temperature matrix, and taking the single batteries corresponding to the elements as an independent gradient; Under the condition that the elements in the third temperature matrix are not equal to a preset threshold value, reserving the elements from the first temperature matrix; Blurring each element to be removed in the first temperature matrix according to the Gaussian convolution kernel matrix, and supplementing the blurring to the first temperature matrix to obtain a new first temperature matrix; Returning to the step of sharpening each element in the first temperature matrix according to the convolution kernel matrix according to the acquired new first temperature matrix to acquire a second temperature matrix; Under the condition that the preset cycle times are reached, taking the residual monomer batteries corresponding to the elements as an independent gradient; And obtaining all single batteries with independent gradients so as to finish classification of the series battery packs.
  2. 2. The method of classifying as set forth in claim 1, wherein the matrix arrangement of the series battery packs to obtain the first temperature matrix of the series battery packs includes: performing matrix arrangement on the series battery packs; discharging each single battery to the minimum allowable voltage; Charging each single battery to a preset voltage according to a preset charging speed; Acquiring a temperature value of the outer surface of the central point of each single battery; and taking the temperature value of each single battery as an element of a first temperature matrix, and acquiring the first temperature matrix corresponding to the series battery packs arranged in the matrix.
  3. 3. The method according to claim 2, wherein the obtaining a first temperature matrix using the temperature value of each of the unit cells as an element of the first temperature matrix includes: a first temperature matrix is obtained according to equation (1), ,(1) Wherein, the In the case of a first temperature matrix, For the number of columns of series-connected battery packs, The number of columns arranged in series with the battery packs.
  4. 4. The classification method of claim 1, wherein sharpening each element in the first temperature matrix according to a convolution kernel matrix to obtain a second temperature matrix comprises: a convolution kernel matrix is obtained according to equation (2), ,(2) Wherein, the Is a convolution kernel matrix.
  5. 5. The method of classifying as claimed in claim 4, wherein sharpening each element in the first temperature matrix according to a convolution kernel matrix to obtain a second temperature matrix comprises: the value of each element in the second temperature matrix is obtained by the formula (3) according to the convolution kernel matrix, ,(3) Wherein, the In a second temperature matrix The value of the element(s), For the row in which the cells are in the first temperature matrix, For the single battery in the first temperature matrix In at least one of these four cases, the corresponding content A kind of electronic device All 0.
  6. 6. The method of classifying as set forth in claim 5, wherein binarizing each element in the second temperature matrix to obtain a third temperature matrix includes: A binarization threshold value is set according to formula (4), ,(4) Wherein, the Is a value of the binary threshold value, Is the maximum value of the elements in the second temperature matrix, Is the minimum of the elements in the second temperature matrix.
  7. 7. The method of classifying as claimed in claim 6, wherein binarizing each element in the second temperature matrix to obtain a third temperature matrix includes: Binarizing each element in the second temperature matrix according to formula (5), ,(5) Wherein, the In a third temperature matrix The value of the element.
  8. 8. The classification method according to claim 1, wherein blurring each element to be removed in the first temperature matrix according to a gaussian convolution kernel matrix and supplementing the first temperature matrix to obtain a new first temperature matrix includes: A gaussian convolution kernel matrix is obtained according to equation (6), ,(6) Wherein, the Is a gaussian convolution kernel matrix.
  9. 9. The method of classifying as set forth in claim 8, wherein blurring each element to be removed from the first temperature matrix according to a gaussian convolution kernel matrix and supplementing the first temperature matrix to obtain a new first temperature matrix includes: The values of the complementary elements are obtained by equation (7) according to the gaussian convolution kernel matrix, ,(7) Wherein, the To be supplemented with The value of the element(s), For the row in which the cells are in the first temperature matrix, For the single battery in the first temperature matrix In at least one of these four cases, the corresponding content A kind of electronic device All 0.
  10. 10. A battery cascade classification system, characterized in that the classification system comprises a processor for performing the classification method according to any of claims 1 to 9.

Description

Classification method and system for gradient utilization of batteries Technical Field The invention relates to the technical field of battery classification, in particular to a classification method and a classification system for gradient utilization of batteries. Background Battery recycling, echelon utilization, and material recycling constitute an emerging industry. The retired power battery can be utilized in a gradient manner after being recovered, the capacity can be generally used for an energy storage power station in a range of 60% -80%, the capacity can be generally used for household storage in a range of 20% -60%, and raw materials can be recovered when the capacity is lower than 20%, however, the classification method needs to perform capacity measurement for a plurality of times, and besides a large amount of time is spent, waste of electric energy can be generated, so that the classification of the retired battery is still a problem to be solved in the gradient utilization of the battery at present. Disclosure of Invention The embodiment of the invention aims to provide a classification method and a classification system for battery echelon utilization. The method and the system realize classification through the temperature test of one-time charge and discharge operation, do not need capacity measurement, avoid wasting a great deal of time and electric energy, and can eliminate the influence of the temperature of the single batteries in matrix arrangement by sharpening convolution to highlight the temperature difference among the single batteries. In order to achieve the above object, an embodiment of the present invention provides a classification method for battery cascade utilization, the classification method including: Performing matrix arrangement on the series battery packs to obtain a first temperature matrix of the series battery packs; Sharpening each element in the first temperature matrix according to the convolution kernel matrix to obtain a second temperature matrix; performing binarization operation on each element in the second temperature matrix to obtain a third temperature matrix; judging whether the elements in the third temperature matrix are equal to a preset threshold value or not; under the condition that the elements in the third temperature matrix are equal to a preset threshold value, removing the elements from the first temperature matrix, and taking the single batteries corresponding to the elements as an independent gradient; Under the condition that the elements in the third temperature matrix are not equal to a preset threshold value, reserving the elements from the first temperature matrix; Blurring each element to be removed in the first temperature matrix according to the Gaussian convolution kernel matrix, and supplementing the blurring to the first temperature matrix to obtain a new first temperature matrix; Returning to the step of sharpening each element in the first temperature matrix according to the convolution kernel matrix according to the acquired new first temperature matrix to acquire a second temperature matrix; Under the condition that the preset cycle times are reached, taking the residual monomer batteries corresponding to the elements as an independent gradient; And obtaining all single batteries with independent gradients so as to finish classification of the series battery packs. Optionally, the matrix arrangement is performed on the series battery packs, to obtain a first temperature matrix of the series battery packs, including: performing matrix arrangement on the series battery packs; discharging each single battery to the minimum allowable voltage; Charging each single battery to a preset voltage according to a preset charging speed; Acquiring a temperature value of the outer surface of the central point of each single battery; and taking the temperature value of each single battery as an element of a first temperature matrix, and acquiring the first temperature matrix corresponding to the series battery packs arranged in the matrix. Optionally, taking the temperature value of each unit cell as an element of a first temperature matrix, and acquiring the first temperature matrix includes: a first temperature matrix is obtained according to equation (1), ,(1) Wherein, the In the case of a first temperature matrix,For the number of columns of series-connected battery packs,The number of columns arranged in series with the battery packs. Optionally, sharpening each element in the first temperature matrix according to a convolution kernel matrix to obtain a second temperature matrix, including: a convolution kernel matrix is obtained according to equation (2), ,(2) Wherein, the Is a convolution kernel matrix. Optionally, sharpening each element in the first temperature matrix according to a convolution kernel matrix to obtain a second temperature matrix, including: the value of each element in the second temperature matrix is obtained b