CN116559699A - Decommissioned battery detection and classification method, device, electronic equipment, storage medium - Google Patents

Abstract

The invention provides a retired battery detection classification method, a retired battery detection classification device, electronic equipment and a storage medium, and relates to the field of battery measurement, wherein the method comprises the following steps: performing charge and discharge test on the retired battery in a preset time period; acquiring integral information of a plurality of preset acquisition moments of the retired battery in the process of carrying out the charge and discharge test, wherein the integral information comprises a voltage value and a current value of the retired battery at the preset acquisition moments; transmitting preset laser signals to a plurality of preset acquisition points on the retired battery, acquiring feedback laser signals of the preset acquisition points at the acquisition time, and generating a plurality of local information at the acquisition time according to the feedback laser signals; and classifying the retired battery by utilizing the local information and the whole information. The method can obtain local information of the retired battery at a preset acquisition point while obtaining other indexes except the voltage value and the current value, so that the health state of the battery is better revealed.

Description

Decommissioned battery detection and classification method, device, electronic equipment, storage medium

technical field

The invention relates to the field of battery measurement, in particular to a method, device, electronic equipment, and storage medium for detecting and classifying retired batteries.

Background technique

Due to the diversity of the internal design of the battery and the diversity of the previous use scenarios, the internal health status of the battery at the time of decommissioning is quite different. In order to realize the scientific secondary utilization of decommissioned batteries, decommissioned batteries are usually detected and classified. The traditional detection method is to obtain the electrical index or characteristics of the battery by detecting the voltage of the retired battery or performing a charge and discharge test on the retired battery, and classify the retired battery accordingly.

In the prior art, the decommissioned batteries are usually classified by detecting electrical output signals or performing charge and discharge tests. On the one hand, the indicators obtained through charge and discharge tests and electrical output signal detection are relatively simple, usually voltage, current, and calculated impedance. These indicators cannot fully reveal the health status of decommissioned batteries. On the other hand, these indicators only characterize the overall performance of the battery and cannot reveal the health status of different parts of the battery.

Contents of the invention

The following is an overview of the topics described in detail in this article. This summary is not intended to limit the scope of the claims.

In order to solve at least one of the above-mentioned problems, the embodiments of the present application propose a decommissioned battery detection and classification method, device, electronic equipment, and storage medium, which can classify decommissioned batteries according to the overall information of decommissioned batteries and the partial information of different parts.

According to the first aspect of the present application, a decommissioned battery detection and classification method is proposed, including: performing a charge and discharge test on the decommissioned battery within a preset time period; Collect the overall information at the time, the overall information includes the voltage value and current value of the decommissioned battery at the preset collection time; emit a preset laser signal to a plurality of preset collection points on the decommissioned battery, at the Collect feedback laser signals of multiple preset collection points at the collection time, and generate multiple local information at the collection time according to the feedback laser signals; use the local information and the overall information to analyze the decommissioned battery sort.

According to the decommissioned battery detection and classification method of the first aspect of the present application, by setting a plurality of preset collection points on the decommissioned battery, transmitting preset laser signals to these preset collection points, and collecting the laser reflection signals of these preset collection points and Calculating local information can obtain other indicators besides voltage value and current value, and at the same time obtain local information of decommissioned batteries at preset collection points, so it can better reveal the health status of batteries and increase the accuracy of classification.

In some embodiments, the generating a plurality of local information at the acquisition time according to the feedback laser signal includes: obtaining laser attribute information according to the feedback laser signal, and the laser attribute information includes: phase information, wavelength information , at least one of amplitude information; obtaining the local information from the laser property information according to a selection rule.

In some embodiments, the generating and classifying the decommissioned batteries by using the local information and the overall information includes: using a preset algorithm to convert a plurality of the local information into multiple information according to the overall information. a first parameter; the first parameter characterizes the temperature state of the decommissioned battery at a corresponding preset collection point; when the maximum value among the multiple first parameters is less than a first threshold, classify the decommissioned battery is the first category; when the maximum value of the plurality of first parameters is greater than or equal to the first threshold and less than or equal to the second threshold, the decommissioned battery is classified into the second category; when a plurality of the first When the first parameter, the maximum value of the parameters, is greater than the second threshold, the decommissioned battery is classified into the third category.

In some embodiments, the generating and classifying the decommissioned batteries by using the local information and the overall information includes: using a preset algorithm to convert a plurality of the local information into multiple information according to the overall information. a first parameter; the first parameter characterizes the temperature state of the decommissioned battery at a corresponding preset collection point; when the maximum value among the multiple first parameters and the minimum value among the multiple first parameters When the difference between is greater than or equal to the third threshold, the decommissioned battery is classified into the fourth category.

In some embodiments, the generating and classifying the decommissioned battery by using the local information and the overall information includes: using a preset algorithm to convert one of the local information and the overall information into a second parameter; The second parameter characterizes the stress change of one of the preset collection points of the decommissioned battery; acquiring the second parameter of at least one of the preset collection points, and when the second parameter is greater than a fourth threshold, The decommissioned batteries are classified into the fifth category.

In some embodiments, the generating and classifying the decommissioned battery by using the local information and the overall information includes: converting the local information into an image to obtain a first sub-image; Stitching a plurality of the first sub-images obtained from the information to obtain the first image; converting the overall information into images to obtain the second image; inputting the first image and the second image into the pre-trained first model, so that the first model outputs a damage index; when the damage index is greater than a preset fifth threshold, classify the decommissioned battery into the sixth category.

In some embodiments, according to the decommissioned battery detection and classification method according to any one of claims 1 to 6, in the charging test of the charge and discharge test, the charging power is greater than the rated charging power of the decommissioned battery.

According to the second aspect of the embodiment of the present application, a decommissioned battery detection and classification device is also proposed, including a charging and discharging module for performing a charging and discharging test on decommissioned batteries within a preset time period; an overall information acquisition module for obtaining all The overall information of the decommissioned battery at multiple preset collection times during the charging and discharging test process, the overall information includes the voltage value and current value of the decommissioned battery at the preset collection time; a local information generation module, It is used to transmit preset laser signals to multiple preset collection points on the decommissioned battery, collect feedback laser signals of multiple preset collection points at the collection time, and generate the A plurality of partial information at a time is collected; a classification module is configured to classify the decommissioned batteries by using the partial information and the overall information.

According to the third aspect of the embodiments of the present application, an electronic device is also provided, including a memory, a processor, a communication bus, a communication interface, and a computer program stored in the memory and operable on the processor. The communication bus uses To realize the connection and communication between the processor and the memory; when the processor executes the computer program, the decommissioned battery detection and classification method as described in any one of the above is realized.

According to the fourth aspect of the embodiments of the present application, a storage medium is also proposed, the storage medium is a readable storage medium, and the readable storage medium stores a computer program, and the computer program is used to make the computer execute: The decommissioned battery detection and classification method described in any one of the above.

It can be understood that the beneficial effects of the above-mentioned second aspect to the fourth aspect compared with the related technology are the same as those of the above-mentioned first aspect compared with the related technology. Please refer to the relevant description in the above-mentioned first aspect. This will not be repeated here.

Additional features and advantages of the application will be set forth in the description which follows, and, in part, will be obvious from the description, or may be learned by practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Description of drawings

FIG. 1 is a schematic diagram of an implementation framework of a method for detecting a decommissioned battery according to an embodiment of the present application.

FIG. 2 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

FIG. 3 is a schematic diagram of a method for setting preset collection points according to an embodiment of the present application.

Fig. 4 is a schematic diagram of a method for setting preset collection points according to another embodiment of the present application.

Fig. 5 is a schematic diagram of a method for setting preset collection points according to another embodiment of the present application.

FIG. 6 is a flow chart of a method for detecting and classifying retired batteries according to an embodiment of the present application.

Fig. 7 is a schematic diagram of a decommissioned battery detection device according to an embodiment of the present application.

Detailed ways

In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. However, it will be apparent to those skilled in the art that embodiments of the present application may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, optical circuits, and methods are omitted so as not to obscure the description of the embodiments of the present application with unnecessary detail.

It should be noted that although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than in the flowchart. The terms "first", "second" and the like in the specification and claims and the above drawings are used to distinguish similar objects, and not necessarily used to describe a specific sequence or sequence.

For the convenience and brevity of expression, some basic optical path adjustment devices are omitted in this embodiment, such as concave lenses, convex lenses, mirrors, optical fiber channels, and optical repeaters. everywhere, and the setting method thereof does not affect the realization of the beneficial effects of the embodiments of the present application. In addition, some known circuits, connection structures, circuit internal structures, processor and chip structures, known sensing devices and light measuring devices are also omitted, and the application does not limit the above structures and does not affect the embodiments of the application Beneficial effects are achieved.

This application can be applied to a sorting device for decommissioned batteries. Referring to FIG. 1, FIG. 1 is a schematic diagram of a battery sorting device 10 according to an embodiment of the present application. The device includes a charge-discharge connector 11, a charge-discharge tester 12, a classification control device 13, Optical fiber sensor 14 and optical sensor control device 15. The charge-discharge connector 11 is connected to the pole or power-taking point of the decommissioned battery 1 , so that the charge-discharge tester 12 can discharge electricity from the decommissioned battery 1 or charge electricity into the decommissioned battery 1 through the charge-discharge connector 11 . The classification control device 13 is connected to the charge and discharge tester 12 so that the classification control device 13 can control the charge and discharge tester to perform a charge and discharge test on the decommissioned battery 1 and obtain the charge and discharge data generated by the decommissioned battery 1 in response to the charge and discharge test. At least one optical fiber sensor 14 controlled by the optical sensor control device 15 is installed on the surface of the decommissioned battery 1 (including the exposed outer surface and/or the inner surface that can be touched after disassembly) at a preset collection point, and can emit laser light The optical sensor controller 15 can obtain preset parameters according to the feedback laser signal from the optical sensor, so that the classification control device 13 connected to the optical sensor controller 15 can receive these preset parameters. There can be one or more preset collection points, and the location and number of preset collection points need to be set according to the different types of decommissioned batteries 1 .

When using the optical fiber sensor 14 to test the stress, the optical fiber sensor 14 can be clamped with a clamp so that it fits on the surface of the decommissioned battery 1, so as to measure the stress signal of the battery during charging and discharging more accurately and stably.

The optical sensing control device 15 may include an optical fiber terminal box and an optical device, pull out at least one optical fiber from the optical fiber terminal box, and connect to the optical fiber sensor 14, and the optical fiber terminal box is used to couple the laser signal emitted by the optical device into Each fiber optic sensor 14 transmits the feedback laser signal returned by the fiber optic sensor 14 to the corresponding optical device (optical detection device) to obtain various parameters. The optical device includes or is externally connected with a laser and various optical measuring devices, and the laser can be one of semiconductor lasers, solid-state lasers (microchip lasers), gas lasers (excimer lasers), and dye lasers. The optical measurement device is used to measure one or more of the mode, wavelength, phase, amplitude, and polarization information of the laser signal and the feedback laser signal, and feed the information back to the classification control device 13 .

The classification control device 13 is any host computer, which can be various types of computers with computing power according to different embodiments and implementation conditions. For example, in some embodiments, the decommissioned battery detection and classification method of the present application needs to be completed through a trained model Classification, based on the requirements for computing power, at this time the classification control device 13 can be one of the consumer-grade computer systems of handheld computers, notebook computers, and ultra-mobile personal computers (ultra-mobile personal computer, UMPC), or it can be an independent A physical server, or a server cluster or a distributed system composed of multiple physical servers. However, in some other embodiments, the classification is implemented by a preset algorithm. From the perspective of cost saving, the classification control device 13 can be an industrial-grade embedded computing device, a smart phone, or an industrial terminal.

After the above-mentioned decommissioned battery sorting device, special sorting and conveying equipment can be set up, and the decommissioned batteries 1 can be sorted and transported to a predetermined position by manipulators, multi-directional conveyor belts, manual sorting, etc. When placing the charging and discharging connector 11 and the optical fiber sensor 14 at the corresponding position of the decommissioned battery 1, a manipulator can be used with a fixture to place it at a predetermined position, and the position identification can be completed by visual identification through a preset model, or through manual setting identification Identification, such as setting a two-dimensional code, a magnetic label or an ArUco mark at a predetermined position, enables the manipulator to identify the position where the charging and discharging connector 11 or the fiber optic sensor 14 needs to be placed by means of various visual or non-visual, contact or non-contact identification methods.

The purpose of this application is to provide a decommissioned battery detection and classification method, device, electronic equipment, and storage medium. While performing electrochemical tests on the battery to obtain the overall information of the decommissioned battery, the optical fiber sensor attached to the battery surface is used to monitor the battery. Real-time monitoring of temperature changes and stress changes to obtain local information such as temperature and stress at preset collection points. Or obtain one or more of the mode, wavelength, phase, amplitude, and polarization information of the feedback laser signal obtained by the fiber optic sensor as local information. Combine the overall information and local information to get the accurate status of the decommissioned battery, and classify the decommissioned battery through the preset classification method or the trained model.

The above-mentioned decommissioned battery inspection and classification method in this application can be applied to the classification of various types of rechargeable batteries, such as winding batteries, laminated batteries, or all-tab batteries, etc. The shape of the battery shell can be square batteries, cylindrical batteries, etc. Batteries, pouch batteries, blade batteries. In some possible embodiments of the present application, different preset collection points are set according to different types of batteries.

Referring to Fig. 3 and Fig. 4, Fig. 3 is a schematic diagram of the preset collection point setting position of the decommissioned battery classification method according to the embodiment of the present application, and Fig. 4 is a kind of preset collection point setting of the decommissioned battery classification method according to another embodiment of the present application Location map. In the cylindrical battery of the embodiment of Fig. 3, preset laser light can be emitted by fiber optic sensors at the first collection point 301, the second collection point 302, and the third collection point 303 arranged on the side of the cylindrical battery. signal and collect the feedback laser signal. In the square battery of the embodiment of Fig. 4, the fourth collection point 401, the fifth collection point 402 of the pole, and the sixth collection point 403, the seventh collection point 404, and the eighth collection point on the side of the square battery can be A fiber optic sensor is installed on 405 to emit preset laser signals to collect feedback laser signals. In the prismatic battery, since the heating point is higher than that of the cylindrical battery, some preset collection points can be set closer to the pole to detect the overall heating degree of the prismatic battery. For example, the ninth collection point 410 can be added. In some embodiments, in order to better measure the stress data of the prismatic battery, the fiber optic sensor installed at the seventh collection point 404 should be tightened with a clamp so that the fiber optic sensor fits the surface of the prismatic battery casing, thereby improving the accuracy of the measured stress data. sensitivity.

Referring to FIG. 5 , FIG. 5 is a schematic diagram of a preset collection point setting position of the decommissioned battery classification method according to the embodiment of the present application. In the embodiment where the decommissioned battery in FIG. 5 is a blade battery, in addition to the tenth collection point 501 and the eleventh collection point 502 set on the pole, the collection points can be set along the arrangement direction of the battery slices inside the blade battery. For example, the twelfth collection point 503 , the thirteenth collection point 504 , the fourteenth collection point 505 , and the fifteenth collection point 506 . In some possible embodiments, in addition to the two collection points set on the pole, each collection point can correspond to a battery slice set in the blade battery housing, so the total number of preset collection points is the battery of the blade battery Add 2 to the number of slices. It should be noted that because the internal conditions of some blade batteries are unknown, the collection points are not set according to the number of battery slices in some embodiments of the blade battery, but are evenly arranged on the casing along the length of the blade battery.

This application can also set different preset collection point positions according to other battery classifications. For example, for square lead-acid batteries, considering the heating characteristics of lead-acid batteries, the number of collection points set on the side of the battery (the surface without poles) should be reduced to simplify the process. .

In some embodiments, the decommissioned battery is a pouch battery, and the sixteenth collection point can be set inside the pouch battery casing, so that the sixteenth collection point can directly collect the information of the battery liquid of the pouch battery.

The following mainly describes the decommissioned battery detection and classification method of the present application.

Referring to FIG. 6 , FIG. 6 is a flowchart of a method for classifying retired batteries according to an embodiment of the present application. As shown in Figure 6, the method includes:

S100: Perform a charge and discharge test on the decommissioned battery within a preset time period;

S200: Obtain the overall information of the decommissioned battery at multiple preset collection moments during the charging and discharging test; the overall information includes the voltage value and current value of the decommissioned battery at the preset collection time;

S300: Transmit preset laser signals to multiple preset collection points on the decommissioned battery, collect feedback laser signals of multiple preset collection points at the collection time, and generate the Collect multiple local information at the moment;

S400: Classify the decommissioned batteries by using the local information and the overall information.

Specifically, in S100 , the charging and discharging test can be performed by connecting the terminal of the decommissioned battery in a pressing manner through a device such as the charging and discharging connector 11 . According to the requirements of different decommissioned batteries, the decommissioned batteries are charged and discharged with preset voltage and current.

In order to improve the test efficiency and test the decommissioned battery better, it is required to increase the charging and discharging speed of the decommissioned battery in the charge and discharge test, and in some embodiments, the charging power is greater than the rated charging power of the decommissioned battery. Specifically, the decommissioned battery may be charged with high power within a preset time period, for example, within 5 minutes, and the overall information of the decommissioned battery may be acquired in S200 described later. In the discharge process, a larger load can also be used or the pole of the decommissioned battery can be soaked in a connected dilute electrolyte to increase the discharge rate. By using a faster charging and discharging speed, obviously, in some embodiments, directly connecting the poles through the charging and discharging tester 12 can also achieve the same purpose. On the one hand, the decommissioned batteries can be subjected to greater pressure, which facilitates more accurate classification of decommissioned batteries in subsequent steps. On the other hand, because non-decommissioned batteries need to consider the service life and battery health, the health test of non-decommissioned batteries cannot generate greater pressure. In the process of sorting retired batteries, because the charging and discharging time is short, short-term high current and high voltage charging and discharging have little effect on the service life and health of retired batteries, and even if they are damaged, they will not cause too much loss. , and the use of high current charging and discharging can more expose the problems of decommissioned batteries.

In S200, specifically, in the charging and discharging test, different charging and discharging strategies will be adopted at different collection times. For example, different charging currents used at different acquisition moments in the charging process. In the charging test of some embodiments, the first charging current is used in the first minute, the charging is stopped in the second minute, and the second charging current is used in the third minute. The first charging current is 1.5 times the rated charging power of the decommissioned battery, and the second charging current is 1.2 times the rated charging power of the decommissioned battery. The purpose is to obtain the overall information of the decommissioned battery under different charging current states and the local information corresponding to the overall information described later.

The overall information refers to the different current values (i ₁ , i ₂ , ..., _in ) and corresponding voltage values (u ₁ , u ₂ , ..., u _n ) The value processed by the preset algorithm. Since different charge and discharge strategies are used in the charge and discharge test, these two sets of current and voltage values can fully reflect the overall state of the retired battery. Therefore, the overall information of the decommissioned battery can be obtained according to the above-mentioned current and voltage values and using a preset processing method. Specifically, the following charge and discharge parameters may be obtained according to the above charge and discharge experiments, and at least two of them may be selected as overall information. That is, internal resistance, discharge rate, charge acceptance capability, average current value, average voltage value, maximum current value, maximum voltage value, and the like.

In some embodiments, the preset algorithm may be to select a plurality of the above-mentioned charging and discharging parameters to calculate a weighted average to generate overall information, and the overall information is used as a dimensionless parameter to evaluate the overall health status of the decommissioned battery.

In some other embodiments, the preset algorithm may be based on the statistics of a certain type of battery, for example, there are multiple retired batteries of the same type, and there is at least one normal battery of the same type as a reference sample. At this time, multiple charging and discharging parameters of multiple decommissioned batteries can be selected to calculate the information entropy. When the information entropy of a decommissioned battery is greater than the preset entropy threshold, it is judged that the overall health status of the decommissioned battery is abnormal. The information entropy may be Shannon entropy or other information entropy based on Shannon entropy, such as multi-scale entropy. Using the entropy algorithm can use more information as an associated quantity to evaluate the overall health status of retired batteries, but it is not applicable in the case of disordered retired battery models.

In some other embodiments, one or more of the above charging and discharging parameters may also be directly selected, and entered into the calculation described later as the overall information.

In S300, for example, the optical sensor control device 15 can be used to generate laser light, and the optical fiber sensor 14 can be used to convert the preset laser signal into a preset laser signal, and the preset laser signal can be sent to multiple preset collection points on the decommissioned battery. The collection point may be any one or more of those in FIG. 3 , FIG. 4 and FIG. 5 . Then the optical fiber sensor 14 receives the feedback laser signal generated by the preset laser signal corresponding to the preset collection point on the decommissioned battery, and transmits the feedback laser signal to the optical sensing control device 15 . The optical sensing control device 15 converts the feedback laser signal into a digital signal and submits it to the classification control device for analysis, so as to generate the local information of the preset collection point. When there are multiple preset collection points, multiple pieces of local information are generated. It is easy to understand that the preset laser signals sent to different preset collection points may be different, and different preset laser signals should be selected according to specific measurement schemes.

It is easy to understand that by feeding back the laser signal, the phase information, wavelength information, and amplitude information of the feedback laser signal generated by the preset collection point of the decommissioned battery can be obtained. In addition, other laser signals such as polarization signal and mode signal can also be obtained. A plurality of local information at the acquisition time can be generated according to the feedback laser signal. For example, the temperature information of the preset collection point can be calculated according to the wavelength information. The amplitude information is used to calculate the temperature information of a preset collection point extending into the decommissioned battery, such as the sixteenth collection point. The accuracy of the temperature information calculated using the wavelength is further improved by the phase information. If there is a fixture at the preset collection point, when the fiber optic sensor is close to the surface of the decommissioned battery, stress information can also be obtained. In some embodiments, the state of health of the decommissioned battery can also be judged by obtaining the change of the refractive index of the electrolyte through the phase signal. That is, the phase information, wavelength information, and amplitude information of the feedback laser signal can reflect the characteristics of the decommissioned battery at the preset collection point. Therefore, laser attribute information can be obtained according to the feedback laser signal, and the laser attribute information includes: at least one of phase information, wavelength information, and amplitude information; local information can be obtained from the laser attribute information according to selection rules. The specific content required by the local information will be selected according to different implementation modes.

Finally, S400 may be executed to classify the decommissioned batteries by using local information and overall information.

Specifically, classification can be accomplished by using a pre-trained first model.

The first model includes at least a sequentially connected convolution group, filter group and scoring group. The convolutional set includes at least one first convolutional layer, the filter set includes at least one first filter, and the scoring set includes at least one first scorer. The filtering weights in the filter bank can be modulated according to different decommissioned battery models. The first scorer also includes a first pooling layer and a second pooling layer connected to the convolution group.

Dividing the overall information and the local information into different vector sets, the overall information includes at least one of the internal resistance, discharge rate, charge receiving capacity, average current value, average voltage value, maximum current value, and maximum voltage value of the decommissioned battery; and The local information of each preset collection point is taken as a vector set, and the local information may include at least one of phase information, wavelength information, amplitude information, polarization signal, and mode signal. A feature of the first model of the present application is to transform the overall information and local information into visual data, and then use the visual data recognition method for recognition. Specifically, different information is combined. When a certain value becomes larger, the grayscale data of its color is increased. For example, when the maximum voltage value is small, it is converted into a light gray color block, and when the value is large Converted to a dark gray patch. Thus, the local information is converted into the first sub-image, and the first image is obtained after splicing a plurality of color blocks. The first sub-image represents an image of a certain preset collection point at a certain collection time, and the first image represents images of all preset collection points at a certain collection time.

Therefore, a "visual" comparison can be performed on the transformed pictures, so the first model of the present application can not be limited to full space learning, computer vision analysis, convolutional neural network, deep neural network, fully connected neural network, recurrent neural network, Vector machine models, etc.

Similarly, the second image can also be obtained by converting the overall information into an image according to the above steps.

Each vector set is sequentially input into the first convolutional layer of the first model, wherein the vector set corresponding to the overall information is output to the first pooling layer after passing through the convolution group, and the vector set corresponding to the local information is passed through convolution The output of the group is output to the second pooling layer. The first pooling layer and the second pooling layer each independently consist of at least one of the following group: max pooling layer, fractional max pooling layer, average pooling layer and L2-norm pooling layer. Then normalize the results of the first pooling layer and the second pooling layer. It should be noted that when normalizing, the results output by the first pooling layer and the results output by the second pooling layer have the same different weights.

The first scorer includes decision tree, multiple additive regression tree algorithm, clustering algorithm, principal component analysis algorithm, nearest neighbor analysis algorithm, linear discriminant analysis algorithm, quadratic discriminant analysis algorithm, support vector machine algorithm, evolutionary method algorithm, projection search tracking algorithm or a collection thereof. And the first scorer includes at least one fully connected layer and one classification logistic regression cost layer.

It is easy to understand that after inputting the overall information and multiple local information into the pre-trained first model, the first model will output different damage indices corresponding to different preset collection points.

Decommissioned batteries can be classified according to the above-mentioned damage index classification thresholds.

In the above-mentioned embodiments, it is mentioned that the first model is used to complete the classification but it is obvious that the visual conversion may not be performed, but the classification of decommissioned batteries can also be completed by using other classification models such as knowledge graphs or cluster analysis.

In some other embodiments, the classification may also be accomplished in the following manner.

S411: Use a preset algorithm to convert multiple local information into multiple first parameters according to the overall information; the first parameter represents the temperature state of the decommissioned battery at the corresponding preset collection point;

S412: When the maximum value among the multiple first parameters is less than the first threshold, classify the decommissioned battery into the first category;

S413: When the maximum value among the plurality of first parameters is greater than or equal to the first threshold and less than or equal to the second threshold, classify the decommissioned battery into the second category;

S414: Classify the decommissioned battery into the third category when the maximum first parameter among the multiple first parameters is greater than the second threshold.

In some embodiments, the following steps are further included.

S421: Use a preset algorithm to convert multiple partial information into multiple first parameters according to the overall information; the first parameter represents the temperature state of the decommissioned battery at the corresponding preset collection point;

S422: When the difference between the maximum value among the multiple first parameters and the minimum value among the multiple first parameters is greater than or equal to the third threshold, classify the decommissioned battery into the fourth category.

In other embodiments, it may further include:

S431: Using a preset algorithm to convert a local information and an overall information into a second parameter; the second parameter represents the stress change of a preset collection point of the decommissioned battery;

S432: Obtain a second parameter of at least one preset collection point, and when the second parameter is greater than a fourth threshold, classify the decommissioned battery into the fifth category.

In an embodiment using a machine learning model for classification, it may also include:

S441: Image the local information to obtain a first sub-image;

S442: Concatenate multiple first sub-images obtained according to the local information to obtain a first image;

S443: Image the overall information to obtain a second image;

S444: Input the first image and the second image into a pre-trained first model, so that the first model outputs a damage index;

S445: When the damage index is greater than the preset fifth threshold, classify the decommissioned battery into the sixth category.

The following describes S400 in the implementation manner of the present application with reference to an embodiment.

[Example 1]

Embodiment 1 can be applied to the decommissioned battery sorting device shown in FIG. 1 . The decommissioned battery is a cylindrical battery as shown in Figure 3. The operator clamps the charging and discharging connectors 11 on the first collection point 301 and the second collection point 302 respectively, and places an optical fiber sensor 14 on the third collection point 303 . The charging and discharging tester 12 collects the open circuit voltage U ₀ of the decommissioned battery, and then performs a charging test within a preset time period t ₀ , and in a specific embodiment, t ₀ is equal to 1 minute. The current rate of the charging test is 1.5C, and C refers to the nominal current of the retired battery. For example, taking the 18650 cylindrical battery with a nominal capacity of 2.5Ah as an example, the current of the charging test is 3.75A. Obtain the voltage value u ₁ and current data i ₁ (i ₁ =3.75A) when charging for 1 minute.

The charging time t ₀ , the open circuit voltage U ₀ , the voltage value u ₁ and the circuit data i ₁ are sent to the classification control device 13 . These data are considered aggregate information.

While performing the charge and discharge test, the optical sensing control device 15 transmits a preset laser signal through the optical fiber sensor 14 on the third collection point 303, and receives the returned feedback laser signal. According to the selection rule of Embodiment 1, the feedback laser The laser attribute information of the signal includes wavelength information and intensity information, and the intensity information can be obtained by calculating the amplitude information, which is processed by the optical sensing control device 15 and converted into digital local information, and then transmitted to the classification control device 13 .

The preset laser signal is a laser with a wavelength in the range of 1520-1580nm, and the fiber sensor 14 may be a Fiber Bragg Gratings sensor (Fiber Bragg Gratings sensors).

As the temperature of the decommissioned battery changes during charging and discharging, the laser attribute information (such as wavelength information and amplitude information) of the feedback laser signal in the FBG sensor changes. In this embodiment, the local information is the temperature of the preset collection point, and the temperature is calculated through the feedback laser signal of the optical fiber sensor corresponding to the preset collection point.

The classification control device 13 processes the obtained overall information and local information according to a preset algorithm, ie formula 1.

Wherein, Y is a state parameter characterizing the state of the decommissioned battery, T=T ₀ +Δλ/k _T , and k _T is an empirical constant, which can be obtained through measurement fitting, or the sensitivity of the optical sensor 14 to temperature. Δλ characterizes the wavelength change. This state parameter can be regarded as the first parameter.

In some embodiments, the temperature T can also be calculated from the light intensity, for example, the temperature can be obtained by calculating the luminous flux of the anti-Stokes Raman scattered light and the luminous flux of the back Stokes Raman scattered light. can also be used in this formula.

If Y is less than 0.5Ω·℃, the retired battery is classified into the first category, if Y is greater than or equal to 0.5Ω·℃ and less than 0.7Ω·℃, the retired battery is classified into the second category, if Y is greater than or equal to 0.7Ω·℃ , the decommissioned batteries are classified into the third category.

[Example 2]

Embodiment 2 can be applied to the decommissioned battery sorting device shown in FIG. 1 . The decommissioned battery is a square battery as shown in Figure 4. Make the manipulator identify the battery type, and set optical fibers at the fourth collection point 401, the fifth collection point 402 on the pole, and the sixth collection point 403, the seventh collection point 404 and the eighth collection point 405 on the side of the square battery sensor 14, and at the same time, the manipulator connects the charging and discharging connector 11 to the fourth collection point 401 and the fifth collection point 402 respectively by pressing. A plurality of fiber optic sensors 14 are respectively placed at the sixth collection point 403 , the seventh collection point 404 and the eighth collection point 405 , and at the same time, fiber optic sensors are also arranged at the fourth collection point 401 and the fifth collection point 402 . Wherein, a clamp may be applied at the sixth collection point 403 to make the optical fiber sensor 14 close to the decommissioned battery. The above method can be completed by a technician manipulating the manipulator, or it can be automatically performed by the manipulator recognizing the point.

The charging and discharging tester 12 first collects the open circuit voltage U ₀ of the decommissioned battery, and then performs a charging test within a preset time period t ₁ , and in a specific embodiment, t ₁ is equal to 30 seconds. The current rate of the charging test is 2C, and C refers to the nominal current of the retired battery. For example, taking a 100Ah square battery as an example, the charging current is 200A. Obtain the voltage value u ₁ and current data i ₁ (i ₁ =200A) when charging for 30 seconds.

While performing the charge and discharge test, the optical sensing control device 15 emits a preset laser signal through the fiber optic sensor 14 on the sixth collection point 403, the seventh collection point 404, and the eighth collection point 405, and receives the returned feedback laser The signal, the feedback laser signal includes wavelength information and intensity information, the intensity information can be obtained by calculating the amplitude information, processed by the optical sensing control device 15 and converted into digital local information, and then transmitted to the classification control device 13 .

The preset laser signal is a laser with a wavelength in the range of 1520-1580nm, and the fiber sensor 14 may be a Fiber Bragg Gratings sensor (Fiber Bragg Gratings sensors).

As the temperature of the decommissioned battery changes during charging and discharging, the feedback laser signal (e.g., wavelength, intensity) in the Fiber Bragg FBG sensor changes. Finally, it is obtained by the classification control device 13 as local information.

The algorithm preset by the classification control device 13 , that is, formula 2, processes the obtained overall information and local information.

Through the above formula 2, the state parameter Y _i is obtained in the same manner as in Embodiment 1. Where T _i is the temperature T ₁ , T ₂ , T ₃ , T 4 , T ₅ at the fourth collection point 401, the fifth collection point 402, the sixth collection point 403, the seventh collection point ₄₀₄ and the eighth collection point 405 . Thus, different Y ₁ , Y ₂ , Y ₃ , Y ₄ , Y ₅ are obtained. Taking the maximum value Y _MAX as the first parameter, it is obtained that when Y _MAX is less than 0.01Ω·℃, the retired battery is classified into the first category, and if Y _MAX is greater than or equal to 0.01Ω·℃ and less than or equal to 0.015Ω·℃, it is retired Batteries are classified into the second category, and into the third category if Y _MbX is greater than or equal to 0.015Ω·°C. If the difference between the highest temperature and the lowest temperature of the five positions T ₁ , T ₂ , T ₃ , T ₄ , and T ₅ is greater than 5°C, the decommissioned battery will be classified into the fourth category, and the sensor position with the highest temperature will be recorded It is the abnormal position of the decommissioned battery. If the stress change F at the sensor 4 position is greater than 100N, the decommissioned battery is classified into the fifth category. In this embodiment, the stress change F can be regarded as the second parameter, and 100N can be regarded as the fourth threshold value. The preset algorithm, that is, the stress change measurement formula is schematically shown as:

Where Δλ represents the amount of wavelength change, Characterizes the amount of phase change.

[Example 3]

Embodiment 3 can be applied to the decommissioned battery sorting device shown in FIG. 1 . The decommissioned battery is a blade battery as shown in Figure 5.

Make the manipulator identify the battery type, and at the tenth collection point 501, the eleventh collection point 502 on the pole, and the twelfth collection point 503, the thirteenth collection point 504, and the fourteenth collection point on the side of the blade battery Fiber optic sensors 14 are installed at 505 and the fifteenth collection point 506 respectively, and at the same time, the manipulator connects the charging and discharging connector 11 to the tenth collection point 501 and the eleventh collection point 502 respectively by pressing. And a plurality of optical fiber sensors 14 are respectively placed in the twelfth collection point 503, the thirteenth collection point 504, the fourteenth collection point 505 and the fifteenth collection point 506, meanwhile, at the tenth collection point 501, the eleventh collection point Point 502 is also provided with a fiber optic sensor. The above method can be completed by a technician manipulating the manipulator, or it can be automatically performed by the manipulator recognizing the point.

The charging and discharging tester 12 first collects the open-circuit voltage U ₀ of the decommissioned battery, and then performs a charging test within a preset time period (t ₂ , t ₃ , t _{4 .} . . ). In a specific embodiment, t ₂ is equal to 60 Second. The current rate of the charging test is 2C, C refers to the nominal current of the retired battery, t ₃ is equal to 30 seconds, the current rate of the charging test is 0C, t ₄ is equal to 60 seconds, and the current rate of the charging test is 3C.

While performing the charge and discharge test, the optical sensing control device 15 passes through the tenth collection point 501, the eleventh collection point 502, the twelfth collection point 503, the thirteenth collection point 504, the fourteenth collection point 505, And the fiber optic sensor 14 at the fifteenth collection point 506 emits a preset laser signal and receives the returned feedback laser signal. According to the selection rules of Embodiment 3, the laser attribute information of the feedback laser signal includes wavelength information, amplitude information, and phase information. , mode information, etc., after being processed by the optical sensing control device 15 and converted into digital local information, it is transmitted to the classification control device 13 .

The classification control device 13 runs the trained convolutional neural network model, and the classification control device 13 first converts the digitized feedback laser signals of different collection points at different times (t ₂ , t ₃ , t ₄ ...) according to the preset color The conversion rules are converted into color blocks, and the color blocks are combined to generate the first image. Then (t ₂ , t ₃ , t ₄ ...) the current value (i ₂ , i ₃ , i ₄ ...) and voltage value (u ₂ , u ₃ , u ₄ ...) at the same moment in accordance with the preset The color conversion rules are converted into color blocks, and the multiple color blocks are combined to generate the second image. The first image and the second image are respectively input into the trained convolutional neural network model to obtain multiple damage indices. Different damage indices characterize the damage at different acquisition moments (t ₂ , t ₃ , t ₄ . . . ).

The maximum value among the multiple loss indices is taken, and when the damage index is greater than the preset fifth threshold, the decommissioned battery is classified into the sixth category. It should be noted that the fifth threshold needs to be determined according to different first models and different samples when training the first model, so the fifth threshold is not a fixed value.

Referring to FIG. 2, the method of the embodiment of the present application can be used in an electronic device 20 as shown in FIG. The computer program running on the processor 22, the communication bus 23 is used to realize the connection and communication between the processor 22 and the memory 21; when the processor 22 executes the computer program, it implements the Any decommissioned battery detection and classification method. The aforementioned communication bus 23 may include not only a data bus, but also a power bus, a control bus, and a status signal bus. However, for the sake of clarity, various buses are marked as communication bus 23 in the figure. The method disclosed in the embodiment of the present application above can be applied to the processor 23, or implemented by the processor 23. The processor 1310 may be an integrated circuit. The chip has signal processing capability. In some implementation processes, some or all steps of the above methods may be implemented by hardware integrated logic circuits in the processor 23 or instructions in the form of software. The processor 23 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, an off-the-shelf programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The processor 23 may implement or execute various methods, steps and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor 23 may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in the embodiments of the present application may be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software module can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory or registers, and other mature storage media in the field. The storage medium is located in the memory 21, for example, the processor 23 can read the information in the memory 21, and complete some or all steps of the above method in combination with its hardware.

Referring to FIG. 7 , the present application also proposes a decommissioned battery detection and classification device 30, including: a charging and discharging module for charging and discharging a decommissioned battery within a preset time period; an overall information acquisition module for obtaining decommissioned batteries in The overall information of multiple preset acquisition moments during the charging and discharging test process, the overall information includes the voltage value and current value of the decommissioned battery at the preset acquisition time; the local information generation module is used to emit preset laser signals to the decommissioned battery Multiple preset collection points, collect the feedback laser signals of multiple preset collection points at the collection time, and generate multiple local information at the collection time according to the feedback laser signals; the classification module is used to use local information and overall information to classify decommissioned batteries sort.

The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program is executed by hardware (such as a processor, etc.), so as to be executed by any device in the embodiment of the present application Part or all of the steps of any one of the methods. The computer-readable storage medium includes volatile or nonvolatile, removable or Non-removable media. Computer-readable storage media include but are not limited to RAM (Random Access Memory, random access memory), ROM (Read-Only Memory, read-only memory), EEPROM (Electrically Erasable Programmable read only memory, electrically erasable programmable read-only memory), Flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory, compact disk read-only memory), digital versatile disk (DVD) or other optical disk storage, magnetic cartridges, tapes, magnetic disk storage or other magnetic storage devices, or can Any other medium used to store desired information and which can be accessed by a computer.

Those skilled in the art should understand that all or some of the steps, systems, and functional modules/units in the methods disclosed above can be implemented as software (implemented by computer program code executable by a computing device), firmware, hardware, and appropriate combinations thereof. In a hardware implementation, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be composed of several physical components. Components cooperate to execute. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit .

It should also be understood that references to "one embodiment" or "some embodiments" described in the description of the embodiments of the present application mean that specific features described in conjunction with the embodiments of the present application are included in one or more embodiments of the embodiments of the present application. , structure or characteristics. Thus, appearances of the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in other embodiments," etc. in various places in this specification are not necessarily All refer to the same embodiment, but mean "one or more but not all embodiments" unless specifically stated otherwise. The terms "including", "comprising", "having" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in the description of the embodiments of the application, unless otherwise clearly defined, set Words such as , installation, and connection should be understood in a broad sense, and those skilled in the art can reasonably determine the specific meanings of the above words in the embodiments of the present application in combination with the specific content of the technical solution.

The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those of ordinary skill in the art, various modifications can be made without departing from the gist of the present invention. kind of change.

Claims (10)

1. A method for detecting and classifying decommissioned batteries, comprising: Conduct charge and discharge tests on decommissioned batteries within a preset period of time; Acquiring the overall information of multiple preset collection moments of the decommissioned battery during the charging and discharging test, the overall information including the voltage value and current value of the decommissioned battery at the preset collection time; transmitting the preset The laser signal is sent to a plurality of preset collection points on the decommissioned battery, and the feedback laser signals of a plurality of the preset collection points are collected at the collection time, and a plurality of feedback laser signals at the collection time are generated according to the feedback laser signal. local information; Classifying the decommissioned batteries by using the local information and the overall information. 2. The decommissioned battery detection and classification method according to claim 1, wherein the generation of a plurality of local information at the acquisition time according to the feedback laser signal comprises: Obtain laser attribute information according to the feedback laser signal, where the laser attribute information includes: at least one of phase information, wavelength information, and amplitude information; The local information is obtained from the laser attribute information according to a selection rule. 3. The method for detecting and classifying decommissioned batteries according to claim 2, wherein said generating and classifying said decommissioned batteries using said partial information and said overall information comprises: using a preset algorithm to convert a plurality of the partial information into a plurality of first parameters respectively according to the overall information; the first parameters characterize the temperature state of the decommissioned battery at a corresponding preset collection point; When the maximum value of the plurality of first parameters is less than a first threshold, classifying the decommissioned battery into the first category; When the maximum value of the plurality of first parameters is greater than or equal to the first threshold and less than or equal to a second threshold, classifying the decommissioned battery into the second category; When the maximum first parameter among the plurality of first parameters is greater than the second threshold, the decommissioned battery is classified into the third category. 4. The method for detecting and classifying decommissioned batteries according to claim 2, wherein said generating and classifying said decommissioned batteries using said partial information and said overall information comprises: using a preset algorithm to convert a plurality of the partial information into a plurality of first parameters respectively according to the overall information; the first parameters characterize the temperature state of the decommissioned battery at a corresponding preset collection point; When the difference between the maximum value among the multiple first parameters and the minimum value among the multiple first parameters is greater than or equal to a third threshold, the decommissioned battery is classified into the fourth category. 5. The method for detecting and classifying decommissioned batteries according to claim 2, wherein said generating and classifying said decommissioned batteries using said local information and said overall information comprises: Using a preset algorithm to convert one of the local information and the overall information into a second parameter; the second parameter represents the stress change of a preset collection point of the decommissioned battery; The second parameter of at least one preset collection point is acquired, and when the second parameter is greater than a fourth threshold, the decommissioned battery is classified into the fifth category. 6. The decommissioned battery detection and classification method according to claim 2, wherein said generating and classifying said decommissioned batteries by using said local information and said overall information comprises: Image the local information to obtain a first sub-image; stitching a plurality of the first sub-images obtained according to the local information to obtain a first image; Image the overall information to obtain a second image; inputting the first image and the second image into a pre-trained first model, so that the first model outputs a damage index; When the damage index is greater than the preset fifth threshold, the decommissioned battery is classified into the sixth category. 7. The decommissioned battery detection and classification method according to any one of claims 1 to 6, characterized in that, in the charging test of the charge and discharge test, the charging power is greater than the rated charging power of the decommissioned battery. 8. A decommissioned battery detection and classification device, characterized in that it includes The charge and discharge module is used to conduct charge and discharge tests on decommissioned batteries within a preset period of time; An overall information acquisition module, configured to acquire overall information of the decommissioned battery at multiple preset collection moments during the charging and discharging test, the overall information including the voltage value of the decommissioned battery at the preset collection time and current value; A local information generation module, configured to emit preset laser signals to multiple preset collection points on the decommissioned battery, collect feedback laser signals from multiple preset collection points at the collection time, and The laser signal generates a plurality of local information at the acquisition moment; A classification module, configured to classify the decommissioned batteries by using the local information and the overall information. 9. An electronic device, comprising a memory, a processor, a communication bus, a communication interface and a computer program stored on the memory and operable on the processor, characterized in that, The communication bus is used to realize connection and communication between the processor and the memory; When the processor executes the computer program, the decommissioned battery detection and classification method according to any one of claims 1 to 7 is realized. 10. A storage medium, the storage medium is a readable storage medium, characterized in that, the readable storage medium stores a computer program, and the computer program is used to make a computer execute: as claimed in any one of claims 1 to 7. A method for detecting and classifying retired batteries.