Lithium battery health state rapid detection device and method
By combining a conveyor belt and a high-voltage discharge module with sound acquisition and neural network recognition technology, the problems of slow speed, high cost, and safety in existing battery health status detection have been solved, enabling rapid and accurate assessment and efficient sorting of lithium battery health status.
Patent Information
- Application Number
- CN202411192603.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-29
AI Technical Summary
Existing battery health status detection devices and methods suffer from problems such as slow detection speed, high cost, high difficulty, and safety concerns.
A conveyor belt and a constant temperature and soundproof enclosure are used to send lithium batteries to a high-voltage discharge module for high-voltage discharge. The discharge sound is collected by a sound acquisition device, and the health status of the lithium batteries is identified by a neural network. Feature extraction and analysis are performed using spectrograms and pre-emphasis algorithms to build a dataset for training.
It enables rapid and accurate assessment of the health status of lithium batteries, improves testing speed, reduces costs, and enhances safety and testing efficiency. It is suitable for factory testing of new batteries and tiered sorting of used batteries.
Smart Images

Figure CN119044807A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of battery health status (remaining capacity, health, performance status, etc.) detection device / detection method, and specifically to a lithium battery health status rapid detection device and method. Background Art
[0002] Prior art CN115130370A discloses a method for creating a battery health status detection model, a detection method and a detection device thereof. The detection method includes obtaining the discharge sound emitted when the battery is boosted; inputting the discharge sound emitted when the battery is boosted into a trained battery health status detection model to evaluate the health status of the battery to obtain the battery health status information. A battery health status detection device includes: a battery boosting device, which is used to boost the battery to be tested until a high-voltage discharge is generated and a discharge sound is emitted; a sound collection device, which is used to collect the discharge sound emitted when the battery is boosted; an evaluation module, in which the trained battery health status detection model is written into the detection module, which is used to evaluate the health status of the battery according to the discharge sound emitted when the battery is boosted to obtain the battery health status information.
[0003] However, existing battery health status detection devices and methods still have problems such as slow detection speed, high cost, great difficulty, and insecurity. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a device and method for quickly detecting the health status of a lithium battery. The lithium battery is quickly transported to a detection platform by a conveying device, and a neural network method can be used to quickly and accurately identify the high-voltage discharge sound generated by the lithium battery after discharging with the tip electrode, thereby accurately judging the health status of the lithium battery and realizing rapid and accurate sorting of the battery health status. It can quickly and accurately evaluate the health status (remaining capacity, health, performance status, etc.) of the lithium battery, providing effective support for improving the detection speed of lithium batteries, reducing costs, and ensuring quality.
[0005] In order to achieve the above object, the technical solution adopted by the present invention is:
[0006] A lithium battery health status rapid detection device comprises a conveyor belt (1), a lifting platform (2), a constant temperature soundproof box (3), a control panel (4), a left travel switch (5), a battery tray (6), a lithium battery (7), a photoelectric switch (8), a baffle (9), a right travel switch (10), a workbench (11), a high voltage power supply (12), and a health status evaluation machine (13); the device is characterized in that: the conveyor belt and the constant temperature soundproof box are arranged on a frame, the left travel switch and the right travel switch are arranged at the left and right ends of the conveyor belt respectively, the lithium battery is placed in the battery tray, the constant temperature soundproof box is installed in the middle of the conveyor belt, and the constant temperature soundproof box is installed in the middle of the conveyor belt. A control panel is arranged on the top of the sound box, a lifting platform driven by four synchronously vertically moving screw rods and slide rails is installed inside the constant temperature sound insulation box, a high-voltage discharge module is arranged at the bottom of the constant temperature sound insulation box, and the high-voltage discharge module includes a tip electrode (34) and a copper foil (33), a sound collection device (31) and an air pipe (32) are arranged at the rear end of the constant temperature sound insulation box, an ultraviolet lamp (34) is arranged at the right end of the constant temperature sound insulation box, a front and rear photoelectric switch is arranged in the middle of the conveyor belt, a baffle is installed at the right end of the conveyor belt, a workbench is arranged on the right side of the conveyor belt, and a high-voltage power supply machine connected to the tip electrode and a health status assessment machine are placed on the workbench.
[0007] Furthermore, the running speed of the conveyor belt is 0.1m / s, the length of the conveyor belt is 0.7m, and the running speed of the lifting platform is 0.06m / s; the lithium battery is placed in the limit hole of the battery tray, and the left stroke switch pressed by the battery tray starts to start, and the bottom of the battery tray presses the left stroke switch; the battery tray and battery bracket used to fix the lithium battery adopt a detachable structure, which is convenient for replacing other types of battery trays, placing different types of batteries or placing multiple batteries at the same time; with the transmission of the conveyor belt, the battery tray transports the lithium battery to the constant temperature soundproof box. The constant temperature soundproof box adopts a three-layer structure of high-purity aluminum material and two layers of sound insulation felt, and is installed above the conveyor belt. The rear side of the conveyor belt is provided with a limit hole for the sound collection device and a small hole for connecting the wires.
[0008] Furthermore, after three seconds, when the battery tray moves into the constant temperature soundproof box, the photoelectric switch detects the lithium battery, stops the conveyor belt, and simultaneously the lifting platform and its shielding door (21) move downward; the high-voltage discharge module uses four synchronous vertically moving screw rods and slide rails to drive the lifting platform to move up and down in the constant temperature soundproof box, driving the tip electrode, copper foil, ultraviolet lamp, and air pipe to move up and down, and the downward movement time is 2.5 seconds. After the distance between the tip electrode and the battery electrode is small enough, until the tip electrode is parallel to the positive electrode of the lithium battery, and then the high-voltage power supply is turned on for high-voltage discharge; the diameter of the discharge tip electrode is about 10 mm, the cone angle is about 30°, the tip electrode is connected to the 0-30KV DC high-voltage power supply through a wire, the tip electrode is fixed to the bottom of the lifting platform, the copper foil is pressed against the copper foil sheet by a non-metallic pressure block and fixed to the bottom of the lifting platform, the air pipe and the ultraviolet lamp are arranged on the right side of the lifting platform, and are all fixed by threads.
[0009] Furthermore, the control line of the ultraviolet lamp (35) is led out from the rear side of the constant temperature soundproof box, and the intensity of ultraviolet light is adjusted to irradiate the surface of the lithium battery, and short-wave ultraviolet light is used to irradiate the positive electrode of the battery to accelerate the escape of electrons inside the battery.
[0010] Furthermore, in the constant temperature soundproof box, an appropriate limiting hole is opened on the rear side of the constant temperature soundproof box to place the sound collection device horizontally. The sound collection device quickly collects the discharge sound of the lithium battery with high sensitivity. After one minute of collection, the frequency response range of the sound collection device is 50-20kHz and the maximum sampling frequency is 48000Hz, which effectively restores the digital signal in transmission to the original signal with high fidelity. The stored audio format is .wav format audio, and the PCM encoding format used also effectively guarantees the original characteristics of the sound signal.
[0011] Furthermore, after the sound collection is completed, the lifting platform rises, and the rising time is 2.5 seconds. Then, when the battery tray moves through the conveyor belt for four seconds to reach the baffle (9), the conveyor belt at the bottom of the baffle is provided with a right stroke switch, and the baffle is fixedly connected to the partition in the conveyor belt. When the battery tray hits the baffle, the lithium battery is separated to complete the unloading of the lithium battery, and then the battery tray is driven by the conveyor belt to be sent back to its original position.
[0012] A method for detecting a lithium battery health status rapid detection device, comprising:
[0013] The health status assessment module adopts a health status assessment machine. The health status assessment module uses the feature extraction + method of the spectrogram as the identification feature vector based on the time-frequency analysis of the discharge sound, so as to prepare for the subsequent neural network deployment; when the discharge sound frequency is very high, the sound transmission in the air medium will cause serious energy loss of the high-frequency sound part. Sound pre-emphasis is a signal processing method that compensates for the high-frequency component of the sound signal, thereby improving the high-frequency resolution and output signal-to-noise ratio of the sound, making the sound characteristics more obvious. The energy of the discharge sound signal is mostly concentrated in the high-frequency band. Therefore, the sound spectrogram generated by the high-voltage discharge of lithium batteries 7 in different health states is extracted, and the discharge sound is pre-emphasized. According to the pre-emphasis algorithm and FFT sound signal analysis, the sound is analyzed in the time-frequency domain;
[0014] A variety of different types of lithium batteries were used to make samples for high-voltage discharge to obtain different data models. The samples included cylindrical lithium battery samples, soft-pack lithium battery samples, square-shell lithium battery samples and button lithium battery samples. Cylindrical lithium batteries include 18650 lithium batteries, 26680 lithium batteries, and 32700 lithium batteries. Square-shell lithium batteries include steel-shell lithium batteries and plastic-shell lithium batteries.
[0015] Furthermore, the health status assessment module establishes a spectrogram dataset. The spectrogram scale selects a frame length of 25ms and a frequency range of 0-25kHz. In order to better combine with the subsequent neural network model, the grayscale image is mapped to an RGB three-channel color spectrogram, and the discharge sound spectrograms generated by high-voltage discharge of lithium batteries in different health states are drawn. The spectrogram batch generation algorithm is used to construct a spectrogram dataset.
[0016] The health status assessment module extracts the features of the collected high-voltage discharge sound signal and feeds it into the neural network for training. The specific process is as follows: 1) Collect the discharge sound generated by the high-voltage discharge of the lithium battery, and collect a 60s sound signal file each time; 2) Cut the lithium battery discharge sound and perform time-frequency analysis, and use the spectrogram features of the discharge sound of the lithium battery with different health levels as a feature extraction method to achieve rapid conversion of sound signal-spectrogram and then feed it into the neural network model for recognition. Through the spectrogram batch generation algorithm, according to the frame length of 25ms and the frame shift of 8.5ms, the .wav format sound wave sequence is converted one by one into a spectrogram image to construct a spectrogram data set; 3) Build a neural network, feed the spectrogram data set into the neural network for training, and estimate the health status of the lithium battery. By continuously increasing the spectrogram data set and training time, the model parameters are optimized to improve the accuracy of the lithium battery health status assessment.
[0017] Furthermore, the steps for establishing a high-voltage discharge sound database of lithium batteries in different health states are as follows: 1) selecting lithium batteries of the same brand, type and batch in different health states, and dividing the lithium batteries into an experimental group and a test group; 2) performing a health state detection on the selected lithium batteries through a battery capacity classification test system, and marking and classifying the lithium batteries in different health states; 3) using a high-voltage discharge module to perform high-voltage discharge on the classified lithium batteries, and combining a sound collection device to collect the generated high-voltage discharge sounds; 4) accurately extracting lithium battery state change information through a sound processing and recognition circuit; 5) matching lithium batteries in different health states with the collected high-voltage discharge sounds one by one to form a lithium battery high-voltage discharge sound database.
[0018] The invention discloses a rapid detection device and method for the health status of a lithium battery. The lithium battery is transported to a designated position after three seconds via a conveyor belt. A high-voltage power supply is used to generate a high voltage at the tip electrode. A large potential difference is generated by the tip electrode being a certain distance away from the battery electrode. After the electric field strength reaches the breakdown field strength, a high-voltage discharge occurs. The high-voltage discharge sound is collected for one minute. A spectrogram is used in combination with an artificial intelligence neural network to detect and identify the high-voltage discharge sound. The high-voltage discharge sound information generated by the lithium battery in one minute is fully utilized to diagnose the health status of the lithium battery. After three minutes of diagnosis, the remaining service life of the battery is accurately evaluated, which solves the problems of slow speed, high cost, high difficulty, and unsafety of the existing detection methods. This detection device and method can be used for qualified detection of new batteries leaving the factory and for the cascade sorting and utilization of waste batteries. The capacity of lithium batteries can be accurately detected within five minutes, the detection efficiency and safety of battery sorting can be improved, the comprehensive utilization level of lithium battery resources can be improved, and the quality of cascaded lithium battery products can be guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] Figure 1 This is a schematic diagram of the main structure of the lithium battery health status rapid detection device of the present invention;
[0020] Figure 2 This is a schematic diagram of the top view of the structure of the lithium battery health status rapid detection device of the present invention;
[0021] Figure 3 This is a schematic diagram of the main structure of the lithium battery health status rapid detection device of the present invention.
[0022] In the figure: conveyor belt 1, lifting platform 2, constant temperature sound insulation box 3, control panel 4, left travel switch 5, battery tray 6, lithium battery 7, through-beam photoelectric switch 8, baffle 9, right travel switch 10, workbench 11, high-voltage power supply 12, health status assessment machine 13, shielding door 21, sound collection device 31, air pipe 32, copper foil 33, tip electrode 34, ultraviolet lamp 35. DETAILED DESCRIPTION
[0023] In order to make the technical solution and advantages of the present invention clearer, the technical solution of the present invention will be further described in detail in detail with reference to the accompanying drawings. It can be understood that the specific embodiments described here are only partial embodiments of the present invention, which are only used to explain the present invention, not to limit the present invention. It should be noted that, for the convenience of description, only the parts / structures related to the present invention are shown in the accompanying drawings, and other related parts can refer to the general design. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other to obtain new embodiments.
[0024] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, unless otherwise defined, the technical terms or scientific terms used in the description of the present invention should be the common meanings understood by ordinary technicians in the field to which the present invention belongs.
[0025] The present invention will be further described in detail below in conjunction with the accompanying drawings.
[0026] like Figure 1-3 As shown, a lithium battery health status rapid detection device includes a conveyor belt 1, a lifting platform 2, a constant temperature soundproof box 3, a control panel 4, a left travel switch 5, a battery tray 6, a lithium battery 7, a photoelectric switch 8, a baffle 9, a right travel switch 10, a workbench 11, a high-voltage power supply 12, and a health status assessment machine 13; it is characterized in that: the conveyor belt 1 and the constant temperature soundproof box 3 are arranged on a frame, the left travel switch 5 and the right travel switch 10 are respectively arranged at the left and right ends of the conveyor belt 1, the lithium battery 7 is placed in the battery tray 6, the constant temperature soundproof box 3 is installed in the middle of the conveyor belt 1, and the control panel is arranged on the top of the constant temperature soundproof box 3 4. A lifting platform 2 driven by four synchronously vertically moving screw rods and slide rails is installed inside the constant temperature and sound insulation box 3. A high-voltage discharge module is arranged at the bottom of the constant temperature and sound insulation box 3. The high-voltage discharge module includes a tip electrode 34 and a copper foil 33. A sound collection device 31 and an air pipe 32 are arranged at the rear end of the constant temperature and sound insulation box 3. An ultraviolet lamp 35 is arranged at the right end of the constant temperature and sound insulation box 3. A front and rear photoelectric switch 8 is arranged in the middle of the conveyor belt 1. A baffle 9 is installed at the right end of the conveyor belt 1. A workbench 11 is arranged on the right side of the conveyor belt 1. A high-voltage power supply machine 12 connected to the tip electrode 34 and a health status assessment machine 13 are placed on the workbench 11.
[0027] The running speed of the conveyor belt 1 is 0.1 m / s, the length of the conveyor belt 1 is 0.7 m, and the running speed of the lifting platform 2 is 0.06 m / s.
[0028] The lithium battery 7 is placed in the limiting hole of the battery tray 6, and the left travel switch 5 pressed by the battery tray 6 starts to start, and the bottom of the battery tray 6 presses the left travel switch 5. The battery tray 6 and the battery bracket used to fix the lithium battery 7 adopt a detachable structure, which is convenient for replacing other types of battery trays 6, placing different types of batteries or placing multiple batteries at the same time, so that the lithium battery 7 status rapid detection device achieves the purpose of universality and high efficiency.
[0029] As the conveyor belt 1 is driven, the battery tray 6 transports the lithium battery 7 to the constant temperature soundproof box 3. The constant temperature soundproof box 3 adopts a three-layer structure of high-purity aluminum material and two layers of soundproof felt, and is installed above the conveyor belt 1. The rear side of the conveyor belt 1 is provided with a limit hole for the sound collection device 31 and four small holes for connecting wires. It ensures the constant temperature and sound insulation of the constant temperature soundproof box 3, reduces the influence of external noise on the high-voltage discharge sound, reduces the influence of ambient temperature on the state of the lithium battery, and improves the accuracy of the estimation of the health state of the lithium battery 7.
[0030] After three seconds, when the battery tray 6 moves into the constant temperature soundproof box 3, the photoelectric switch 8 detects the lithium battery 7, stops the conveyor belt 1, and moves the lifting platform 2 and its shielding door 21 downward; the high-voltage discharge module uses four synchronous vertically moving screw rods and slide rails to drive the lifting platform 2 to move up and down in the constant temperature soundproof box 3, driving the tip electrode 34, copper foil 33, ultraviolet lamp 35, and air pipe 32 to move up and down. After the downward movement time is 2.5 seconds, the distance between the tip electrode and the battery electrode is small enough until the tip electrode 34 is parallel to the positive electrode of the lithium battery, and then the high-voltage power supply 12 is turned on for high-voltage discharge; the diameter of the discharge tip electrode 34 is 10mm, the cone angle is 30°, the tip electrode 34 is connected to the 0-30KV DC high-voltage power supply 12 through a wire, the tip electrode 34 is fixed at the bottom of the lifting platform 2, the copper foil 33 is pressed against the copper foil by a non-metallic pressure block and fixed to the bottom of the lifting platform 2, the air pipe 32 and the ultraviolet lamp 35 are arranged on the right side of the lifting platform 2, and are both fixed by threads.
[0031] The control line of the ultraviolet lamp 35 is led out from the rear side of the constant temperature soundproof box 3, and the intensity of the ultraviolet light is adjusted to irradiate the surface of the lithium battery 7. Short-wave ultraviolet light is used to irradiate the positive electrode of the battery (the positive electrode material is aluminum, and ultraviolet light with a wavelength lower than 296nm is required) to accelerate the escape of electrons inside the battery, and study the influence of the photoelectric effect on the high-voltage discharge sound of the lithium battery, as well as the changes in the discharge sound of lithium batteries in different health states under the same ultraviolet light irradiation.
[0032] In the constant temperature soundproof box 3, an appropriate limiting hole is opened on the rear side of the constant temperature soundproof box 3 to place the sound collection device 31 horizontally. The sound collection device 31 quickly collects the discharge sound of the lithium battery 7 with high sensitivity. After one minute of collection, the frequency response range of the sound collection device 31 is 50-20kHz and the maximum sampling frequency is 48000Hz, which effectively restores the digital signal in transmission to the original signal with high fidelity. The audio format stored is .wav format audio, and the PCM encoding format used also effectively guarantees the original characteristics of the sound signal.
[0033] After the sound collection is completed, the lifting platform 2 rises, and the rising time is 2.5 seconds. Then, when the battery tray 6 moves through the conveyor belt 1 for four seconds to reach the baffle 9, the conveyor belt 1 at the bottom of the baffle 9 is provided with a right travel switch 10, and the baffle 9 is fixedly connected to the partition in the conveyor belt 1. When the battery tray 6 hits the baffle 9, the lithium battery 7 is separated to complete the unloading of the lithium battery 7, and then the battery tray 6 is driven by the conveyor belt 1 to be sent back to its original position.
[0034] A detection method for a lithium battery health status rapid detection device, wherein a health status assessment module adopts a health status assessment machine 13, and the health status assessment module uses a spectrogram as a feature extraction + method as an identification feature vector based on a time-frequency analysis of the discharge sound, so as to prepare for the subsequent neural network deployment; when the discharge sound frequency is very high, the sound transmission in the air medium will cause serious energy loss in the high-frequency sound part, and sound pre-emphasis is a signal processing method for compensating for the high-frequency component of the sound signal, thereby improving the high-frequency resolution and output signal-to-noise ratio of the sound, making the sound characteristics more obvious, and the energy of the discharge sound signal is mostly concentrated in the high-frequency band, so the sound spectrogram generated by the high-voltage discharge of lithium batteries 7 in different health states is extracted, and the discharge sound is pre-emphasized, and the sound is analyzed in the time-frequency domain according to the pre-emphasis algorithm and FFT sound signal analysis.
[0035] Furthermore, a variety of different types of lithium batteries were used to make samples for high-voltage discharge to obtain different data models. The samples included cylindrical lithium battery samples, soft-pack lithium battery samples, square-shell lithium battery samples and button lithium battery samples. Among them, cylindrical lithium batteries include 18650 lithium batteries, 26680 lithium batteries, and 32700 lithium batteries. Square-shell lithium batteries include steel-shell lithium batteries and plastic-shell lithium batteries.
[0036] Furthermore, the health status assessment module establishes a spectrogram dataset. The spectrogram scale selects a frame length of 25ms and a frequency range of 0-25kHz. In order to better combine with the subsequent neural network model, the grayscale image is mapped into an RGB three-channel color spectrogram, and the discharge sound spectrograms generated by high-voltage discharge of lithium batteries in different health states are drawn. The spectrogram dataset is constructed through the spectrogram batch generation algorithm.
[0037] Furthermore, the health status assessment module extracts features from the collected high-voltage discharge sound signal and feeds it into the neural network for training. The specific process is as follows: 1) Collect the discharge sound generated by the high-voltage discharge of the lithium battery, and collect a 60s sound signal file each time; 2) Perform time-frequency analysis on the lithium battery discharge sound after cropping, and use the spectrogram features of the discharge sound of the lithium battery with different health levels as a feature extraction method to achieve rapid conversion of the sound signal-spectrogram and then feed it into the neural network model for recognition. Through the spectrogram batch generation algorithm, according to the frame length of 25ms and the frame shift of 8.5ms, the .wav format sound wave sequence is converted one by one into a spectrogram image to construct a spectrogram data set; 3) Build a neural network, feed the spectrogram data set into the neural network for training, and estimate the health status of the lithium battery. By continuously increasing the spectrogram data set and training time, the model parameters are optimized to improve the accuracy of the lithium battery health status assessment.
[0038] Furthermore, the steps for establishing a high-voltage discharge sound database of lithium batteries in different health states are as follows: 1) selecting lithium batteries of the same brand, type and batch in different health states (remaining capacity, health, performance status, etc.), and dividing the lithium batteries into an experimental group and a test group; 2) performing a health status test on the selected lithium batteries through a battery capacity classification test system, and marking and classifying lithium batteries in different health states; 3) performing high-voltage discharge on the classified lithium batteries using a high-voltage discharge module, and collecting the generated high-voltage discharge sounds in combination with a sound collection device; 4) accurately extracting lithium battery state change information through a sound processing and recognition circuit; 5) matching lithium batteries in different health states with the collected high-voltage discharge sounds one by one to form a lithium battery high-voltage discharge sound database.
[0039] The invention discloses a rapid detection device and method for the health status of a lithium battery. The lithium battery is transported to a designated position after three seconds via a conveyor belt. A high-voltage power supply is used to generate a high voltage at the tip electrode. A large potential difference is generated by the tip electrode being a certain distance away from the battery electrode. After the electric field strength reaches the breakdown field strength, a high-voltage discharge occurs. The high-voltage discharge sound is collected for one minute. A spectrogram is used in combination with an artificial intelligence neural network to detect and identify the high-voltage discharge sound. The high-voltage discharge sound information generated by the lithium battery in one minute is fully utilized to diagnose the health status of the lithium battery. After three minutes of diagnosis, the remaining service life of the battery is accurately evaluated, which solves the problems of slow speed, high cost, high difficulty, and unsafety of the existing detection methods. This detection device and method can be used for qualified detection of new batteries leaving the factory and for the cascade sorting and utilization of waste batteries. The capacity of lithium batteries can be accurately detected within five minutes, the detection efficiency and safety of battery sorting can be improved, the comprehensive utilization level of lithium battery resources can be improved, and the quality of cascaded lithium battery products can be guaranteed.
[0040] The above-mentioned implementation modes are for explanation of the present invention rather than limitation of the present invention. It can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A lithium battery health status rapid detection device, comprising a conveyor belt (1), a lifting platform (2), a constant temperature soundproof box (3), a control panel (4), a left travel switch (5), a battery tray (6), a lithium battery (7), a photoelectric switch (8), a baffle (9), a right travel switch (10), a workbench (11), a high voltage power supply (12), and a health status assessment machine (13); characterized in that: The conveyor belt and the constant temperature soundproof box are arranged on the frame. The left and right ends of the conveyor belt are respectively provided with a left travel switch and a right travel switch. The lithium battery is placed in the battery tray. The constant temperature soundproof box is installed in the middle of the conveyor belt. The top of the constant temperature soundproof box is provided with a control panel. The interior of the constant temperature soundproof box is provided with a lifting platform driven by four synchronous vertically moving screw rods and slide rails. The bottom of the constant temperature soundproof box is provided with a high-voltage discharge module. The high-voltage discharge module includes a tip electrode (34) and a copper foil (33). The rear end of the constant temperature soundproof box is provided with a sound collection device (31) and an air pipe (32). The right end of the constant temperature soundproof box is provided with an ultraviolet lamp (34). The middle of the conveyor belt is provided with a front and rear photoelectric switch. The right end of the conveyor belt is provided with a baffle. A workbench is provided on the right side of the conveyor belt. A high-voltage power supply machine connected to the tip electrode and a health status assessment machine are placed on the workbench.
2. A lithium battery health status rapid detection device as claimed in claim 1, characterized in that: The running speed of the conveyor belt is 0.1m / s, the length of the conveyor belt is 0.7m, and the running speed of the lifting platform is 0.06m / s; the lithium battery is placed in the limit hole of the battery tray, and the left stroke switch pressed by the battery tray starts to start, and the bottom of the battery tray presses the left stroke switch; the battery tray and battery bracket used to fix the lithium battery adopt a detachable structure, which is convenient for replacing other types of battery trays, placing different types of batteries or placing multiple batteries at the same time; with the transmission of the conveyor belt, the battery tray transports the lithium battery to the constant temperature soundproof box. The constant temperature soundproof box adopts a three-layer structure of high-purity aluminum material and two layers of sound insulation felt, and is installed above the conveyor belt. The rear side of the conveyor belt is provided with a limit hole for the sound collection device and a small hole for connecting the wires.
3. A lithium battery health status rapid detection device as claimed in claim 2, characterized in that: After three seconds, when the battery tray moves into the constant temperature soundproof box, the photoelectric switch detects the lithium battery, stops the conveyor belt, and moves the lifting platform and its shielding door (21) downward. The high-voltage discharge module uses four synchronous vertically moving screw rods and slide rails to drive the lifting platform to move up and down in the constant temperature soundproof box, driving the tip electrode, copper foil, ultraviolet lamp, and air pipe to move up and down. After the downward movement time is 2.5 seconds, the distance between the tip electrode and the battery electrode is small enough until the tip electrode is parallel to the positive electrode of the lithium battery, and then the high-voltage power supply is turned on for high-voltage discharge. The diameter of the discharge tip electrode is about 10 mm, and the cone angle is about 30°. The tip electrode is connected to a 0-30KV DC high-voltage power supply through a wire. The tip electrode is fixed to the bottom of the lifting platform. The copper foil is pressed against the copper foil sheet by a non-metallic pressure block and fixed to the bottom of the lifting platform. The air pipe and ultraviolet lamp are arranged on the right side of the lifting platform and are all fixed by threads.
4. A lithium battery health status rapid detection device as claimed in claim 3, characterized in that: The ultraviolet lamp (35) leads a control line from the rear side of the constant temperature soundproof box, adjusts the intensity of ultraviolet light to irradiate the surface of the lithium battery, and uses short-wave ultraviolet light to irradiate the positive electrode of the battery to accelerate the escape of electrons inside the battery.
5. A lithium battery health status rapid detection device as claimed in claim 4, characterized in that: In the constant temperature soundproof box, an appropriate limiting hole is opened on the rear side of the constant temperature soundproof box to place the sound collection device horizontally. The sound collection device quickly collects the discharge sound of the lithium battery with high sensitivity. After one minute of collection, the frequency response range of the sound collection device is 50-20kHz and the maximum sampling frequency is 48000Hz, which effectively restores the digital signal in transmission to the original signal with high fidelity. The audio format stored is .wav format audio, and the PCM encoding format used also effectively guarantees the original characteristics of the sound signal.
6. A lithium battery health status rapid detection device as claimed in claim 5, characterized in that: After the sound collection is completed, the lifting platform rises, and the rising time is 2.5 seconds. Then, when the battery tray moves through the conveyor belt for four seconds to reach the baffle (9), the conveyor belt at the bottom of the baffle is provided with a right stroke switch, and the baffle is fixedly connected to the partition in the conveyor belt. When the battery tray hits the baffle, the lithium battery is separated to complete the unloading of the lithium battery, and then the battery tray is driven by the conveyor belt to return to its original position.
7. A method for detecting a lithium battery health status rapid detection device, comprising a lithium battery health status rapid detection device as claimed in claim 6, comprising: The health status assessment module adopts a health status assessment machine. The health status assessment module uses the feature extraction + method of the spectrogram as the identification feature vector based on the time-frequency analysis of the discharge sound, so as to prepare for the subsequent neural network deployment; when the discharge sound frequency is very high, the sound transmission in the air medium will cause serious energy loss of the high-frequency sound part. Sound pre-emphasis is a signal processing method that compensates for the high-frequency component of the sound signal, thereby improving the high-frequency resolution and output signal-to-noise ratio of the sound, making the sound characteristics more obvious. The energy of the discharge sound signal is mostly concentrated in the high-frequency band. Therefore, the sound spectrogram generated by the high-voltage discharge of lithium batteries 7 in different health states is extracted, and the discharge sound is pre-emphasized. According to the pre-emphasis algorithm and FFT sound signal analysis, the sound is analyzed in the time-frequency domain; A variety of different types of lithium batteries were used to make samples for high-voltage discharge to obtain different data models. The samples included cylindrical lithium battery samples, soft-pack lithium battery samples, square-shell lithium battery samples and button lithium battery samples. Cylindrical lithium batteries include 18650 lithium batteries, 26680 lithium batteries, and 32700 lithium batteries. Square-shell lithium batteries include steel-shell lithium batteries and plastic-shell lithium batteries.
8. The detection method according to claim 7, characterized in that The health status assessment module establishes a spectrogram dataset. The spectrogram scale selects a frame length of 25ms and a frequency range of 0-25kHz. In order to better combine with the subsequent neural network model, the grayscale image is mapped to an RGB three-channel color spectrogram, and the discharge sound spectrograms generated by high-voltage discharge of lithium batteries in different health states are drawn. The spectrogram batch generation algorithm is used to construct a spectrogram dataset. The health status assessment module extracts the features of the collected high-voltage discharge sound signals and feeds them into the neural network for training. The specific process is as follows: 1) Collect the discharge sound generated by the high-voltage discharge of the lithium battery, and collect 60s of sound signal files each time; 2) After cutting the discharge sound of the lithium battery, perform time-frequency analysis, generate spectrogram features of the discharge sound of the lithium battery with different health levels as a feature extraction method, realize the rapid conversion of sound signal-spectrogram and send it to the neural network model for recognition, and use the spectrogram batch generation algorithm to convert the .wav format sound wave sequence one by one into the spectrogram image according to the scale of 25ms frame length and 8.5ms frame shift to construct a spectrogram data set; 3) Build a neural network, feed the spectrogram data set into the neural network for training, estimate the health status of the lithium battery, and improve the accuracy of the lithium battery health status assessment by continuously increasing the spectrogram data set and training time, optimizing the model parameters; The steps for establishing a high-voltage discharge sound database of lithium batteries in different health states are as follows: 1) select lithium batteries of the same brand, type and batch in different health states, and divide the lithium batteries into an experimental group and a test group; 2) perform health state detection on the selected lithium batteries through a battery capacity classification test system, and mark and classify the lithium batteries in different health states; 3) use a high-voltage discharge module to perform high-voltage discharge on the classified lithium batteries, and combine a sound collection device to collect the generated high-voltage discharge sound; 4) accurately extract lithium battery state change information through a sound processing and recognition circuit; 5) The lithium batteries in different health states are matched one by one with the collected high-voltage discharge sounds to form a lithium battery high-voltage discharge sound database.
Citation Information
Patent Citations
High-voltage electronic air refreshing device
CN102423502A
Lithium battery film thickness and surface defect detection method and device based on conveying belt
CN109827968A
Battery health state detection model creation method, detection method and detection equipment thereof
CN115130370A
Lithium battery state rapid detection method based on high-voltage discharge sound recognition
CN117538755A
RFID detection device and RFID channel machine
CN212041515U
