An artificial intelligence recycling method for full-scene

Abstract

This application relates to the field of battery recycling technology, providing an AI-powered recycling method for all scenarios. The method is applied to a battery recycling system, which includes a physical dismantling station and a chemical reaction station in sequence. The method includes the following steps: acquiring battery information of multiple batteries to be recycled and the remaining processing time of the chemical reaction station; for each battery to be recycled, inputting the battery information into a preset dismantling time prediction model to obtain the predicted dismantling time; selecting the battery whose preset dismantling time is less than the current remaining processing time, and whose difference between the current remaining processing time and the preset dismantling time is the smallest, as the target battery to be recycled; and making the target battery the next dismantling object at the physical dismantling station. This method can improve battery recycling efficiency.

Description

Technical Field

[0001] This application relates to the field of battery recycling technology, and in particular to an artificial intelligence recycling method applicable to all scenarios. Background Technology

[0002] In current mainstream power battery recycling bases, the operating mode is usually a linear assembly line designed based on "deterministic input." In pursuit of economies of scale, factories have built rigid automated dismantling lines and hydrometallurgical extraction tanks with fixed formulas for specific single vehicle models. However, with the explosive growth of the new energy industry, the actual retirement wave has not arrived according to the pre-set single specification, but has instead exhibited extremely fragmented characteristics. Faced with a large number of mixed retired batteries, including unclassified retired ride-hailing vehicle battery packs, pouch batteries of unknown origin from electric two-wheelers, and even bulk cells transported from consumer electronics dismantling stations.

[0003] How to efficiently process these non-standard, heterogeneous batteries has become an urgent problem to be solved. Summary of the Invention

[0004] This application provides an AI-powered recycling method applicable to all scenarios, which can improve the recycling efficiency of batteries.

[0005] To achieve the above objectives, this application adopts the following technical solution: In a first aspect, this application discloses an AI-based recycling method for all scenarios. This method is applied to a battery recycling system, which includes a physical dismantling station and a chemical reaction station in sequence according to the process. The method includes the following steps: obtaining battery information of multiple batteries to be recycled and the current remaining processing time of the chemical reaction station; for each battery to be recycled, inputting the battery information of the battery to be recycled into a preset dismantling time prediction model to obtain the predicted dismantling time of the battery to be recycled; the preset dismantling time prediction model is determined based on the historical dismantling time of the physical dismantling station for dismantling historical batteries to be recycled; selecting the battery to be recycled whose preset dismantling time is less than the current remaining processing time and whose difference between the current remaining processing time and the preset dismantling time is the smallest as the target battery to be recycled; the target battery to be recycled is the next dismantling object of the physical dismantling station.

[0006] Furthermore, after identifying the target battery to be recycled, the method further includes: acquiring the load characteristic parameters of the physical dismantling station in real time when performing dismantling operations on the target battery to be recycled at the physical dismantling station, and determining whether the physical dismantling station is in an abnormal dismantling state based on the load characteristic parameters; adjusting the dismantling mode of the physical dismantling station when it is in an abnormal dismantling state, and determining the expected remaining dismantling time; acquiring the current reaction state parameters and preset safety threshold parameters of the chemical reaction station, and determining the tolerable waiting time of the chemical reaction station based on the current reaction state parameters and preset safety threshold parameters; and controlling the chemical reaction station to reduce the reaction rate when the expected remaining dismantling time is greater than or equal to the tolerable waiting time.

[0007] More specifically, in some preferred embodiments, the physical disassembly station is an ultrasonic cutting component, and the load characteristic parameters include the acoustic impedance value of the ultrasonic cutting component. Determining whether the physical disassembly station is in an abnormal disassembly state based on the load characteristic parameters includes: calculating the rate of change of the acoustic impedance value; when the rate of change of the acoustic impedance value exhibits a non-linear jump and exceeds a preset rate of change threshold, it is determined that the physical disassembly station is in an abnormal disassembly state; otherwise, it is determined that the physical disassembly station is not in an abnormal disassembly state.

[0008] Based on the above, this application further proposes that the disassembly mode includes a conventional cutting mode and a micro-disturbance grinding mode, adjusting the disassembly mode of the physical disassembly station, and determining the estimated remaining disassembly time, including: sending a first control command to the physical disassembly station to switch the physical disassembly station from the conventional cutting mode to the micro-disturbance grinding mode; obtaining the unit volume processing time parameter of the physical disassembly station under the micro-disturbance grinding mode; and using the product of the remaining disassembly volume of the target battery to be recycled and the unit volume processing time parameter as the estimated remaining disassembly time.

[0009] In some preferred embodiments, the current reaction state parameters include the current redox potential value, the current reaction temperature, and the reactant concentration. The preset safety threshold parameters include the redox potential critical threshold. The tolerable waiting time of the chemical reaction station is determined based on the current reaction state parameters and the preset safety threshold parameters, including: determining the current chemical reaction rate constant based on the current reaction temperature and reactant concentration; determining the rise time required for the current redox potential value to rise to the redox potential critical threshold based on the chemical reaction rate constant; and determining the rise time as the tolerable waiting time.

[0010] As a technological improvement, the chemical reaction station includes a cooling jacket for temperature regulation and a main stirring motor for driving the stirring components. Controlling the chemical reaction station to reduce the reaction rate includes: sending a second control command to the chemical reaction station; the second control command instructing the chemical reaction station to increase the flow rate of the cooling medium entering the cooling jacket by a preset increase amount; and sending a third control command to the chemical reaction station; the third control command instructing the chemical reaction station to reduce the driving frequency of the main stirring motor by a preset decrease amount.

[0011] Based on the above, after controlling the chemical reaction station to reduce the reaction rate, the method further includes: redetermining the tolerable waiting time according to the reduced reaction rate of the chemical reaction station; when the expected remaining dismantling time is greater than or equal to the redetermined tolerable waiting time, and the difference between the expected remaining dismantling time and the redetermined tolerable waiting time is less than a preset time threshold, adding a preset dose of reducing agent to the chemical reaction station; the reducing agent is used to reduce the products of the chemical reaction in the chemical reaction station; when the expected remaining dismantling time is greater than or equal to the redetermined tolerable waiting time, and the difference between the expected remaining dismantling time and the redetermined tolerable waiting time is greater than or equal to the preset time threshold, controlling the physical dismantling station to stop dismantling the target battery to be recycled, and redetermining the target battery to be recycled among multiple batteries to be recycled.

[0012] As a further improvement, before acquiring the battery information of multiple batteries to be recycled, the method further includes: acquiring the terminal voltage of each battery to be recycled and a first correspondence; the first correspondence includes a one-to-one correspondence between multiple terminal voltage ranges and multiple discharge load resistance values; taking the discharge load resistance value corresponding to the terminal voltage range of the battery to be recycled in the first correspondence as the target discharge load resistance value of the battery to be recycled, and discharging the battery to be recycled based on the target discharge load resistance value; during the discharge process of the battery to be recycled, monitoring the temperature change rate of the terminal of the battery to be recycled in real time, and when the temperature change rate exceeds a preset temperature change rate threshold, disconnecting the load of the battery to be recycled and triggering a preset cooling mechanism.

[0013] In one embodiment, before acquiring the terminal voltage of each of the plurality of batteries to be recycled, the method further includes: acquiring the electrochemical impedance spectroscopy (EIS), ultrasonic flaw detection data, and health index of each of the plurality of initial batteries including the plurality of batteries to be recycled; the health index being the ratio of the current capacity to the designed capacity of the initial battery; for each of the plurality of initial batteries, determining the battery feature vector of the initial battery based on the EIS and ultrasonic flaw detection data; determining the similarity between the battery feature vector of the initial battery and the battery feature vector of the reusable battery; designating the initial battery with a similarity greater than a preset similarity threshold and a health index greater than a preset health index as a usable battery; and designating the initial battery with a similarity less than or equal to a preset similarity threshold, or a health index less than or equal to a preset health index, as the battery to be recycled.

[0014] In one implementation, triggering a preset cooling mechanism includes: forcibly cooling the battery to be recycled with air; and spraying coolant onto the battery to be recycled.

[0015] Secondly, this application also discloses a battery recycling system for all scenarios. The battery recycling system includes a physical dismantling station and a chemical reaction station in sequence. The system also includes an acquisition device and a processing device. The acquisition device is used to acquire battery information of multiple batteries to be recycled and the current remaining processing time of the chemical reaction station. The processing device is used to input the battery information of each battery to be recycled into a preset dismantling time prediction model to obtain the predicted dismantling time for that battery. The preset dismantling time prediction model is determined based on the historical dismantling times of the physical dismantling station. The processing device is used to select the battery to be recycled whose preset dismantling time is less than the current remaining processing time, and whose difference between the current remaining processing time and the preset dismantling time is the smallest, as the target battery to be recycled. This target battery to be recycled becomes the next dismantling object of the physical dismantling station.

[0016] Beneficial Effects: This application discloses an AI-powered recycling method applicable to all scenarios. By acquiring battery information of the batteries to be recycled and the remaining processing time of the chemical reaction station, and using a preset dismantling time prediction model to predict the dismantling time of each battery, it intelligently selects the batteries to be recycled whose preset dismantling time is less than the current remaining processing time and has the smallest difference as the target dismantling object. This method effectively solves the problem that existing recycling systems cannot handle non-standard and heterogeneous battery inputs, avoids automated robot downtime due to inability to match preset templates, and avoids value dissipation caused by the lack of rapid health grading capabilities. By dynamically optimizing the dismantling sequence, this application can significantly improve recycling efficiency, reduce operational risks, and ensure smooth connection of subsequent chemical reaction processes, thereby overcoming the shortcomings of existing technologies such as high safety risks, large resource losses, and low chemical recycling efficiency. It achieves adaptive processing of mixed retired batteries and has significant and superior technical effects. Attached Figure Description

[0017] Figure 1 A flowchart illustrating an AI-powered recycling method applicable to all scenarios, as provided in this application; Figure 2 A flowchart illustrating an AI-powered recycling method applicable to all scenarios, as provided in this application; Figure 3 A flowchart illustrating an AI-powered recycling method applicable to all scenarios, as provided in this application; Figure 4 This application provides an architectural diagram of an AI-powered recycling system for all scenarios. Detailed Implementation

[0018] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0019] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0020] Traditional battery recycling systems face challenges when dealing with non-standard, heterogeneous, and mixed-chemical batteries with unknown residual values, including functional failures, value loss, safety risks, and reduced chemical recycling efficiency. For example, visual recognition systems frequently report errors due to their inability to match preset templates, causing automated robots to stop; production lines lacking rapid health grading capabilities are forced to use crude crushing methods, directly pulverizing high-value-added modules into low-value powder; and the inability to identify highly active damaged batteries leads to flash explosions caused by electrolyte evaporation within the crushing chamber; the mixing of different chemical systems causes a surge in the amount of impurity removal agents used in subsequent chemical extraction processes, ultimately resulting in metal salts with purity below battery-grade standards, trapping the entire production line in a vicious cycle of high energy consumption, low output, and high risk.

[0021] In this regard, such as Figure 1 As shown, this application proposes an AI-based recycling method for all scenarios. This method is applied to a battery recycling system, which includes a physical dismantling station and a chemical reaction station in sequence. The method includes the following steps: S101. Obtain battery information for multiple batteries to be recycled and the current remaining processing time of the chemical reaction station.

[0022] S102. For each of the multiple batteries to be recycled, input the battery information of the battery to be recycled into the preset dismantling time prediction model to obtain the predicted dismantling time of the battery to be recycled.

[0023] The preset dismantling time prediction model is determined based on the historical dismantling time of historical batteries to be recycled at the physical dismantling station.

[0024] S103. The battery to be recycled that has the smallest difference between the preset dismantling time and the current remaining processing time and the preset dismantling time is selected as the target battery to be recycled; the target battery to be recycled is the next dismantling object of the physical dismantling station.

[0025] This application introduces an intelligent decision-making mechanism to achieve adaptive processing of different types of batteries to be recycled, effectively solving many problems existing in the prior art and significantly improving the safety, efficiency and economy of battery recycling.

[0026] To better understand the technical solution proposed in this application, some key terms and implementation environments involved will be explained first.

[0027] A "battery recycling system" is a comprehensive system integrating multiple stations such as physical dismantling and chemical reactions, designed for the efficient and safe recycling of used batteries. This system can handle various types and specifications of retired batteries and adjust its processing strategy according to the specific condition of each battery.

[0028] The "physical dismantling station" is an important component of the battery recycling system, primarily responsible for the mechanical dismantling of battery packs or modules to separate different components such as cells, casings, and connectors. This station can employ various dismantling technologies, such as mechanical cutting, ultrasonic dismantling, and laser cutting.

[0029] The "chemical reaction station" is another crucial link in the battery recycling system. It is mainly responsible for chemically processing the physically disassembled battery cells or active materials to extract valuable metal elements. This station typically involves processes such as hydrometallurgy and pyrometallurgy, and requires precise control of reaction conditions to ensure recycling efficiency and product purity.

[0030] "Battery information for batteries to be recycled" can include battery type (such as lithium-ion batteries, nickel-metal hydride batteries, etc.), chemical system (such as lithium iron phosphate, ternary materials, etc.), size, weight, health status (such as remaining capacity, internal resistance, etc.), historical usage data, etc. This information can be obtained by scanning the QR code or RFID tag on the battery, or through the battery management system (BMS) interface.

[0031] "Remaining processing time at the chemical reaction station" refers to the remaining time required for the chemical reaction station to complete the current processing work in the current batch processing task. This time can be calculated by real-time monitoring of the operating status, processing progress, and preset process parameters of the chemical reaction station.

[0032] The "Preset Dismantling Time Prediction Model" is a prediction model trained on historical data, used to predict the dismantling time of a specific battery to be recycled at the physical dismantling station. This model can employ machine learning algorithms, such as regression analysis and neural networks, to learn the relationship between battery information and actual dismantling time in historical dismantling data.

[0033] The core of the AI-powered recycling method for all scenarios proposed in this application lies in optimizing the battery recycling process through an intelligent decision-making mechanism, thereby improving overall efficiency and safety.

[0034] First, this method requires obtaining battery information for multiple batteries to be recycled and the remaining processing time at the chemical reaction station. For example, a scanning device can be installed at the entrance of the battery recycling system to automatically identify the unique identifier of each battery entering the system and retrieve its corresponding battery information from a database, including battery type, chemical system, size, and health status. Simultaneously, by interacting with the control system of the chemical reaction station, the remaining time for the batch tasks currently being processed at that station can be obtained in real time.

[0035] Secondly, for each of the multiple batteries to be recycled, the battery information is input into a pre-defined dismantling time prediction model to obtain the predicted dismantling time. For example, a machine learning model trained on historical dismantling data can be used. This model takes information such as battery type, size, and health status as input and outputs the estimated dismantling time for that battery at the physical dismantling station. This model learns and establishes a mapping relationship between battery characteristics and dismantling time by analyzing the actual time spent dismantling historical batteries at the physical dismantling station.

[0036] Finally, the battery to be recycled that has a preset dismantling time shorter than the current remaining processing time, and has the smallest difference between the current remaining processing time and the preset dismantling time, is selected as the target battery for recycling. This target battery is then designated as the next battery to be recycled at the physical dismantling station. For example, the system iterates through the predicted dismantling time of all batteries to be recycled and compares it with the current remaining processing time at the chemical reaction station. Provided that the predicted dismantling time is shorter than the current remaining processing time, the battery with the smallest difference between the two is selected. This selection strategy aims to maximize the collaborative efficiency between the physical dismantling station and the chemical reaction station, avoiding resource waste caused by station waiting.

[0037] The AI-powered recycling method proposed in this application, applicable to all scenarios, effectively solves many problems existing in traditional battery recycling systems through intelligent scheduling strategies. In traditional systems, the lack of identification and processing capabilities for different types of batteries, as well as the lack of effective coordination mechanisms between workstations, leads to low recycling efficiency and high safety risks. For example, when a large influx of non-standard batteries occurs, the system cannot adjust its dismantling strategy according to battery characteristics, potentially resulting in low dismantling efficiency or even equipment damage. Simultaneously, the lack of coordination between physical dismantling workstations and chemical reaction workstations may cause one workstation to stagnate due to waiting, resulting in resource waste.

[0038] This application introduces a pre-defined dismantling time prediction model, which accurately predicts the dismantling time of the batteries to be recycled at the physical dismantling station based on the battery information. This allows the system to plan ahead and avoid blind dismantling. More importantly, by comparing the predicted dismantling time with the current remaining processing time at the chemical reaction station and selecting the battery with the smallest difference as the next to be dismantled, intelligent collaboration between the physical dismantling station and the chemical reaction station is achieved. This collaborative mechanism ensures that the physical dismantling station can continuously provide suitable materials to the chemical reaction station, avoiding the chemical reaction station from being stalled due to insufficient materials or experiencing processing pressure due to material accumulation.

[0039] Compared to existing technologies, the core innovation of this application lies in its forward-looking scheduling and collaborative optimization capabilities. Traditional systems are often linear and rigid, unable to adapt to the varied and complex input of retired batteries. This application, by introducing predictive models and intelligent matching mechanisms, enables the battery recycling system to dynamically adjust its processing strategy based on real-time data, achieving "flexible" production. For example, when faced with a mixed stream of retired batteries, this application can predict the dismantling time of each battery based on its characteristics and, combined with the processing capacity of downstream chemical reaction stations, intelligently select the optimal dismantling sequence. This optimization not only improves overall recycling efficiency and reduces energy consumption, but more importantly, by avoiding unnecessary downtime and waiting, it significantly enhances the system's operational stability and reduces safety risks caused by improper handling. Therefore, this application represents a significant advancement in improving the economic, environmental, and social benefits of battery recycling.

[0040] like Figure 2 As shown, this application proposes an AI-powered recycling method applicable to all scenarios, which, after identifying the target battery to be recycled, further includes: S201. When performing dismantling operations on the target battery to be recycled at the physical dismantling station, the load characteristic parameters of the physical dismantling station are obtained in real time, and the physical dismantling station is determined to be in an abnormal dismantling state based on the load characteristic parameters.

[0041] S202. When the physical disassembly station is in an abnormal disassembly state, adjust the disassembly mode of the physical disassembly station and determine the estimated remaining disassembly time.

[0042] S203. Obtain the current reaction status parameters and preset safety threshold parameters of the chemical reaction station, and determine the tolerable waiting time of the chemical reaction station based on the current reaction status parameters and preset safety threshold parameters.

[0043] S204. When the expected remaining dismantling time is greater than or equal to the tolerable waiting time, control the chemical reaction station to reduce the reaction rate.

[0044] Specifically, real-time acquisition of load characteristic parameters at the physical dismantling station refers to continuously monitoring key operational indicators of the physical dismantling station during the dismantling of target batteries for recycling using sensors or other monitoring equipment. These load characteristic parameters reflect the actual working status and stress conditions of the physical dismantling station. Determining whether the physical dismantling station is in an abnormal dismantling state based on these load characteristic parameters aims to promptly detect abnormal situations such as jamming, overload, efficiency reduction, or equipment failure that may occur during the dismantling process.

[0045] When a physical disassembly station is in an abnormal disassembly state, adjusting the disassembly mode of that station can be understood as switching the operating strategy or work method of the physical disassembly station according to the specific circumstances of the abnormality, in order to adapt to the current disassembly difficulty or resolve the abnormal problem. For example, it can switch from a conventional, efficient disassembly mode to a more refined, gentler, or more fault-tolerant mode. At the same time, it is necessary to reassess and determine the estimated remaining disassembly time required to complete the remaining disassembly work under the new disassembly mode, in order to provide a basis for subsequent decision-making.

[0046] In practical applications, obtaining the current reaction status parameters and preset safety threshold parameters of the chemical reaction station refers to continuously monitoring various key process parameters within the chemical reaction station, such as temperature, pressure, and concentration, and combining this with pre-set safety operating limits or critical values. Based on these parameters, the tolerable waiting time for the chemical reaction station is determined. The purpose is to assess the maximum time that the chemical reaction station can safely and stably wait for the physical dismantling station to complete its work under the current conditions, avoiding reaction runaway or product quality degradation due to excessive waiting time.

[0047] Furthermore, when the remaining dismantling time is expected to be greater than or equal to the tolerable waiting time, the reaction rate of the chemical reaction station is reduced. The purpose is to slow down the reaction process of the chemical reaction station by actively intervening in its operation, thereby extending its safe waiting time to match the dismantling delay caused by abnormal conditions at the physical dismantling station. This avoids adverse effects on the chemical reaction station due to delays at the physical dismantling station, ensuring the coordination and safety of the entire recycling process.

[0048] This application's solution effectively addresses the problem of unforeseen anomalies during dismantling by introducing a real-time monitoring and feedback mechanism for the physical dismantling station. Specifically, when the physical dismantling station is dismantling the target battery to be recycled, the system acquires its load characteristic parameters in real time. If these parameters indicate an abnormal dismantling state, such as increased dismantling difficulty or equipment malfunction, the system responds immediately. First, it adjusts the dismantling mode of the physical dismantling station, for example, switching to a more flexible mode adapted to abnormal situations, to attempt to resolve or alleviate the current dismantling difficulties, and recalculates the estimated remaining dismantling time required to complete the remaining dismantling. Simultaneously, to ensure the continuity and safety of the entire recycling process, the system synchronously acquires the current reaction status parameters and preset safety threshold parameters of the chemical reaction station, and accurately assesses the maximum safe waiting time the chemical reaction station can wait under current conditions, i.e., its tolerable waiting time. By comparing the estimated remaining dismantling time with the tolerable waiting time, if the estimated remaining dismantling time exceeds the safe waiting limit of the chemical reaction station, the system will proactively control the chemical reaction station to reduce its reaction rate. This collaborative control mechanism allows the physical dismantling station more time to handle anomalies, while avoiding potential reaction runaway or safety hazards that may occur in the chemical reaction station due to prolonged waiting, thus ensuring the smooth operation and safety of the entire battery recycling process.

[0049] Through the above technical solutions, this application can effectively address various abnormal situations that may occur at the physical dismantling station during actual operation, significantly improving the robustness and safety of the battery recycling process. Real-time monitoring of load characteristic parameters and timely detection of dismantling anomalies enable the system to respond quickly and adjust dismantling strategies, avoiding prolonged shutdowns or equipment damage caused by abnormal situations. More importantly, by dynamically assessing the tolerable waiting time at the chemical reaction station and proactively controlling the reaction rate at the chemical reaction station based on the anticipated delays at the physical dismantling station, the negative impacts of delays at the physical dismantling station on the chemical reaction station, such as reactant degradation, temperature runaway, or safety risks, are effectively avoided. This not only ensures the stable operation of the chemical reaction station but also optimizes the synergy of the entire recycling process, thereby improving the overall efficiency and safety of battery recycling and reducing potential operational risks.

[0050] As shown in Figure 3, this application further proposes that the aforementioned physical disassembly station is an ultrasonic cutting assembly, and the aforementioned load characteristic parameters include the acoustic impedance value of the aforementioned ultrasonic cutting assembly. Determining whether the aforementioned physical disassembly station is in an abnormal disassembly state based on the aforementioned load characteristic parameters includes: S301. Calculate the rate of change of acoustic impedance value.

[0051] S302. When the rate of change of the above acoustic impedance value exhibits a nonlinear jump and exceeds the preset rate of change threshold, it is determined that the above physical disassembly station is in an abnormal disassembly state; otherwise, it is determined that the above physical disassembly station is not in an abnormal disassembly state.

[0052] Specifically, the physical disassembly station is set up with an ultrasonic cutting assembly. An ultrasonic cutting assembly is a device that uses high-frequency ultrasonic vibrations to cut materials. Its characteristics include high cutting precision and a small heat-affected zone, making it suitable for the fine disassembly of complex materials such as batteries. During ultrasonic cutting, the interaction between the cutting head and the material to be cut affects the propagation characteristics of the ultrasonic waves, which is reflected in the acoustic impedance value of the ultrasonic cutting assembly. The load characteristic parameter is specifically defined as the acoustic impedance value of the ultrasonic cutting assembly. Acoustic impedance is the resistance encountered by sound waves when propagating in a medium; it reflects the density and speed of sound of the medium. During ultrasonic cutting operations, the acoustic impedance value changes when the cutting head comes into contact with the battery material, cuts it, or encounters abnormal conditions (such as sudden changes in material hardness, cutting head wear, jamming, etc.).

[0053] In practical applications, the operating status of the physical disassembly station is monitored by calculating the rate of change of acoustic impedance. The rate of change of acoustic impedance refers to the speed at which the acoustic impedance changes over time. During normal cutting, the rate of change of acoustic impedance usually fluctuates within a relatively stable range. When abnormal situations occur, such as sudden obstruction of the cutting head or drastic changes in the internal structure of the material, the acoustic impedance value will experience rapid and drastic nonlinear jumps. A nonlinear jump refers to a significant, nonlinear, and drastic fluctuation in the rate of change of acoustic impedance that exceeds the normal fluctuation range within a short period of time. The preset rate of change threshold is a critical value determined based on historical data and experimental results, used to distinguish between normal fluctuations and abnormal jumps. When the rate of change of acoustic impedance exceeds this preset rate of change threshold, the physical disassembly station is judged to be in an abnormal disassembly state.

[0054] This application's solution concretizes the physical disassembly station into an ultrasonic cutting component, using its acoustic impedance value as a load characteristic parameter, enabling more precise and sensitive capture of microscopic changes during the disassembly process. When the ultrasonic cutting component is operating, its acoustic impedance value directly reflects the coupling state and energy transfer efficiency between the cutting head and the battery material. When the cutting process proceeds smoothly, the rate of change of the acoustic impedance value remains within a certain range. However, once anomalies occur, such as wear of the cutting head leading to decreased cutting efficiency, abnormal hard points in the battery's internal structure causing cutting obstruction, or jamming between the cutting head and the battery, these anomalies will alter the ultrasonic energy transfer path, resulting in significant, non-linear, and rapid jumps in the acoustic impedance value. By monitoring and calculating the rate of change of the acoustic impedance value in real time and comparing it with a preset rate of change threshold, these anomalies can be identified promptly and accurately, thereby avoiding equipment damage or safety accidents caused by failure to detect anomalies in a timely manner.

[0055] Through the above technical solution, this application provides a more accurate and sensitive method for detecting abnormal conditions at physical disassembly stations. Compared to detection methods that rely on generalized load characteristic parameters, using the acoustic impedance value of the ultrasonic cutting component as a load characteristic parameter and monitoring the nonlinear jumps in its rate of change can identify potential abnormalities in the disassembly process earlier and more accurately, such as cutting head wear, material jamming, or internal structural abnormalities. This helps improve the reliability and real-time performance of abnormality detection, reduces the risk of false alarms and missed alarms, and thus effectively ensures the continuity, safety, and efficiency of physical disassembly operations, avoiding equipment damage or production interruptions caused by failure to handle abnormalities in a timely manner.

[0056] This application further proposes the above-mentioned steps for adjusting the disassembly mode of the physical disassembly station and determining the estimated remaining disassembly time, including: The disassembly modes include conventional cutting mode and micro-disturbance grinding mode. The disassembly mode of the physical disassembly station is adjusted, and the estimated remaining disassembly time is determined, including: Send a first control command to the physical dismantling station to switch the physical dismantling station from the conventional cutting mode to the micro-disturbance grinding mode; obtain the unit volume processing time parameter of the physical dismantling station in the micro-disturbance grinding mode; and use the product of the remaining dismantling volume of the target battery to be recycled and the unit volume processing time parameter as the estimated remaining dismantling time.

[0057] Specifically, the disassembly mode can be understood as the operating method or strategy adopted by the physical disassembly station when disassembling the target battery to be recycled. The conventional cutting mode refers to the mode in which the physical disassembly station, under normal operating conditions, uses preset, highly efficient cutting parameters and paths to quickly disassemble the battery, aiming to maximize disassembly efficiency. The micro-disturbance grinding mode is an auxiliary or emergency mode activated when an anomaly occurs at the physical disassembly station. Its characteristics include using a lower cutting speed, a smaller cutting depth, or a finer grinding path to reduce impact on the battery and potential risks. It also allows the system to continue limited disassembly operations under abnormal conditions, aiming to complete the disassembly task as much as possible while ensuring safety or to buy time for subsequent processing.

[0058] In practical applications, the first control command is a signal or command issued by the system to the physical disassembly station, instructing the physical disassembly station to switch from the currently executing conventional cutting mode to the micro-disturbance grinding mode. This command can be automatically generated and sent by the central control unit after determining that the physical disassembly station is in an abnormal disassembly state based on real-time monitored load characteristic parameters (such as the rate of change of acoustic impedance).

[0059] The unit volume processing time parameter refers to the time required for a physical dismantling station to process a unit volume of target recyclable battery material under micro-perturbation grinding mode. This parameter can be obtained through experimental calibration, historical data analysis, or real-time monitoring, and its purpose is to provide basic data for calculating the estimated remaining dismantling time. For example, the unit volume processing time of different types and states of batteries under micro-perturbation grinding mode can be tested and recorded in advance to form a parameter library for system query.

[0060] Furthermore, the remaining dismantling volume of the target battery to be recycled refers to the volume of the portion of the target battery to be recycled that has not yet been dismantled when the physical dismantling station switches to the micro-perturbation grinding mode. This volume can be estimated through initial battery information, the volume of the dismantled portion, or measured in real time through visual recognition, 3D scanning, etc. The estimated remaining dismantling time refers to the time required to complete the dismantling of the remaining portion of the target battery to be recycled after the physical dismantling station switches to the micro-perturbation grinding mode. This time is obtained by multiplying the remaining dismantling volume by the unit volume processing time parameter in the micro-perturbation grinding mode. Its purpose is to provide a relatively accurate time estimate so that the system can better coordinate the work between the physical dismantling station and the chemical reaction station.

[0061] This application's solution introduces two dismantling modes: a conventional cutting mode and a micro-disturbance grinding mode. When the physical dismantling station experiences an abnormal dismantling condition, it can flexibly switch from the high-efficiency conventional cutting mode to the safer and more precise micro-disturbance grinding mode. This mode switching allows the physical dismantling station to adopt a more robust strategy in the face of abnormal situations, rather than simply shutting down or responding with a single mode, thus minimizing risk while maintaining the continuity of the dismantling operation as much as possible. Furthermore, by obtaining the unit volume processing time parameter in the micro-disturbance grinding mode and combining it with the remaining dismantling volume of the target battery to be recycled, the estimated remaining dismantling time is accurately calculated, enabling the system to obtain a more reliable time estimate. This accurate estimated time is crucial for subsequent judgment on whether to control the chemical reaction station to reduce the reaction rate, avoiding unnecessary stagnation or overload of the chemical reaction station due to inaccurate time estimation. This effectively solves the problems of lack of flexibility in dismantling mode adjustment and inaccurate estimated dismantling time under abnormal dismantling conditions.

[0062] Through the above technical solution, this application can intelligently switch to a micro-disturbance grinding mode when an anomaly occurs at the physical dismantling station, effectively reducing dismantling risks and improving operational safety, while avoiding production interruptions caused by complete shutdown. Furthermore, by accurately calculating the estimated remaining dismantling time based on the unit volume processing time parameter under the micro-disturbance grinding mode, the system can obtain a more accurate time estimate, thus providing a reliable basis for subsequent coordinated control with the chemical reaction station. This significantly improves the robustness and coordination of the entire battery recycling system, ensuring efficient and safe operation under abnormal conditions.

[0063] In some embodiments of this application, when determining the tolerable waiting time for a chemical reaction station, it is necessary to comprehensively consider multiple reaction state parameters and preset safety threshold parameters. Specifically, the determination of the tolerable waiting time for a chemical reaction station can be carried out in the following manner.

[0064] Current reaction status parameters include the current redox potential value, current reaction temperature, and reactant concentration. Safety threshold parameters include the critical redox potential threshold. Based on the current reaction status parameters and preset safety threshold parameters, the tolerable waiting time for the chemical reaction station is determined, including: Based on the current reaction temperature and reactant concentration, determine the current chemical reaction rate constant; based on the chemical reaction rate constant, determine the rise time required for the current redox potential value to rise to the redox potential critical threshold; determine the rise time as the tolerable waiting time.

[0065] Among these, the current reaction state parameters refer to physical or chemical quantities that are directly related to the reaction process and safety and are monitored or acquired in real time during a chemical reaction at the chemical reaction station. Specifically, the current redox potential reflects the extent and trend of the redox reaction in the reaction system, and its changes can indicate the reaction activity and potential risks; the current reaction temperature is a key factor affecting the chemical reaction rate, directly related to the reaction speed and thermodynamic equilibrium; and the reactant concentration determines the initial conditions of the reaction and the rate of reactant consumption. Accurate acquisition of these parameters is crucial for assessing the reaction state.

[0066] Preset safety threshold parameters refer to critical values ​​that are preset to ensure the safe and stable operation of a chemical reaction station. Specifically, the redox potential critical threshold is a preset redox potential value. When the redox potential of the reaction system reaches or exceeds this threshold, it may mean that the reaction is out of control, side reactions are aggravated, or there are safety hazards, requiring intervention measures.

[0067] When determining the current rate constant of a chemical reaction, chemical kinetic models such as the Arrheniuse equation can be used, combined with experimental data or theoretical calculations, to perform dynamic calculations based on real-time monitoring of the current reaction temperature and reactant concentration. This rate constant reflects the inherent rate of the chemical reaction under the current conditions.

[0068] Furthermore, based on the established chemical reaction rate constant, and combined with the reaction kinetic equations, the time required for the current redox potential value to rise from its current state to a preset redox potential critical threshold can be predicted; this rise time represents the tolerable waiting time for the chemical reaction station to reach a potentially hazardous state under the current reaction conditions.

[0069] This application's solution introduces the current redox potential value, current reaction temperature, and reactant concentration as parameters of the current reaction state, and sets a critical redox potential threshold as a preset safety threshold parameter, enabling a more comprehensive and accurate assessment of the real-time status of the chemical reaction station. By dynamically determining the chemical reaction rate constant based on the current reaction temperature and reactant concentration, and predicting the time required for the redox potential value to reach the critical threshold based on this rate constant, this time is used as the tolerable waiting time. This multi-parameter dynamic prediction method ensures that the determination of the tolerable waiting time is no longer a static preset value, but can be adjusted and optimized in real time according to the actual reaction progress, thus more accurately reflecting the actual safety margin of the chemical reaction station.

[0070] The above technical solution enables dynamic and accurate assessment of the tolerable waiting time at the chemical reaction station. Compared to relying solely on a single parameter or fixed threshold, this solution comprehensively considers multiple key parameters such as reaction temperature, reactant concentration, and redox potential, and introduces dynamic calculation of the chemical reaction rate constant, making the determined tolerable waiting time closer to actual reaction conditions. This helps provide a more reliable and safe upper limit for the waiting time at the chemical reaction station when anomalies occur at the physical dismantling station and an extension of dismantling time is required, thereby avoiding uncontrolled chemical reactions or safety risks due to excessively long waiting times, effectively improving the safety and controllability of the entire battery recycling process.

[0071] This application further proposes steps for controlling the reaction rate at the aforementioned chemical reaction station, including: The chemical reaction station includes a cooling jacket for temperature regulation and a main stirring motor for driving the stirring components. Controlling the chemical reaction station to reduce the reaction rate includes: Send a second control command to the chemical reaction station; the second control command is used to instruct the chemical reaction station to increase the flow rate of the cooling medium into the cooling jacket by a preset increase amount; send a third control command to the chemical reaction station; the third control command is used to instruct the chemical reaction station to reduce the drive frequency of the main stirring motor by a preset decrease amount.

[0072] Specifically, the aforementioned chemical reaction station is a crucial link in the battery recycling system, where the chemical dissolution or conversion reactions of battery materials take place. To precisely control the reaction process, this chemical reaction station is equipped with a cooling jacket for temperature regulation and a main stirring motor for driving the stirring components. The cooling jacket typically surrounds the outside of the reactor and absorbs the heat of reaction by introducing a cooling medium, thereby lowering the reaction temperature; the main stirring motor drives the stirring components to ensure thorough mixing of the reactants, improving reaction efficiency and uniformity.

[0073] When it is necessary to reduce the chemical reaction rate, the solution of this application achieves this by sending a second control command and a third control command. The second control command instructs the chemical reaction station to increase the flow rate of the cooling medium into the cooling jacket by a preset increment. This means that more cooling medium (such as cooling water or cooling oil) is pumped into the cooling jacket, thereby more effectively removing the heat of reaction and reducing the temperature of the reaction system. Reaction temperature is one of the key factors affecting the chemical reaction rate, and lowering the temperature usually significantly slows down the reaction rate. The preset increment can be set according to the specific reaction characteristics and the required rate reduction to achieve precise temperature control.

[0074] Simultaneously, the third control command instructs the chemical reaction station to reduce the drive frequency of the main stirring motor by a preset reduction amount. The drive frequency of the main stirring motor directly affects the rotational speed of the stirring components, thereby affecting the mixing intensity and mass transfer efficiency of the reactants in the reactor. Reducing the drive frequency weakens the stirring effect, resulting in a decrease in the contact frequency and mass transfer rate between reactants, thus indirectly slowing down the overall chemical reaction rate. The preset reduction amount can also be adjusted according to actual needs to achieve the desired stirring effect and reaction rate. Through the synergistic effect of these two control commands, an effective and controllable reduction in the reaction rate of the chemical reaction station can be achieved.

[0075] The solution presented in this application effectively reduces the reaction rate at the chemical reaction station by controlling two key kinetic factors. First, by increasing the flow rate of the cooling medium in the cooling jacket, the temperature of the reaction system can be rapidly and effectively reduced. According to the Arrhenius equation, the chemical reaction rate has an exponential relationship with temperature; a decrease in temperature significantly reduces the effective supply of activation energy, thereby drastically slowing down the reaction rate. Second, by reducing the drive frequency of the main stirring motor, the stirring intensity is weakened, which directly affects the diffusion and mixing efficiency of reactants in the liquid phase. For multiphase reactions or diffusion-controlled reactions, a decrease in stirring intensity limits the effective collision opportunities between reactant molecules, thus reducing the macroscopic reaction rate. It is precisely by simultaneously controlling these two core parameters—reaction temperature and reactant mixing efficiency—that the chemical reaction station can achieve a precise and controllable reduction in the reaction rate, adapting to abnormal situations at the upstream physical dismantling station.

[0076] Through the above technical solution, this application provides a precise and efficient reaction rate control mechanism for chemical reaction stations. Compared to coarse adjustments based solely on a single parameter (such as temperature or stirring), this application achieves more refined and stable regulation of the reaction rate by synergistically controlling the cooling medium flow rate of the cooling jacket and the drive frequency of the main stirring motor. This not only effectively addresses potential anomalies at the upstream physical dismantling station, preventing overheating, increased side reactions, or safety hazards caused by excessively fast reaction rates, but also minimizes the impact on battery recycling efficiency and product quality while ensuring process continuity. Therefore, the solution in this application significantly improves the flexibility and intelligence of the battery recycling system, enhances its ability to cope with emergencies, and ensures the stable operation and safety of the entire recycling process.

[0077] This application further proposes a scheme to take further measures based on the actual situation after controlling the chemical reaction station to reduce the reaction rate, so as to manage the battery recycling process more flexibly and safely.

[0078] After controlling the above-mentioned chemical reaction stations to reduce the reaction rate, the method also includes: Based on the reduced reaction rate at the chemical reaction station, the tolerable waiting time is redefined. When the estimated remaining dismantling time is greater than or equal to the redefined tolerable waiting time, and the difference between the estimated remaining dismantling time and the redefined tolerable waiting time is less than a preset time threshold, a preset dose of reducing agent is added to the chemical reaction station. The reducing agent is used to reduce the products of the chemical reaction in the chemical reaction station. When the estimated remaining dismantling time is greater than or equal to the redefined tolerable waiting time, and the difference between the estimated remaining dismantling time and the redefined tolerable waiting time is greater than or equal to a preset time threshold, the physical dismantling station is controlled to stop dismantling the target battery to be recycled, and the target battery to be recycled among multiple batteries to be recycled is redefined.

[0079] Specifically, after the reaction rate at a chemical reaction station is reduced, its internal reaction kinetics and material conversion rates will change. Therefore, it is necessary to reassess the maximum safe waiting time under the current state based on the reduced reaction rate, i.e., to redetermine the tolerable waiting time. This aims to obtain a more accurate and real-time safe waiting window to guide subsequent decisions. The redefined tolerable waiting time can be calculated using a pre-defined chemical kinetic model based on factors such as the reduced reaction rate, current reactant concentration, temperature, and preset safety threshold parameters.

[0080] Furthermore, when the estimated remaining dismantling time is still greater than or equal to the redefined tolerable waiting time, but the difference between the two is small (i.e., less than the preset time threshold), it indicates that although a delay exists at the physical dismantling station, it is still within a controllable range, and the risk can be avoided through appropriate intervention. At this time, a preset dose of reducing agent can be added to the chemical reaction station. The role of this reducing agent is to reduce the products of the chemical reaction in the chemical reaction station. For example, if harmful or unstable intermediate products are generated during the chemical reaction, the reducing agent can convert them into more stable or harmless substances, thereby effectively extending the safe operating time of the chemical reaction station, giving the physical dismantling station more time to complete its work, and avoiding unnecessary downtime.

[0081] As a further safeguard, when the estimated remaining dismantling time is greater than or equal to the redefined tolerable waiting time, and the difference between the two is significant (i.e., greater than or equal to the preset time threshold), it indicates that the delay at the physical dismantling station has severely exceeded the safe tolerance of the chemical reaction station. Continued waiting could lead to unacceptable risks for the chemical reaction station. In this case, to prioritize the safety and stability of the chemical reaction station, the system will control the physical dismantling station to cease dismantling operations on the current target battery to be recycled. Subsequently, the system will reassess the information of all batteries to be recycled and, based on the latest situation (including the status of the chemical reaction station and the availability of the physical dismantling station), re-determine a target battery to be recycled that is more suitable for the current conditions, in order to restore the efficiency and safety of the overall recycling process.

[0082] This application's solution, by introducing a multi-level adaptive control strategy, effectively addresses the problem that simply reducing the reaction rate of the chemical reaction station may be insufficient to handle all complex situations when anomalies occur at the physical dismantling station, leading to prolonged delays. First, by redefining the tolerable waiting time, the system can obtain a more accurate safe waiting time window based on the actual state of the chemical reaction station after reducing its reaction rate, thus providing a reliable basis for subsequent decision-making. Second, when the difference between the estimated remaining dismantling time and the redefined tolerable waiting time is small, the chemical reaction process can be actively intervened by adding a reducing agent to consume harmful products, thereby extending the safe operating time of the chemical reaction station. This provides an opportunity for the physical dismantling station to complete its work, avoiding downtime caused by minor delays and improving the system's continuous operation capability. Finally, when the difference between the estimated remaining dismantling time and the redefined tolerable waiting time is large, the system can decisively stop the current operation at the physical dismantling station and reselect the target battery to be recycled. This is a more thorough risk avoidance measure, ensuring the absolute safety of the chemical reaction station, preventing potential serious accidents, and allowing the system to replan the operation under new and safer conditions, thereby maintaining the stability and controllability of the overall recycling process.

[0083] Through the above technical solution, this application provides a more refined and adaptive battery recycling process management mechanism. This mechanism not only effectively addresses various abnormal situations that may occur at the physical dismantling station, but also maximizes recycling efficiency while ensuring the safety of the chemical reaction station through a tiered response strategy. Specifically, by adjusting the tolerable waiting time in real time, conditionally adding reducing agents, and decisively stopping and rescheduling dismantling tasks when necessary, this application significantly improves the robustness and safety of the battery recycling system, reduces risks caused by poor coordination between stations, optimizes resource utilization efficiency, and avoids unnecessary downtime and material loss.

[0084] This application further proposes that, prior to obtaining battery information for multiple batteries to be recycled, the method includes: Obtain the terminal voltage of each of the multiple batteries to be recycled and a first correspondence relationship; the first correspondence relationship includes a one-to-one correspondence between multiple terminal voltage ranges and multiple discharge load resistance values; take the discharge load resistance value corresponding to the terminal voltage range of the battery to be recycled in the first correspondence relationship as the target discharge load resistance value of the battery to be recycled, and discharge the battery to be recycled based on the target discharge load resistance value. During the discharge process of the battery to be recycled, the temperature change rate of the battery terminals is monitored in real time. When the temperature change rate exceeds the preset temperature change rate threshold, the load of the battery to be recycled is disconnected and the preset cooling mechanism is triggered.

[0085] Specifically, before recycling batteries, it is necessary to first obtain the terminal voltage of each battery to be recycled. This terminal voltage can be measured in real time using a voltage measuring device. Simultaneously, a preset first correspondence needs to be obtained. This first correspondence, pre-established and stored in the system, contains a one-to-one correspondence between multiple terminal voltage ranges and multiple discharge load resistance values. For example, when the battery's terminal voltage is within a specific range, the system will recommend a matching discharge load resistance value to ensure the safety and efficiency of the discharge process.

[0086] After obtaining the terminal voltage of the battery to be recycled, the system uses a first correspondence to find the voltage range within which the terminal voltage falls and determines the corresponding discharge load resistance value. This determined discharge load resistance value is used as the target discharge load resistance value for the battery to be recycled. Subsequently, based on this target discharge load resistance value, the battery to be recycled is discharged. The discharge process can be achieved by connecting the battery to a load resistor with a corresponding resistance value to safely dissipate the residual electrical energy in the battery.

[0087] In practical applications, during the discharge process of batteries to be recycled, it is necessary to monitor the temperature change rate of the battery terminals in real time. This monitoring can be achieved by placing temperature sensors near the battery terminals, which transmit real-time temperature data to the control system for calculation. When the monitored temperature change rate exceeds a preset threshold, it indicates a potential risk of overheating. In this case, to prevent thermal runaway, the system immediately disconnects the load from the battery, stops discharging, and triggers a preset cooling mechanism. This preset cooling mechanism can include, but is not limited to, forced air cooling, liquid cooling, or coolant spraying to rapidly reduce the battery temperature and ensure operational safety.

[0088] This application's solution effectively addresses the safety issues arising from residual battery charge and potential thermal runaway risks in traditional methods by introducing a pre-discharge and thermal management mechanism before the battery enters the physical dismantling station. First, by acquiring the terminal voltage of the battery to be recycled and combining it with a preset first correspondence, the system can match the most suitable discharge load resistance value to each battery, avoiding the problems of excessively rapid or insufficient discharge that may occur with fixed load discharge, thus achieving efficient and safe initial discharge. Second, during the discharge process, the system monitors the temperature change rate of the terminals in real time and sets a preset temperature change rate threshold, enabling the system to promptly detect abnormal battery heating. Once the temperature change rate exceeds the threshold, the load is immediately disconnected and a cooling mechanism is triggered, rapidly suppressing heat accumulation and effectively preventing thermal runaway. This significantly improves the safety of the battery recycling process and creates a safe and stable operating environment for subsequent physical dismantling and chemical reaction stations.

[0089] Through the aforementioned technical solution, this application introduces proactive safety management measures at the initial stage of battery recycling. Intelligent matching of the discharge load ensures that the batteries to be recycled are discharged in a controlled and efficient manner, avoiding safety hazards caused by improper discharge. Simultaneously, the combination of real-time temperature monitoring and an emergency cooling mechanism provides robust safety assurance for the battery recycling process, significantly reducing the risk of thermal runaway, fire, or explosion during the pre-treatment stage. This not only protects the personal safety of operators but also prevents equipment damage and production interruptions, thereby improving the reliability and economic efficiency of the entire battery recycling system.

[0090] This application further proposes a method for classifying initial batteries before obtaining the terminal voltage of the aforementioned batteries to be recycled, the method comprising: The process involves acquiring the electrochemical impedance spectroscopy (EIS) data, ultrasonic flaw detection data, and health index for each initial cell from a pool of multiple initial cells to be recycled. The health index is the ratio of the current capacity to the designed capacity of the initial cell. For each initial cell, the cell feature vector is determined based on its EIS and ultrasonic flaw detection data. The similarity between the cell feature vector of the initial cell and the cell feature vector of a reusable cell is determined. Initial cells with a similarity greater than a preset similarity threshold and a health index greater than a preset health index are designated as reusable cells. Initial cells with a similarity less than or equal to a preset similarity threshold, or a health index less than or equal to a preset health index, are designated as cells to be recycled.

[0091] Specifically, electrochemical impedance spectroscopy (EIS) data refers to the impedance response data of a battery at different frequencies obtained through EIS testing. It reflects the internal electrochemical processes, material properties, and aging state of the battery, revealing information such as ohmic impedance, charge transfer impedance, and diffusion impedance. Its purpose is to assess the battery's internal health and potential defects. Ultrasonic testing data can be understood as information about the battery's internal structure obtained through ultrasonic detection technology. For example, it can detect physical defects such as bubbles, cracks, and delamination, aiming to assess the battery's mechanical integrity and structural stability. In practical applications, the health index refers to the ratio of the battery's current capacity to its design capacity. For example, a battery with a design capacity of 100 Ah has a health index of 80% if its current capacity is 80 Ah. Its purpose is to quantify the overall performance degradation of the battery.

[0092] Furthermore, the battery feature vector is a multi-dimensional data representation formed by feature extraction and fusion of electrochemical impedance spectroscopy data and ultrasonic flaw detection data. Its purpose is to comprehensively and quantitatively describe the internal state and physical structure characteristics of the battery. Similarity refers to the degree of matching between the battery feature vector of the initial battery, calculated using a specific algorithm (e.g., cosine similarity, Euclidean distance, etc.), and the pre-established battery feature vector of a reusable battery. Its purpose is to determine the similarity between the initial battery and known healthy, reusable batteries. The preset similarity threshold and preset health index are judgment criteria set based on empirical data, industry standards, or specific application requirements. Their purpose is to serve as quantitative boundaries distinguishing reusable batteries from batteries awaiting recycling.

[0093] The proposed solution comprehensively and multidimensionally assesses the internal electrochemical state, physical structural integrity, and overall performance degradation of a battery by integrating the electrochemical impedance spectroscopy (EIS), ultrasonic flaw detection data, and health index of the initial battery. Because the EIS reveals the battery's internal electrochemical characteristics, the ultrasonic flaw detection data reflects physical structural defects, and the health index directly quantifies the battery's usable capacity, the battery feature vector constructed from these data accurately characterizes the battery's true condition.

[0094] By comparing the similarity between the feature vectors of the initial batteries and those of reusable batteries, and combining this with a health index assessment, batteries with good performance, intact structure, and remaining reusable can be effectively identified and separated from the batteries awaiting recycling. This pre-screening mechanism avoids sending still-valuable batteries directly into the recycling process, thus solving the potential resource waste problem in the aforementioned basic scheme.

[0095] Through the aforementioned technical solution, this application enables intelligent and refined classification of initial batteries, significantly improving battery resource utilization efficiency. Specifically, by introducing multi-dimensional evaluation indicators such as electrochemical impedance spectroscopy, ultrasonic flaw detection data, and health index, and combining them with similarity analysis of battery feature vectors, batteries with tiered utilization value can be accurately identified, avoiding unnecessary recycling of these batteries, thereby extending the battery's lifespan and reducing resource consumption. Furthermore, this solution ensures that only batteries that truly meet the end-of-life standards enter the subsequent discharge and recycling stages, optimizing the efficiency and economy of the entire battery recycling process and providing an effective way to maximize the value of batteries throughout their entire lifecycle.

[0096] This application proposes a battery recycling system for all scenarios. The battery recycling system includes a physical dismantling station and a chemical reaction station in sequence according to the process. The battery recycling system also includes an acquisition device and a processing device. The acquisition device is used to acquire battery information of multiple batteries to be recycled and the current remaining processing time of the chemical reaction station. The processing device is used to input the battery information of each battery to be recycled into a preset dismantling time prediction model to obtain the predicted dismantling time of the battery to be recycled. The preset dismantling time prediction model is determined based on the historical dismantling time of the physical dismantling station. The processing device is used to select the battery to be recycled whose preset dismantling time is less than the current remaining processing time and whose difference between the current remaining processing time and the preset dismantling time is the smallest as the target battery to be recycled. The target battery to be recycled is the next dismantling object of the physical dismantling station.

[0097] The acquisition device can be configured in various forms, and its main function is to acquire battery information of multiple batteries to be recycled and the current remaining processing time of the chemical reaction station. In one implementation, the acquisition device may include one or more sensors, scanning devices, and a communication module. For example, the scanning device can be used to read QR codes or RFID tags on the batteries to be recycled to obtain their basic information; the sensors can be used to detect the physical dimensions, weight, etc. of the batteries; and the communication module can interact with the battery management system (BMS) or the factory's central control system to obtain more detailed battery health status, historical usage data, and real-time operating parameters and processing progress of the chemical reaction station. In some embodiments, the acquisition device can be designed as an automated identification unit integrated at the entrance of the physical dismantling station, or it can exist as a standalone inspection station.

[0098] The processing device can be configured to perform complex computational and decision-making tasks. Its main functions include inputting battery information of the batteries to be recycled into a preset dismantling time prediction model to obtain a predicted dismantling time, and determining the target batteries to be recycled based on the predicted dismantling time and the current remaining processing time at the chemical reaction station. In one implementation, the processing device may include one or more computing units such as microprocessors, central processing units (CPUs), graphics processing units (GPUs), field-programmable gate arrays (FPGAs), or application-specific integrated circuits (ASICs), as well as corresponding memory and communication interfaces. The processing device can run a preset software program containing the dismantling time prediction model. This model can be a model trained based on machine learning algorithms (e.g., support vector machines, neural networks, or decision trees), established by analyzing the relationship between battery information and actual dismantling times in historical dismantling data. The processing device receives data from the acquisition device, performs predictive calculations, and determines the next battery to be recycled at the physical dismantling station based on preset logic (i.e., selecting the battery whose predicted dismantling time is less than the current remaining processing time and whose difference is the smallest). In some embodiments, the processing device may be a stand-alone industrial control computer or a module integrated into the central control unit of the battery recycling system.

[0099] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An AI-powered recycling method applicable to all scenarios, characterized in that, The method is applied to a battery recycling system, which includes a physical dismantling station and a chemical reaction station in sequence. The method includes the following steps: Obtain battery information and the current remaining processing time of multiple batteries to be recycled at the chemical reaction station; For each of the multiple batteries to be recycled, the battery information of the battery to be recycled is input into a preset dismantling time prediction model to obtain the predicted dismantling time of the battery to be recycled; the preset dismantling time prediction model is determined based on the historical dismantling time of the physical dismantling station for dismantling historical batteries to be recycled. The battery to be recycled that has a preset dismantling time that is less than the current remaining processing time and has the smallest difference between the current remaining processing time and the preset dismantling time is designated as the target battery to be recycled; the target battery to be recycled is the next dismantling object of the physical dismantling station.

2. The AI-powered recycling method for all scenarios according to claim 1, characterized in that, After identifying the target batteries to be recycled, the method further includes: When the physical dismantling station performs dismantling operations on the target battery to be recycled, the load characteristic parameters of the physical dismantling station are acquired in real time, and the physical dismantling station is determined to be in an abnormal dismantling state based on the load characteristic parameters. When the physical disassembly station is in an abnormal disassembly state, adjust the disassembly mode of the physical disassembly station and determine the estimated remaining disassembly time; The current reaction status parameters and preset safety threshold parameters of the chemical reaction station are obtained, and the tolerable waiting time of the chemical reaction station is determined based on the current reaction status parameters and preset safety threshold parameters. When the estimated remaining dismantling time is greater than or equal to the tolerable waiting time, the chemical reaction station is controlled to reduce the reaction rate.

3. The AI-powered recycling method for all scenarios according to claim 2, characterized in that, The physical disassembly station is an ultrasonic cutting assembly. The load characteristic parameters include the acoustic impedance value of the ultrasonic cutting assembly. Determining whether the physical disassembly station is in an abnormal disassembly state based on the load characteristic parameters includes: Calculate the rate of change of the acoustic impedance value; When the rate of change of the acoustic impedance value exhibits a non-linear jump and exceeds a preset rate of change threshold, it is determined that the physical disassembly station is in an abnormal disassembly state; otherwise, it is determined that the physical disassembly station is not in an abnormal disassembly state.

4. The AI-powered recycling method for all scenarios according to claim 2, characterized in that, The disassembly modes include a conventional cutting mode and a micro-disturbance grinding mode. Adjusting the disassembly mode of the physical disassembly station and determining the estimated remaining disassembly time includes: Send a first control command to the physical disassembly station to switch the physical disassembly station from the conventional cutting mode to the micro-disturbance grinding mode; Obtain the unit volume processing time parameter of the physical disassembly station under the micro-disturbance grinding mode; The product of the remaining dismantling volume of the target battery to be recycled and the processing time per unit volume is used as the estimated remaining dismantling time.

5. The AI-powered recycling method for all scenarios according to claim 2, characterized in that, The current reaction state parameters include the current redox potential value, the current reaction temperature, and the reactant concentration. The preset safety threshold parameters include the critical redox potential threshold. The tolerable waiting time for the chemical reaction station is determined based on the current reaction state parameters and the preset safety threshold parameters, including: Determine the current chemical reaction rate constant based on the current reaction temperature and the reactant concentration; Based on the chemical reaction rate constant, determine the rise time required for the current redox potential value to rise to the redox potential critical threshold. The rise time is determined as the tolerable waiting time.

6. The AI-powered recycling method for all scenarios according to claim 2, characterized in that, The chemical reaction station includes a cooling jacket for temperature regulation and a main stirring motor for driving the stirring components. Controlling the chemical reaction station to reduce the reaction rate includes: Send a second control command to the chemical reaction station; the second control command is used to instruct the chemical reaction station to increase the flow rate of the cooling medium into the cooling jacket by a preset increment. A third control command is sent to the chemical reaction station; the third control command is used to instruct the chemical reaction station to reduce the drive frequency of the main stirring motor by a preset reduction amount.

7. The AI-powered recycling method for all scenarios according to claim 2, characterized in that, After controlling the chemical reaction station to reduce the reaction rate, the method further includes: Based on the reaction rate after the chemical reaction station is reduced, the tolerable waiting time is re-determined; When the estimated remaining dismantling time is greater than or equal to the re-determined tolerable waiting time, and the difference between the estimated remaining dismantling time and the re-determined tolerable waiting time is less than a preset time threshold, a preset dose of reducing agent is added to the chemical reaction station; the reducing agent is used to reduce the products of the chemical reaction in the chemical reaction station. When the estimated remaining dismantling time is greater than or equal to the re-determined tolerable waiting time, and the difference between the estimated remaining dismantling time and the re-determined tolerable waiting time is greater than or equal to a preset time threshold, the physical dismantling station is controlled to stop dismantling the target battery to be recycled, and the target battery to be recycled among multiple batteries to be recycled is re-determined.

8. The AI-powered recycling method for all scenarios according to claim 1, characterized in that, Before obtaining battery information for multiple batteries to be recycled, the method further includes: Obtain the terminal voltage of each of the multiple batteries to be recycled and a first correspondence relationship; the first correspondence relationship includes a one-to-one correspondence between multiple terminal voltage ranges and multiple discharge load resistance values; The discharge load resistance value corresponding to the terminal voltage range of the battery to be recycled in the first correspondence is taken as the target discharge load resistance value of the battery to be recycled, and the battery to be recycled is discharged based on the target discharge load resistance value. During the discharge process of the battery to be recycled, the temperature change rate of the battery terminals is monitored in real time. When the temperature change rate exceeds the preset temperature change rate threshold, the load of the battery to be recycled is disconnected and a preset cooling mechanism is triggered.

9. The AI-powered recycling method for all scenarios according to claim 8, characterized in that, Before obtaining the terminal voltage of each of the multiple batteries to be recycled, the method further includes: Obtain the electrochemical impedance spectroscopy data, ultrasonic flaw detection data, and health index of each initial cell in a plurality of initial cells including the plurality of cells to be recycled; the health index is the ratio of the current capacity to the design capacity of the initial cell. For each of the multiple initial cells, the cell feature vector of the initial cell is determined based on the electrochemical impedance spectroscopy data and ultrasonic flaw detection data of the initial cell. Determine the similarity between the battery feature vector of the initial battery and the battery feature vector of the reusable battery; Initial batteries with a similarity greater than a preset similarity threshold and a health index greater than a preset health index are selected as usable batteries. Initial batteries with a similarity less than or equal to a preset similarity threshold, or with a health index less than or equal to a preset health index, are designated as the batteries to be recycled.

10. The AI-powered recycling method for all scenarios according to claim 8, characterized in that, Trigger the preset cooldown mechanism, including: The batteries to be recycled are subjected to forced air cooling; Coolant is sprayed onto the batteries to be recycled.

Images