Battery core welding method, device, electronic equipment, storage medium and program product

Abstract

The embodiments of the present disclosure provide a battery cell welding method, device, electronic device, readable storage medium, and computer program product, relating to the field of battery cell manufacturing technology. The method includes: obtaining a battery cell to be welded, performing at least one welding operation on the battery cell to be welded, until the welding status of the battery cell to be welded is successful. The welding operation includes: obtaining first welding state information and a value prediction model for the current welding operation, determining the welding energy required for the current welding operation, welding the battery cell to be welded, determining the welding state in the second welding state information, and if the welding fails, updating the value prediction model, using the second welding state information as the next first welding state information, and using the updated value prediction model as the value prediction model corresponding to the next welding operation. The embodiments of the present disclosure achieve adjustment of welding parameters based on welding state feedback, can adapt to various complex working conditions, and improve battery cell welding efficiency.

Description

Technical Field

[0001] The present disclosure relates to the technical field of battery cell manufacturing, and in particular to a battery cell welding method, device, electronic device, storage medium, and program product. Background Art

[0002] With the development of the battery cell manufacturing industry, ultrasonic welding is widely used in battery cell manufacturing. Battery cell welding technology is a core process in battery manufacturing. Through laser welding, ultrasonic welding, resistance welding, and other methods, the positive and negative electrodes of the battery cell and other components are precisely connected to ensure the integrity and electrical performance of the battery's internal circuit. The process requires strict control of welding quality, thermal effects, and the environment to ensure the consistency, safety, and reliability of the battery. Heat is a key factor in the battery cell welding process. It achieves the melting and bonding of battery cell materials through the highly concentrated and precisely controlled laser beam.

[0003] Conventional ultrasonic welding usually uses fixed welding parameters, and the welding energy depends on manual experience. However, due to differences in workpiece material, size, environment and other factors, fixed welding energy is difficult to adapt to different working conditions. When the working conditions change, it will cause poor weld quality and lead to low battery cell welding efficiency. Summary of the Invention

[0004] The embodiments of the present disclosure provide a battery cell welding method, device, electronic device, computer-readable storage medium, and computer program product, aiming to solve the technical problem that fixed parameters are difficult to adapt to different working conditions, resulting in low welding efficiency.

[0005] In a first aspect, a battery core welding method is provided, the method comprising: obtaining a battery core to be welded;

[0006] Performing at least one welding operation on the battery cell to be welded until the welding status of the battery cell to be welded is successful;

[0007] Welding operations include:

[0008] Acquire first welding state information corresponding to a current welding operation of the battery cell to be welded; the first welding state information includes welding state and welding power;

[0009] Obtaining a value prediction model corresponding to a current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation;

[0010] Using welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value;

[0011] If the welding state in the second welding state information is welding failure, predicting the welding energy required for the next welding operation from at least one candidate energy based on the second welding state information using a value prediction model; and updating the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value.

[0012] The second welding state information is used as the first welding state information for the next time, and the updated value prediction model is used as the value prediction model corresponding to the next welding operation.

[0013] Optionally, obtaining a value prediction model corresponding to the current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation, includes:

[0014] Obtaining a value prediction model corresponding to the current welding operation, using the welding power in the first welding state information as an input to the value prediction model, and obtaining a value function corresponding to at least one candidate energy;

[0015] Determine the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies;

[0016] The candidate energy with the highest probability is selected from the at least one candidate energy as the welding energy required for the current welding operation.

[0017] Optionally, determining the selection probability corresponding to each candidate energy based on the cost function corresponding to each candidate energy and the sum of the cost functions of the candidate energies includes:

[0018] Determining first probability parameters corresponding to each candidate energy based on the value function and preset temperature parameters corresponding to each candidate energy; the preset temperature parameters are used to simulate actual temperature conditions, and the temperature parameters gradually decrease as the number of welding operations increases;

[0019] The selection probability corresponding to each candidate energy is determined based on the first probability parameter corresponding to each candidate energy and the sum of the first probability parameters corresponding to the candidate energy.

[0020] Optionally, based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value, the value prediction model is updated, including:

[0021] Determining a corresponding first value function based on the welding energy required for the current welding operation through a preset value prediction model;

[0022] Determining a corresponding second value function based on the welding energy required for the next welding operation using a preset value prediction model;

[0023] Based on the first value function, the second value function and the reward value, the value prediction model is updated.

[0024] Optionally, updating the value prediction model based on the first value function, the second value function, and the reward value includes:

[0025] Determining a first evaluation value corresponding to the current welding operation based on the first value function and welding parameters corresponding to the current welding operation;

[0026] determining a second evaluation value corresponding to the next welding operation based on the second value function and welding parameters corresponding to the next welding operation; and obtaining a target evaluation value based on the second evaluation value and the reward value;

[0027] obtaining an error value based on a difference between the target evaluation value and the first evaluation value;

[0028] Update the value prediction model based on the error value.

[0029] Optionally, welding the battery cell to be welded using welding energy to obtain corresponding second welding state information and a reward value includes:

[0030] Using welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information;

[0031] A corresponding reward value is determined based on the second welding state information through a preset reward function.

[0032] Optionally, if the welding state in the second welding state information is welding failure, then using a value prediction model and based on the second welding state information, predicting the welding energy required for the next welding operation from at least one candidate energy includes:

[0033] If the welding state in the second welding state information is welding failure, the welding power in the second welding state information is used as an input of the value prediction model to obtain a value function corresponding to at least one candidate energy;

[0034] Determine the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies;

[0035] A candidate energy with the highest probability is selected from the at least one candidate energy as the welding energy required for the next welding operation.

[0036] In a second aspect, a core welding device is provided, the device comprising:

[0037] A cell acquisition module is used to obtain cells to be welded;

[0038] A cell welding module is used to perform at least one welding operation on the cell to be welded until the welding state of the cell to be welded is successful;

[0039] Cell welding module, also used for:

[0040] Acquire first welding state information corresponding to a current welding operation of the battery cell to be welded; the first welding state information includes welding state and welding power;

[0041] Obtaining a value prediction model corresponding to a current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation;

[0042] Using welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value;

[0043] If the welding state in the second welding state information is welding failure, predicting the welding energy required for the next welding operation from at least one candidate energy based on the second welding state information using a value prediction model; and updating the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value;

[0044] The second welding state information is used as the first welding state information for the next time, and the updated value prediction model is used as the value prediction model corresponding to the next welding operation.

[0045] According to a third aspect, an electronic device is provided, comprising:

[0046] A memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any one of the methods in the first aspect of the present disclosure.

[0047] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the battery core welding method shown in any one of the first aspects of the present disclosure is implemented.

[0048] In a fifth aspect, a computer program product is provided, comprising a computer program, characterized in that when the computer program is executed by a processor, the steps of any one of the methods in the first aspect of the present disclosure are implemented.

[0049] The technical solutions provided by the embodiments of the present disclosure have the following beneficial effects:

[0050] The battery cell welding method provided by the present disclosure performs at least one welding operation on the battery cell to be welded, including: obtaining first welding state information corresponding to the current welding operation of the battery cell to be welded, including welding state and welding power, and obtaining a value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, welding the battery cell to be welded using the welding energy, and obtaining corresponding second welding state information and a reward value; if the welding state in the second welding state information is welding failure, then using the value prediction model, based on the second welding state information, predicting the welding energy required for the next welding operation from at least one candidate energy; based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, updating the value prediction model, using the second welding state information as the next first welding state information, and using the updated value prediction model as the value prediction model corresponding to the next welding operation; for each welding operation, updating the value prediction model according to the feedback information corresponding to the welding operation, and being able to adjust parameter selection in time according to the feedback information, thereby effectively improving welding quality and welding efficiency.

[0051] Furthermore, at least one candidate welding energy and the corresponding value function are predicted by using the value prediction model corresponding to the welding operation, and the selection probability of each candidate welding energy is determined based on the value function. Since the value prediction model will be adjusted according to the feedback information of the welding operation, the selection probability corresponding to each candidate energy will also change according to the adjusted value function. The selection probability obtained in this way can make the selected welding energy more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following briefly introduces the drawings required for describing the embodiments of the present disclosure.

[0053] Figure 1 A schematic diagram of an application scenario of a battery core welding method provided by an embodiment of the present disclosure;

[0054] Figure 2 A schematic diagram of a process for welding a battery cell according to an embodiment of the present disclosure;

[0055] Figure 3 A schematic flow chart of a welding operation in a battery core welding method provided in an embodiment of the present disclosure;

[0056] Figure 4 A simplified flowchart of a battery core welding method provided in an embodiment of the present disclosure;

[0057] Figure 5 A schematic flow chart of an example of a battery core welding method provided in an embodiment of the present disclosure;

[0058] Figure 6 A schematic structural diagram of a battery core welding device provided in an embodiment of the present disclosure;

[0059] Figure 7 A schematic structural diagram of an electronic device applicable to a battery core welding method provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0060] The following describes embodiments of the present disclosure in conjunction with the accompanying drawings. It should be understood that the embodiments described below in conjunction with the accompanying drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present disclosure and do not constitute a limitation on the technical solutions of the embodiments of the present disclosure.

[0061] Those skilled in the art will understand that, unless otherwise stated, the singular forms "a", "an", "said", and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present disclosure mean that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the present technical field. It should be understood that when we say that an element is "connected" or "coupled" to another element, the element can be directly connected or coupled to the other element, or it can refer to the establishment of a connection relationship between the element and the other element through an intermediate element. In addition, the "connection" or "coupling" used here can include wireless connection or wireless coupling. The terms "or", "and / or", "including at least one of the following", etc. used in the present disclosure can be interpreted as inclusive, or mean any one or any combination. For example, "including at least one of the following: A, B, C" means "any one of the following: A; B; C; A and B; A and C; B and C; A and B and C", and for another example, "A, B or C" or "A, B and / or C" means "any one of the following: A; B; C; A and B; A and C; B and C; A and B and C".

[0062] In the embodiments of the present disclosure, the term "module" or "unit" refers to a computer program or portion of a computer program that has a predetermined function and works together with other related components to achieve a predetermined goal. The ... that can be implemented in whole or in part using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0063] In order to make the objectives, technical solutions and advantages of the present disclosure more clear, the embodiments of the present disclosure will be further described in detail below with reference to the accompanying drawings.

[0064] First, the technical terms involved in this disclosure are introduced and explained:

[0065] Welding energy: Welding energy refers to the energy required to melt the metal and form the weld during welding. It is an important parameter for measuring the heat input during the welding process and has a significant impact on welding quality, efficiency, and cost.

[0066] Welding power: Welding power refers to the energy supplied by the welding equipment to the welding arc or heat source, such as a laser, during the welding process. The required welding energy directly influences the selection and adjustment of welding power. Welding operations require optimizing welding power usage by adjusting welding parameters to control welding energy based on the specific requirements of the welding task, achieving efficient, high-quality welds.

[0067] Ultrasonic welding is a common welding method used extensively in battery cell manufacturing. Traditional ultrasonic welding typically employs fixed welding parameters (such as energy and time) that rely on manual experience. However, due to variations in workpiece material, size, and environmental factors, fixed parameters are difficult to adapt to diverse operating conditions, leading to poor weld quality and high energy consumption. Problems with existing technologies include: fixed welding parameters are unable to adapt to varying operating conditions, resulting in poor weld quality stability; parameter setting relies primarily on manual experience, lacking scientific theoretical guidance, resulting in low efficiency and a lack of versatility; and the inability to adjust welding parameters based on real-time feedback, resulting in low energy efficiency.

[0068] The battery cell welding method, device, electronic device, computer-readable storage medium, and computer program product provided in the present disclosure are intended to solve at least one of the above technical problems in the prior art.

[0069] In response to at least one of the above-mentioned technical problems or areas that need improvement in the related art, the present disclosure proposes a battery cell welding method, device, electronic device, computer-readable storage medium and computer program product. The battery cell welding method provided by the solution performs at least one welding operation on the battery cell to be welded, including: obtaining first welding state information corresponding to the current welding operation of the battery cell to be welded, including welding state and welding power, and obtaining a value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, using the welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value; if the second If the welding status in the welding status information is welding failure, the value prediction model is used to predict the welding energy required for the next welding operation from at least one candidate energy based on the second welding status information. Based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, the value prediction model is updated, and the second welding status information is used as the first welding status information for the next time. The updated value prediction model is used as the value prediction model corresponding to the next welding operation. For each welding operation, the value prediction model is updated according to the feedback information corresponding to the welding operation, and the parameter selection can be adjusted in time according to the feedback information, thereby effectively improving the welding quality and welding efficiency.

[0070] Furthermore, at least one candidate welding energy and the corresponding value function are predicted by using the value prediction model corresponding to the welding operation, and the selection probability of each candidate welding energy is determined based on the value function. Since the value prediction model will be adjusted according to the feedback information of the welding operation, the selection probability corresponding to each candidate energy will also change according to the adjusted value function. The selection probability obtained in this way can make the selected welding energy more accurate.

[0071] The following describes several exemplary embodiments to illustrate the technical solutions of the embodiments of the present disclosure and the technical effects produced by the technical solutions of the present disclosure. It should be noted that the following embodiments can refer to, draw on, or combine with each other, and the same terms, similar features, and similar implementation steps in different embodiments will not be repeated.

[0072] Figure 1 A schematic diagram of an application scenario of the battery cell welding method provided in an embodiment of the present disclosure, wherein the application environment may include a terminal 101 equipped with a welding control system, and the terminal 101 can be connected to an ultrasonic welder to control the ultrasonic welder to perform welding operations on the battery cells to be welded.

[0073] Specifically, the welding control system in the terminal 101 obtains the battery cell to be welded, and performs at least one welding operation on the battery cell to be welded until the welding status of the battery cell to be welded is successful welding, wherein the welding operation includes: obtaining first welding status information corresponding to the current welding operation of the battery cell to be welded, the first welding status information including welding status and welding power, obtaining a value prediction model corresponding to the current welding operation, based on the first welding status information and the value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, using the welding energy to weld the battery cell to be welded, and obtaining corresponding second welding status information and a reward value. If the welding status in the second welding status information is welding failure, the welding energy required for the next welding operation is predicted from at least one candidate energy based on the second welding status information through the value prediction model, and based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, updating the value prediction model, using the second welding status information as the first welding status information for the next time, and using the updated value prediction model as the value prediction model corresponding to the next welding operation.

[0074] The above application scenario is only an example and does not limit the application scenario of the battery core welding method disclosed herein.

[0075] Those skilled in the art will appreciate that the terminal may be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a laptop computer, a digital broadcast receiver, a MID (Mobile Internet Devices), a PDA (Personal Digital Assistant), a desktop computer, a smart home appliance, a vehicle-mounted terminal (such as a vehicle-mounted navigation terminal, a vehicle-mounted computer, etc.), a smart speaker, a smart watch, etc. The terminal and the server may be connected directly or indirectly via wired or wireless communication, but is not limited thereto.

[0076] The server may include a server equipped with a computer capable of processing database operations. The server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It may also be a cloud server or server cluster that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The embodiments of the present invention can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, assisted driving, etc. The specific requirements can also be determined based on the actual application scenario requirements and are not limited here.

[0077] In some possible implementations, taking the execution subject as a welding control system as an example, the embodiment of the present disclosure provides a cell welding method, such as Figure 2 As shown, the following steps may be included:

[0078] S201, obtaining a battery cell to be welded.

[0079] Specifically, when the production process of the battery cell reaches the welding step, the battery cell is used as the battery cell to be welded in the embodiment, and the basic information of the battery cell to be welded is obtained. The basic information includes: battery cell type, size, tab material and position, welding method and parameters, heat-affected zone, safety specifications, welding quality standards, and battery cell electrical characteristics and production environment, etc. Based on the basic information, the welding method corresponding to the battery cell to be welded is selected, which can make the battery cell more reliable and stable after welding. In the specific implementation process, the production steps of the battery cell include electrode preparation, winding or lamination, assembly, liquid injection, sealing, aging test and other steps; welding usually occurs in the assembly stage, after the battery cell is assembled and before liquid injection and sealing. This process ensures that the battery cell can be effectively connected to other battery cells in the battery pack and external equipment.

[0080] S202 , performing at least one welding operation on the battery cell to be welded until the welding status of the battery cell to be welded is successful.

[0081] Among them, after each welding of the battery cell to be welded, the welding status of the battery cell will be detected. If the welding fails, the parameters will continue to be adjusted for welding. If the welding is successful, the welding operation will be ended.

[0082] Specifically, the welding operation includes determining corresponding welding parameters based on the battery cells to be welded, welding the battery cells to be welded based on the welding parameters, and obtaining the battery cell status after welding to determine whether to continue the welding operation. In this way, the welding resources can be stopped in time after success to avoid waste of resources.

[0083] Among them, the above welding operation, such as Figure 3 As shown, the following steps are included:

[0084] S211, obtaining first welding state information corresponding to a current welding operation of a battery cell to be welded.

[0085] Among them, the first welding status information includes welding status and welding power; the welding status is used to indicate whether the welding of the above-mentioned battery cell to be welded is successful; the welding power includes the initial condition of the battery cell to be welded during the first welding operation, and is used to indicate the condition of the battery cell after the corresponding welding operation is executed in the second and subsequent welding operations.

[0086] Specifically, if it is the first welding operation, the first welding state information corresponding to the current welding operation is the initial value, wherein the initial values ​​of the welding state and the welding power can be determined based on the historical welding data of the battery cells of the same batch of the battery cells to be welded, or can be input by the operator in the welding control system for the battery cells to be welded. If it is not the first welding operation, the first welding state information is obtained after the last welding operation is performed.

[0087] S212, obtaining a value prediction model corresponding to the current welding operation, and determining the welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation.

[0088] The value prediction model may include an RBF neural network (Radial Basis Function Neural Network), the input of the value prediction model is the first welding state information, and the output is a value function corresponding to at least one candidate welding energy.

[0089] Specifically, the first welding state information corresponding to the current welding operation is input into the value prediction model corresponding to the current welding operation to obtain a value function corresponding to at least one candidate energy. Based on each candidate energy and the value function, the welding energy required for the current welding operation is selected. The method of selecting welding energy from the candidate energies may include ranking-based selection and probability-based selection, and may also receive the operator's selection operation for energy to determine the welding energy. The method of selecting welding energy is relatively flexible, which makes the welding method disclosed in the present invention highly adaptable.

[0090] S213 , welding the battery cell to be welded using welding energy to obtain corresponding second welding state information and a reward value.

[0091] The second welding state information may also include welding state and welding power. The reward value may be pre-set and may vary as the number of welding operations increases.

[0092] Specifically, after determining the welding energy, the ultrasonic welder can be controlled to use the welding energy to weld the battery cell to be welded, obtain the second welding status information, determine the welding power corresponding to the welding energy, and detect whether the battery cell to be welded is successful, and obtain the reward value corresponding to the current welding. If the welding is successful, the welding operation is stopped. If the welding fails, the next welding operation is continued based on the reward value and the second welding status information.

[0093] During the specific implementation process, the welding control system is also constantly optimized and updated. The sum of the reward values ​​corresponding to the welding operations in the welding steps is an important basis for system optimization. Generally, the goal is to achieve a higher reward value. The reward value can be expressed based on the following formula:

[0094] G t = r t+1 +γ r t+2 +γ 2 r t+3 +…+γ n-1 r t+n

[0095] Among them, G t represents the reward value, t represents the number of welding operations, r t is the immediate reward for the t-th welding operation, γ∈[0,1] is the discount factor, the smaller the value is, the more emphasis is placed on immediate rewards, and the closer it is to 1, the more emphasis is placed on future rewards, and n is a natural number.

[0096] S214. If the welding status in the second welding status information is welding failure, the welding energy required for the next welding operation is predicted from at least one candidate energy based on the second welding status information through the value prediction model; based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, the value prediction model is updated.

[0097] Specifically, if the welding status in the second welding status information is welding failure, the value prediction model is used, and the second welding status information is used as the input of the model to predict the welding energy required for the next welding operation from at least one candidate energy. Based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the corresponding reward value, the value prediction model corresponding to the current welding operation is updated to obtain the value prediction model to be used for the next welding operation. Based on the selection and prediction of the two welding energies and the reward value of the timely feedback of the welding operation, the value prediction model is updated in real time, which can make the selection of welding energy more accurate, reduce the number of trial and error and resource waste, and reduce welding costs.

[0098] S215: Use the second welding state information as the first welding state information for the next time, and use the updated value prediction model as the value prediction model corresponding to the next welding operation.

[0099] Specifically, only one welding is performed in one welding operation, and only the welding corresponding to the current welding energy is performed, and the second welding state information obtained is used as the first welding state information for the next welding operation. The welding energy required for the next welding operation is not executed, but is only used to update the value prediction model; each time a welding operation is performed and the welding is not successful, the value prediction model will be updated, so each time a welding operation is performed, the updated value prediction model must be determined to make the determination of the welding energy more accurate.

[0100] In some possible implementations, in step S212, obtaining a value prediction model corresponding to the current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation include:

[0101] (1) Obtaining a value prediction model corresponding to the current welding operation, using the welding power in the first welding state information as input to the value prediction model, and obtaining a value function corresponding to at least one candidate energy;

[0102] (2) Determine the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies;

[0103] (3) Selecting the candidate energy with the highest probability from at least one candidate energy as the welding energy required for the current welding operation.

[0104] The candidate energies are all energies that can be used on the battery cells to be welded, and can be preset based on input from an operator.

[0105] Specifically, a value prediction model corresponding to the current welding operation is obtained, and the welding power in the first welding state information is used as the input of the value prediction model to obtain the value functions corresponding to at least one candidate energy, and the sum of the value functions of the candidate energies is obtained. Based on the value function and the sum of each candidate energy, the selection probability corresponding to each candidate energy is determined, and the candidate energy with the highest probability of being selected from at least one candidate energy is used as the welding energy required for the current welding operation.

[0106] In the specific implementation process, the value function is used to evaluate the expected return of taking a specific action in a given state. In this embodiment, the specific action is welding energy. The value function of welding energy can be expressed as:

[0107]

[0108] Among them, π is the strategy, that is, how to choose the welding energy, q π(s, a) represents the value function (expected return) of selecting welding energy a when the welding power in the first welding state information is s, t represents the number of welding operations, R t represents the reward value obtained by using welding energy a in the t-th welding operation. γ∈[0,1] is a discount factor, which is used to measure the importance of future rewards relative to current rewards. The closer it is to 1, the more important the future rewards are. The closer it is to 0, the less important the future rewards are.

[0109] In some possible implementations, determining the selection probability corresponding to each candidate energy based on the cost function corresponding to each candidate energy and the sum of the cost functions of the candidate energies in the above steps includes:

[0110] (1) Determining a first probability parameter corresponding to each candidate energy based on the value function and the preset temperature parameter corresponding to each candidate energy;

[0111] (2) Determine the selection probability corresponding to each candidate energy based on the first probability parameter corresponding to each candidate energy and the sum of the first probability parameters corresponding to the candidate energy.

[0112] The preset temperature parameters are used to simulate the actual temperature conditions, and the temperature parameters gradually decrease as the number of welding operations increases.

[0113] Specifically, based on the value function and preset temperature parameters corresponding to each candidate energy, the first probability parameter corresponding to each candidate energy is determined. Based on the first probability parameter corresponding to each candidate energy and the sum of the first probability parameters corresponding to the candidate energy, the selection probability corresponding to each candidate energy is determined. The selection probabilities corresponding to each candidate energy are compared to select the welding energy.

[0114] In the specific implementation process, when reinforcement learning is applied to the embodiments of the present disclosure, in order to optimize the selection probability of an action as much as possible and help learn a good strategy more quickly, a probability weight can be assigned to each action, and the selection probability can be obtained based on a softmax selection strategy (soft maximization strategy). The function of the selection probability can be expressed by the following formula:

[0115]

[0116] Among them, P (a i ) is the candidate energy a when the welding power is s i The probability of selection, a i is the welding energy that can be selected for the i-th option, Q(s, a i ) is the welding energy a taken under state s i The value of >0 is the temperature control parameter, when When the value is large, the selection probability of all actions is close to uniform distribution; when When the value is small, the probability distribution is sharper and the probability of high-value actions being selected is higher. Setting the temperature parameter can make the decay over time more consistent with the actual situation, avoid tedious parameter trials, and effectively balance the relationship between exploration and utilization.

[0117] The above formula can also be written in logarithmic form:

[0118]

[0119] Among them, P (a i ) is to select the candidate energy a in state s i The probability of selection, a i is the welding energy that can be selected for the i-th option, Q(s, a i ) is the welding energy a taken under state s i The value of >0 is the temperature control parameter, is the initial temperature parameter, and n is the time step, i.e. the number of welding operations.

[0120] In some possible implementations, in the above steps, updating the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value includes:

[0121] (1) Determining a corresponding first value function based on the welding energy required for the current welding operation through a preset value prediction model;

[0122] (2) determining a corresponding second value function based on the welding energy required for the next welding operation through a preset value prediction model;

[0123] (3) Based on the first value function, the second value function and the reward value, update the value prediction model.

[0124] Specifically, the welding energy required for the current welding is obtained, and the energy corresponding to the next welding is predicted. The value function of the welding energy corresponding to the two welding operations is predicted through the value prediction model. Based on the first value function corresponding to the welding energy selected for the current welding operation, the second value function corresponding to the next welding operation, and the reward value, the welding energy and the welding power in the first welding state information are substituted into the first value function, and the predicted next welding energy and the welding power in the second welding state information are substituted into the second value function. The error value is obtained based on the first value function and the second value function, and the value prediction model is updated based on the error value.

[0125] In some possible implementations, in the above steps, updating the value prediction model based on the first value function, the second value function, and the reward value includes:

[0126] (1) determining a first evaluation value corresponding to the current welding operation based on the first value function and the welding parameters corresponding to the current welding operation;

[0127] (2) determining a second evaluation value corresponding to the next welding operation based on the second value function and the welding parameters corresponding to the next welding operation;

[0128] (3) Obtaining a target evaluation value based on the second evaluation value and the reward value;

[0129] (4) obtaining an error value based on the difference between the target evaluation value and the first evaluation value;

[0130] (5) Update the value prediction model based on the error value.

[0131] Specifically, the welding parameters corresponding to the current welding operation are substituted into the first value function to obtain the corresponding first evaluation value; the welding parameters corresponding to the next welding operation are substituted into the second value function to obtain the second evaluation value; based on the second evaluation value and the reward value, the target evaluation value is obtained; based on the difference between the target evaluation value and the first evaluation value, the error value is obtained; with the goal of minimizing the error value, the corresponding update parameters are obtained; and the value prediction model is updated based on the update parameters.

[0132] In the specific implementation process, the above steps can be implemented based on the SARSA algorithm (State-Action-Reward-State-Action), where:

[0133]

[0134] Where a is the welding energy currently used, is the learning rate, which determines how much new information affects existing estimates, The larger the value, the greater the impact of new information on the Q value; The smaller the value, the slower the update of Q value. R is the reward value for performing welding with welding energy a. is a discount factor that determines the importance of the reward of future welding operations relative to the current reward, Q(s t+1 , a t+1 ) is the next welding status information is s t+1 When welding energy a t+1 The estimated value of. In the above formula, Q(s, a) is the first value function, Q(s t+1 , a t+1) is the second value function, R is the reward value, and the welding parameters are substituted into The above-mentioned target evaluation value can be obtained.

[0135] In the specific implementation process, the difference between the target evaluation value and the first evaluation value (prediction value) can be expressed as:

[0136]

[0137] in, It is the TD error (Temporal-Difference Error). Specifically, the TD error can be expressed as follows:

[0138]

[0139] in, It is the TD error, which represents the difference between the value function predicted at the t-th welding operation and the actual value and reward value at the t+1-th welding operation. The goal is to minimize the error value, which can make the value function closer to the actual value.

[0140] In the specific implementation process, the value function can be parameterized. The update of the parameters is equivalent to the update of the value function. At this time, the approximate expression of the state-action value function is:

[0141]

[0142] Among them, w is the approximator parameter, and each parameter value determines a value function accordingly.

[0143] Specifically, updating the value prediction model can be regarded as updating w in the above formula, and performing gradient descent update on the above TD error to obtain the updated parameters. The update rule is:

[0144]

[0145] in, is the TD error, is the gradient of the above loss function with respect to w.

[0146] In some possible implementations, in step S213, welding the battery cell to be welded using welding energy to obtain corresponding second welding state information and a reward value includes:

[0147] (1) welding the battery cell to be welded using welding energy to obtain corresponding second welding state information;

[0148] (2) Determine the corresponding reward value based on the second welding state information through a preset reward function.

[0149] Specifically, the reward function may be pre-set, and the reward value corresponding to the welding energy obtained for each welding operation, based on the reward function, is different. The welding energy is used to weld the battery cell to be welded, and corresponding second welding state information is obtained. The corresponding reward value is determined based on the second welding state information using the preset reward function.

[0150] In the specific implementation process, Figure 4 As shown in the figure, the reward R is usually defined by the environment. The ultrasonic welder selects the corresponding welding energy A for welding based on the welding power S. The environment determines the reward R based on the welding energy A and the new state S', combined with the rules or goals of the welding environment, and the feedback of the welding quality. In reinforcement learning, the reward is set through environmental design. It provides positive and negative feedback based on the behavior and task goals of the agent to guide the agent to learn the correct strategy. The size and frequency of the reward need to be adjusted according to the characteristics of the task.

[0151] In some possible implementations, in step S214, if the welding status in the second welding status information is welding failure, then using a value prediction model and based on the second welding status information, predicting the welding energy required for the next welding operation from at least one candidate energy includes:

[0152] (1) If the welding state in the second welding state information is welding failure, the welding power in the second welding state information is used as an input of the value prediction model to obtain a value function corresponding to at least one candidate energy;

[0153] (2) Determine the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies;

[0154] (3) Selecting the candidate energy with the highest probability from the at least one candidate energy as the welding energy required for the next welding operation.

[0155] The second welding state information also includes welding state and welding power.

[0156] Specifically, if the welding status in the second welding status information is welding failure, the welding power in the second welding status information is used as the input of the value prediction model to obtain the value function corresponding to at least one candidate energy. Based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies, the selection probability corresponding to each candidate energy is determined, and the candidate energy with the highest probability is selected from at least one candidate energy as the welding energy required for the next welding operation.

[0157] In the above embodiment, at least one welding operation is performed on the battery cell to be welded, including: obtaining first welding state information corresponding to the current welding operation of the battery cell to be welded, including welding state and welding power, and obtaining a value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, using the welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value; if the welding state in the second welding state information is welding failure, then using the value prediction model, based on the second welding state information, predicting the welding energy required for the next welding operation from at least one candidate energy; based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, updating the value prediction model, using the second welding state information as the next first welding state information, and using the updated value prediction model as the value prediction model corresponding to the next welding operation; for each welding operation, the value prediction model is updated according to the feedback information corresponding to the welding operation, and parameter selection can be adjusted in time according to the feedback information, effectively improving welding quality and welding efficiency.

[0158] Furthermore, at least one candidate welding energy and the corresponding value function are predicted by using the value prediction model corresponding to the welding operation, and the selection probability of each candidate welding energy is determined based on the value function. Since the value prediction model will be adjusted according to the feedback information of the welding operation, the selection probability corresponding to each candidate energy will also change according to the adjusted value function. The selection probability obtained in this way can make the selected welding energy more accurate.

[0159] In one example, the battery core welding method of the present disclosure is as follows: Figure 5 As shown, this may include:

[0160] Obtain a battery cell to be welded, and perform at least one welding operation on the battery cell to be welded until the welding status of the battery cell to be welded is successful.

[0161] Among them, welding operations include:

[0162] Acquire first welding state information corresponding to a current welding operation of the battery cell to be welded; the first welding state information includes welding state and welding power;

[0163] Obtaining a value prediction model corresponding to the current welding operation, using the welding power in the first welding state information as an input to the value prediction model, and obtaining a value function corresponding to at least one candidate energy;

[0164] Determining first probability parameters corresponding to each candidate energy based on the value function and preset temperature parameters corresponding to each candidate energy; the preset temperature parameters are used to simulate actual temperature conditions, and the temperature parameters gradually decrease as the number of welding operations increases;

[0165] determining a selection probability corresponding to each candidate energy based on the first probability parameter corresponding to each candidate energy and the sum of the first probability parameters corresponding to the candidate energy;

[0166] Selecting a candidate energy with the highest probability from among the at least one candidate energy as the welding energy required for the current welding operation;

[0167] The welding energy is used to weld the battery cell to be welded, and corresponding second welding state information and a reward value are obtained.

[0168] If the welding status in the second welding status information is welding success, the welding operation on the battery cell to be welded is ended.

[0169] If the welding state in the second welding state information is welding failure, the welding power in the second welding state information is used as an input of the value prediction model to obtain a value function corresponding to at least one candidate energy;

[0170] Based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies, the selection probability corresponding to each candidate energy is determined; and the candidate energy with the highest probability is selected from at least one candidate energy as the welding energy required for the next welding operation.

[0171] Determine a corresponding first value function based on the welding energy required for the current welding operation using a preset value prediction model; determine a corresponding second value function based on the welding energy required for the next welding operation using a preset value prediction model;

[0172] Determining a first evaluation value corresponding to the current welding operation based on a first value function and welding parameters corresponding to the current welding operation; determining a second evaluation value corresponding to the next welding operation based on a second value function and welding parameters corresponding to the next welding operation; obtaining a target evaluation value based on the second evaluation value and a reward value; and obtaining an error value based on a difference between the target evaluation value and the first evaluation value to update the value prediction model;

[0173] The second welding state information is used as the first welding state information for the next time, and the updated value prediction model is used as the value prediction model corresponding to the next welding operation.

[0174] The above-mentioned battery cell welding method, by performing at least one welding operation on the battery cell to be welded, includes: obtaining first welding state information corresponding to the current welding operation of the battery cell to be welded, including welding state and welding power, and obtaining a value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, using the welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value; if the welding state in the second welding state information is welding failure, then using the value prediction model, based on the second welding state information, predicting the welding energy required for the next welding operation from at least one candidate energy; based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, updating the value prediction model, using the second welding state information as the next first welding state information, and using the updated value prediction model as the value prediction model corresponding to the next welding operation; for each welding operation, updating the value prediction model according to the feedback information corresponding to the welding operation, and being able to adjust parameter selection in time according to the feedback information, thereby effectively improving welding quality and welding efficiency.

[0175] Furthermore, at least one candidate welding energy and the corresponding value function are predicted by using the value prediction model corresponding to the welding operation, and the selection probability of each candidate welding energy is determined based on the value function. Since the value prediction model will be adjusted according to the feedback information of the welding operation, the selection probability corresponding to each candidate energy will also change according to the adjusted value function. The selection probability obtained in this way can make the selected welding energy more accurate.

[0176] The present disclosure provides a battery core welding device, such as Figure 6 As shown, the battery core welding device 60 may include: a battery core acquisition module 601 and a battery core welding module 602, wherein:

[0177] A cell acquisition module 601 is used to acquire cells to be welded;

[0178] The cell welding module 602 is configured to perform at least one welding operation on the cell to be welded until the welding state of the cell to be welded is successful;

[0179] The battery core welding module 602 is also used for:

[0180] Acquire first welding state information corresponding to a current welding operation of the battery cell to be welded; the first welding state information includes welding state and welding power;

[0181] Obtaining a value prediction model corresponding to a current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation;

[0182] Using welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value;

[0183] If the welding state in the second welding state information is welding failure, predicting the welding energy required for the next welding operation from at least one candidate energy based on the second welding state information using a value prediction model; and updating the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value.

[0184] The second welding state information is used as the first welding state information for the next time, and the updated value prediction model is used as the value prediction model corresponding to the next welding operation.

[0185] As an optional embodiment, in the device, the cell welding module 602 is specifically used to:

[0186] Obtaining a value prediction model corresponding to a current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation, including:

[0187] Obtaining a value prediction model corresponding to the current welding operation, using the welding power in the first welding state information as an input to the value prediction model, and obtaining a value function corresponding to at least one candidate energy;

[0188] Determine the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies;

[0189] The candidate energy with the highest probability is selected from the at least one candidate energy as the welding energy required for the current welding operation.

[0190] As an optional embodiment, in the device, the cell welding module 602 is specifically used to:

[0191] Based on the value functions corresponding to the candidate energies and the sum of the value functions of the candidate energies, the selection probability corresponding to the candidate energies is determined, including:

[0192] Determining first probability parameters corresponding to each candidate energy based on the value function and preset temperature parameters corresponding to each candidate energy; the preset temperature parameters are used to simulate actual temperature conditions, and the temperature parameters gradually decrease as the number of welding operations increases;

[0193] The selection probability corresponding to each candidate energy is determined based on the first probability parameter corresponding to each candidate energy and the sum of the first probability parameters corresponding to the candidate energy.

[0194] As an optional embodiment, in the device, the cell welding module 602 is specifically used to:

[0195] Based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value, the value prediction model is updated, including:

[0196] Determining a corresponding first value function based on the welding energy required for the current welding operation through a preset value prediction model;

[0197] Determining a corresponding second value function based on the welding energy required for the next welding operation using a preset value prediction model;

[0198] Based on the first value function, the second value function and the reward value, the value prediction model is updated.

[0199] As an optional embodiment, in the device, the cell welding module 602 is specifically used to:

[0200] Based on the first value function, the second value function and the reward value, the value prediction model is updated, including:

[0201] Determining a first evaluation value corresponding to the current welding operation based on the first value function and welding parameters corresponding to the current welding operation;

[0202] determining a second evaluation value corresponding to the next welding operation based on the second value function and welding parameters corresponding to the next welding operation; and obtaining a target evaluation value based on the second evaluation value and the reward value;

[0203] obtaining an error value based on a difference between the target evaluation value and the first evaluation value;

[0204] Update the value prediction model based on the error value.

[0205] As an optional embodiment, in the device, the cell welding module 602 is specifically used to:

[0206] The welding energy is used to weld the battery cell to be welded, and corresponding second welding state information and reward value are obtained, including:

[0207] Using welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information;

[0208] A corresponding reward value is determined based on the second welding state information through a preset reward function.

[0209] As an optional embodiment, in the device, the cell welding module 602 is specifically used to:

[0210] If the welding state in the second welding state information is welding failure, then using a value prediction model and based on the second welding state information, predicting welding energy required for a next welding operation from at least one candidate energy, including:

[0211] If the welding state in the second welding state information is welding failure, the welding power in the second welding state information is used as an input of the value prediction model to obtain a value function corresponding to at least one candidate energy;

[0212] Determine the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies;

[0213] A candidate energy with the highest probability is selected from the at least one candidate energy as the welding energy required for the next welding operation.

[0214] The battery cell welding device provided by the present disclosure performs at least one welding operation on the battery cell to be welded, including: obtaining first welding state information corresponding to the current welding operation of the battery cell to be welded, including welding state and welding power, and obtaining a value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, using the welding energy to weld the battery cell to be welded, and obtaining corresponding second welding state information and a reward value; if the welding state in the second welding state information is welding failure, then using the value prediction model, based on the second welding state information, predicting the welding energy required for the next welding operation from at least one candidate energy; based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, updating the value prediction model, using the second welding state information as the next first welding state information, and using the updated value prediction model as the value prediction model corresponding to the next welding operation; for each welding operation, updating the value prediction model according to the feedback information corresponding to the welding operation, and being able to adjust parameter selection in time according to the feedback information, thereby effectively improving welding quality and welding efficiency.

[0215] Furthermore, at least one candidate welding energy and the corresponding value function are predicted by using the value prediction model corresponding to the welding operation, and the selection probability of each candidate welding energy is determined based on the value function. Since the value prediction model will be adjusted according to the feedback information of the welding operation, the selection probability corresponding to each candidate energy will also change according to the adjusted value function. The selection probability obtained in this way can make the selected welding energy more accurate.

[0216] The apparatus of the embodiments of the present disclosure can execute the methods provided by the embodiments of the present disclosure, and their implementation principles are similar and have corresponding technical effects. The actions performed by each module in the apparatus of each embodiment of the present disclosure correspond to the steps in the methods of each embodiment of the present disclosure. For detailed functional descriptions of each module of the apparatus, please refer to the descriptions of the corresponding methods shown above, and will not be repeated here.

[0217] An embodiment of the present disclosure provides an electronic device (computer device / equipment / system), including a memory, a processor, and a computer program stored in the memory, and the processor executes the above computer program to implement the steps of the method provided in any optional embodiment of the present disclosure. Compared with the prior art, the present invention can achieve the following: by performing at least one welding operation on the battery cell to be welded, including: obtaining the first welding state information corresponding to the current welding operation of the battery cell to be welded, including the welding state and the welding power, and obtaining the value prediction model corresponding to the current welding operation, determining the welding energy required for the current welding operation from at least one candidate energy, using the welding energy to weld the battery cell to be welded, and obtaining the corresponding second welding state information and the reward value; if the welding state in the second welding state information is welding failure, then through the value prediction model, based on the second welding state information, predicting the welding energy required for the next welding operation from at least one candidate energy; based on the welding energy required for the current welding operation, the welding energy required for the next welding operation and the reward value, updating the value prediction model, using the second welding state information as the first welding state information for the next time, and using the updated value prediction model as the value prediction model corresponding to the next welding operation; for each welding operation, updating the value prediction model according to the feedback information corresponding to the welding operation, and being able to adjust the parameter selection in time according to the feedback information, and effectively improving the welding quality and welding efficiency.

[0218] In an alternative embodiment, an electronic device is provided, such as Figure 7 As shown, Figure 7 The electronic device 7000 shown includes: a processor 7001 and a memory 7003. The processor 7001 and the memory 7003 are connected, for example, via a bus 7002. Optionally, the electronic device 7000 may further include a transceiver 7004, which may be used for data exchange between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the number of transceivers 7004 is not limited to one, and the structure of the electronic device 7000 does not constitute a limitation on the embodiments of the present disclosure.

[0219] Processor 7001 can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic device, transistor logic device, hardware component, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the present disclosure. Processor 7001 can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0220] Bus 7002 may include a path for transmitting information between the above components. Bus 7002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. Bus 7002 may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0221] The memory 7003 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store computer programs and can be read by a computer, without limitation herein.

[0222] The memory 7003 is used to store the computer program for executing the embodiments of the present disclosure, and the execution is controlled by the processor 7001. The processor 7001 is used to execute the computer program stored in the memory 7003 to implement the steps shown in the above method embodiments.

[0223] The electronic equipment includes but is not limited to: a terminal or a server capable of adjusting the welding operation using the above-mentioned battery core welding method.

[0224] An embodiment of the present disclosure provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0225] It should be noted that the computer-readable storage medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0226] The embodiments of the present disclosure further provide a computer program product, including a computer program, which can implement the steps and corresponding contents of the aforementioned method embodiments when executed by a processor.

[0227] In the description and claims of the present disclosure and the accompanying drawings, the terms "first," "second," "third," "fourth," "1," "2," and the like (if any) are used to distinguish similar objects and are not necessarily used to describe a particular order or sequential sequence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present disclosure described herein can be practiced in an order other than that shown or described.

[0228] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0229] It should be understood that, although the flowcharts of the embodiments of the present disclosure indicate the various operation steps by arrows, the order of implementation of these steps is not limited to the order indicated by the arrows. Unless otherwise clearly stated herein, in some implementation scenarios of the embodiments of the present disclosure, the implementation steps in each flowchart can be performed in other orders as required. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages based on the actual implementation scenario. Some or all of these sub-steps or stages can be executed at the same time, and each sub-step or stage in these sub-steps or stages can also be executed at different times. In scenarios where the execution times are different, the order of execution of these sub-steps or stages can be flexibly configured as required, and the embodiments of the present disclosure do not limit this.

[0230] The above description is only an optional implementation method for some implementation scenarios of the present disclosure. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the solution of the present disclosure, other similar implementation methods based on the technical ideas of the present disclosure also fall within the protection scope of the embodiments of the present disclosure.

Claims

1. A battery core welding method, characterized in that: include: Get the battery cells to be welded; Performing at least one welding operation on the battery cell to be welded until the welding state of the battery cell to be welded is successful; The welding operation includes: Acquire first welding state information corresponding to a current welding operation of the battery cell to be welded; the first welding state information includes welding state and welding power; Obtaining a value prediction model corresponding to a current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation; Using the welding energy to weld the battery core to be welded, and obtaining corresponding second welding state information and a reward value; If the welding state in the second welding state information is welding failure, predicting the welding energy required for the next welding operation from at least one candidate energy based on the second welding state information using the value prediction model; and updating the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value; The second welding state information is used as the first welding state information for the next time, and the updated value prediction model is used as the value prediction model corresponding to the next welding operation.

2. The battery core welding method according to claim 1, characterized in that: The acquiring of a value prediction model corresponding to the current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation, includes: Obtaining a value prediction model corresponding to the current welding operation, using the welding power in the first welding state information as input to the value prediction model, and obtaining a value function corresponding to each of the at least one candidate energy; Determining a selection probability corresponding to each candidate energy based on the cost function corresponding to each candidate energy and the sum of the cost functions of the candidate energies; The candidate energy with the highest probability among the at least one candidate energy is selected as the welding energy required for the current welding operation.

3. The battery core welding method according to claim 2, characterized in that: The determining of the selection probability corresponding to each candidate energy based on the value function corresponding to each candidate energy and the sum of the value functions of the candidate energies includes: Determining first probability parameters corresponding to each candidate energy based on the value function and a preset temperature parameter corresponding to each candidate energy; the preset temperature parameter is used to simulate actual temperature conditions, and the temperature parameter gradually decreases as the number of welding operations increases; The selection probability corresponding to each candidate energy is determined based on the first probability parameter corresponding to each candidate energy and the sum of the first probability parameters corresponding to the candidate energy.

4. The battery core welding method according to claim 2, characterized in that: The updating of the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value includes: Determining a corresponding first value function based on the welding energy required for the current welding operation using a preset value prediction model; Determining a corresponding second value function based on the welding energy required for the next welding operation using a preset value prediction model; The value prediction model is updated based on the first value function, the second value function, and the reward value.

5. The battery core welding method according to claim 4, characterized in that: The updating of the value prediction model based on the first value function, the second value function, and the reward value includes: determining a first evaluation value corresponding to the current welding operation based on the first value function and welding parameters corresponding to the current welding operation; determining a second evaluation value corresponding to the next welding operation based on the second value function and welding parameters corresponding to the next welding operation; and obtaining a target evaluation value based on the second evaluation value and the reward value; obtaining an error value based on a difference between the target evaluation value and the first evaluation value; The value prediction model is updated based on the error value.

6. The battery core welding method according to claim 1, characterized in that: The step of welding the battery core to be welded using the welding energy to obtain corresponding second welding state information and a reward value includes: Using the welding energy to weld the battery core to be welded, and obtaining corresponding second welding state information; A corresponding reward value is determined based on the second welding state information through a preset reward function.

7. The battery core welding method according to claim 1, characterized in that: If the welding state in the second welding state information is welding failure, predicting the welding energy required for the next welding operation from at least one candidate energy based on the second welding state information by using the value prediction model includes: If the welding state in the second welding state information is welding failure, the welding power in the second welding state information is used as an input of the value prediction model to obtain a value function corresponding to the at least one candidate energy; Determining a selection probability corresponding to each candidate energy based on a value function corresponding to each candidate energy and a sum of the value functions of the candidate energies; A candidate energy with the highest probability is selected from the at least one candidate energy as the welding energy required for the next welding operation.

8. A battery core welding device, characterized in that: include: A cell acquisition module is used to obtain cells to be welded; A cell welding module, configured to perform at least one welding operation on the cell to be welded until the welding state of the cell to be welded is successful; The battery core welding module is also used for: Acquire first welding state information corresponding to a current welding operation of the battery cell to be welded; the first welding state information includes welding state and welding power; Obtaining a value prediction model corresponding to a current welding operation, and determining a welding energy required for the current welding operation from at least one candidate energy based on the first welding state information and the value prediction model corresponding to the current welding operation; Using the welding energy to weld the battery core to be welded, and obtaining corresponding second welding state information and a reward value; If the welding state in the second welding state information is welding failure, predicting the welding energy required for the next welding operation from at least one candidate energy based on the second welding state information using the value prediction model; and updating the value prediction model based on the welding energy required for the current welding operation, the welding energy required for the next welding operation, and the reward value; The second welding state information is used as the first welding state information for the next time, and the updated value prediction model is used as the value prediction model corresponding to the next welding operation.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the battery core welding method according to any one of claims 1 to 7 are implemented.

11. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images