Electrochemical directional cycle lithium extraction and deintercalation balance method, device, equipment and medium

Abstract

This invention discloses an electrochemical directional cycling method, apparatus, equipment, and medium for lithium extraction / deintercalation balancing. Based on operating parameters set for the electrolyzer, lithium extraction / deintercalation is performed, and the required deintercalation time for the lithium intercalation cell electrode plates is calculated. The operating parameters are input into a pre-trained lithium intercalation time calculation model, which outputs the lithium intercalation time corresponding to complete lithium intercalation. The time difference between the deintercalation and intercalation times is calculated, and this time difference, along with the operating parameters, is input into a pre-trained reducing agent calculation model, which outputs the required reducing agent dosage for the lithium intercalation cell electrode plates. The lithium intercalation and deintercalation balance is controlled based on the required reducing agent dosage. This method can balance the lithium extraction / deintercalation rates and improve lithium extraction efficiency.

Description

Technical Field

[0001] This invention relates to the field of battery materials technology, and more specifically, to a method, apparatus, equipment, and medium for lithium extraction and deintercalation equilibrium in electrochemical directional cycling. Background Technology

[0002] With the depletion of non-renewable energy sources, new energy vehicles, as a typical representative of new energy development and utilization, have experienced rapid growth in recent years. The development and utilization of new energy vehicles is an inevitable trend, and they will eventually surpass the market share of traditional fuel vehicles, gradually replacing them. Lithium, as an essential energy metal for the power systems of new energy vehicles, will also see a dramatic increase in market demand. The efficient, clean, and low-cost mining of lithium resources is crucial for the sustainable development of the new energy vehicle industry. With the depletion of high-quality lithium ore resources, high-grade ore that meets current processing standards is becoming increasingly scarce, and the cost of lithium extraction from ore is also relatively high. Compared to ore extraction, lithium resources in brine are abundant, and lithium exists in ionic form, giving it a natural cost advantage in lithium extraction.

[0003] Electrodialysis, as a green process technology, specifically utilizes electrochemical directional cyclic deintercalation and intercalation technology. At the anode, an oxidation reaction occurs, extracting lithium from lithium iron phosphate; at the cathode, a reduction reaction occurs, intercalating lithium from the brine into the lithium iron phosphate. Ideally, for every lithium ion extracted at the anode, one should be intercalated at the cathode, with both extraction and intercalation rates matching. However, in actual reactions, the cathode intercalation process is significantly affected by factors such as brine viscosity and ion concentration. This results in a much slower cathode intercalation process compared to the anode deintercalation, leading to a mismatch between the deintercalation and intercalation capacities of the anode and cathode. Furthermore, the two electrodes require continuous polarity switching, which, with repeated switching, gradually reduces the lithium absorption capacity and slows down the lithium extraction efficiency.

[0004] Therefore, it is urgent to calculate the time required for the material to be fully lithium-intercalated, balance the deintercalation rate based on the lithium extraction time, and solve the problem that the lithium absorption capacity gradually decreases and the lithium extraction efficiency becomes slower as the number of deintercalation and extraction cycles increases. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes an electrochemical directional cycling method, apparatus, equipment, and medium for balancing lithium extraction and deintercalation rates, thereby improving lithium extraction efficiency.

[0006] This invention provides an electrochemically oriented cycling method for lithium extraction / deintercalation equilibrium, the method comprising:

[0007] Based on the working parameters set for the electrolytic cell, lithium extraction and deintercalation are performed, and the required delithiation time for the electrode plate of the lithium intercalation cell is calculated.

[0008] The working index parameters are input into the pre-trained lithium intercalation time calculation model, and the lithium intercalation time corresponding to full lithium intercalation is output.

[0009] The time difference is calculated based on the delithiation time and the lithium insertion time. The time difference and the working index parameters are input into the pre-trained reducing agent calculation model, and the required reducing agent dosage for the lithium insertion slot electrode plate is output.

[0010] Control the lithium insertion and delithiation balance according to the required reducing agent dosage.

[0011] Preferably, the training process of the lithium intercalation time calculation model or the reducing agent calculation model specifically includes:

[0012] Obtain a training sample set including input data and target values;

[0013] The training sample set is input into the neural network recognition model to be trained for training. The model parameters are adjusted by gradient descent algorithm according to the preset training accuracy value, training speed and number of iterations. The trained model is used as the target model, which is specifically the lithium intercalation time calculation model or the reducing agent calculation model.

[0014] The number of input layer nodes, hidden layer nodes, and output layer nodes of the neural network recognition module are determined by the number of working index parameters in the input data.

[0015] As a preferred embodiment, before using the trained model as the target model, the method further includes:

[0016] The input data from the obtained verification sample set is input into the trained neural network recognition model to obtain the model output value corresponding to the input data;

[0017] Calculate the relative error between the model output value of the input data and the target value corresponding to the validation sample set;

[0018] The trained model is validated based on the calculated relative error, and the validated trained model is output.

[0019] Preferably, the step of validating the trained model based on the calculated relative error and outputting a validated trained model specifically includes:

[0020] The percentage of qualified input data in the verification sample set whose statistical relative error is within a first preset range.

[0021] When the statistically determined qualified rate is not within the second preset range, the trained model is deemed not to meet the requirements. The number of samples in the training sample set is increased, and the training sample set is input into the neural network recognition model for further training.

[0022] When the statistically determined qualified rate is within the second preset range, the trained model is deemed to meet the requirements, and the trained model is output.

[0023] Preferably, the step of inputting the training sample set into the neural network recognition model to be trained for training specifically includes:

[0024] A neural network recognition model is used to calculate the input data in the training sample set to obtain the output value of the model. The weights of each neuron connected in the neural network recognition model are corrected based on the error between the output value and the corresponding target value. The corrected weights are then passed to each layer of the neuron layer by layer using the backpropagation algorithm. The error is continuously corrected iteratively until the number of iterations reaches a preset value or the sum of squares of the error between the model's output value and the corresponding target value is within a preset range.

[0025] When the number of iterations reaches the preset value and the sum of squared errors between the model's output value and the corresponding target value is not within the preset range, the number of hidden layers of the neural network recognition model is increased, and the training sample set is re-inputted into the neural network recognition model to be trained for training until the sum of squared errors between the model's output value and the corresponding target value is within the preset range.

[0026] The model training is complete when the sum of the squared errors between the model's output value and the corresponding target value is within a preset range.

[0027] Preferably, the neural network recognition model employs a backpropagation neural network;

[0028] The neural network recognition model uses the Sigmoid function as the training function.

[0029] The neural network recognition model uses the tansig function as the hidden layer transfer function.

[0030] The neural network recognition model uses the Softmax function as the output layer transfer function.

[0031] Preferably, the operating parameters include temperature, lithium ion concentration in the brine, capacity of the lithium-absorbing material coated on the electrode plate, particle diffusion coefficient of the lithium-absorbing material, constant current supply current, and constant voltage supply voltage.

[0032] Preferably, when the target model is a lithium intercalation duration calculation model, the neural network recognition model has 6 nodes in the input layer, 5 nodes in the hidden layer, and 1 node in the output layer.

[0033] The process of obtaining the training sample set for the target model specifically includes:

[0034] Record the first index parameter used in each group of experiments and the corresponding complete lithium intercalation time when lithium extraction electrode plates of several different materials are used for single lithium extraction experiments.

[0035] The first index parameter recorded is used as the input data of the sample set, and the recorded full lithium intercalation time is used as the target value corresponding to the input data in the sample set.

[0036] The sample set is divided according to a preset ratio to obtain a training sample set and a validation sample set.

[0037] Furthermore, the lithium extraction electrode plates made of different materials include lithium titanate lithium extraction electrode plates, lithium phosphate lithium extraction electrode plates, lithium silicate lithium extraction electrode plates, and lithium manganese oxide lithium extraction electrode plates.

[0038] The first recorded parameters include temperature, lithium ion concentration in the brine, capacity of the lithium-absorbing material coated on the electrode plate, particle diffusion coefficient of the lithium-absorbing material, constant current supply current, and constant voltage supply voltage.

[0039] As a preferred embodiment, when the target model is the reducing agent calculation model, the neural network recognition model has 4 nodes in the input layer, 3 nodes in the hidden layer, and 1 node in the output layer.

[0040] The process of obtaining the training sample set for the target model specifically includes:

[0041] Record the second index parameter and corresponding reaction time used in each group of experiments when lithium extraction electrode plates with different reducing agent dosages are used for single lithium extraction experiments;

[0042] The recorded second indicator parameter and reaction time are used as input data for the sample set, and the recorded reducing agent dose is used as the target value corresponding to the input data in the sample set.

[0043] The sample set is divided according to a preset ratio to obtain a training sample set and a validation sample set.

[0044] Preferably, the recorded second index parameters include temperature, constant current supply current, and constant voltage supply voltage.

[0045] This invention also provides an electrochemically oriented cycling lithium extraction / deintercalation equilibrium device, the device comprising:

[0046] The data acquisition module is used to perform lithium extraction and deintercalation based on the working parameters set for the electrolyzer, and to calculate the delithiation time required for the electrode plates of the lithium intercalation cell.

[0047] The duration calculation module is used to input the working index parameters into the pre-trained lithium intercalation duration calculation model and output the lithium intercalation duration corresponding to full lithium intercalation.

[0048] The dosage calculation module is used to calculate the duration difference based on the delithiation duration and the lithium insertion duration, input the duration difference and the working index parameters into the pre-trained reducing agent calculation model, and output the reducing agent dosage required for the lithium insertion slot electrode plate.

[0049] The balance control module is used to control the lithium insertion and delithiation balance according to the required dosage of reducing agent.

[0050] This invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements an electrochemically oriented cycling lithium extraction-deintercalation equilibrium method as described in any of the above embodiments.

[0051] This invention also provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform an electrochemically oriented cycling lithium extraction / deintercalation equilibrium method as described in any of the preceding embodiments.

[0052] This invention provides an electrochemically oriented cyclic lithium extraction / deintercalation balancing method, apparatus, equipment, and medium. Based on the operating parameters set for the electrolyzer, lithium extraction / deintercalation is performed, and the required deintercalation time for the lithium intercalation cell electrode plates is calculated. The operating parameters are input into a pre-trained lithium intercalation time calculation model, which outputs the lithium intercalation time corresponding to complete lithium intercalation. The time difference between the deintercalation and intercalation times is calculated, and this time difference, along with the operating parameters, is input into a pre-trained reducing agent calculation model, which outputs the required reducing agent dosage for the lithium intercalation cell electrode plates. The lithium intercalation and deintercalation balance is controlled based on the required reducing agent dosage. This method can balance the lithium extraction / deintercalation rates and improve lithium extraction efficiency. Attached Figure Description

[0053] Figure 1 This is a schematic flowchart of an electrochemical directional cycling lithium extraction and deintercalation equilibrium method provided in an embodiment of the present invention;

[0054] Figure 2 This is a schematic diagram of the target model training process provided in an embodiment of the present invention;

[0055] Figure 3 This is a schematic diagram of the model verification process provided in an embodiment of the present invention;

[0056] Figure 4This is a schematic diagram of the structure of the neural network recognition model provided in an embodiment of the present invention;

[0057] Figure 5 This is a schematic diagram of the structure of an electrochemically oriented cyclic lithium extraction / deintercalation equilibrium device provided in an embodiment of the present invention;

[0058] Figure 6 This is a schematic diagram of the structure of a terminal device provided in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] Example 1

[0061] See Figure 1 This is a schematic flowchart of an electrochemical directional cycling lithium extraction and deintercalation equilibrium method provided in an embodiment of the present invention, the method including steps S1 to S4;

[0062] S1, based on the working parameters set for the electrolytic cell, perform lithium extraction and deintercalation, and calculate the delithiation time required for the electrode plate of the lithium intercalation cell;

[0063] S2, input the working index parameters into the pre-trained lithium intercalation time calculation model, and output the lithium intercalation time corresponding to full lithium intercalation;

[0064] S3, calculate the time difference based on the delithiation time and the lithium insertion time, input the time difference and the working index parameters into the pre-trained reducing agent calculation model, and output the reducing agent dosage required for the lithium insertion slot electrode plate;

[0065] S4 controls the lithium insertion and delithiation balance based on the required amount of reducing agent.

[0066] In the specific implementation of this embodiment, it is necessary to pre-train the lithium intercalation time calculation model and the reducing agent calculation model. The two models can be constructed using neural networks and trained using training data.

[0067] The input to the lithium intercalation time calculation model is the operating parameters of the electrolyzer, and the output is the lithium intercalation time corresponding to full lithium intercalation.

[0068] The input to the reducing agent calculation model is the operating parameters of the electrolyzer and the reduction time, and the output is the amount of reducing agent required for the lithium intercalation cell electrode plate;

[0069] It should be noted that the construction and training processes of the lithium intercalation duration calculation model and the reducing agent calculation model can employ model training methods, including machine learning, neural network learning, and other model training methods.

[0070] The reduction time of the reducing agent calculation model is obtained by the difference between the delithiation time required by the lithium intercalation cell electrode plate and the lithium intercalation time corresponding to complete lithium intercalation. This difference is caused by the imbalance between the delithiation rate and the lithium intercalation rate. Therefore, it is necessary to add a reducing agent to balance the delithiation rate and the lithium intercalation rate, so as to solve the problem that the lithium absorption capacity gradually decreases as the number of delithiation and intercalation increases, resulting in a slower and slower lithium extraction efficiency, and thus improve the lithium extraction efficiency.

[0071] In actual implementation, during normal deintercalation in the electrolytic cell, lithium extraction and deintercalation are performed in the electrolytic cell based on the aforementioned working index parameters, and the required deintercalation time t2 for the electrode plate of the lithium intercalation cell is calculated.

[0072] Obtain the working parameters of the electrolyzer, input the working parameters into the pre-trained lithium intercalation time calculation model for calculation, and obtain the lithium intercalation time t1 required for complete lithium intercalation;

[0073] The time difference Δt is calculated based on the delithiation time t2 and the lithium insertion time t1. Δt = t1 - t2. The time difference is the reduction time of the reducing agent that needs to be added to the electrolytic cell.

[0074] By inputting the electrolytic cell's operating parameters and the time difference Δt into the reducing agent calculation model, the required dosage of reducing agent for the lithium intercalation cell electrode plate can be obtained.

[0075] Add reducing agent to the electrolytic cell according to the dosage of reducing agent to ensure that the lithium insertion rate and the lithium extraction rate are balanced, so as to avoid the problem of increasingly slower lithium extraction efficiency due to inconsistent insertion and extraction rates, thereby improving lithium extraction efficiency.

[0076] Example 2

[0077] In another embodiment of the present invention, the training process of the lithium intercalation time calculation model and the reducing agent calculation model is similar, both employing the target model training method, specifically including:

[0078] A neural network recognition model is used as the target model to be trained.

[0079] Obtain a training sample set containing input data and target values. The training sample set contains several sets of samples containing input data and corresponding target values.

[0080] The training sample set is input into the neural network recognition model to be trained. The model parameters are adjusted using the gradient descent algorithm according to the preset training accuracy, training speed and number of iterations. The model is trained using the input data and the corresponding target value. The neural network recognition model approximates a function, and the output of the model is linked to the input to obtain the correspondence between the input data and the target value.

[0081] By constructing the model, we can establish the correspondence between working index parameters and lithium intercalation time, the correspondence between time difference and working index parameters and reducing agent dosage. Combined with the high-speed computing power of the computer, we can quickly and accurately obtain the lithium intercalation time and reducing agent dosage based on the model input.

[0082] The trained model is used as the target model.

[0083] It should be noted that the specific model training process and model architecture can adopt convolutional neural networks, recurrent neural networks, etc., and the model training methods can utilize machine learning, transfer learning, and reinforcement learning.

[0084] Preferably, Mini-batch, SGD, AdaGrad, momentum, RMSprop, adam, and learning rate decay methods can be used to improve model training speed; regularization, drop-out, early-stopping, and batch-norm can be applied by changing the number of features, the number of training samples, the model depth, the weights of a certain layer of the model, etc.

[0085] It should be noted that the number of input layer nodes, hidden layer nodes, and output layer nodes of the neural network recognition module are determined based on the number of working indicator parameters to improve model accuracy.

[0086] Example 3

[0087] In yet another embodiment provided by the present invention, see Figure 2 This is a schematic diagram of the target model training process provided in an embodiment of the present invention. The method includes:

[0088] Step S201: Obtain a training sample set including input data and target values;

[0089] Step S202: Input the training sample set into the neural network recognition model to be trained for training;

[0090] Step S203: Obtain a verification sample set including input data and target values. The verification sample set contains several sets of samples containing input data and corresponding target values.

[0091] Step S204: Validate the trained neural network recognition model based on the validation sample set;

[0092] Step S205: Determine whether the verification passed;

[0093] If not, return to step S201;

[0094] If so, proceed to step S206;

[0095] S206, Output the validated training model;

[0096] S207, The trained model is used as the target model.

[0097] Preferably, the trained neural network recognition model is validated based on a validation sample set. The specific validation process includes:

[0098] The input data from the obtained verification sample set is input into the trained neural network recognition model to obtain the model output value t3 corresponding to the input data;

[0099] Calculate the error Δt2 between the model output value t3 of the input data and the target value t4 corresponding to the validation sample set. Δt2 = |t3 - t4|. The relative error is the ratio of the error to the target value, i.e., the relative error Δ = |t3 - t4| / t4.

[0100] The relative error magnitude can be used to determine whether the trained neural network recognition model is qualified.

[0101] Even if the model is unqualified, continue model training;

[0102] Under the condition that the model is qualified, the output model is used as the lithium intercalation time calculation model or the reducing agent calculation model.

[0103] The model trained on the training sample set is validated by a validation sample set. The model that passes the accuracy validation is output as the target model, thereby improving the accuracy of the model.

[0104] Example 4

[0105] In yet another embodiment provided by the present invention, see Figure 3 This is a schematic diagram of the model verification process provided in an embodiment of the present invention; the model verification process includes the following steps:

[0106] Step S301: Calculate the relative error between the model output value of the input data and the target value corresponding to the verification sample set, that is, calculate the error Δt2 between the model output value t3 of the input data and the target value t4 corresponding to the verification sample set, Δt2=|t3-t4|, and the relative error is the ratio of the error to the target value, that is, the relative error Δ=|t3-t4| / t4;

[0107] Step S302: Determine whether a certain relative error is within a preset range, that is, determine whether the relative error of a certain sample is greater than 3%.

[0108] If so, proceed to step S303;

[0109] If not, proceed to step S304;

[0110] Step S303: Determine the sample as a non-conforming sample and proceed to step S305; that is, if the relative error of a sample is greater than 3%, then the sample is determined to be a non-conforming sample.

[0111] Step S304: Determine the sample as a qualified sample and execute step S304; that is, if the relative error of a sample is not greater than 3%, then the sample is determined to be a qualified sample.

[0112] Step S305: Calculate the pass rate, that is, the proportion of pass samples in all samples of the verification sample set.

[0113] Step S306: Determine whether the pass rate is within the second preset range, that is, determine whether the pass rate is not less than 95%.

[0114] If not, proceed to step S307;

[0115] If so, proceed to step S308;

[0116] Step S307: Determine that the trained model does not meet the requirements, and proceed to step S309;

[0117] Step S308: Determine if the trained model meets the requirements, and output the trained model.

[0118] S309, The sample set with the increased number of samples is input into the neural network recognition model for further training, and step S202 in the previous embodiment is executed;

[0119] It should be noted that in this embodiment, the first preset range is [0, 0.03]. In other embodiments, the first preset range can be set to other values.

[0120] It should be noted that in this embodiment, the second preset range is [0.95, 1]. In other embodiments, the second preset range can be set to other values.

[0121] The 30 sets of data that were not used in the training were used to form a validation sample set. The pass rate of qualified samples in the validation sample set was calculated. Based on the pass rate, it was determined whether the trained model met the requirements. This can determine the overall accuracy of the trained model under multiple sample data and avoid the existence of sample error.

[0122] Example 5

[0123] In another embodiment of the present invention, the training process in the neural network recognition model specifically includes:

[0124] First, 70 randomly selected training sample sets are input, and a neural network recognition model is used to calculate the input data to obtain the output value of the model. The purpose of neural network learning is to continuously correct the weights of each neuron connected inside the neural network by the error between the actual output value of the neural network and the target value of the inverted object, and to approach the target by continuously reducing the error.

[0125] Meanwhile, the backpropagation algorithm is used to pass the corrected weights layer by layer to each layer of the neuron, continuously iterating and correcting the error, so that the sum of squared errors is continuously reduced.

[0126] After the backpropagation algorithm reaches the preset maximum number of iterations, it is determined whether the sum of squared errors between the model's output value and the corresponding target value is within a preset range.

[0127] If not, it indicates that after the preset upper limit of the number of iterations, the sum of squared errors between the model output value and the corresponding target value is not within the preset range, and the model accuracy still does not meet the requirements. At this time, the number of hidden layers of the neural network recognition model is increased, and the training sample set in the sample set is re-input into the neural network recognition model to be trained for training. The sum of squared errors is re-judged with respect to the preset range until the sum of squared errors between the model output value and the corresponding target value is within the preset range.

[0128] The model training is complete when the sum of the squared errors between the model's output value and the corresponding target value is within a preset range.

[0129] If the sum of squared errors still does not meet the requirements after reaching the specified number of iterations during model training, the number of hidden layers is increased and the network is trained repeatedly until the sum of squared errors between the model's output value and the corresponding target value is within the preset range, the model reaches the preset accuracy requirement, and the model training is completed.

[0130] Example 6

[0131] In yet another embodiment provided by the present invention, see Figure 4 This is a schematic diagram of the structure of the neural network recognition model provided in the embodiment of the present invention;

[0132] The neural network recognition model uses a backpropagation neural network. The computational principle of the backpropagation neural network model is to train the model by function approximation. It has a strong nonlinear mapping capability and a flexible network structure. Moreover, the theory of this network is relatively mature at present and it is suitable for the construction of multi-factor input models.

[0133] Figure 3 The layer consists of an input layer (Layer L1), a hidden layer (Layer L2), and an output layer (Layer L3). The input data x1, x2, and x3 from the input layer are transmitted through the neurons in the hidden layer to obtain activation values ​​a1, respectively. (2) a2 (2) and a3 (2) Then it is output from the output layer.

[0134] In constructing the neural network recognition model, the Sigmoid function is used as the training function, the tansig function is used as the hidden layer transfer function, and the purelin function is used as the output layer transfer function.

[0135] The expression for the Sigmoid function is: By using hidden layers and the Sigmoid activation function, it is possible to complete the learning of non-linearly separable neural network recognition models.

[0136] Where x is the number of nodes and C is the number of output nodes, the output values ​​of multi-class classification can be converted into a probability distribution in the range of [0, 1] and 1 by using the Softmax function.

[0137] Example 7

[0138] In another embodiment of the present invention, during the electrochemical lithium extraction process, the factors affecting the time required for complete lithium insertion of the electrode plate include temperature T, lithium ion concentration C in the brine, capacity S of the lithium-absorbing material coated on the electrode plate, ion diffusion coefficient f of the lithium-absorbing material, constant current supply I, and constant voltage supply V. By obtaining these influencing factors as working index parameters, precise control of lithium extraction and deintercalation in electrochemical directional cycling can be achieved.

[0139] Example 8

[0140] In another embodiment of the present invention, when constructing the lithium intercalation duration calculation model, the input layer of the neural network recognition model is set with 6 nodes, the hidden layer with 5 nodes, and the output layer with 1 node; the process of obtaining the training sample set specifically includes:

[0141] When conducting single lithium extraction experiments using lithium extraction electrode plates made of several different materials, the index parameters used in each group of experiments and the corresponding complete lithium intercalation time are recorded. The recorded first index parameter is used as the input data of the sample set, and the recorded complete lithium intercalation time is used as the target value corresponding to the input data in the sample set. A total of 100 sets of sample data are collected through the experiment. 70 sets are randomly selected from the sample set to form a training sample set, and the remaining 30 sets of sample data are used as a validation sample set.

[0142] By using experimental results of lithium extraction electrode plates made of different materials under different index parameters as a sample set, the data breadth of the sample set can be increased, which can improve the calculation accuracy of the model under lithium extraction electrode plates made of different materials.

[0143] Example 9

[0144] In another embodiment of the present invention, 25 groups of lithium extraction electrode plates containing lithium titanate, lithium phosphate, lithium silicate, and lithium manganese oxide materials were selected for single lithium extraction experiments. During the experiments, parameters such as temperature T, lithium ion concentration C in the brine, capacity S of the lithium-absorbing material coated on the electrode plate, ion diffusion coefficient f of the lithium-absorbing material, constant current I, and constant voltage V were recorded for each group during the lithium extraction process. The complete lithium intercalation time was also recorded. A sample set was formed with temperature T, lithium ion concentration C in the brine, capacity S of the lithium-absorbing material coated on the electrode plate, ion diffusion coefficient f of the lithium-absorbing material, current I, and voltage V as inputs and complete lithium intercalation time as output. The recorded sample set was used to train the lithium intercalation time calculation model.

[0145] Example 10

[0146] In another embodiment of the present invention, when the target model is the reducing agent calculation model, the neural network recognition model has 6 nodes in the input layer, 5 nodes in the hidden layer, and 1 node in the output layer. The process of obtaining the training sample set specifically includes:

[0147] One hundred lithium extraction experiments were conducted on lithium extraction electrode plates with different reducing agent dosages. The 100 experiments were divided into groups with dosages of 0.1 mol, 0.2 mol, 0.3 mol…10 mol. The index parameters and corresponding reaction times for each group were recorded when different reducing agent dosages were used. The recorded index parameters and reaction times were used as input data for the sample set, and the recorded reducing agent dosage was used as the target value corresponding to the input data in the sample set. A total of 100 sample sets were collected through the experiments. 70 sets were randomly selected from the sample set to form a training sample set, and the remaining 30 sets were used as a validation sample set.

[0148] By using experimental results of lithium extraction electrode plates with different reducing agent dosages under different index parameters as a sample set, the data breadth of the sample set can be increased, which can improve the calculation accuracy of the model under lithium extraction electrode plates of different materials.

[0149] In the specific implementation of the present invention, the following parameters are obtained: temperature T, lithium ion concentration C in the brine, capacity S of the lithium-absorbing material coated on the electrode plate, ion diffusion coefficient f of the lithium-absorbing material, constant current supply I, and constant voltage supply V. Specific parameters are shown in Table 1.

[0150] Table 1 Work Indicator Parameters

[0151] index Temperature C Voltage V Current I diffusion coefficient f Capacity S <![CDATA[Li in brine + concentration]]> parameter 28℃ 0.3v 20A 2.2cm / s 278mAh / g 0.2g / L

[0152] The lithium insertion time calculation model calculated the complete lithium insertion time of the electrode plate to be t1 = 6.4h, the required delithiation time of the electrode plate in the lithium insertion tank to be t2 = 4.9h, the calculated time difference Δt = 1.5h, and the required amount of reducing agent to be 2.13mol was calculated by the reducing agent calculation model.

[0153] This invention addresses the problem of existing electrochemical lithium extraction technologies where the lithium insertion rate is slower than the lithium removal rate, leading to incomplete lithium insertion and a decreasing capacity and efficiency of the lithium insertion electrode plate after multiple lithium extraction operations. It proposes an electrochemical directional cycling method for lithium extraction and deintercalation balance. Through model calculations, the deintercalation time of the deintercalation electrode plate in the lithium extraction cell is quantified to achieve complete lithium insertion, allowing for targeted adjustments to achieve relative balance and improve lithium extraction efficiency.

[0154] Example 11

[0155] In another embodiment of the present invention, 100 groups of reducing agents are used for experiments. The recorded index parameters include the temperature T, constant current supply I, and constant voltage supply V during the experimental reaction process. The second index parameter, which includes the temperature T, constant current supply I, and constant voltage supply V, and the reaction time of the experiment are used as the model input, and the reducing agent dosage is used as the model output to train the reducing agent dosage calculation model.

[0156] Example 12

[0157] See Figure 5 This is a schematic diagram of an electrochemically oriented cycling lithium extraction / deintercalation balancing device provided in an embodiment of the present invention. The device includes: a data acquisition module, a duration calculation module, a dosage calculation module, and a balancing control module.

[0158] The data acquisition module is used to perform lithium extraction and deintercalation based on the working index parameters set for the electrolytic cell, and to calculate the delithiation time required for the electrode plate of the lithium intercalation cell.

[0159] The duration calculation module is used to input the working index parameters into the pre-trained lithium intercalation duration calculation model and output the lithium intercalation duration corresponding to full lithium intercalation.

[0160] The dosage calculation module is used to calculate the duration difference based on the delithiation duration and the lithium insertion duration, input the duration difference and the working index parameters into the pre-trained reducing agent calculation model, and output the reducing agent dosage required for the lithium insertion slot electrode plate.

[0161] The balance control module is used to control the lithium insertion and delithiation balance according to the required reducing agent dosage.

[0162] The electrochemically oriented cycling lithium extraction and deintercalation balancing device provided in this embodiment can perform all the steps and functions of the electrochemically oriented cycling lithium extraction and deintercalation balancing method provided in any of the above embodiments. The specific functions of the device will not be described in detail here.

[0163] Example 13

[0164] See Figure 6 This is a schematic diagram of a terminal device provided in an embodiment of the present invention. The terminal device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an electrochemically directed cycling lithium extraction / deintercalation balancing program. When the processor executes the computer program, it implements the steps in the various embodiments of the electrochemically directed cycling lithium extraction / deintercalation balancing method described above, for example... Figure 1 The steps S1 to S4 are shown. Alternatively, when the processor executes the computer program, it implements the functions of each module in the above-described device embodiments.

[0165] For example, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the electrochemically directed cycling lithium extraction / deintercalation balancing device. For example, the computer program can be divided into several modules, the specific functions of which have been described in detail in the electrochemically directed cycling lithium extraction / deintercalation balancing method provided in any of the above embodiments; therefore, the specific functions of the device will not be repeated here.

[0166] The electrochemically oriented cycling lithium extraction / deintercalation balancing device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the schematic diagram is merely an example of an electrochemically oriented cycling lithium extraction / deintercalation balancing device and does not constitute a limitation on such a device. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, the device may also include input / output devices, network access devices, buses, etc.

[0167] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electrochemically oriented cycling lithium extraction / deintercalation balancing device, connecting various parts of the device via various interfaces and lines.

[0168] The memory can be used to store the computer program and / or modules. The processor, by running or executing the computer program and / or modules stored in the memory and calling the data stored in the memory, realizes various functions of the electrochemically oriented cycling lithium extraction and deintercalation balancing device. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0169] The integrated module of the electrochemical directional cycling lithium extraction / deintercalation balancing device, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0170] It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications are also considered to be within the scope of protection of this invention.

Claims

1. A method for lithium extraction / deintercalation equilibrium through electrochemical directional cycling, characterized in that, The method includes: Based on the working parameters set for the electrolytic cell, lithium extraction and deintercalation are performed, and the required delithiation time for the electrode plate of the lithium intercalation cell is calculated. The working index parameters are input into the pre-trained lithium intercalation time calculation model, and the lithium intercalation time corresponding to full lithium intercalation is output. The time difference is calculated based on the delithiation time and the lithium insertion time. The time difference and the working index parameters are input into the pre-trained reducing agent calculation model, and the required reducing agent dosage for the lithium insertion slot electrode plate is output. The lithium insertion and delithiation balance is controlled according to the required dosage of reducing agent.

2. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 1, characterized in that, The training process of the lithium intercalation time calculation model or the reducing agent calculation model specifically includes: Obtain a training sample set including input data and target values; The training sample set is input into the neural network recognition model to be trained. The model parameters are adjusted by gradient descent algorithm according to the preset training accuracy value, training speed and number of iterations. The trained model is used as the target model, specifically the lithium intercalation time calculation model or the reducing agent calculation model; The number of input layer nodes, hidden layer nodes, and output layer nodes of the neural network recognition model are determined by the number of working index parameters in the input data.

3. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 2, characterized in that, Before using the trained model as the target model, the method further includes: The input data from the obtained verification sample set is input into the trained neural network recognition model to obtain the model output value corresponding to the input data; Calculate the relative error between the model output value of the input data and the target value corresponding to the validation sample set; The trained model is validated based on the calculated relative error, and the validated trained model is output.

4. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 3, characterized in that, The process of validating the trained model based on the calculated relative error and outputting a valid trained model includes: The percentage of qualified input data in the verification sample set whose statistical relative error is within a first preset range. When the statistically determined qualified rate is not within the second preset range, the trained model is deemed not to meet the requirements. The number of samples in the training sample set is increased, and the training sample set is input into the neural network recognition model for further training. When the statistically determined qualified rate is within the second preset range, the trained model is deemed to meet the requirements, and the trained model is output.

5. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 2, characterized in that, The step of inputting the training sample set into the neural network recognition model to be trained for training specifically includes: A neural network recognition model is used to calculate the input data in the training sample set to obtain the output value of the model. The weights of each neuron connected in the neural network recognition model are corrected based on the error between the output value and the corresponding target value. The corrected weights are then passed to each layer of the neuron layer by layer using the backpropagation algorithm. The error is continuously corrected iteratively until the number of iterations reaches a preset value or the sum of squares of the error between the model's output value and the corresponding target value is within a preset range. When the number of iterations reaches the preset value and the sum of squared errors between the model's output value and the corresponding target value is not within the preset range, the number of hidden layers of the neural network recognition model is increased, and the training sample set is re-inputted into the neural network recognition model to be trained for training until the sum of squared errors between the model's output value and the corresponding target value is within the preset range. The model training is complete when the sum of the squared errors between the model's output value and the corresponding target value is within a preset range.

6. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 2, characterized in that, The neural network recognition model employs a backpropagation neural network. The neural network recognition model uses the Sigmoid function as the training function. The neural network recognition model uses the tansig function as the hidden layer transfer function. The neural network recognition model uses the purelin function as the output layer transfer function.

7. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 1, characterized in that, The operating parameters include temperature, lithium ion concentration in the brine, capacity of the lithium-absorbing material coated on the electrode plate, particle diffusion coefficient of the lithium-absorbing material, constant current supply current, and constant voltage supply voltage.

8. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 2, characterized in that, When the target model is a lithium intercalation duration calculation model, the neural network recognition model has 6 nodes in the input layer, 5 nodes in the hidden layer, and 1 node in the output layer. The process of obtaining the training sample set for the target model specifically includes: Record the first index parameter used in each group of experiments and the corresponding complete lithium intercalation time when lithium extraction electrode plates of several different materials are used for single lithium extraction experiments. The first index parameter recorded is used as the input data of the sample set, and the recorded full lithium intercalation time is used as the target value corresponding to the input data in the sample set. The sample set is divided according to a preset ratio to obtain a training sample set and a validation sample set.

9. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 8, characterized in that, Lithium extraction electrode plates made of different materials include lithium titanate lithium extraction electrode plates, lithium phosphate lithium extraction electrode plates, lithium silicate lithium extraction electrode plates, and lithium manganese oxide lithium extraction electrode plates. The first recorded parameters include temperature, lithium ion concentration in the brine, capacity of the lithium-absorbing material coated on the electrode plate, particle diffusion coefficient of the lithium-absorbing material, constant current supply current, and constant voltage supply voltage.

10. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 2, characterized in that, When the target model is the reducing agent calculation model, the neural network recognition model has 4 nodes in the input layer, 3 nodes in the hidden layer, and 1 node in the output layer. The process of obtaining the training sample set for the target model specifically includes: Record the second index parameter and corresponding reaction time used in each group of experiments when lithium extraction electrode plates with different reducing agent dosages are used for single lithium extraction experiments; The recorded second indicator parameter and reaction time are used as input data for the sample set, and the recorded reducing agent dose is used as the target value corresponding to the input data in the sample set. The sample set is divided according to a preset ratio to obtain a training sample set and a validation sample set.

11. The electrochemical directional cycling method for lithium extraction / deintercalation equilibrium according to claim 10, characterized in that, The second set of recorded parameters includes temperature, constant current supply, and constant voltage supply.

12. An electrochemically directional cycling lithium extraction / deintercalation equilibrium device, characterized in that, The device includes: The data acquisition module is used to perform lithium extraction and deintercalation based on the working parameters set for the electrolyzer, and to calculate the delithiation time required for the electrode plates of the lithium intercalation cell. The duration calculation module is used to input the working index parameters into the pre-trained lithium intercalation duration calculation model and output the lithium intercalation duration corresponding to full lithium intercalation. The dosage calculation module is used to calculate the duration difference based on the delithiation time and the lithium insertion time, input the duration difference and the working index parameters into the pre-trained reducing agent calculation model, and output the reducing agent dosage required for the lithium insertion slot electrode plate. The balance control module is used to control the lithium insertion and delithiation balance according to the required dosage of reducing agent.

13. A terminal device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the electrochemically oriented cycling lithium extraction / deintercalation equilibrium method as described in any one of claims 1 to 11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the electrochemically oriented cycling lithium extraction / deintercalation equilibrium method as described in any one of claims 1 to 11.

Images