Intelligent oven control system and method based on deep learning and computational fluid dynamics

Abstract

The invention discloses an intelligent oven control system and method based on deep learning and computational fluid dynamics, and the system comprises a deep learning model which is used for predicting the operation conditions of an oven based on the pole piece drying requirements; the calculation fluid dynamic model is used for simulating fluid flow and heat transfer in the drying oven based on the prediction result of the deep learning model and the geometrical shape and material attributes of the drying oven, obtaining airflow distribution, temperature distribution, humidity distribution and pole piece moisture content in the drying oven and verifying the prediction result based on the deep learning model; the method is used for adjusting oven operation conditions in real time by adopting a model prediction control algorithm according to a set drying process target. The intelligent oven control system provided by the invention can adjust the operating conditions of the oven in time, so that the pole piece is uniformly heated, and the optimal drying performance, energy efficiency and product quality are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of lithium battery preparation, and in particular to an intelligent oven control system and method based on deep learning and computational fluid dynamics. Background Art

[0002] In the lithium battery cell manufacturing process, trace moisture has an important impact on the performance of lithium-ion battery electrodes, and baking can significantly remove moisture from the pole pieces. The traditional coating and baking process is to quickly pass the pole pieces through a hot air oven, which has problems such as high energy consumption, uneven heat source in the oven, insufficient hot air circulation, and insufficient temperature control accuracy, resulting in uneven temperature of the pole pieces, residual moisture and impurities, affecting the safety and reliability of lithium batteries. Moreover, uneven temperature leads to different evaporation rates of water, and the binder in the slurry moves, affecting the performance and consistency of lithium batteries. In addition, the oven cannot detect the temperature and humidity of the battery cells in real time, and cannot detect and solve problems in time, affecting the quality and production efficiency of the battery. These problems limit the production efficiency and quality of lithium batteries. The development of a new coating oven system is of great significance to improving the production efficiency and quality of lithium batteries. Summary of the invention

[0003] Based on the technical problems existing in the background technology, the present invention proposes an intelligent oven control system and method based on deep learning and computational fluid dynamics. The control system predicts the oven operating conditions based on the deep learning algorithm, and simulates the fluid flow and heat transfer in the oven based on computational fluid dynamics. It can adjust the operating conditions of the oven in real time to make the pole pieces heated evenly and optimize energy consumption.

[0004] To achieve the above object, the technical solution adopted by the present invention is as follows: In one aspect, the present invention provides an intelligent oven control system based on deep learning and computational fluid dynamics, comprising: A sensor system that detects the temperature, humidity, and air flow inside the oven and feeds them into the deep learning model and computational fluid dynamics model; A deep learning model is used to predict the oven operating conditions based on the electrode drying requirements and output to the computational fluid dynamics model; A computational fluid dynamics model is used to simulate the fluid flow and heat transfer in the oven based on the prediction results of the deep learning model and the geometric shape and material properties of the oven, obtain the airflow distribution, temperature distribution, humidity distribution and moisture content of the pole piece in the oven, verify the prediction results based on the deep learning model, and output the oven operating conditions that meet the requirements; The output module is used to adjust the oven operating conditions in real time and output them based on the drying process objectives and the output results of the computational fluid dynamics model.

[0005] Preferably, the sensor system includes a temperature sensor, a humidity sensor and an air flow sensor for detecting the temperature, humidity and air flow inside the oven; The sensor system is arranged around the pole piece, inside the pole piece and at the oven air inlet.

[0006] In this preferred solution, by setting sensors around the pole piece, inside the pole piece and at the air inlet of the oven, the measurement data is made more comprehensive, providing data support for subsequent deep learning models and computational fluid dynamics models.

[0007] Preferably, the deep learning model is trained based on historical data. The training process of the deep learning model is as follows: Collect the historical data of oven and electrode drying, including oven temperature, humidity, air flow, initial humidity of the electrode, moisture content of the electrode after drying, and electrode uniformity, and normalize the data; The normalized data set was input into the constructed deep learning model to obtain the prediction results of the oven operating conditions. The difference between the prediction results and the actual oven operating conditions was calculated based on the loss function, and the model parameters were optimized using the Adam optimizer. Save the model parameters with the lowest loss to obtain the trained deep learning model; The deep learning model includes: The input layer is used to receive oven data and electrode quality indicators. The oven data includes: oven temperature, humidity, air flow rate and electrode initial humidity; the electrode quality indicators include the moisture content of the electrode after drying and the electrode uniformity; the air flow rate is calculated by air flow; The encoder layer is used to extract features from the input data using a convolutional neural network encoder; LSTM layer to model the temporal relationship between input data and oven operating conditions; A decoder layer for predicting oven operating conditions using a fully connected neural network decoder based on the learned features; the oven operating conditions include oven temperature, humidity, and air flow rate.

[0008] Preferably, the convolutional neural network encoder comprises three convolutional layers, each having 32, 64 and 128 filters, each convolutional layer is followed by a maximum pooling layer and two fully connected layers, wherein the two fully connected layers have 128 and 64 neurons, respectively; The LSTM layer has 128 units; The decoder layer includes 2 fully connected layers and 1 output layer; wherein the 2 fully connected layers have 64 and 32 neurons respectively, and the output layer has 1 neuron.

[0009] Preferably, the computational fluid dynamics model simulates the fluid flow and heat transfer in the oven to obtain the airflow distribution, temperature distribution and humidity distribution in the oven. The specific implementation process is as follows: A1. Establish the geometric model of the oven, including the size, shape and internal structure of the oven; A2. Establishing a fluid flow region according to the oven geometric model and dividing it into unit grids; A3. Selection k-ε Turbulence models as computational fluid dynamics models; A4. Using the prediction results of the deep learning model as the initial parameters of the computational fluid dynamics model, setting the convective heat transfer coefficient of the electrode, the initial water content of the electrode, the material properties of the electrode, the pressure of the oven exhaust port, and the heat transfer coefficient of the oven wall; A5. Perform simulation calculations to obtain the air flow distribution, temperature distribution, humidity distribution and moisture content of the electrode inside the oven.

[0010] In this preferred embodiment, k-ε As a computational fluid dynamics model, the turbulence model takes into account the oven structure and simulates the airflow distribution, temperature distribution, humidity distribution and moisture content of the electrode inside the oven. It can more accurately simulate the gas flow and temperature distribution inside the oven, making it easier to verify whether the oven conditions meet the drying requirements.

[0011] Preferably, the computational fluid dynamics model verifies the prediction result based on the deep learning model, and the specific process is as follows: Setting monitoring points in the computational fluid dynamics model and setting optimal parameter standards, when the monitoring point data is consistent with the set optimal parameter standards or the difference is less than a threshold value, the drying requirements are met; The monitoring points are arranged around the pole piece, inside the pole piece and at the oven air inlet, and the monitoring point data are data simulated by the computational fluid dynamics model, including the temperature, humidity, air flow and speed, and the moisture content of the pole piece at the monitoring points.

[0012] Preferably, the output module adjusts the oven operating conditions in real time based on a model predictive control algorithm, and the specific process is as follows: B1. Establish a mathematical model of the electrode drying process to describe the heat transfer, mass transfer and fluid dynamics process during the electrode drying process; B2. Inputting the prediction results of the deep learning model verified by the computational fluid dynamics model into the established mathematical model to predict the influence of the oven operating conditions on the electrode drying process during the electrode drying process; B3. According to the prediction results of the mathematical model and the set drying process goals, the current control action is optimized and output; the current control action includes adjusting the oven temperature, humidity and air flow.

[0013] In this preferred embodiment, the optimization objectives of the drying process are further considered, such as minimum energy consumption, maximum drying efficiency, etc., and the oven conditions are further adjusted to improve energy efficiency while ensuring the best drying performance.

[0014] In a second aspect, the present invention provides an intelligent oven control method of the above-mentioned intelligent oven control system based on deep learning and computational fluid dynamics, comprising: C1. Obtain the temperature, humidity, air flow rate and initial humidity data of the oven, as well as the required moisture content and uniformity data of the electrode after drying, and perform normalization processing; C2. Input the normalized data into the pre-trained deep learning model to obtain the predicted oven operating conditions; C3. Input the predicted oven operating conditions into the constructed computational fluid dynamics model, simulate the fluid flow and heat transfer in the oven, obtain the airflow distribution, temperature distribution, humidity distribution and electrode moisture content in the oven, verify the prediction results based on the deep learning model, and output the oven operating conditions that meet the requirements; C4. Based on the set drying process target and the output result of the computational fluid dynamics model, the oven operating conditions are adjusted in real time and output.

[0015] The beneficial effects of the present invention are: (1) The present invention provides an intelligent oven control system based on deep learning and computational fluid dynamics. The oven operating conditions are predicted by training a deep learning model, and the fluid flow and heat transfer in the oven are simulated by a computational fluid dynamics model, and the predicted oven operating conditions are then verified. Based on the control system, the optimal oven operating conditions can be obtained, and the oven operating conditions can be adjusted in real time so that the electrode is heated evenly, thereby ensuring the best drying performance and improving product quality.

[0016] (2) This control system can achieve faster drying time and higher production efficiency, reducing costs.

[0017] (3) This control system can easily adapt to different materials and drying processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of the control principle of the intelligent oven control system provided by the present invention; Figure 2 The schematic diagram of the model control prediction algorithm provided by the present invention; Figure 3 A flow chart of the intelligent oven control method provided by the present invention; Figure 4The figure is a comparison chart of the temperature consistency of the pole pieces of the intelligent oven of the present invention and the traditional oven. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0020] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0021] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0022] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection with an intermediate.

[0023] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0024] It should be emphasized here that the step marks mentioned below are not intended to limit the order of the steps, but it should be understood that the steps can be executed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be executed simultaneously.

[0025] In a first aspect, the first embodiment of the present invention provides an intelligent oven control system based on deep learning and computational fluid dynamics, see Figure 1 , the control system includes: The sensor system includes temperature, humidity and air flow sensors. The sensor system is used to detect the temperature, humidity and air flow inside the oven and provide them to the deep learning model and the computational fluid dynamics model. It should be noted that the air flow sensor is set near the pole piece, inside the pole piece and at the oven air inlet.

[0026] The deep learning model is used to predict the optimal oven operating conditions based on the electrode drying requirements and output them to the computational fluid dynamics model; the electrode drying requirements include the moisture content and uniformity of the electrode after drying; the oven operating conditions include: oven internal temperature, air flow and oven humidity.

[0027] The computational fluid dynamics model (CFD) is used to simulate the fluid flow and heat transfer in the oven based on the oven operating conditions, taking into account the geometry and material properties of the oven, to obtain the airflow distribution, temperature distribution and humidity distribution in the oven, and output the oven operating conditions that meet the drying requirements; the airflow distribution and temperature distribution can be used to determine whether the oven operating conditions meet the drying requirements of the electrode; the humidity distribution can be used to determine whether the moisture in the oven can be discharged from the oven.

[0028] The data storage system is used to store historical data and real-time data collected by the sensor system for deep learning model training and CFD model simulation.

[0029] The output module is used to adjust the operating conditions of the oven in real time and output them based on the drying process objectives and the output results of the computational fluid dynamics model to ensure that the operating conditions of the oven can meet the drying requirements of the electrode and the drying process objectives. The operating conditions of the oven include: temperature, humidity and air flow in the oven.

[0030] It should be known that the oven includes a shell, a heating system, an airflow system, an oven cavity, a conveying system, a control system, an exhaust system and an alarm device. These systems work in coordination to ensure the normal operation and efficient drying of the coating oven.

[0031] Furthermore, the oven shell and the oven cavity are made of stainless steel, which has good sealing, high temperature resistance and corrosion resistance.

[0032] Furthermore, the heating system adopts electric heating or steam heating, and different heating methods are selected according to drying requirements.

[0033] Furthermore, before the production process, the airflow system and the exhaust system are simulated by CFD models to determine the structural components and ensure the uniformity of the temperature and hot air flow field.

[0034] Furthermore, the conveying system transports the pole pieces into the oven, including rollers, motors and other components.

[0035] In this embodiment, the deep learning model is trained based on the historical data of the oven, as follows: Collect historical oven data, including oven temperature, air flow, initial electrode humidity, and moisture content and electrode uniformity after drying. Normalize the data and input it into the deep learning model for training.

[0036] The deep learning model structure is as follows: 1. Input layer: The input layer receives the historical data of the oven, including: oven temperature (T), humidity, air flow rate (AFR) and initial humidity of the electrode (EH), electrode quality indicators: moisture content (WC) and electrode uniformity (EU) after drying. It should be noted that the air flow rate (AFR) is the volume flow of air passing through the pipeline per unit time, which is calculated by the air flow rate.

[0037] 2. Encoder layer: A convolutional neural network (CNN) encoder is used to extract features from the input data. In this embodiment, the CNN has 3 convolutional layers with 32, 64, and 128 filters, each followed by a maximum pooling layer and 2 fully connected layers with 128 and 64 neurons, respectively.

[0038] 3. LSTM layer: Use the long short-term memory (LSTM) layer to model the temporal relationship between input data and oven operating conditions. This includes: the impact of oven operation (temperature, humidity, and air flow) on input data (temperature, electrode humidity, and electrode moisture content); the impact of oven operation on input data will continue to future time points; there is a delay in oven operation, and changes in operating conditions may take some time to affect input data; the impact of changes in oven operating conditions on input data is not linear, such as the oven operation increases by 5°C, while the temperature monitored in the oven may only increase by 3°C.

[0039] The LSTM layer consists of 1 LSTM layer with 128 units.

[0040] 4. Decoder layer: A fully connected neural network (FCNN) decoder is used to predict the oven operating conditions based on the learned features. The FCNN consists of 2 fully connected layers with 64 and 32 neurons respectively, and 1 output layer with 1 neuron to predict the optimal oven operating conditions, including oven temperature, air flow rate, and oven humidity.

[0041] 5. Loss function: The mean square error (MSE) loss function is used to measure the difference between the predicted results and the actual oven operating conditions.

[0042] 6. Optimizer: Use Adam optimizer to optimize deep learning model parameters.

[0043] The deep learning model training process is as follows: S1. Data preprocessing: The collected historical data are normalized and divided into training set, validation set and test set (80% for training, 10% for validation and 10% for testing).

[0044] S2. Model training: Input the normalized training set into the constructed deep learning model to obtain the prediction results. Calculate the difference between the prediction results and the actual oven operating conditions based on the loss function, and use the Adam optimizer to optimize the model parameters. Set the model training times to 100 and the batch size to 32.

[0045] S3, model evaluation: save the model parameters with the lowest MSE loss to obtain the trained deep learning model.

[0046] S4. Model testing: Use the test set to test the final trained deep learning model to evaluate its performance.

[0047] In this embodiment, a computational fluid dynamics model (CFD) simulates the fluid flow and heat transfer in the oven to obtain the air flow distribution, temperature distribution and humidity distribution in the oven, as follows: A1. Establish the geometric model of the oven, including the size, shape and internal structure of the oven.

[0048] A2. Grid generation: Establish the fluid flow area according to the geometric model of the oven and divide it into small unit grids for numerical calculation. The size of the grid has a great influence on the simulation results.

[0049] A3. Physical model selection: Select k-ε Turbulence model, since the simulation involves temperature changes, the energy equation is turned on and a humidity convection transfer model is established.

[0050] A4. Use the prediction results of the deep learning model as the initial parameters of the fluid dynamics model, including: air flow, oven temperature and humidity; set the convective heat transfer coefficient of the electrode, the initial water content of the electrode, and the material properties of the electrode; set the pressure of the oven exhaust port; and set the heat transfer coefficient of the oven wall.

[0051] A5. Perform simulation calculations and process the simulation results to obtain the air flow distribution, temperature distribution and humidity distribution inside the oven.

[0052] The CFD model simulates the temperature field, flow field, and moisture content through the parameters predicted by deep learning. Some monitoring points are set in the CFD model, and the optimal parameter standards are set. When the data at the monitoring point is consistent with the optimal parameter standard or the difference is small, the data at the monitoring point will be output and the recorded data will be saved for updating and improving the CFD model.

[0053] The monitoring points can be set near the electrode, inside the electrode and at the air inlet of the oven. The monitoring point data is the data simulated by the CFD model, including parameters such as the temperature, humidity, air flow and speed, and moisture content of the electrode at the monitoring point.

[0054] In the present invention, a CFD model is used to verify the prediction results of the deep learning model. CFD calculates the temperature field and velocity field in the oven under the input parameters, and determines whether the process parameters are output to the control system by whether the data obtained from the monitoring point meets the drying requirements. The CFD model needs data to be improved, such as turbulence models, turbulence parameters, etc., in order to be closer to actual working conditions.

[0055] In this embodiment, the output module adjusts the operating conditions of the oven in real time based on the model predictive control algorithm (MPC). Figure 2 , specifically: B1. Model establishment: Establish a mathematical model of the electrode drying process to describe the physical processes such as heat transfer, mass transfer, fluid dynamics, etc. during the electrode drying process.

[0056] B2. Parameter identification: Input the prediction results (including temperature, humidity, and air flow) of the deep learning model verified by the CFD model into the established mathematical model to predict the influence of parameters such as temperature, humidity, and air flow in the oven on the drying of the electrode during the electrode drying process.

[0057] B3. Based on the MPC prediction results, calculate the current control action to achieve the optimization goals of the drying process, such as minimum energy consumption, maximum drying efficiency, etc.

[0058] B4. Execute control actions, such as adjusting oven temperature, air flow, humidity and other parameters to optimize the drying process.

[0059] The present invention uses an MPC algorithm to enable the oven to achieve optimal drying conditions, ensuring that the material is dried to the required dryness level, while also minimizing energy consumption and the risk of overheating or insufficient drying.

[0060] It should be noted that for environments that exceed operating conditions, alarm information is issued in a timely manner and fed back to the control system for processing, and a new round of deep learning prediction and CFD simulation is carried out.

[0061] Based on the above-mentioned intelligent oven control system, the second embodiment of the present invention provides an intelligent oven control method. Figure 3 ,include: C1. Obtain the oven temperature, air flow and initial electrode humidity data, as well as the required electrode moisture content and electrode uniformity data after drying, and perform normalization processing; C2. Input the normalized data into a pre-trained deep learning model to obtain the predicted optimal oven operating conditions, which include oven temperature, air flow, and humidity; C3. Input the predicted optimal oven operating conditions into the constructed CFD model, simulate the fluid flow and heat transfer in the oven, obtain the airflow distribution, temperature distribution and electrode moisture content in the oven, verify the prediction results of the deep learning model and output the optimal oven operating conditions to the model predictive control algorithm; C4. Based on the set drying process goals and the output results of the computational fluid dynamics model, the oven operating conditions are adjusted and output in real time.

[0062] The third embodiment of the present invention compares the traditional oven method with the above-mentioned intelligent oven method, and the process is as follows: D1. Take the electrodes from the same batch, 300 mm wide, to ensure that the material properties of the electrodes and the initial conditions are the same.

[0063] D2. Place the electrode pieces in a conventional oven and the intelligent oven of the present invention respectively, bake them in the oven for the same time, and measure the temperature of the electrode pieces in the oven.

[0064] D3. The pole piece temperature is set to 365K as the standard temperature. A high-precision temperature probe is used to measure the pole piece temperature every 50mm and record the temperature. The probe accuracy is 0.01K. The measurement results are as follows: Figure 4 shown.

[0065] from Figure 4 It can be seen that the temperature on both sides of the electrode is lower than the middle temperature, the temperature fluctuation range of the smart oven is smaller, and the smart oven can make the electrode achieve better temperature consistency.

[0066] The above is only an embodiment of the present invention, and its description is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent application. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present application, which all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be based on the attached claims.

Claims

1. An intelligent oven control system based on deep learning and computational fluid dynamics, characterized in that: include: A sensor system that detects the temperature, humidity, and air flow inside the oven and feeds them into the deep learning model and computational fluid dynamics model; A deep learning model is used to predict the oven operating conditions based on the electrode drying requirements and output to the computational fluid dynamics model; A computational fluid dynamics model is used to simulate the fluid flow and heat transfer in the oven based on the prediction results of the deep learning model and the geometric shape and material properties of the oven, obtain the airflow distribution, temperature distribution, humidity distribution and moisture content of the pole piece in the oven, verify the prediction results based on the deep learning model, and output the oven operating conditions that meet the requirements; The output module is used to adjust the oven operating conditions in real time and output them based on the drying process objectives and the output results of the computational fluid dynamics model.

2. The intelligent oven control system based on deep learning and computational fluid dynamics according to claim 1, characterized in that: The sensor system includes a temperature sensor, a humidity sensor and an air flow sensor, which are used to detect the temperature, humidity and air flow inside the oven; The sensor system is arranged around the pole piece, inside the pole piece and at the oven air inlet.

3. The intelligent oven control system based on deep learning and computational fluid dynamics according to claim 1, characterized in that: The deep learning model is trained based on historical data. The training process of the deep learning model is as follows: Collect the historical data of oven and electrode drying, including oven temperature, humidity, air flow, initial humidity of the electrode, moisture content of the electrode after drying, and electrode uniformity, and normalize the data; The normalized data set was input into the constructed deep learning model to obtain the prediction results of the oven operating conditions. The difference between the prediction results and the actual oven operating conditions was calculated based on the loss function, and the model parameters were optimized using the Adam optimizer. Save the model parameters with the lowest loss to obtain the trained deep learning model; The deep learning model includes: The input layer is used to receive oven data and electrode quality indicators. The oven data includes: oven temperature, humidity, air flow rate and electrode initial humidity; the electrode quality indicators include the moisture content of the electrode after drying and the electrode uniformity; the air flow rate is calculated by air flow; The encoder layer is used to extract features from the input data using a convolutional neural network encoder; LSTM layer to model the temporal relationship between input data and oven operating conditions; A decoder layer for predicting oven operating conditions using a fully connected neural network decoder based on the learned features; the oven operating conditions include oven temperature, humidity, and air flow rate.

4. The intelligent oven control system based on deep learning and computational fluid dynamics according to claim 3, characterized in that: The convolutional neural network encoder includes 3 convolutional layers with 32, 64 and 128 filters respectively, each convolutional layer is followed by a maximum pooling layer and 2 fully connected layers, wherein the 2 fully connected layers have 128 and 64 neurons respectively; The LSTM layer has 128 units; The decoder layer includes 2 fully connected layers and 1 output layer; wherein the 2 fully connected layers have 64 and 32 neurons respectively, and the output layer has 1 neuron.

5. The intelligent oven control system based on deep learning and computational fluid dynamics according to claim 3, characterized in that: The computational fluid dynamics model simulates the fluid flow and heat transfer in the oven to obtain the airflow distribution, temperature distribution and humidity distribution in the oven. The specific implementation process is as follows: A1. Establish the geometric model of the oven, including the size, shape and internal structure of the oven; A2. Establishing a fluid flow region according to the oven geometric model and dividing it into unit grids; A3. Selection k-ε Turbulence models as computational fluid dynamics models; A4. Using the prediction results of the deep learning model as the initial parameters of the computational fluid dynamics model, setting the convective heat transfer coefficient of the electrode, the initial water content of the electrode, the material properties of the electrode, the pressure of the oven exhaust port, and the heat transfer coefficient of the oven wall; A5. Perform simulation calculations to obtain the air flow distribution, temperature distribution, humidity distribution and electrode moisture content inside the oven.

6. The intelligent oven control system based on deep learning and computational fluid dynamics according to claim 5, characterized in that: The computational fluid dynamics model verifies the prediction results based on the deep learning model, and the specific process is as follows: Setting monitoring points in the computational fluid dynamics model and setting optimal parameter standards, when the monitoring point data is consistent with the set optimal parameter standards or the difference is less than a threshold value, the drying requirements are met; The monitoring points are arranged around the pole piece, inside the pole piece and at the oven air inlet, and the monitoring point data are data simulated by the computational fluid dynamics model, including the temperature, humidity, air flow and speed, and the moisture content of the pole piece at the monitoring points.

7. The intelligent oven control system based on deep learning and computational fluid dynamics according to claim 5, characterized in that: The output module adjusts the oven operating conditions in real time based on the model predictive control algorithm. The specific process is as follows: B1. Establish a mathematical model of the electrode drying process to describe the heat transfer, mass transfer and fluid dynamics process during the electrode drying process; B2. Inputting the prediction results of the deep learning model verified by the computational fluid dynamics model into the established mathematical model to predict the influence of the oven operating conditions on the electrode drying process during the electrode drying process; B3. According to the prediction results of the mathematical model and the set drying process goals, the current control action is optimized and output; the current control action includes adjusting the oven temperature, humidity and air flow.

8. The intelligent oven control method of the intelligent oven control system based on deep learning and computational fluid dynamics according to any one of claims 1 to 7, characterized in that: include: C1. Obtain the temperature, humidity, air flow rate and initial humidity data of the oven, as well as the required moisture content and uniformity data of the electrode after drying, and perform normalization processing; C2. Input the normalized data into the pre-trained deep learning model to obtain the predicted oven operating conditions; C3. Input the predicted oven operating conditions into the constructed computational fluid dynamics model, simulate the fluid flow and heat transfer in the oven, obtain the airflow distribution, temperature distribution, humidity distribution and electrode moisture content in the oven, verify the prediction results based on the deep learning model, and output the oven operating conditions that meet the requirements; C4. Based on the set drying process target and the output result of the computational fluid dynamics model, the oven operating conditions are adjusted in real time and output.

Images