A method and system for intelligent control of exhaust gas treatment for battery recycling

Abstract

The application discloses a kind of waste gas treatment intelligent control method and system for battery recycling, method includes: the response signal that sensor is launched is obtained by being arranged on battery waste gas treatment device, and according to the response signal, generate sensor network;Data acquisition is carried out to the waste gas treatment device by sensor network, the feature extraction of the collected sensing data is carried out, and based on the sensor network, combined with the extracted waste gas data features, construct dynamic adsorption biochemical catalytic model;Simulation operation is carried out by the dynamic adsorption biochemical catalytic model, and real-time working condition prediction is carried out according to the collected simulation operation data, and the waste gas prediction result in preset future time is obtained;Based on the dynamic adsorption biochemical catalytic model, according to the waste gas prediction result carries out multi-objective optimization, obtains the control decision corresponding to each waste gas treatment device, and according to the control decision, the waste gas treatment device is controlled.

Description

Technical Field

[0001] This invention relates to the field of battery waste gas recovery technology, and in particular to an intelligent control method and system for waste gas treatment in battery recycling. Background Technology

[0002] Currently, waste gas treatment technologies generated during lithium battery recycling are still in the exploratory and developmental stage, lacking mature technologies that have undergone long-term validation. For example, existing technologies may not be able to completely remove harmful substances when treating volatile organic compounds (VOCs) in waste gas, leading to non-compliance with emission standards. Furthermore, because battery recycling is a centralized process, numerous waste gas treatment devices exist within the centralized battery recycling process. Currently, it is impossible to integrate and control these large-scale centralized treatment devices, resulting in inconsistent control of VOCs treated by different devices. This leads to low resource utilization and waste gas recovery rates, and may also cause secondary pollution.

[0003] The integration and optimization of existing waste gas treatment processes are still relatively low, lacking real-time monitoring and intelligent adjustment and control capabilities for the entire waste gas treatment process, making it difficult to achieve efficient and energy-saving waste gas treatment. Summary of the Invention

[0004] This invention provides an intelligent control method for waste gas treatment in battery recycling, which solves the technical problem in the prior art that VOCs treated by different devices cannot be uniformly controlled, making it difficult to achieve efficient and energy-saving battery waste gas treatment.

[0005] To address the aforementioned technical problems, embodiments of the present invention provide an intelligent control method for waste gas treatment in battery recycling, comprising:

[0006] The system acquires response signals emitted by sensors installed on the battery exhaust gas treatment device and generates a sensor network based on the response signals; wherein each battery exhaust gas treatment device is equipped with at least one sensor, and each sensor emits a response signal corresponding to its identity information.

[0007] Data is collected from the waste gas treatment device through a sensor network. The collected sensor data is then feature-extracted. Based on the sensor network and the extracted waste gas data features, a dynamic adsorption biochemical catalysis model is constructed. Each dynamic adsorption biochemical catalysis model corresponds to one waste gas treatment device.

[0008] The dynamic adsorption biochemical catalysis model is used for simulation operation, and the collected simulation operation data is used for real-time operating condition prediction to obtain the waste gas prediction results within a preset future time.

[0009] Based on the dynamic adsorption biochemical catalysis model, multi-objective optimization is performed according to the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device, and the waste gas treatment device is controlled according to the control decisions.

[0010] As a preferred embodiment, the step of acquiring the response signals emitted by the sensors installed on the battery exhaust gas treatment device, and generating a sensor network based on the response signals, specifically includes:

[0011] A docking signal is transmitted to a preset range via a signal transmitting device, so that each sensor installed on the battery exhaust gas treatment device receives the docking signal, generates a response signal, and sends it outward.

[0012] The response signal is acquired, and the identity information of each sensor is parsed based on the response signal.

[0013] Based on the identity information of each sensor, the battery exhaust gas treatment device set up by each sensor is identified, and node information of the corresponding battery exhaust gas treatment device is generated according to the type of exhaust gas treated by each battery exhaust gas treatment device.

[0014] Based on the node information and the sensors, a sensor network is constructed.

[0015] As a preferred embodiment, the step of collecting data from the waste gas treatment device via a sensor network, extracting features from the collected sensor data, and constructing a dynamic adsorption biochemical catalysis model based on the sensor network and the extracted waste gas data features includes:

[0016] The sensor network is used to monitor the battery exhaust gas treatment devices at each node, and to obtain the component data and equipment process parameters of the battery exhaust gas treated by each device.

[0017] According to the Kalman filter algorithm, abnormal data are removed from the component data and equipment process parameters;

[0018] The composition data and equipment process parameters after removing abnormal data are normalized, and the features of the normalized composition data and equipment process parameters are extracted through a preset sliding window to obtain the exhaust gas data features.

[0019] Based on the sensor network, time-series features are added to the exhaust gas data characteristics corresponding to each battery exhaust gas treatment device to construct a dynamic adsorption biochemical catalysis model.

[0020] As a preferred embodiment, the step of adding time-series features to the waste gas data characteristics corresponding to each battery waste gas treatment device based on the sensor network to construct a dynamic adsorption biochemical catalysis model specifically includes:

[0021] Based on the sensor network and according to the node information of each battery exhaust gas treatment device, the structural data of the node information of each battery exhaust gas treatment device is pulled out, and modeling is performed according to the structural data to obtain the adsorption layer structure model of each battery exhaust gas treatment device.

[0022] Based on the preset sliding window, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device;

[0023] Based on the characteristics of the waste gas data with added time features and the node information of each battery waste gas treatment device, an adsorption-diffusion model is constructed.

[0024] The adsorption layer structure model and the adsorption diffusion model are coupled, and preset data-driven methods are added to the coupled model to construct a dynamic adsorption biochemical catalysis model corresponding to each battery exhaust gas treatment device.

[0025] As a preferred embodiment, the simulation operation using the dynamic adsorption biochemical catalysis model, and the real-time operating condition prediction based on the collected simulation data, to obtain the predicted exhaust gas results for a preset future time period, specifically includes:

[0026] The dynamic adsorption biochemical catalysis model is used to simulate the corresponding battery exhaust gas treatment device. During the simulation, data is collected in real time from the running dynamic adsorption biochemical catalysis model to obtain the simulation operation data of each dynamic adsorption biochemical catalysis model.

[0027] The collected simulation data is input into a preset LSTM neural network, and the dynamic adsorption biochemical catalysis model is used to predict the real-time operating conditions within a preset future time, thereby obtaining the waste gas prediction results within the preset future time.

[0028] As a preferred embodiment, the method for constructing the preset LSTM neural network specifically includes:

[0029] Historical simulation data of the dynamic adsorption biochemical catalysis model were obtained, and the historical simulation data were preprocessed.

[0030] Feature selection is performed on the preprocessed historical simulation data to obtain key historical data features;

[0031] Based on the dimensions of the key historical data features, an initial LSTM neural network is constructed, and the loss function and optimization algorithm are set.

[0032] The initial LSTM neural network is trained based on the loss function and optimization algorithm, and the trained LSTM neural network is evaluated through a preset test set to obtain a qualified LSTM neural network as the preset LSTM neural network.

[0033] As a preferred embodiment, the step of performing multi-objective optimization based on the dynamic adsorption biochemical catalysis model and the waste gas prediction results to obtain control decisions corresponding to each waste gas treatment device, and controlling the waste gas treatment device according to the control decisions, specifically includes:

[0034] Based on the dynamic adsorption biochemical catalysis model, the energy consumption target, removal rate target, and solvent recovery rate target are determined.

[0035] By constraining the energy consumption target, removal rate target, and solvent recovery rate target, multi-objective optimization conditions are obtained.

[0036] The waste gas prediction results are used as a particle swarm, and based on the multi-objective optimization conditions, the fitness of each particle in the particle swarm is calculated, and the optimal solutions of individual particles and the global particle swarm are updated; wherein, each particle corresponds to the waste gas prediction result of each dynamic adsorption biochemical catalysis model, and the position and velocity of each particle are respectively represented as a point in its solution space and the direction of motion.

[0037] Based on the optimal solutions for individual particles and the optimal solution for the global particle swarm, the velocity and position of each particle are updated until a preset iteration condition is met. Then, the optimal position of the global particle swarm is output as the optimization result.

[0038] Based on the optimization results, the optimization objects and optimization objectives of the dynamic adsorption biochemical catalysis model are determined, thereby generating control decisions corresponding to each waste gas treatment device.

[0039] Each of the aforementioned control decisions is sent to each waste gas treatment device, and the waste gas treatment device is controlled based on the corresponding control decision.

[0040] Accordingly, the present invention also provides an intelligent control system for waste gas treatment for battery recycling, comprising: a control host computer and a plurality of battery waste gas treatment devices connected to the control host computer;

[0041] The battery exhaust gas treatment device is used to collect and adsorb the exhaust gas produced by the battery recycling equipment, and to collect exhaust gas data through sensors installed therein.

[0042] The host computer for control includes: a response module, a model module, a simulation module, and an optimization module;

[0043] The response module is used to acquire response signals emitted by sensors installed on the battery exhaust gas treatment device, and generate a sensor network based on the response signals; wherein each battery exhaust gas treatment device is equipped with at least one sensor, and each sensor emits a response signal corresponding to its identity information.

[0044] The model module is used to collect data from the waste gas treatment device through a sensor network, extract features from the collected sensor data, and construct a dynamic adsorption biochemical catalysis model based on the sensor network and the extracted waste gas data features; wherein, each dynamic adsorption biochemical catalysis model corresponds to one waste gas treatment device.

[0045] The simulation module is used to perform simulation operation through the dynamic adsorption biochemical catalysis model, and to perform real-time operating condition prediction based on the collected simulation operation data to obtain the waste gas prediction result within a preset future time.

[0046] The optimization module is used to perform multi-objective optimization based on the dynamic adsorption biochemical catalysis model and the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device, and to control the waste gas treatment device according to the control decisions.

[0047] Accordingly, the present invention also provides a terminal device comprising 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 intelligent control method for exhaust gas treatment for battery recycling as described in any of the above.

[0048] Accordingly, the present invention also provides a computer-readable storage medium comprising 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 intelligent control method for exhaust gas treatment for battery recycling as described in any of the above claims.

[0049] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0050] The technical solution of this invention acquires the response signals emitted by sensors installed on the battery exhaust gas treatment device, thereby generating a corresponding sensor network. Based on this sensor network, corresponding data is collected to generate corresponding exhaust gas data characteristics to construct a corresponding dynamic adsorption biochemical catalysis model. This enables the simulation and real-time operation prediction of the dynamic adsorption biochemical catalysis model, thereby obtaining the exhaust gas prediction results within a preset future time. Finally, multi-objective optimization is performed to realize the control decisions corresponding to each exhaust gas treatment device, thereby achieving cluster control of each exhaust gas treatment device. This ensures the integration and optimization of the system's exhaust gas treatment process, and improves real-time monitoring and intelligent adjustment and control capabilities. Attached Figure Description

[0051] Figure 1 : A flowchart of an intelligent control method for waste gas treatment in battery recycling provided in an embodiment of the present invention;

[0052] Figure 2 : This is a structural diagram of the intelligent control system for waste gas treatment provided in an embodiment of the present invention;

[0053] Figure 3 : This is a structural diagram of the control host computer provided in an embodiment of the present invention. Detailed Implementation

[0054] 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. Example 1

[0055] Please refer to Figure 1 The present invention provides an intelligent control method for waste gas treatment in battery recycling, which is executed by a host computer and includes the following steps S101-S104:

[0056] Step S101: Obtain the response signals emitted by the sensors installed on the battery exhaust gas treatment device, and generate a sensor network based on the response signals; wherein, each battery exhaust gas treatment device is equipped with at least one sensor, and each sensor emits a response signal corresponding to its identity information.

[0057] As a preferred embodiment, the step of acquiring the response signal emitted by the sensor installed on the battery exhaust gas treatment device and generating a sensor network based on the response signal specifically includes:

[0058] A docking signal is transmitted to a preset range via a signal transmitting device, so that each sensor installed on the battery exhaust gas treatment device receives the docking signal, generates a response signal, and sends it outward; the response signal is acquired, and the identity information of each sensor is parsed based on the response signal; based on the identity information of each sensor, the battery exhaust gas treatment device to which each sensor is installed is identified, and node information of the corresponding battery exhaust gas treatment device is generated according to the type of exhaust gas treated by each battery exhaust gas treatment device; based on the node information and the sensors, a sensor network is constructed.

[0059] In this embodiment, the signal transmitting device is located in the host computer, which controls and executes the signal transmitting device. By transmitting a docking signal to a preset range through the signal transmitting device, each sensor on the battery exhaust gas treatment device can receive the docking signal, generate a response signal, and send the response signal outward, thereby ensuring that all sensors within the preset range can receive and process the signal.

[0060] In this embodiment, after receiving the response signals from each corresponding sensor, the host computer analyzes the identity information of each sensor, including but not limited to the type of sensor, the type of waste gas data collected, and the type and model of the battery waste gas treatment device. Each battery waste gas treatment device includes at least one or more of the following: a dynamic adsorption layer, a biocatalytic filter bed, an electrochemical reactor, and a solvent recovery device. Sensors include, but are not limited to, VOCs concentration sensors, temperature and humidity sensors, pressure sensors, and gas composition mass spectrometers. Each battery waste gas treatment device includes at least one sensor.

[0061] In this embodiment, by using the identity information of each sensor, the battery exhaust gas treatment device installed by each sensor can be identified, thereby generating node information for the corresponding battery exhaust gas treatment device. This includes the type, model, and operating status data of the battery exhaust gas treatment device in that node, as well as the model, data type, and data value of the sensor corresponding to the battery exhaust gas treatment device. Furthermore, by using the information of each node, a corresponding sensor network can be constructed.

[0062] Step S102: Data is collected from the waste gas treatment device through a sensor network, features are extracted from the collected sensor data, and a dynamic adsorption biochemical catalysis model is constructed based on the sensor network and the extracted waste gas data features; wherein, each dynamic adsorption biochemical catalysis model corresponds to a waste gas treatment device.

[0063] As a preferred embodiment, the step of collecting data from the waste gas treatment device through a sensor network, extracting features from the collected sensor data, and constructing a dynamic adsorption biochemical catalysis model based on the sensor network and the extracted waste gas data features includes:

[0064] The sensor network monitors the battery exhaust gas treatment devices at each node, acquiring component data and equipment process parameters of the battery exhaust gas treated by each device. Anomalies in the component data and equipment process parameters are removed using a Kalman filter algorithm. The component data and equipment process parameters after anomaly removal are then normalized, and features are extracted from the normalized data using a preset sliding window to obtain exhaust gas data features. Based on the sensor network, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device, thereby constructing a dynamic adsorption biochemical catalysis model.

[0065] In this embodiment, the operating status of the battery exhaust gas treatment device is monitored in real time through a sensor network deployed at each node (i.e., the battery exhaust gas treatment device). The sensor network can acquire key component data (such as VOCs concentration, particulate matter concentration, etc.) and process parameters of the equipment (such as temperature, pressure, flow rate, etc.) during the exhaust gas treatment process.

[0066] In this embodiment, the Kalman filter algorithm is used to process the collected component data and equipment process parameters. Kalman filtering is a statistical estimation-based filtering method that can effectively identify and remove outliers or noise from the data. By dynamically estimating the state of the battery exhaust gas treatment device and the overall system, and combining measured and / or predicted values, the impact of noise on the data is reduced. It is understood that measured values ​​can be evaluated using the collected data, and predicted values ​​can be obtained by fitting the measured values ​​and calculating corresponding weights.

[0067] In this embodiment, the component data and equipment process parameters after removing outlier data are normalized to a range of [0,1] or [-1,1]. Normalization eliminates dimensional differences between parameters, facilitating subsequent processing. The normalized data is then processed using a preset sliding window. Sliding window technology extracts local features from the data while reducing data dimensionality. For example, a fixed-size window can be set, and statistical features (such as mean and variance) of the data within the window are extracted each time the window is slid, thus obtaining the characteristics of the exhaust gas data.

[0068] In this embodiment, based on a sensor network, time-series features are added to the exhaust gas data characteristics corresponding to each battery exhaust gas treatment device. These time-series features reflect the changing trends of the data over time, helping the model capture dynamic changes. Then, by combining the normalized feature data and the time-series features, a dynamic adsorption-biochemical catalysis model is constructed. This model can simulate the dynamic changes in adsorption, biochemical reactions, and catalysis during the exhaust gas treatment process, thereby achieving accurate prediction and control of the exhaust gas treatment process. It can be understood that through sensor network monitoring, Kalman filtering algorithm, data normalization and feature extraction, and the construction of the dynamic adsorption-biochemical catalysis model, intelligent monitoring and optimization of the battery exhaust gas treatment process are achieved, resulting in significant economic and environmental benefits.

[0069] As a preferred embodiment, the step of adding time-series features to the exhaust gas data characteristics corresponding to each battery exhaust gas treatment device based on the sensor network to construct a dynamic adsorption biochemical catalysis model specifically includes:

[0070] Based on the sensor network and according to the node information of each battery exhaust gas treatment device, the structural data of the node information of each battery exhaust gas treatment device is pulled out, and modeling is performed according to the structural data to obtain the adsorption layer structure model of each battery exhaust gas treatment device.

[0071] Based on the preset sliding window, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device;

[0072] Based on the characteristics of the waste gas data with added time features and the node information of each battery waste gas treatment device, an adsorption-diffusion model is constructed.

[0073] The adsorption layer structure model and the adsorption diffusion model are coupled, and preset data-driven methods are added to the coupled model to construct a dynamic adsorption biochemical catalysis model corresponding to each battery exhaust gas treatment device.

[0074] In this embodiment, sensor network monitoring and structural data modeling are implemented. Based on the sensor network, node information of each battery exhaust gas treatment device is monitored. The sensor network collects real-time operational data of the exhaust gas treatment devices through sensor nodes deployed at different locations. Then, based on the structural data of each node, an adsorption layer structure model is constructed for each battery exhaust gas treatment device. The structural data includes physical parameters of the adsorption layer (such as adsorbent type, porosity, specific surface area, etc.) and process parameters (such as exhaust gas flow rate, temperature, etc.) to simulate the physical and chemical properties of the adsorption layer. Furthermore, time-series feature extraction is performed. Using a preset sliding window technique, time-series features are added to the exhaust gas data features (such as VOCs concentration, particulate matter concentration, etc.) corresponding to each battery exhaust gas treatment device. Additionally, the sliding window technique extracts statistical features (such as mean, variance, etc.) within the window by sliding a fixed-size window across the time-series data, thereby capturing the dynamic changes of the data over time.

[0075] In this embodiment, an adsorption-diffusion model is constructed by combining time-series waste gas data features with node information from each battery waste gas treatment device. This model describes the diffusion process of waste gas in the adsorption layer, considering both the adsorption capacity of the adsorbent and the diffusion characteristics of the waste gas. The adsorption-diffusion model can predict the distribution and concentration changes of waste gas in the adsorption layer, providing a theoretical basis for dynamic control. Furthermore, the coupled model and data-driven approach are optimized by coupling the adsorption layer structure model and the adsorption-diffusion model to form a comprehensive dynamic adsorption-biochemical catalysis model. By integrating the physical structure of the adsorption layer and the diffusion process, the coupled model can more accurately simulate the dynamic behavior of waste gas treatment. In addition, preset data-driven methods (such as historical operating data and fault diagnosis data) are added to the coupled model to further optimize its predictive ability and adaptability.

[0076] In this embodiment, the adsorption layer structure model can be constructed based on CAD (accuracy ±0.1mm); the adsorption diffusion model can be constructed and generated by simulating the pore diffusion process of the adsorbent using the finite element method, based on Darcy's law and the Langmuir equation (adsorption isotherm equation) and combined with the characteristics of the waste gas data.

[0077] Step S103: Simulate the operation using the dynamic adsorption biochemical catalysis model and predict the operating conditions in real time based on the collected simulation data to obtain the predicted waste gas results for the preset future time.

[0078] As a preferred embodiment, the step of performing simulation operation through the dynamic adsorption biochemical catalysis model and making real-time operating condition predictions based on the collected simulation operation data to obtain the predicted exhaust gas results within a preset future timeframe specifically includes:

[0079] The dynamic adsorption biochemical catalysis model is used to simulate the corresponding battery exhaust gas treatment device. During the simulation, data is collected in real time from the running dynamic adsorption biochemical catalysis model to obtain the simulation operation data of each dynamic adsorption biochemical catalysis model.

[0080] The collected simulation data is input into a preset LSTM neural network, and the dynamic adsorption biochemical catalysis model is used to predict the real-time operating conditions within a preset future time, thereby obtaining the waste gas prediction results within the preset future time.

[0081] In this embodiment, the corresponding battery waste gas treatment device is simulated using a pre-constructed dynamic adsorption biochemical catalysis model. During the simulation, data on the model's operation are collected in real time, including key indicators such as the adsorption layer structure parameters of the waste gas treatment device, changes in waste gas composition, temperature, and pressure, thereby obtaining simulation operation data for each dynamic adsorption biochemical catalysis model.

[0082] In this embodiment, simulation data is input into a pre-defined LSTM neural network. The LSTM network is a long short-term memory network capable of effectively processing time-series data and capturing dynamic changes and long-term dependencies within the data. Within a pre-defined future timeframe, the LSTM network predicts the real-time operating conditions of the dynamic adsorption biochemical catalysis model and outputs predicted waste gas parameters for that future timeframe, including key parameters such as waste gas component concentration and treatment efficiency.

[0083] As a preferred embodiment, the method for constructing the preset LSTM neural network specifically includes:

[0084] Historical simulation data of the dynamic adsorption biochemical catalysis model were obtained, and the historical simulation data were preprocessed.

[0085] Feature selection is performed on the preprocessed historical simulation data to obtain key historical data features;

[0086] Based on the dimensions of the key historical data features, an initial LSTM neural network is constructed, and the loss function and optimization algorithm are set.

[0087] The initial LSTM neural network is trained based on the loss function and optimization algorithm, and the trained LSTM neural network is evaluated through a preset test set to obtain a qualified LSTM neural network as the preset LSTM neural network.

[0088] In this embodiment, before inputting the simulation data into the LSTM network, data preprocessing, including normalization and feature extraction, is required during data preprocessing and model training. Normalization scales the data to a uniform range, reducing dimensional differences between different features. Feature extraction analyzes the time-series characteristics of the data to extract features significant for prediction, thereby improving model training efficiency and prediction accuracy. Specifically, waste gas data emitted during production is collected, including real-time concentrations of various VOCs and related production parameters such as equipment operating status and raw material usage. Environmental parameters such as temperature, humidity, and air pressure, which may affect VOC emissions and diffusion, are also collected. The collected data is then cleaned, removing missing and outlier values. The data is then standardized or normalized to eliminate dimensional differences between variables and improve model training performance. Features that significantly affect VOC concentration fluctuations are analyzed and selected, such as environmental factors like temperature, humidity, and air pressure, through methods like correlation analysis and importance analysis.

[0089] In this embodiment, the structure of the LSTM model is defined, including an input layer, hidden layers, and an output layer. The dimension of the input layer should be consistent with the selected number of features. The output layer predicts the VOCs concentration fluctuations over a preset future time period, preferably 30 minutes. In the hidden layer, an appropriate activation function (such as the ReLU function) is selected to avoid the gradient vanishing problem and improve the model's nonlinear fitting ability. Additionally, a Dropout layer can be added to prevent overfitting.

[0090] In this embodiment, the preprocessed dataset is divided into a training set and a test set, and the LSTM model is trained using the training set. During training, an appropriate loss function (such as mean squared error) and an optimization algorithm (such as the Adam optimizer) are set to minimize the loss function. Simultaneously, suitable hyperparameters, such as learning rate, number of training epochs, and batch size, need to be selected to improve the model's training efficiency and prediction accuracy. The optimal combination of hyperparameters can be determined through experimentation or by using hyperparameter optimization algorithms. Furthermore, to enhance the model's adaptability and predictive ability, some improved LSTM model structures can be considered, for example, introducing an attention mechanism into the LSTM model to better handle complex environmental adaptation and data information learning.

[0091] In this embodiment, during the training of the LSTM network, optimization algorithms (including but not limited to the Adam optimizer, etc.) are used to adjust the network weights and biases to minimize prediction error. Simultaneously, methods such as cross-validation are used to validate the model, ensuring that it has good generalization ability on unseen data.

[0092] In this embodiment, the trained LSTM model is evaluated using a test set. Evaluation metrics may include mean squared error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The validated and optimized LSTM model is applied to a practical battery production exhaust gas treatment system, receiving current time-series data in real time and outputting VOC concentration fluctuation predictions for the next 30 minutes. These exhaust gas predictions can provide a basis for adjusting the operation of exhaust gas treatment equipment, allowing for proactive measures to reduce VOC emissions.

[0093] Step S104: Based on the dynamic adsorption biochemical catalysis model, perform multi-objective optimization according to the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device, and control the waste gas treatment device according to the control decisions.

[0094] As a preferred embodiment, the step of performing multi-objective optimization based on the dynamic adsorption biochemical catalysis model and the waste gas prediction results to obtain control decisions corresponding to each waste gas treatment device, and controlling the waste gas treatment device according to the control decisions, specifically includes:

[0095] Based on the dynamic adsorption biochemical catalysis model, the energy consumption target, removal rate target, and solvent recovery rate target are determined.

[0096] By constraining the energy consumption target, removal rate target, and solvent recovery rate target, multi-objective optimization conditions are obtained.

[0097] The waste gas prediction results are used as a particle swarm, and based on the multi-objective optimization conditions, the fitness of each particle in the particle swarm is calculated, and the optimal solutions of individual particles and the global particle swarm are updated; wherein, each particle corresponds to the waste gas prediction result of each dynamic adsorption biochemical catalysis model, and the position and velocity of each particle are respectively represented as a point in its solution space and the direction of motion.

[0098] Based on the optimal solutions for individual particles and the optimal solution for the global particle swarm, the velocity and position of each particle are updated until a preset iteration condition is met. Then, the optimal position of the global particle swarm is output as the optimization result.

[0099] Based on the optimization results, the optimization objects and optimization objectives of the dynamic adsorption biochemical catalysis model are determined, thereby generating control decisions corresponding to each waste gas treatment device.

[0100] Each of the aforementioned control decisions is sent to each waste gas treatment device, and the waste gas treatment device is controlled based on the corresponding control decision.

[0101] In this embodiment, multi-objective optimization conditions are determined. Based on a dynamic adsorption biochemical catalysis model, energy consumption, removal rate, and solvent recovery rate targets are identified. These targets reflect key performance indicators in the waste gas treatment process. For example, the energy consumption target can be used to assess the reduction of energy consumption in the waste gas treatment process; the removal rate target can be used to assess the removal efficiency of pollutants (such as VOCs) in the waste gas; and the solvent recovery rate target can be used to assess the solvent recovery and utilization rate. These targets are then constrained to form multi-objective optimization conditions, which guide the subsequent optimization process and ensure that the optimization results simultaneously meet multiple performance indicators.

[0102] In this embodiment, the exhaust gas prediction result is used as the initial input to the particle swarm. Each particle corresponds to an exhaust gas prediction result of the dynamic adsorption biochemical catalysis model, and its position and velocity are represented as a point in the solution space and its direction of motion, respectively. The parameters for initializing the particle swarm include: the number of particles, inertia weight, individual learning factor, swarm learning factor, and upper and lower bounds of the search space. The number of particles is determined according to the problem size and complexity. The inertia weight, individual learning factor, and swarm learning factor all control the particle's motion behavior, and the upper and lower bounds of the search space are used to limit the particle's position range.

[0103] In this embodiment, based on multi-objective optimization conditions, the fitness of each particle in the particle swarm is calculated. The fitness function is typically a multi-objective function that comprehensively considers energy consumption, removal rate, and solvent recovery rate. In each iteration: the individual particle optimal solution (pbest) is updated: if the current particle's fitness is better than its historical best value, its individual optimal position is updated; and the global particle swarm optimal solution (gbest) is updated: if the current particle's fitness is better than the global best value, its global optimal position is updated. This allows the velocity and position of each particle to be updated based on the individual optimal solution and the global optimal solution. The above fitness calculation and particle update process is repeated until a preset iteration condition is met (such as the maximum number of iterations or fitness convergence), and finally, the optimal position of the global particle swarm is output as the optimization result.

[0104] In this embodiment, based on the optimization results, the optimization objects and optimization targets of the dynamic adsorption biochemical catalysis model are determined, control decisions corresponding to each waste gas treatment device are generated, and then the control decisions are sent to each waste gas treatment device. Based on the corresponding control decisions, the device is controlled in real time to optimize the waste gas treatment process.

[0105] Implementing the above embodiments has the following effects:

[0106] The technical solution of this invention acquires the response signals emitted by sensors installed on the battery exhaust gas treatment device, thereby generating a corresponding sensor network. Based on this sensor network, corresponding data is collected to generate corresponding exhaust gas data characteristics to construct a corresponding dynamic adsorption biochemical catalysis model. This enables the simulation and real-time operation prediction of the dynamic adsorption biochemical catalysis model, thereby obtaining the exhaust gas prediction results within a preset future time. Finally, multi-objective optimization is performed to realize the control decisions corresponding to each exhaust gas treatment device, thereby achieving cluster control of each exhaust gas treatment device. This ensures the integration and optimization of the system's exhaust gas treatment process, and improves real-time monitoring and intelligent adjustment and control capabilities. Example 2

[0107] Please see Figure 2 The present invention also provides an intelligent control system for waste gas treatment for battery recycling, comprising: a control host computer 01 and a plurality of battery waste gas treatment devices 02 connected to the control host computer 01.

[0108] The battery exhaust gas treatment device 02 is used to collect and adsorb the exhaust gas produced by the battery recycling equipment, and to collect exhaust gas data through sensors installed therein.

[0109] Please see Figure 3 The control host computer 01 includes: a response module 201, a model module 202, a simulation module 203, and an optimization module 204;

[0110] The response module 201 is used to acquire the response signals emitted by the sensors installed on the battery exhaust gas treatment device 02, and generate a sensor network based on the response signals; wherein, each battery exhaust gas treatment device 02 is provided with at least one sensor, and each sensor emits a response signal corresponding to its identity information.

[0111] The model module 202 is used to collect data from the waste gas treatment device 02 through a sensor network, extract features from the collected sensor data, and construct a dynamic adsorption biochemical catalysis model based on the sensor network and the extracted waste gas data features; wherein, each dynamic adsorption biochemical catalysis model corresponds to one waste gas treatment device 02.

[0112] The simulation module 203 is used to perform simulation operation through the dynamic adsorption biochemical catalysis model, and to perform real-time operating condition prediction based on the collected simulation operation data to obtain the waste gas prediction result within a preset future time.

[0113] The optimization module 204 is used to perform multi-objective optimization based on the dynamic adsorption biochemical catalysis model and the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device 02, and to control the waste gas treatment device 02 according to the control decisions.

[0114] As a preferred embodiment, the step of acquiring the response signals emitted by the sensors installed on the battery exhaust gas treatment device 02, and generating a sensor network based on the response signals, specifically includes:

[0115] A docking signal is transmitted to a preset range via a signal transmitting device, so that each sensor installed on the battery exhaust gas treatment device 02 receives the docking signal, generates a response signal, and sends it outward.

[0116] The response signal is acquired, and the identity information of each sensor is parsed based on the response signal.

[0117] Based on the identity information of each sensor, the battery exhaust gas treatment device 02 set by each sensor is identified, and according to the type of exhaust gas treated by each battery exhaust gas treatment device 02, the node information of the corresponding battery exhaust gas treatment device 02 is generated.

[0118] Based on the node information and the sensors, a sensor network is constructed.

[0119] As a preferred embodiment, the step of collecting data from the waste gas treatment device 02 via a sensor network, extracting features from the collected sensor data, and constructing a dynamic adsorption biochemical catalysis model based on the sensor network and the extracted waste gas data features includes:

[0120] The sensor network is used to monitor the battery exhaust gas treatment devices 02 at each node, and to obtain the component data and equipment process parameters of the battery exhaust gas treated by each battery exhaust gas treatment device 02.

[0121] According to the Kalman filter algorithm, abnormal data are removed from the component data and equipment process parameters;

[0122] The composition data and equipment process parameters after removing abnormal data are normalized, and the features of the normalized composition data and equipment process parameters are extracted through a preset sliding window to obtain the exhaust gas data features.

[0123] Based on the sensor network, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device 02 to construct a dynamic adsorption biochemical catalysis model.

[0124] As a preferred embodiment, based on the sensor network, the waste gas data characteristics corresponding to each battery waste gas treatment device 02 are added with time-series features to construct a dynamic adsorption biochemical catalysis model, specifically including:

[0125] Based on the sensor network and according to the node information of each battery exhaust gas treatment device 02, the structural data corresponding to the node information of each battery exhaust gas treatment device 02 is pulled, and modeling is performed according to the structural data to obtain the adsorption layer structure model corresponding to each battery exhaust gas treatment device 02.

[0126] Based on the preset sliding window, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device 02;

[0127] Based on the characteristics of the exhaust gas data with added time features and the node information of each battery exhaust gas treatment device 02, an adsorption-diffusion model is constructed.

[0128] The adsorption layer structure model and the adsorption diffusion model are coupled, and preset data driving is added to the coupled model to construct a dynamic adsorption biochemical catalytic model corresponding to each battery exhaust gas treatment device 02.

[0129] As a preferred embodiment, the simulation operation using the dynamic adsorption biochemical catalysis model, and the real-time operating condition prediction based on the collected simulation data, to obtain the predicted exhaust gas results for a preset future time period, specifically includes:

[0130] The dynamic adsorption biochemical catalysis model is used to simulate the corresponding battery exhaust gas treatment device 02. During the simulation, data of the running dynamic adsorption biochemical catalysis model is collected in real time to obtain the simulation operation data of each dynamic adsorption biochemical catalysis model.

[0131] The collected simulation data is input into a preset LSTM neural network, and the dynamic adsorption biochemical catalysis model is used to predict the real-time operating conditions within a preset future time, thereby obtaining the waste gas prediction results within the preset future time.

[0132] As a preferred embodiment, the method for constructing the preset LSTM neural network specifically includes:

[0133] Historical simulation data of the dynamic adsorption biochemical catalysis model were obtained, and the historical simulation data were preprocessed.

[0134] Feature selection is performed on the preprocessed historical simulation data to obtain key historical data features;

[0135] Based on the dimensions of the key historical data features, an initial LSTM neural network is constructed, and the loss function and optimization algorithm are set.

[0136] The initial LSTM neural network is trained based on the loss function and optimization algorithm, and the trained LSTM neural network is evaluated through a preset test set to obtain a qualified LSTM neural network as the preset LSTM neural network.

[0137] As a preferred embodiment, the step of performing multi-objective optimization based on the dynamic adsorption biochemical catalysis model and the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device 02, and controlling the waste gas treatment device 02 according to the control decisions, specifically includes:

[0138] Based on the dynamic adsorption biochemical catalysis model, the energy consumption target, removal rate target, and solvent recovery rate target are determined.

[0139] By constraining the energy consumption target, removal rate target, and solvent recovery rate target, multi-objective optimization conditions are obtained.

[0140] The waste gas prediction results are used as a particle swarm, and based on the multi-objective optimization conditions, the fitness of each particle in the particle swarm is calculated, and the optimal solutions of individual particles and the global particle swarm are updated; wherein, each particle corresponds to the waste gas prediction result of each dynamic adsorption biochemical catalysis model, and the position and velocity of each particle are respectively represented as a point in its solution space and the direction of motion.

[0141] Based on the optimal solutions for individual particles and the optimal solution for the global particle swarm, the velocity and position of each particle are updated until a preset iteration condition is met. Then, the optimal position of the global particle swarm is output as the optimization result.

[0142] Based on the optimization results, the optimization objects and optimization objectives of the dynamic adsorption biochemical catalysis model are determined, thereby generating control decisions corresponding to each waste gas treatment device 02.

[0143] Each of the control decisions is sent to each waste gas treatment device 02, and the waste gas treatment device 02 is controlled based on the corresponding control decision.

[0144] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0145] Implementing the above embodiments has the following effects:

[0146] The technical solution of this invention acquires the response signals emitted by sensors installed on the battery exhaust gas treatment device, thereby generating a corresponding sensor network. Based on this sensor network, corresponding data is collected to generate corresponding exhaust gas data characteristics to construct a corresponding dynamic adsorption biochemical catalysis model. This enables the simulation and real-time operation prediction of the dynamic adsorption biochemical catalysis model, thereby obtaining the exhaust gas prediction results within a preset future time. Finally, multi-objective optimization is performed to realize the control decisions corresponding to each exhaust gas treatment device, thereby achieving cluster control of each exhaust gas treatment device. This ensures the integration and optimization of the system's exhaust gas treatment process, and improves real-time monitoring and intelligent adjustment and control capabilities. Example 3

[0147] Accordingly, the present invention also provides a terminal device, comprising: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the intelligent control method for exhaust gas treatment for battery recycling as described in any of the above embodiments.

[0148] The terminal device in this embodiment includes a processor, a memory, and a computer program and computer instructions stored in the memory and executable on the processor. When the processor executes the computer program, it implements the various steps described in Embodiment 1 above, for example... Figure 1 The steps S101 to S104 are shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiment, such as simulation module 203.

[0149] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units 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 terminal device. For example, the simulation module 203 is used to perform simulation operation through the dynamic adsorption biochemical catalysis model and to perform real-time operating condition prediction based on the collected simulation operation data to obtain the predicted exhaust gas result within a preset future time.

[0150] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal 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 a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than illustrated, or combine certain components, or different components. For example, the terminal device may also include input / output devices, network access devices, buses, etc.

[0151] 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. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0152] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile terminal, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital card (SD card), flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0153] Wherein, if the modules / units integrated in the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of 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 the computer program is executed by a processor, it can implement the steps of the various method embodiments described above. Wherein, the computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content contained in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals. Example 4

[0154] Accordingly, the present invention also provides a computer-readable storage medium comprising 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 intelligent control method for exhaust gas treatment for battery recycling as described in any of the above embodiments.

[0155] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A smart control method for waste gas treatment in battery recycling, characterized in that, include: The system acquires response signals emitted by sensors installed on the battery exhaust gas treatment device and generates a sensor network based on the response signals; wherein each battery exhaust gas treatment device is equipped with at least one sensor, and each sensor emits a response signal corresponding to its identity information. Data is collected from the waste gas treatment device through a sensor network. The collected sensor data is then feature-extracted. Based on the sensor network and the extracted waste gas data features, a dynamic adsorption biochemical catalysis model is constructed. Each dynamic adsorption biochemical catalysis model corresponds to one waste gas treatment device. The dynamic adsorption biochemical catalysis model is used for simulation operation, and the collected simulation operation data is used for real-time operating condition prediction to obtain the waste gas prediction results within a preset future time. Based on the dynamic adsorption biochemical catalysis model, multi-objective optimization is performed according to the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device, and the waste gas treatment devices are controlled according to the control decisions; The step of collecting data from the waste gas treatment device via a sensor network, extracting features from the collected sensor data, and constructing a dynamic adsorption biochemical catalytic model based on the sensor network and the extracted waste gas data features includes: The sensor network is used to monitor the battery exhaust gas treatment devices at each node, and to obtain the component data and equipment process parameters of the battery exhaust gas treated by each device. According to the Kalman filter algorithm, abnormal data are removed from the component data and equipment process parameters; The composition data and equipment process parameters after removing abnormal data are normalized, and the features of the normalized composition data and equipment process parameters are extracted through a preset sliding window to obtain the exhaust gas data features. Based on the sensor network, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device to construct a dynamic adsorption biochemical catalysis model. The step of adding time-series features to the exhaust gas data characteristics corresponding to each battery exhaust gas treatment device based on the sensor network to construct a dynamic adsorption biochemical catalysis model specifically includes: Based on the sensor network and according to the node information of each battery exhaust gas treatment device, the structural data corresponding to the node information of each battery exhaust gas treatment device is retrieved, and modeling is performed according to the structural data to obtain the adsorption layer structure model of each battery exhaust gas treatment device. Based on the preset sliding window, time-series features are added to the exhaust gas data features corresponding to each battery exhaust gas treatment device; An adsorption-diffusion model is constructed based on the characteristics of the waste gas data with added time-series features and the node information of each battery waste gas treatment device. The adsorption layer structure model and the adsorption diffusion model are coupled, and preset data driving is added to the coupled model to construct a dynamic adsorption biochemical catalysis model corresponding to each battery exhaust gas treatment device.

2. The intelligent control method for waste gas treatment in battery recycling as described in claim 1, characterized in that, The process of acquiring response signals emitted by sensors installed on the battery exhaust gas treatment device and generating a sensor network based on the response signals specifically includes: A docking signal is transmitted to a preset range via a signal transmitting device, so that each sensor installed on the battery exhaust gas treatment device receives the docking signal, generates a response signal, and sends it outward. The response signal is acquired, and the identity information of each sensor is parsed based on the response signal. Based on the identity information of each sensor, the battery exhaust gas treatment device set up by each sensor is identified, and node information of the corresponding battery exhaust gas treatment device is generated according to the type of exhaust gas treated by each battery exhaust gas treatment device. Based on the node information and the sensors, a sensor network is constructed.

3. The intelligent control method for waste gas treatment in battery recycling as described in claim 2, characterized in that, The process involves simulating the dynamic adsorption biochemical catalysis model and predicting real-time operating conditions based on the collected simulation data to obtain predicted exhaust gas levels for a predetermined future timeframe. Specifically, this includes: The dynamic adsorption biochemical catalysis model is used to simulate the corresponding battery exhaust gas treatment device. During the simulation, data is collected in real time from the running dynamic adsorption biochemical catalysis model to obtain the simulation operation data of each dynamic adsorption biochemical catalysis model. The collected simulation data is input into a preset LSTM neural network, and the dynamic adsorption biochemical catalysis model is used to predict the real-time operating conditions within a preset future time, thereby obtaining the waste gas prediction results within the preset future time.

4. The intelligent control method for waste gas treatment in battery recycling as described in claim 3, characterized in that, The method for constructing the preset LSTM neural network specifically includes: Historical simulation data of the dynamic adsorption biochemical catalysis model were obtained, and the historical simulation data were preprocessed. Feature selection is performed on the preprocessed historical simulation data to obtain key historical data features; Based on the dimensions of the key historical data features, an initial LSTM neural network is constructed, and the loss function and optimization algorithm are set. The initial LSTM neural network is trained based on the loss function and optimization algorithm, and the trained LSTM neural network is evaluated through a preset test set to obtain a qualified LSTM neural network as the preset LSTM neural network.

5. The intelligent control method for waste gas treatment in battery recycling as described in claim 4, characterized in that, The process involves using the dynamic adsorption biochemical catalysis model, performing multi-objective optimization based on the waste gas prediction results to obtain control decisions for each waste gas treatment device, and controlling the waste gas treatment device according to the control decisions. Specifically, this includes: Based on the dynamic adsorption biochemical catalysis model, the energy consumption target, removal rate target, and solvent recovery rate target are determined. By constraining the energy consumption target, removal rate target, and solvent recovery rate target, multi-objective optimization conditions are obtained. The waste gas prediction results are used as a particle swarm, and based on the multi-objective optimization conditions, the fitness of each particle in the particle swarm is calculated, and the optimal solutions of individual particles and the global particle swarm are updated; wherein, each particle corresponds to the waste gas prediction result of each dynamic adsorption biochemical catalysis model, and the position and velocity of each particle are respectively represented as a point in its solution space and the direction of motion. Based on the optimal solutions for individual particles and the optimal solution for the global particle swarm, the velocity and position of each particle are updated until a preset iteration condition is met. Then, the optimal position of the global particle swarm is output as the optimization result. Based on the optimization results, the optimization objects and optimization objectives of the dynamic adsorption biochemical catalysis model are determined, thereby generating control decisions corresponding to each waste gas treatment device. Each of the aforementioned control decisions is sent to each waste gas treatment device, and the waste gas treatment device is controlled based on the corresponding control decision.

6. An intelligent control system for waste gas treatment in battery recycling, applied to the intelligent control method for waste gas treatment in battery recycling as described in any one of claims 1 to 5, characterized in that, include: A host computer for control and several battery exhaust gas treatment devices connected to the host computer for control; The battery exhaust gas treatment device is used to collect and adsorb the exhaust gas produced by the battery recycling equipment, and to collect exhaust gas data through sensors installed therein. The host computer for control includes: a response module, a model module, a simulation module, and an optimization module; The response module is used to acquire response signals emitted by sensors installed on the battery exhaust gas treatment device, and generate a sensor network based on the response signals; wherein each battery exhaust gas treatment device is equipped with at least one sensor, and each sensor emits a response signal corresponding to its identity information. The model module is used to collect data from the waste gas treatment device through a sensor network, extract features from the collected sensor data, and construct a dynamic adsorption biochemical catalysis model based on the sensor network and the extracted waste gas data features; wherein, each dynamic adsorption biochemical catalysis model corresponds to one waste gas treatment device. The simulation module is used to perform simulation operation through the dynamic adsorption biochemical catalysis model, and to perform real-time operating condition prediction based on the collected simulation operation data to obtain the waste gas prediction result within a preset future time. The optimization module is used to perform multi-objective optimization based on the dynamic adsorption biochemical catalysis model and the waste gas prediction results to obtain the control decisions corresponding to each waste gas treatment device, and to control the waste gas treatment device according to the control decisions.

7. A terminal device, characterized in that, The device 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 intelligent control method for exhaust gas treatment for battery recycling as described in any one of claims 1 to 5.

8. 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 containing the computer-readable storage medium to perform the intelligent control method for exhaust gas treatment for battery recycling as described in any one of claims 1 to 5.

Images