A method for treating malodorous and organic waste gases with enhanced absorption by pretreatment

Abstract

This application relates to the technical field of waste gas treatment, and discloses a method for treating malodorous and organic waste gas with enhanced absorption through pretreatment. The method includes: obtaining characteristic parameters of waste gas components through sensors; calculating pretreatment chemical agent formulations using a convolutional neural network; adding chemical agents according to instructions to remove easily treatable components and condition difficult-to-treat components; monitoring absorption liquid parameters to form characteristic indicators; adjusting gas-liquid distribution based on the indicators to construct an optimal environment; introducing the conditioned waste gas into the environment for enhanced absorption to obtain qualified gas. This application aims at malodorous and organic waste gas with complex and variable components, and through the combination of intelligent pretreatment and enhanced absorption, realizes the efficient and collaborative removal of easily treatable and difficult-to-treat components, while reducing the usage amount of chemical agents, improving treatment stability, and meeting the requirements of strict emission standards.

Description

Technical Field

[0001] This application relates to the technical field of waste gas treatment, and particularly to a method for treating malodorous and organic waste gas with enhanced absorption through pretreatment. Background Art

[0002] Malodorous and organic waste gases are common environmental pollutants in industries such as industrial production, livestock farming, sewage treatment, and waste treatment. Existing treatment technologies mainly include activated carbon adsorption method, chemical reagent washing method, biological filter method, and photocatalytic oxidation method, etc. Although the activated carbon adsorption method is simple to operate, its adsorption capacity is limited and the adsorption material needs to be replaced frequently; although the chemical reagent washing method has a relatively high treatment efficiency, it will result in a large amount of reagent dosing; the biological filter method has a low operating cost, but has high requirements for environmental conditions and a long start-up period; although the photocatalytic oxidation method can achieve deep degradation, it has high energy consumption and low efficiency when treating large-volume waste gases.

[0003] However, these traditional technologies face a common key problem: that is, they have insufficient adaptability to the complex and variable components of waste gases. Due to the large fluctuations in the components, concentrations, and physical properties of waste gases, the existing treatment processes with fixed parameters are difficult to effectively cope with, resulting in unreasonable use of reagents, low treatment efficiency, and energy waste. At the same time, most treatment methods have good effects on easily treatable components, but for difficult-to-treat components such as sulfur-containing organic compounds and high-molecular-weight aromatic compounds, there are often lack of effective pretreatment means and targeted enhanced absorption strategies, resulting in unstable treatment effects and difficulty in meeting the increasingly strict environmental protection discharge standards. Summary of the Invention

[0004] This application provides a method for treating malodorous and organic waste gas with enhanced absorption through pretreatment, which is used for treating malodorous and organic waste gas with complex and variable components. Through the combination of intelligent pretreatment and enhanced absorption, it realizes the efficient and synergistic removal of easily treatable and difficult-to-treat components, while reducing the dosage of reagents, improving the treatment stability, and meeting the requirements of strict discharge standards.

[0005] The present application provides a method for treating malodor and organic waste gas by pretreatment enhanced absorption, the method comprising: collecting waste gas component data by means of sensors, monitoring the waste gas components entering the pretreatment tower in real time, and obtaining waste gas component characteristic parameters; calculating the pretreatment agent formula by means of a convolutional neural network model according to the waste gas component characteristic parameters, and generating pretreatment agent addition instructions; adding the pretreatment agent to the pretreatment tower according to the pretreatment agent addition instructions, removing the easy-to-treat components in the waste gas, and performing interface conditioning on the difficult-to-treat components to obtain conditioned waste gas; monitoring the absorption liquid parameters by means of sensors in the enhanced absorption tower, collecting the absorption liquid pH value and concentration data, and forming absorption liquid characteristic indicators; adjusting the gas-liquid distribution in the enhanced absorption tower by means of a fluid dynamics model based on the absorption liquid characteristic indicators and the conditioned waste gas characteristics, and constructing an optimal absorption environment; introducing the conditioned waste gas into the optimal absorption environment, removing the malodor and organic matter in the waste gas by means of enhanced absorption, and obtaining a qualified emission gas.

[0006] In the technical solution provided by the present application, significant beneficial effects have been achieved through the organic combination of multiple technical features. By collecting exhaust gas composition data through sensors and monitoring in real time, dynamic grasp of exhaust gas composition is achieved, avoiding the limitations of traditional fixed parameter processing methods, and laying a data foundation for subsequent precise processing; the convolutional neural network model is used to calculate the pretreatment agent formula, giving full play to the advantages of artificial intelligence algorithms in complex data processing. The three-layer convolution structure design of the neural network model is specifically designed for the multi-dimensional characteristics of exhaust gas characteristic parameters. The progressive feature extraction structure significantly improves the accuracy of agent formula optimization, reduces the amount of agents used and improves the processing efficiency at the same time compared to traditional empirical formulation methods; precise agent addition is performed according to the pretreatment agent addition instructions, which not only efficiently removes easy-to-treat components, but also significantly improves the subsequent absorption efficiency of difficult-to-treat components through interface conditioning technology, making it difficult to remove in traditional processes. The removal rate of components such as methyl mercaptan and dimethyl sulfide has been significantly improved; the sensor monitoring system in the enhanced absorption tower has established a complete set of absorption liquid characteristic indicators, providing a real-time feedback mechanism for the absorption process; based on the absorption liquid characteristic indicators and the characteristics of the exhaust gas after conditioning, the fluid dynamics model realizes the intelligent adjustment of the gas-liquid distribution. The model integrates computational fluid dynamics with real-time monitoring data to create a dynamically responsive gas-liquid contact environment, solving the problems of uneven gas-liquid contact and local short circuit in traditional fixed parameter absorption towers, and improving the gas-liquid contact efficiency; finally, by introducing the exhaust gas after conditioning into the optimal absorption environment for enhanced absorption, efficient removal of odors and organic matter is achieved, the overall treatment efficiency is improved, energy consumption is reduced, and the treatment stability is significantly enhanced, and it has a strong adaptability to fluctuations in exhaust gas composition and concentration. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0008] Figure 1 It is a schematic diagram of an embodiment of the method for treating malodorous and organic waste gas with enhanced absorption by pretreatment in the embodiments of the present application. Specific embodiments

[0009] The embodiments of the present application provide a method for treating malodorous and organic waste gas with enhanced absorption by pretreatment. Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and the above accompanying drawings of the present application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0010] For ease of understanding, the following describes the specific process of the embodiments of the present application. Please refer to Figure 1 , an embodiment of the method for treating malodorous and organic waste gas with enhanced absorption by pretreatment in the embodiments of the present application includes:

[0011] Step S101: Collect waste gas composition data through sensors, monitor the waste gas composition entering the pretreatment tower in real time, and obtain waste gas composition characteristic parameters;

[0012] Step S102: According to the waste gas composition characteristic parameters, use a convolutional neural network model to calculate the pretreatment chemical agent formula and generate a pretreatment chemical agent dosing instruction;

[0013] Step S103: According to the pretreatment chemical agent dosing instruction, add the pretreatment chemical agent into the pretreatment tower to remove the easily treatable components in the waste gas, and at the same time perform interfacial conditioning on the difficult-to-treat components to obtain conditioned waste gas;

[0014] Step S104: Monitor the absorbent parameters through the sensors in the enhanced absorption tower, collect the pH value and concentration data of the absorbent, and form absorbent characteristic indicators;

[0015] Step S105: Based on the characteristic indexes of the absorption liquid and the characteristics of the conditioned waste gas, use the hydrodynamic model to adjust the gas-liquid distribution in the enhanced absorption tower and construct an optimal absorption environment.

[0016] Step S106: Introduce the conditioned waste gas into the optimal absorption environment, and remove the malodorous odor and organic matter in the waste gas through enhanced absorption to obtain the gas that meets the discharge standards.

[0017] It can be understood that the execution subject of this application can be a treatment system for malodorous odor and organic waste gas with pre-treatment and enhanced absorption, or a terminal or a server. Specifically, it is not limited here. In this embodiment of the application, the server is taken as the execution subject for illustration.

[0018] Specifically, waste gas composition data is collected through sensors to monitor the waste gas composition entering the pretreatment tower in real time, and waste gas composition characteristic parameters are obtained. In this step, multiple groups of gas component detectors quantitatively detect hydrogen sulfide, ammonia, methanethiol, dimethyl sulfide, and dimethyl disulfide in the waste gas to obtain waste gas component concentration data. The airflow parameter acquisition unit continuously samples the temperature, humidity, pressure, and flow rate of the waste gas to form waste gas physical property data. The emission source type identification program classifies and marks the waste gas source, and generates waste gas source characteristic values in combination with the industry characteristic database. The time series data processing algorithm performs time series analysis on the waste gas component concentration data to generate a component fluctuation trend map. Data clustering processing performs correlation analysis on the waste gas component concentration data and the waste gas physical property data to construct a multi-dimensional waste gas characteristic correlation matrix. Finally, the waste gas component concentration data, the waste gas physical property data, the waste gas source characteristic values, and the component fluctuation trend map are subjected to data fusion processing to comprehensively form waste gas composition characteristic parameters. According to the waste gas composition characteristic parameters, a convolutional neural network model is used to calculate the pretreatment chemical agent formula, and a pretreatment chemical agent dosing instruction is generated. Waste gas treatment case data is extracted from the historical treatment database and matched with the waste gas composition characteristic parameters to form a waste gas treatment similarity matrix. The waste gas composition characteristic parameters are processed through a three-layer convolutional structure. In the first layer of convolution, a 5×5 convolution kernel is used to extract the basic characteristics of the waste gas components. In the second layer of convolution, a 3×3 convolution kernel is used to capture the interaction characteristics between the components. In the third layer of convolution, a 1×1 convolution kernel is used to integrate the characteristic information. After being processed by the ReLU activation function, a waste gas component characteristic map is generated. The waste gas component characteristic map is sent into the fully connected layer network. Through the linear transformation of the weight matrix and the bias vector, combined with the Softmax function, the reactivity of the waste gas components is quantified, and the waste gas component reactivity index is calculated. Based on the waste gas component reactivity index, a chemical agent-component interaction matrix is constructed, and the component sensitivity parameter and the chemical agent reaction ability parameter are separated through the matrix decomposition algorithm. Accordingly, a chemical agent candidate set is selected from oxidants, acid-base neutralizing agents, complexing agents, and surfactants. The chemical agent candidate set is input into the decision layer of the convolutional neural network. After dimensionality reduction through pooling operations, it is connected with the historical reaction efficiency data through residual connections to generate a chemical agent ratio prediction tensor. Then, the ratio parameters are optimized through backpropagation to form a chemical agent ratio efficiency map. Finally, according to the gradient descent method, the optimal ratio point is extracted from the chemical agent ratio efficiency map, and combined with the cost coefficient matrix for constraint optimization to determine the optimal chemical agent formula, and this formula is converted into a time series control instruction to generate a pretreatment chemical agent dosing instruction.

[0019] According to the dosing instruction of the pretreatment agent, the pretreatment agent is added into the pretreatment tower to remove the easily treatable components in the waste gas, and at the same time, the interface of the difficult-to-treat components is conditioned to obtain the conditioned waste gas. In this step, the pretreatment agent is dosed to different height positions in the pretreatment tower according to the concentration gradient through a multi-point distribution injection system to form a layered structure of the agent concentration. The swirl intensification device tangentially mixes the waste gas and the pretreatment agent to generate a waste gas-agent turbulent contact zone, enabling the hydrogen sulfide and ammonia in the waste gas to react with the agent in a rapid redox reaction to form primary reaction products. The temperature gradient control unit in the tower regulates the reaction temperature of the waste gas-agent, and promotes the catalytic degradation of methyl mercaptan and dimethyl sulfide through the gradient distribution of the temperature field to generate intermediates with enhanced polarity. The interfacial activity regulator adjusts the surface tension of the primary reaction products and the intermediates with enhanced polarity, reduces the transfer resistance of the gas-liquid interface of the difficult-to-treat components, and forms a microemulsion phase. The microemulsion phase is sent to the molecular polarity regulation area, and the molecular polarity of the difficult-to-treat components is changed through the action of the polarity field to increase their water solubility, generating a waste gas stream with enhanced polarity. Finally, the waste gas stream with enhanced polarity is subjected to gas-solid separation treatment to remove the particulate matter and aerosol generated during the reaction process to obtain the conditioned waste gas.

[0020] The parameters of the absorption liquid are monitored through sensors in the enhanced absorption tower, and the pH value and concentration data of the absorption liquid are collected to form the characteristic indexes of the absorption liquid. During this process, the liquid sampling sensor group samples the absorption liquid at regular intervals to obtain the original absorption liquid sample. The pH microelectrode sensor detects the real-time pH value of the original absorption liquid sample, generates pH data in a continuous time series, and constructs a record of the spatial distribution of the pH value in the tower. The conductivity probe measures the total ion concentration in the original absorption liquid sample, records the change trend of the conductivity of the absorption liquid, and obtains the absorption capacity attenuation index. The ultraviolet absorption spectrum analysis quantitatively analyzes the content of dissolved organic matter in the original absorption liquid sample, measures the organic matter concentration value, and forms an organic load data set. The data correlation analysis processor integrates and calculates the spatial distribution record of the pH value, the absorption capacity attenuation index, and the organic load data set to generate the evaluation parameters of the absorption liquid state. The evaluation parameters of the absorption liquid state are compared and analyzed with the historical absorption efficiency data, and the value representing the current performance of the absorption liquid is calculated through the weighted average algorithm to form the characteristic indexes of the absorption liquid.

[0021] Based on the absorption liquid characteristic index and the exhaust gas characteristics after conditioning, the fluid dynamics model is used to adjust the gas-liquid distribution in the enhanced absorption tower to construct the optimal absorption environment. This step uses a multiphase flow analyzer to extract the absorption liquid characteristic index in a refined parameter manner and generate a liquid phase flow characteristic map. The gas flow performance is calculated according to the exhaust gas characteristics after conditioning to form gas phase diffusion pattern data. The liquid phase flow characteristic map and the gas phase diffusion pattern data are input into the computational fluid dynamics analysis system to calculate the gas-liquid contact probability distribution. According to the gas-liquid contact probability distribution, the injection angle and flow distribution of the spray device in the enhanced absorption tower are adjusted to form a gradient distribution absorption area. The boundary layer perturbation of the gradient distribution absorption area is adjusted by the turbulence intensity optimizer to improve the gas-liquid exchange efficiency and generate a multi-dimensional interactive absorption pattern. The multi-dimensional interactive absorption pattern is dynamically detected and calibrated to construct the optimal absorption environment.

[0022] The conditioned exhaust gas is introduced into the optimal absorption environment, and the odor and organic matter in the exhaust gas are removed by enhanced absorption to obtain qualified exhaust gas. In this final step, the porous distributor evenly introduces the conditioned exhaust gas into the optimal absorption environment to form a microbubble flow state. The ion activation device performs polarity enhancement treatment on the absorption liquid in the optimal absorption environment to generate a high-reactivity absorption zone. According to the microbubble flow state and the high-reactivity absorption zone, the temperature gradient in the absorption tower is adjusted to form a temperature-controlled absorption field. The reaction rate data of the exhaust gas and the absorption liquid are collected from the temperature-controlled absorption field to establish a reaction kinetic parameter library. Based on the reaction kinetic parameter library, the residence time in the absorption tower is dynamically controlled through an intelligent feedback mechanism to construct a layered absorption sequence. The purified gas generated by the layered absorption sequence is finally degraded through a catalytic conversion unit to obtain qualified exhaust gas.

[0023] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0024] (1) Quantitatively detect hydrogen sulfide, ammonia, methyl mercaptan, dimethyl sulfide, and dimethyl disulfide in the exhaust gas through multiple sets of gas component detectors to obtain exhaust gas component concentration data;

[0025] (2) The airflow parameter acquisition unit continuously samples the temperature, humidity, pressure and flow rate of the exhaust gas to form the physical characteristics data of the exhaust gas;

[0026] (3) Classify and label the exhaust gas sources based on the emission source type identification procedure, and generate exhaust gas source characteristic values in combination with the industry characteristic database;

[0027] (4) Perform time series analysis on the exhaust gas component concentration data based on the time series data processing algorithm to generate a component fluctuation trend graph;

[0028] (5) Conduct a correlation analysis on the waste gas component concentration data and the waste gas physical property data through data clustering processing, and construct a multi-dimensional waste gas characteristic correlation matrix;

[0029] (6) Perform data fusion processing on the waste gas component concentration data, the waste gas physical property data, the waste gas source characteristic values, and the component fluctuation trend spectrograms to comprehensively form waste gas composition characteristic parameters.

[0030] Specifically, a multi-group gas component detector is used to quantitatively detect hydrogen sulfide, ammonia, methyl mercaptan, dimethyl sulfide, and dimethyl disulfide in the waste gas to obtain waste gas component concentration data. The multi-group gas component detector refers to a detection system composed of multiple gas sensors with different selectivities, and each sensor has a high-sensitivity response to a specific gas component. For example, an electrochemical sensor is used for the detection of hydrogen sulfide, and a current signal is generated by the redox reaction occurring on the electrode surface when the gas reacts with it; for the detection of ammonia, a semiconductor metal oxide sensor is used, and the change in resistance caused by the gas adsorbed on the semiconductor surface is used for detection. These sensors work simultaneously to perform real-time detection on different components, convert the electrical signals into concentration data, and represent the content of each component in ppm.

[0031] The air flow parameter acquisition unit continuously samples the temperature, humidity, pressure, and flow rate of the waste gas to form waste gas physical property data. The air flow parameter acquisition unit consists of a thermocouple temperature sensor, a capacitive humidity sensor, a pressure transmitter, and a thermal gas flowmeter, which respectively measure the temperature (°C), relative humidity (%), pressure (kPa), and flow rate (m / s) of the waste gas. These physical parameters directly affect the waste gas treatment efficiency. For example, temperature affects the reaction rate, humidity affects the absorption of water-soluble gases, and pressure and flow rate affect the residence time of the gas in the tower. These parameters are sampled at a frequency of once per second to form a continuous physical property data stream.

[0032] Classify and label the waste gas source based on the emission source type identification program, and generate waste gas source characteristic values in combination with the industry characteristic library. The emission source type identification program is a data analysis algorithm that divides the waste gas source into different industrial categories, such as chemical plants, pharmaceutical factories, food processing factories, etc., by identifying the characteristic patterns of the waste gas components. This program first extracts the proportional relationship of the waste gas component concentrations, and then performs pattern matching with the typical waste gas characteristics stored in the industry characteristic library. The industry characteristic library is a data set containing the typical waste gas component spectrograms of each industry. By calculating the cosine similarity between the waste gas characteristics and the templates in the library, the most likely source industry of the waste gas is determined, and a waste gas source characteristic value between 0 and 1 is generated, indicating the matching degree of the waste gas with the specific industry characteristics.

[0033] Perform time - series analysis on the exhaust gas component concentration data according to the time - series data processing algorithm to generate a component fluctuation trend map. The time - series data processing algorithm includes moving average, exponential smoothing, and time - window analysis, which are used to process the continuously collected exhaust gas component concentration data. The moving average method averages the concentration data of the nearest N time points to eliminate short - term fluctuation interference; the exponential smoothing method gives higher weights to recent data to better reflect the latest trend; the time - window analysis observes the fluctuation patterns at different time scales. The component fluctuation trend map is a multi - dimensional time - series visualization expression, with the horizontal axis representing time, the vertical axis representing different components, and the color shade representing the concentration change, intuitively showing the change trends and interrelationships of each component over time.

[0034] Perform correlation analysis on the exhaust gas component concentration data and the exhaust gas physical property data through data clustering processing to construct a multi - dimensional exhaust gas property correlation matrix. The data clustering processing uses K - means and hierarchical clustering algorithms to group similar exhaust gas component and physical property data points. The K - means algorithm divides the data into K clusters by minimizing the Euclidean distance from the data points to the cluster centers; hierarchical clustering gradually merges adjacent clusters by calculating the distances between data points. The correlation analysis quantifies the mutual relationship between the exhaust gas component concentration and the physical properties by calculating the Pearson correlation coefficient. The multi - dimensional exhaust gas property correlation matrix is an M×N matrix, where M represents the number of exhaust gas components and N represents the number of physical property parameters, and each element value in the matrix represents the correlation strength between the corresponding component and the physical property.

[0035] Perform data fusion processing on the exhaust gas component concentration data, the exhaust gas physical property data, the exhaust gas source characteristic values, and the component fluctuation trend map to comprehensively form exhaust gas composition characteristic parameters. The data fusion processing uses the Dempster - Shafer evidence theory framework to integrate information from different data sources into a unified set of characteristic parameters. This method first converts each data source into a basic probability assignment function and then calculates the fused result through the DS combination rule. The exhaust gas composition characteristic parameters are a multi - dimensional characteristic vector, including the component composition, physical state, time - varying characteristics, and source information of the exhaust gas, providing complete input information for the subsequent calculation of the pretreatment chemical agent formula.

[0036] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0037] (1) Extract the exhaust gas treatment case data from the historical treatment database, match and compare it with the exhaust gas composition characteristic parameters to form an exhaust gas treatment similarity matrix;

[0038] (2) Process the characteristic parameters of waste gas components through a three-layer convolutional structure. In the first layer of convolution, a 5×5 convolutional kernel is used to extract the basic characteristics of waste gas components. In the second layer of convolution, a 3×3 convolutional kernel is used to capture the interaction characteristics between components. In the third layer of convolution, a 1×1 convolutional kernel is used to integrate the feature information. After being processed by the ReLU activation function, a feature map of waste gas components is generated.

[0039] (3) Feed the feature map of waste gas components into the fully connected layer network. Through the linear transformation of the weight matrix and the bias vector, and combined with the Softmax function, quantify the reactivity of waste gas components, and calculate the reactivity index of waste gas components.

[0040] (4) Construct a reagent-component interaction matrix based on the reactivity index of waste gas components. Through the matrix decomposition algorithm, separate the component sensitivity parameter and the reagent reaction ability parameter, and accordingly screen out the reagent candidate set from oxidants, acid-base neutralizing agents, complexing agents, and surfactants.

[0041] (5) Input the reagent candidate set into the decision layer of the convolutional neural network. After dimensionality reduction through pooling operations, perform residual connection with the historical reaction efficiency data to generate a reagent ratio prediction tensor, and then optimize the ratio parameters through backpropagation to form a reagent ratio efficiency map.

[0042] (6) Extract the optimal ratio point from the reagent ratio efficiency map according to the gradient descent method, perform constraint optimization in combination with the cost coefficient matrix, determine the optimal reagent formula, and convert the optimal reagent formula into a time series control instruction to generate a pretreatment reagent dosing instruction.

[0043] Specifically, extract the waste gas treatment case data from the historical treatment database, match and compare it with the characteristic parameters of waste gas components to form a waste gas treatment similarity matrix. The historical treatment database is a structured data set storing past waste gas treatment records, including waste gas characteristics, treatment methods, reagent formulas, and treatment effects, etc. The matching and comparison process uses the cosine similarity calculation method to compare the current waste gas component characteristic parameters with the cases in the database and calculate the similarity degree. The cosine similarity measures the similarity between two vectors by calculating the cosine value of the angle between them. The closer the value is to 1, the more similar they are. Each element in the waste gas treatment similarity matrix represents the similarity between a certain component of the current waste gas and the corresponding component in the historical case.

[0044] The characteristic parameters of waste gas components are processed through a three-layer convolutional structure to achieve feature extraction and transformation. The convolution operation is a sliding window-based local feature extraction method that captures feature patterns at different scales through convolution kernels of different sizes. The first layer of convolution uses a 5×5 convolution kernel to extract the basic features of waste gas components. The large-sized convolution kernel can capture the component feature relationships in a wider range and extract low-level features from the original feature parameters, such as the component concentration distribution feature and the physical property distribution feature. The second layer of convolution uses a 3×3 convolution kernel to capture the interaction features between components. The medium-sized convolution kernel focuses on extracting finer local feature patterns, such as the cooperative relationship and inhibitory relationship between different components. The third layer of convolution uses a 1×1 convolution kernel to integrate the feature information. The 1×1 convolution kernel is mainly used for information integration and dimensionality reduction between channels, and compresses and integrates the features extracted in the first two layers. Subsequently, it is processed by the ReLU activation function. The form of the ReLU function is f(x)=max(0,x), which can introduce non-linearity and retain the positive activation values, eliminating the negative value interference, and finally generate the waste gas component feature map, which is a multi-dimensional tensor containing the high-level abstract feature representation of waste gas components.

[0045] The waste gas component feature map is fed into the fully connected layer network. Through the linear transformation of the weight matrix and the bias vector, combined with the Softmax function, the reactivity of waste gas components is quantified. The fully connected layer network is a structure in which each neuron in the neural network is connected to all neurons in the previous layer. It performs a linear transformation on the input x through the weight matrix W and the bias vector b: y = Wx + b to achieve the mapping from features to the target output. The Softmax function converts the multi-dimensional vector into a probability distribution form. Through this process, the waste gas components are mapped into the reaction activity space, and the waste gas component reaction activity index is calculated, which is a vector representing the reaction activity degree of each component. The larger the value, the stronger the reactivity of the component.

[0046] Based on the waste gas component reaction activity index, a reagent-component interaction matrix is constructed, and the key parameters are separated through the matrix decomposition algorithm. The reagent-component interaction matrix is a p×q matrix, where p is the number of waste gas components and q is the number of candidate reagents. Each element in the matrix represents the reaction intensity between a specific component and a specific reagent. The matrix decomposition algorithm uses the singular value decomposition (SVD) method to decompose the original interaction matrix M into the product of three matrices, specifically the component sensitivity parameter matrix, the reagent reaction ability parameter matrix, and the singular value matrix, which represent the importance of each feature. The component sensitivity parameter describes the response degree of each component to different types of reagents, and the reagent reaction ability parameter characterizes the processing ability of each reagent to different components. Accordingly, a reagent candidate set is screened from oxidants (such as sodium hypochlorite, hydrogen peroxide), acid-base neutralizing agents (such as sodium hydroxide, sulfuric acid), complexing agents (such as EDTA, citric acid), and surfactants (such as sodium dodecyl sulfate), and the reagents with high reactivity to the current waste gas components are selected.

[0047] Input the set of pharmaceutical candidates into the decision-making layer of the convolutional neural network and perform dimensionality reduction processing through pooling operations. Pooling operation is a feature dimensionality reduction method. The commonly used ones are max pooling and average pooling. By taking the maximum or average value in the local area of the feature map, the data dimension is reduced while key features are retained. The features after dimensionality reduction are connected with the historical reaction efficiency data through residual connection. Residual connection is a special network connection method. By directly adding the input to the output, it alleviates the difficulty of deep network training. The form is y = F(x) + x, where F(x) is the residual function learned by the network. In this way, the features of the pharmaceutical candidate set and the historical treatment effect data are fused to generate a pharmaceutical ratio prediction tensor, which is a multi-dimensional data structure containing the predicted treatment effects of various possible pharmaceutical ratios. Subsequently, the ratio parameters are optimized through the backpropagation algorithm. Backpropagation is the core algorithm for neural network training. By calculating the gradients of the loss function with respect to the parameters of each layer and adjusting the parameters along the gradient direction, continuous iteration is performed until convergence. Finally, a pharmaceutical ratio efficacy map is formed, which is an efficacy surface in a multi-dimensional space. Each point represents a pharmaceutical ratio scheme, and the height of the point represents the expected treatment efficacy.

[0048] Extract the optimal ratio point from the pharmaceutical ratio efficacy map according to the gradient descent method. Gradient descent is an optimization algorithm. By calculating the gradient of the function and moving along the negative gradient direction, the local minimum or global minimum of the function is searched. First, define the objective function J(θ) to represent the negative value of the treatment effect, where θ is the vector of pharmaceutical ratio parameters; then calculate the gradient ∇J(θ) of J(θ) with respect to θ; finally, update the parameters along the negative gradient direction and control the step size of each update. Through multiple iterations, it finally converges to the minimum point of the objective function to obtain the optimal pharmaceutical ratio. Subsequently, combined with the cost coefficient matrix for constrained optimization. The cost coefficient matrix contains the unit cost information of each pharmaceutical. By using the Lagrange multiplier method, the cost constraint is introduced into the optimization objective to minimize the pharmaceutical cost while ensuring the treatment effect. Finally, the optimal pharmaceutical formula is determined, and the optimal pharmaceutical formula is converted into a time series control instruction, which includes information such as the dosage, dosing time point, and dosing order of each pharmaceutical, to generate a pretreatment pharmaceutical dosing instruction.

[0049] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0050] (1) Through a multi-point distribution spraying system, add the pretreatment pharmaceutical to different height positions in the pretreatment tower according to the concentration gradient to form a pharmaceutical concentration stratified structure;

[0051] (2) The tangential mixing of the waste gas and the pretreatment pharmaceutical is carried out by a cyclone intensification device to generate a waste gas-pharmaceutical turbulent contact zone, so that hydrogen sulfide and ammonia in the waste gas react with the pharmaceutical quickly through oxidation-reduction reaction to generate primary reaction products;

[0052] (3) Adjust the reaction temperature of the waste gas and the reagent by using the temperature gradient control unit in the tower, and promote the catalytic degradation of methanethiol and dimethyl sulfide through the gradient distribution of the temperature field to generate intermediates with enhanced polarity.

[0053] (4) Based on the interfacial activity regulator, adjust the surface tension of the primary reaction products and the intermediates with enhanced polarity, reduce the transfer resistance of the gas-liquid interface of the difficult-to-treat components, and form a microemulsion phase.

[0054] (5) Feed the microemulsion phase into the molecular polarity regulation area, change the molecular polarity of the difficult-to-treat components through the action of the polarity field, increase their water solubility, and generate an exhaust gas stream with enhanced polarity.

[0055] (6) Perform gas-solid separation treatment on the exhaust gas stream with enhanced polarity to remove the particulate matter and aerosol generated during the reaction process, and obtain the conditioned exhaust gas.

[0056] Specifically, the pretreatment reagent is added to different height positions in the pretreatment tower according to the concentration gradient through a multi-point distribution injection system to form a reagent concentration stratification structure. The multi-point distribution injection system is a three-dimensional arrangement device composed of multiple nozzles. A number of injection points are arranged along the height direction of the pretreatment tower, and each injection point is equipped with a metering pump and a flow controller with independent control. Adding the pretreatment reagent according to the concentration gradient means that according to the vertical distribution characteristics of the waste gas components in the tower, reagents with different concentrations are added at different heights. High-concentration reagents are added in the bottom area to treat heavy components, medium-concentration reagents are added in the middle area to treat medium components, and low-concentration reagents are added in the top area to treat light components. The reagent concentration stratification structure is a concentration gradient distribution form in the vertical direction, ensuring that all types of waste gas components can contact the reagent with an appropriate concentration and improving the reaction efficiency.

[0057] The waste gas and the pretreatment reagent are tangentially mixed by a swirl intensification device to generate a waste gas-reagent turbulent contact area, enabling the hydrogen sulfide and ammonia in the waste gas to undergo a rapid oxidation-reduction reaction with the reagent to generate primary reaction products. The swirl intensification device is a specially designed gas-liquid mixing device that includes a tangential air inlet and a guide vane, causing the waste gas and the liquid reagent to form a spiral flow path. Tangential mixing refers to the mutual blending of the gas flow and the liquid flow in a non-central axis direction, generating a strong rotational motion through the velocity component in the tangential direction. The waste gas-reagent turbulent contact area is a fluid area where the Reynolds number is higher than the critical value. The Reynolds number is a dimensionless parameter characterizing the fluid motion state, and its calculation formula is Re = ρvd / μ, where ρ is the fluid density, v is the fluid velocity, d is the characteristic length, and μ is the fluid dynamic viscosity. In the turbulent contact area, the hydrogen sulfide in the waste gas undergoes an oxidation reaction with an oxidant (such as hydrogen peroxide) to generate sulfuric acid, and the ammonia undergoes a neutralization reaction with an acidic reagent to generate ammonium salts. These reaction products are collectively referred to as primary reaction products.

[0058] The waste gas-agent reaction temperature is regulated by the temperature gradient control unit inside the tower, and the catalytic degradation of methanethiol and dimethyl sulfide is promoted through the gradient distribution of the temperature field to generate intermediates with enhanced polarity. The temperature gradient control unit is a system composed of temperature sensors and heating / cooling devices distributed at different heights in the pretreatment tower, which can form a preset temperature distribution inside the tower. The temperature field gradient distribution refers to the regular change of temperature along the tower height direction. Usually, the temperature at the bottom is higher and the temperature at the top is lower, forming a temperature decreasing gradient. Catalytic degradation refers to the process in which the molecular structures of sulfur-containing organic compounds such as methanethiol (-S-CH3) and dimethyl sulfide (CH3-S-CH3) are destroyed under the action of a catalyst (such as transition metal ions), generating compounds with higher polarity such as sulfinic acid and sulfonic acid. The intermediate with enhanced polarity refers to a compound whose oxygen-containing functional groups increase and molecular polarity significantly improves after oxidation, such as methylsulfinic acid (CH3-SO-OH) generated after the oxidation of methanethiol.

[0059] Based on the interfacial activity regulator, the surface tension of the primary reaction products and the intermediate with enhanced polarity is adjusted to reduce the transfer resistance of the difficult-to-treat components at the gas-liquid interface and form a microemulsion phase. The interfacial activity regulator is a class of amphiphilic molecules that can reduce the surface tension of the gas-liquid interface, such as non-ionic surfactant polyethylene glycol and anionic surfactant sodium dodecyl sulfate. The surface tension adjustment is a process of reducing the interfacial free energy through the directional arrangement of surfactant molecules at the gas-liquid interface. The gas-liquid interface transfer resistance refers to the mass transfer resistance encountered during the transfer of gaseous pollutants to the liquid phase, mainly due to the boundary layer effect of the gas-liquid two phases. The microemulsion phase is a special dispersion system, a thermodynamically stable system formed by the aqueous phase, oil phase and surfactant, with a particle size usually in the range of 1-100 nm, having high dispersibility and a large specific surface area.

[0060] Feed the microemulsion phase into the molecular polarity regulation zone. By the action of the polarity field, change the polarity of the molecules of the difficult-to-treat components, increase their water solubility, and generate an exhaust gas stream with enhanced polarity. The molecular polarity regulation zone is a specially designed reaction area in the pretreatment tower, equipped with an electric field or a special catalyst, which can affect the molecular charge distribution. The action of the polarity field refers to the process of inducing non-polar or weakly polar molecules to generate temporary dipole moments or enhancing the existing dipole moments through an externally applied electric field or a polar solvent environment. The change in the polarity of the difficult-to-treat component molecules refers to introducing polar groups (such as -OH, -COOH, etc.) into the molecular structure through addition, oxidation and other reactions, or changing the spatial conformation of the original groups, so as to enhance the overall polarity of the molecules. The increase in water solubility is the result of the enhanced interaction between the molecules and water molecules through hydrogen bonding or ion-dipole interactions after the molecular polarity is enhanced. The exhaust gas stream with enhanced polarity is the gas stream after polarity regulation treatment, in which the pollutant components have higher water solubility and are more easily captured in the subsequent absorption process. Perform gas-solid separation treatment on the exhaust gas stream with enhanced polarity to remove the particulate matter and aerosols generated during the reaction process, and obtain the conditioned exhaust gas. Gas-solid separation treatment is a process of separating solid particulate matter in the gas by physical methods. Common methods include gravity sedimentation, inertial separation, filtration, etc. Particulate matter refers to solid particles suspended in the gas, and the particle size is usually greater than 1μm. Aerosol refers to a dispersion system of liquid droplets or solid particles suspended in the gas, and the particle size is usually between 0.01 - 10μm. The conditioned exhaust gas refers to the exhaust gas after passing through each treatment link of the pretreatment tower. Its characteristic is that the easy-to-treat components have been removed, and the physical and chemical properties of the difficult-to-treat components have been adjusted, making it more suitable for subsequent enhanced absorption treatment.

[0061] In a specific embodiment, the process of performing step S104 may specifically include the following steps:

[0062] (1) Regularly sample the absorption liquid through a liquid sampling sensor group installed at different height positions in the enhanced absorption tower, and obtain the original absorption liquid samples at each sampling point;

[0063] (2) Use a pH microelectrode sensor to detect the real-time pH value of the original absorption liquid sample, generate pH data in a continuous time series, and construct a record of the spatial distribution of the pH value in the tower;

[0064] (3) Use a conductivity probe to measure the total ion concentration in the original absorption liquid sample, record the change trend of the absorption liquid conductivity, and obtain the absorption capacity attenuation index;

[0065] (4) Based on ultraviolet absorption spectroscopy analysis, quantitatively analyze the content of dissolved organic matter in the original absorption liquid sample, measure the organic matter concentration value, and form an organic load data set;

[0066] (5) Integrate and calculate the pH value spatial distribution record, absorption capacity attenuation index, and organic load data set through a data correlation analysis processor to generate absorption liquid state evaluation parameters;

[0067] (6) Compare and analyze the absorption liquid state evaluation parameters with historical absorption efficiency data, and calculate the value representing the current absorption liquid performance through a weighted average algorithm to form absorption liquid characteristic indicators.

[0068] Specifically, the absorption liquid is sampled regularly through a liquid sampling sensor group installed at different height positions in the enhanced absorption tower, and the original absorption liquid sample is obtained at each sampling point. The liquid sampling sensor group is a multi-point distributed sampling device, with multiple sampling points evenly arranged along the height direction of the enhanced absorption tower, usually 3 - 5, located in the bottom, middle, and top regions of the tower respectively. Regular sampling refers to the process of automatically collecting absorption liquid samples at preset time intervals (generally 5 - 15 minutes). The original absorption liquid sample refers to the untreated liquid sample directly collected from different positions in the absorption tower, which contains various pollutant information captured during the absorption process.

[0069] Use a pH microelectrode sensor to detect the real-time pH value of the original absorption liquid sample, generate pH data in a continuous time series, and construct a pH value spatial distribution record in the tower. The pH microelectrode sensor is a micro sensor that can accurately measure the acidity and alkalinity of a liquid. It uses the principle of ion-selective electrodes to generate a corresponding potential difference by measuring the hydrogen ion activity and converts it into a pH value. Real-time pH value detection refers to the pH value measurement immediately after sample collection, avoiding possible changes in the sample during storage. The pH data in a continuous time series refers to a series of pH values recorded in chronological order, forming a data set that changes over time. The pH value spatial distribution record in the tower refers to the three-dimensional distribution of pH values at different height positions in the tower, which can intuitively reflect the change of acid-base balance during the absorption process.

[0070] Use a conductivity probe to measure the total ion concentration in the original absorption liquid sample, record the change trend of the absorption liquid conductivity, and obtain the absorption capacity attenuation index. The conductivity probe is a device that measures the conductivity of a liquid. It calculates the conductivity by applying an alternating voltage to the liquid and measuring the current, with the unit of Siemens / meter. The total ion concentration refers to the sum of the concentrations of all positive and negative ions in the liquid, which is directly proportional to the conductivity. The change trend of conductivity refers to the increase or decrease of conductivity over time. Usually, as the use time of the absorption liquid extends, the conductivity will gradually increase. The absorption capacity attenuation index is an index calculated based on the conductivity change rate, which characterizes the degree of weakening of the absorption liquid's ability to capture pollutants. The larger the index value, the more urgent the need for absorption liquid replacement or regeneration.

[0071] Quantitatively analyze the content of dissolved organic matter in the original absorbent sample based on ultraviolet absorption spectroscopy analysis, measure the organic matter concentration value, and form an organic load data set. Ultraviolet absorption spectroscopy analysis is an analytical method based on the absorption characteristics of substances to ultraviolet light, usually measuring the absorbance of samples in the wavelength range of 190 - 400 nm. The content of dissolved organic matter refers to the total amount of organic pollutants captured in the absorbent, usually expressed as total organic carbon (TOC) or chemical oxygen demand (COD). The organic matter concentration value is a specific concentration value calculated according to Lambert-Beer's law based on the absorbance value of the sample at a specific wavelength, and the unit is usually mg / L. The organic load data set is a comprehensive data set containing organic matter concentrations at different sampling points and different time points, reflecting the spatial distribution and temporal variation of the organic matter load in the absorbent.

[0072] Integrate and calculate the pH value spatial distribution record, absorption capacity attenuation index, and organic load data set through a data correlation analysis processor to generate absorbent state evaluation parameters. The data correlation analysis processor is a data fusion and analysis tool that can process multi-source heterogeneous data and mine its internal correlations. The integration calculation includes three steps: data standardization, correlation analysis, and feature extraction. First, standardize data with different dimensions to a unified interval; then calculate the correlation coefficients between parameters to identify mutual influence relationships; finally, extract key features through principal component analysis. The absorbent state evaluation parameters are a set of comprehensive indicators reflecting the current working state of the absorbent, including multiple dimensions such as acid-base balance state, ion saturation degree, and organic load level.

[0073] Compare and analyze the absorbent state evaluation parameters with historical absorption efficiency data, and calculate the value representing the current absorbent performance through a weighted average algorithm to form absorbent characteristic indicators. Historical absorption efficiency data refers to the parameter and efficiency data of past absorption processes recorded in the database, containing absorption effect information under different operating conditions. The comparative analysis is to compare the current absorbent state evaluation parameters with the data under similar operating conditions in historical data to determine the position of the current absorbent state in the historical data distribution. The weighted average algorithm is an average calculation method that takes into account the importance differences of various factors, assigns different weights to different parameters, and calculates the comprehensive result. The absorbent characteristic indicator is a normalized value between 0 and 1, representing the comprehensive performance state of the absorbent, where 1 represents the best state and 0 represents the completely failed state.

[0074] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0075] (1) Refined parameter extraction of the absorbent characteristic indicators through a multiphase flow analyzer to generate a liquid-phase flow characteristic map;

[0076] (2)Calculate the gas flow performance according to the characteristics of the quenched and tempered waste gas to form gas-phase diffusion mode data;

[0077] (3)Input the liquid-phase flow characteristic map and the gas-phase diffusion mode data into the computational fluid dynamics analysis system to calculate the gas-liquid contact probability distribution;

[0078] (4)Adjust the spraying angles and flow rate distributions of the spraying devices in the enhanced absorption tower according to the gas-liquid contact probability distribution to form a gradient distribution absorption area;

[0079] (5)Use a turbulence intensity optimizer to adjust the boundary layer perturbation of the gradient distribution absorption area to improve the gas-liquid exchange efficiency and generate a multi-dimensional interactive absorption pattern;

[0080] (6)Conduct dynamic detection and calibration on the multi-dimensional interactive absorption pattern to construct an optimal absorption environment.

[0081] Specifically, a multi-phase flow analyzer is used to extract refined parameters of the absorption liquid characteristic indicators to generate a liquid-phase flow characteristic map. The multi-phase flow analyzer is a professional device that can characterize the liquid flow characteristics. By measuring the physical parameters of the absorption liquid such as density, viscosity, surface tension, and rheology, the behavior characteristics of the liquid under different flow conditions are obtained. Refined parameter extraction refers to separating the key factors affecting the liquid flow performance from the absorption liquid characteristic indicators, such as the viscosity-temperature relationship curve, surface tension coefficient, density gradient, etc. The liquid-phase flow characteristic map is a multi-dimensional data visualization expression that shows the flow characteristic distribution of the absorption liquid under different temperature, pressure, and flow rate conditions, providing basic data for the subsequent optimization of the gas-liquid contact form. Calculate the gas flow performance according to the characteristics of the quenched and tempered waste gas to form gas-phase diffusion mode data. The gas flow performance calculation is a process of comprehensively analyzing the gas flow rate, pressure, temperature, and component characteristics based on the gas dynamics theory. The calculation process takes into account parameters such as the density, viscosity, and diffusion coefficient of the gas, and is corrected in combination with the waste gas component characteristics (such as molecular weight, polarity, etc.). The gas-phase diffusion mode data is a data set describing the flow and diffusion laws of the gas in the absorption tower, including the gas flow field distribution, concentration gradient distribution, and the change of the diffusion coefficient with the tower height, reflecting the motion characteristics of the gas in different regions.

[0082] Input the liquid-phase flow characteristic map and gas-phase diffusion mode data into the computational fluid dynamics (CFD) analysis system to calculate the gas-liquid contact probability distribution. The CFD analysis system is a fluid flow behavior prediction tool based on numerical simulation, which uses the finite element or finite difference method to solve the Navier-Stokes equations and simulate the flow behavior of gas-liquid two-phase in a complex space. The system discretizes the space inside the absorption tower into a large number of tiny units through grid division, and calculates the mass, momentum, and energy conservation equations in each unit. The gas-liquid contact probability distribution is a spatial distribution function representing the possibility of gas-liquid two-phase contact at each position in the tower. The high-probability region indicates sufficient gas-liquid contact, which is beneficial to the mass transfer process.

[0083] According to the gas-liquid contact probability distribution, adjust the spraying angle and flow rate distribution of the spraying device in the enhanced absorption tower to form a gradient distribution absorption region. The spraying device is a key equipment for distributing liquid in the enhanced absorption tower, which includes multiple spraying layers, and each layer is equipped with several nozzles. The spraying angle refers to the angle between the liquid outlet direction of the nozzle and the horizontal plane, which has a direct impact on the distribution range and shape of liquid droplets. The flow rate distribution refers to controlling the liquid flow rate ratio of the nozzles in each layer and each region to achieve directional enhancement of the absorption capacity of certain regions. The gradient distribution absorption region refers to a spatial structure with uneven liquid distribution density formed in the absorption tower. Usually, the liquid distribution density is increased in the region with high gas-liquid contact probability and appropriately reduced in the region with low probability to optimize resource allocation.

[0084] Use a turbulence intensity optimizer to adjust the boundary layer perturbation of the gradient distribution absorption region to improve the gas-liquid exchange efficiency and generate a multi-dimensional interactive absorption pattern. The turbulence intensity optimizer is a device that can control the turbulence degree of the fluid, such as adjustable baffle plates, flow guiding cones, or packing with special structures. The boundary layer perturbation adjustment refers to the process of changing the flow state of the fluid near the gas-liquid interface, destroying the laminar boundary layer, and enhancing the turbulent mixing. The gas-liquid exchange efficiency refers to the ratio of the mass transferred through the interface per unit time to the theoretical maximum transfer amount, which is affected by the interface area and the mass transfer coefficient. The multi-dimensional interactive absorption pattern refers to a complex gas-liquid contact structure formed through turbulence strengthening and flow field optimization, which has multi-scale and multi-directional interactive characteristics and can adapt to the absorption requirements of different components simultaneously.

[0085] Conduct dynamic detection and calibration on the multi-dimensional interactive absorption pattern to construct an optimal absorption environment. Dynamic detection is to use various sensors distributed in the tower (such as differential pressure sensors, temperature sensors, component concentration sensors, etc.) to monitor the state changes of the absorption process in real time. Calibration is to fine-tune the spraying parameters, flow field structure, etc. through the control system according to the detection results to keep the absorption environment in the best state. The optimal absorption environment refers to the gas-liquid contact state that can achieve the highest mass transfer efficiency and the lowest operating cost under the conditions of specific waste gas component characteristics and absorption liquid characteristics, and is the ideal working point of the entire pretreatment enhanced absorption system.

[0086] Taking the treatment of malodorous waste gas in a certain leather processing factory as an example, the multiphase flow analyzer first analyzes the characteristic indexes of the absorption liquid (0.76), extracts key parameters such as viscosity of 6.8 mPa·s, surface tension of 52 mN / m, and density of 1.12 g / cm³, and generates a liquid-phase flow characteristic map in combination with the temperature influence coefficient. At the same time, according to the characteristics of the waste gas after conditioning (mainly including difficult-to-treat components such as 15 ppm of methyl mercaptan and 8 ppm of dimethyl sulfide), the average gas density of 1.28 kg / m³ and the diffusion coefficient of 0.14 cm² / s are calculated to form gas-phase diffusion mode data. After inputting these two sets of data into the computational fluid dynamics analysis system, the software uses the k-ε turbulence model for numerical simulation, divides more than 500,000 grid cells, converges after 1000 iterative calculations, and obtains the gas-liquid contact probability distribution map in the tower, showing that the contact probability is the highest in the lower-middle area of the tower (in the range of 3-6 meters from the tower bottom), while the contact probability in the top area (9-12 meters from the tower bottom) is significantly insufficient. Based on this analysis result, the parameters of the spraying device in the absorption tower are adjusted: the nozzle angle in the bottom area is adjusted from 60° to 55°, the middle area remains unchanged at 65°, and the top area is increased from 55° to 70°; at the same time, the liquid flow rate distribution ratio is adjusted from the original 3:4:3 to 2:4:4 to form a targeted gradient distribution absorption area. Subsequently, the turbulence intensity optimizer introduces micro-scale turbulence near the gas-liquid contact interface by adjusting the structural parameters of the packing layer. The measurement shows that the Reynolds number of the boundary layer increases from 1800 to 3200, and the mass transfer coefficient is increased by 25%, forming a multi-dimensional interactive absorption pattern with both large-scale flux and micro-scale enhancement. Finally, through the real-time feedback system composed of 20 monitoring points installed in the tower, the state is evaluated and fine-tuned every 30 seconds, and key parameters such as spraying pressure and liquid-gas ratio are dynamically optimized, and finally an optimal absorption environment suitable for the specific waste gas characteristics of this leather factory is constructed, and the removal rate of each component reaches the design standard.

[0087] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0088] (1) Uniformly introduce the conditioned waste gas into the optimal absorption environment through a porous distributor to form a micro-bubble flow state;

[0089] (2) Use an ion activation device to enhance the polarity of the absorption liquid in the optimal absorption environment to generate a highly reactive absorption zone;

[0090] (3) According to the microbubble flow state and the high reaction activity absorption zone, the temperature gradient in the absorption tower is adjusted to form a temperature-controlled absorption field, including: collecting cross-sectional images of the microbubble flow state by a high-speed camera and extracting bubble size distribution data from the cross-sectional images; using an optical detector to scan the polarity field intensity of the high reaction activity absorption zone to generate a reaction activity distribution map; performing spatial matching analysis on the bubble size distribution data and the reaction activity distribution map to calculate the gas-liquid contact area coefficient at each spatial point; according to the gas-liquid contact area coefficient, the temperature requirements of each area of the absorption tower are solved by the heat conduction equation to obtain a temperature control target matrix; according to the temperature control target matrix, the heating power distribution of the temperature control device of the absorption tower wall is adjusted to establish a longitudinal temperature gradient; and the longitudinal temperature gradient is monitored and calibrated in real time by a temperature sensor array to form a temperature-controlled absorption field.

[0091] (4) Collect the reaction rate data of exhaust gas and absorption liquid from the temperature-controlled absorption field and establish a reaction kinetic parameter library;

[0092] (5) Based on the reaction kinetic parameter library, the residence time in the absorption tower is dynamically controlled through an intelligent feedback mechanism to construct a layered absorption sequence;

[0093] (6) The purified gas produced by the layered absorption sequence is finally degraded through a catalytic conversion unit to obtain emission gas that meets the standards.

[0094] Specifically, in the method for treating malodor and organic waste gas by pretreatment and enhanced absorption, introducing the conditioned waste gas into the optimal absorption environment is the key link to achieve efficient treatment. First, the conditioned waste gas is evenly introduced into the optimal absorption environment through a porous distributor to form a microbubble flow state. The porous distributor is a specially designed gas dispersion device, which consists of a porous plate or tube with multiple micropores evenly distributed. The pore size is generally between 0.5-3mm. By controlling the pore size, distribution density and shape, the gas enters the liquid phase in the form of uniform and fine bubbles. The microbubble flow state refers to the movement state and distribution characteristics of a large number of tiny bubbles in the liquid phase, including parameters such as bubble size distribution, rising velocity, and gas phase volume fraction. This flow state can provide a larger gas-liquid contact area.

[0095] The ion activation device is used to enhance the polarity of the absorption liquid in the optimal absorption environment to generate a highly reactive absorption zone. The ion activation device is an electrochemical device that can enhance the polarity of the liquid. It changes the arrangement of water molecules and the degree of ion solvation by applying an electric field or introducing specific ions. The polarity enhancement treatment is to enhance the affinity of the liquid for polar substances by adjusting the dipole moment and hydrogen bond network structure of the liquid molecules. The highly reactive absorption zone refers to the liquid phase area with stronger gas capture ability after treatment. In this area, the absorption rate and capacity of the absorption liquid for the target pollutant are significantly improved.

[0096] According to the characteristics of the microbubble flow state and the highly reactive absorption zone, the temperature gradient inside the absorption tower is adjusted to form a temperature-controlled absorption field, and this process involves multiple key steps. First, a high-speed camera is used to collect cross-sectional images of the microbubble flow state. A high-speed camera is an imaging device capable of capturing clear images of fast-moving objects, with a frame rate usually ranging from 1000 to 10000 frames per second. Cross-sectional image collection refers to the process of using special lighting to illuminate and capture the bubble distribution image on the cross-section at a specific height inside the absorption tower. When extracting the bubble size distribution data from the cross-sectional images, image processing algorithms such as edge detection and circle recognition are adopted to identify the boundary of each bubble, calculate its diameter and area, and form a dataset describing the statistical characteristics of the bubble size.

[0097] The mathematical expression of the bubble size distribution data can be represented by the following formula:

[0098] ;

[0099] where, represents the number density distribution function of bubbles with diameter D; represents the number of bubbles with diameters ranging from D to D + ; represents the width of the diameter interval; represents the total volume of the sampling area.

[0100] An optical detector is used to scan the polarity field intensity of the highly reactive absorption zone to generate a reactivity distribution map. An optical detector is a measuring device based on the principles of fluorescence or Raman spectroscopy, capable of detecting changes in optical properties caused by changes in the polarity of the liquid. Polarity field intensity scanning refers to the process of measuring the liquid polarity intensity at different positions inside the absorption tower and recording the spatial distribution of the ion activation effect. The reactivity distribution map is a three-dimensional dataset representing the reactivity levels of the absorption liquid at various points in space, usually represented by a color gradient indicating the change in activation intensity.

[0101] The polarity field intensity and the reactivity are related as follows:

[0102] ;

[0103] where, represents the reactivity intensity at the spatial point ; represents the polarity field intensity at this point; represents the temperature at this point; are empirical coefficients, representing the proportionality factor, the polarity field intensity influence index, and the temperature suppression factor, respectively.

[0104] Perform a spatial matching analysis on the bubble size distribution data and the reaction activity distribution map, and calculate the gas-liquid contact area coefficient at each spatial point. Spatial matching analysis is the process of corresponding two three-dimensional data sets in the same spatial coordinate system, and data with different resolutions and sampling points are processed through an interpolation algorithm. The gas-liquid contact area coefficient is a parameter representing the size of the gas-liquid interface area per unit volume and is a function of the bubble size and number density.

[0105] The calculation formula for the gas-liquid contact area coefficient \(a_v\) is:

[0106] ;

[0107] where, represents the gas-liquid contact area coefficient at the spatial point ; \(D\) is the bubble diameter; is the bubble size distribution function at this point; is the bubble deformation correction factor considering the influence of reaction activity; and are the considered minimum and maximum bubble diameters respectively.

[0108] According to the gas-liquid contact area coefficient, solve the temperature requirements for each region of the absorption tower through the heat conduction equation to obtain the temperature control target matrix. The heat conduction equation is a partial differential equation describing the law of heat transfer in a medium. Under steady-state conditions, the heat transfer model considering the reaction heat effect can be expressed as:

[0109] ;

[0110] where, represents the thermal conductivity of the medium; is the Laplace operator; is the temperature field function; is the medium density; is the specific heat capacity; is the fluid velocity vector; is the heat effect per unit reaction; is the gas-liquid contact area coefficient. Solve this equation to obtain the temperature values required at each point under the optimal reaction conditions, forming the temperature control target matrix.

[0111] According to the temperature control target matrix, adjust the heating power distribution of the temperature control device on the absorption tower wall to establish a longitudinal temperature gradient. The calculation formula for the power distribution of the temperature control device is:

[0112] ;

[0113] where, represents the heating power of the \(i\)th temperature control region; is the energy conversion efficiency; is the heat exchange area of this region; is the heat transfer coefficient; is the target temperature; is the current temperature. The longitudinal temperature gradient is monitored and calibrated in real time by using a temperature sensor array to form a stable temperature-controlled absorption field, ensuring that the temperature of each region meets the reaction requirements.

[0114] Reaction rate data of the exhaust gas and the absorbent are collected from the temperature-controlled absorption field to establish a reaction kinetics parameter library. The reaction rate data collection is realized by installing component concentration sensors at different height positions in the absorption tower, and the changes of pollutant concentration over time and space are recorded. The reaction kinetics parameter library is a dataset storing the reaction kinetics characteristics under different conditions, including parameters such as reaction rate constant, activation energy, and reaction order, which describe the influence laws of factors such as temperature and concentration on the reaction rate.

[0115] According to the reaction kinetics parameter library, the residence time in the absorption tower is dynamically regulated through an intelligent feedback mechanism to construct a hierarchical absorption sequence. The intelligent feedback mechanism is a control method that adjusts process parameters based on real-time monitoring data. By comparing the difference between the actual treatment effect and the target requirements, the operation state of the system is automatically adjusted. The residence time in the absorption tower refers to the time required for the exhaust gas to enter and leave the absorption tower, which is controlled by adjusting the gas flow rate or the tower height. The hierarchical absorption sequence refers to a sequence of multiple absorption regions with different functions formed along the height direction of the absorption tower. Each region is optimized for specific components to achieve step-by-step purification.

[0116] The purified gas generated by the hierarchical absorption sequence is finally degraded through a catalytic conversion unit to obtain the up-to-standard discharge gas. The catalytic conversion unit is the last barrier of the treatment system, which uses a specific catalyst to carry out deep oxidation or reduction reactions on the residual pollutants to convert harmful substances into harmless substances. The up-to-standard discharge gas refers to the treated gas whose pollutant concentration meets the requirements of relevant environmental protection standards and can be directly discharged into the atmospheric environment.

[0117] In a specific embodiment, the process of dynamically regulating the residence time in the absorption tower through the intelligent feedback mechanism may specifically include the following steps:

[0118] (1) Extract the key reaction rate constants from the reaction kinetics parameter library to generate a multi-component degradation rate spectrum;

[0119] (2) Evaluate the three-dimensional configuration of organic molecules in the multi-component degradation rate spectrum through reaction steric hindrance analysis to form a degradation difficulty classification table;

[0120] (3) Based on the degradation difficulty classification table, calculate the theoretical residence time required for different organic components to establish a time-removal rate correspondence;

[0121] (4)Use a flow control valve to manage the air flow channels inside the absorption tower in a zoned manner, and construct multiple resistance fields;

[0122] (5)Precisely adjust the multiple resistance fields according to the time-removal rate correspondence relationship to form air flow channels with an increasing residence time;

[0123] (6)Treat the waste gas entering the air flow channels with an increasing residence time step by step according to the degradation difficulty, and construct a hierarchical absorption sequence.

[0124] Specifically, extract the key reaction rate constants from the reaction kinetics parameter library to generate a multi-component degradation rate spectrum. The reaction kinetics parameter library is a data set storing the reaction characteristics of pollutants and the absorbent under various reaction conditions, including the influence relationships of conditions such as temperature, pH value, and concentration on the reaction rate. The key reaction rate constant is a numerical parameter characterizing the chemical reaction rate under specific reaction conditions, usually with the unit of L / (mol·s) or mol / (L·s). The multi-component degradation rate spectrum refers to a spectrum formed by arranging the degradation rates of various pollutant components under standard conditions in numerical order, intuitively showing the differences in the reaction activities of different components and providing a data basis for subsequent treatment. Conduct a steric configuration evaluation of the organic molecules in the multi-component degradation rate spectrum through reaction steric hindrance analysis to form a classification table of easy and difficult degradation. Reaction steric hindrance analysis is a method for examining the influence of the molecular spatial structure on the reaction process. By calculating parameters such as the distances between atoms, bond angles, and dihedral angles, the exposure degree and reaction accessibility of the active groups in the molecule are evaluated. The steric configuration evaluation is a process of quantitatively analyzing the three-dimensional structural characteristics of molecules, examining factors such as molecular volume, surface area, and the distribution of hydrophobic regions, and predicting the ease or difficulty of the contact reaction between the molecule and the absorbent. The classification table of easy and difficult degradation is a classification system that divides the organic pollutants in the waste gas into different levels according to the degradation difficulty, usually divided into three categories: easy to degrade, medium difficulty to degrade, and difficult to degrade, providing a theoretical basis for zoned treatment.

[0125] Based on the classification table of easy and difficult degradation, calculate the theoretical residence time required for different organic components, and establish a time-removal rate correspondence relationship. The theoretical residence time refers to the shortest reaction time required to reduce the pollutant concentration to the target value under specific reaction conditions, with the unit of seconds or minutes. The time-removal rate correspondence relationship is a quantitative relationship describing the change in the pollutant removal rate with the residence time, usually represented by a function curve, reflecting the kinetic characteristic differences of different components. This relationship is the core parameter for designing a hierarchical absorption system and determines the spatial distribution and flow control strategy of each treatment area.

[0126] The airflow channels inside the absorption tower are managed in zones by using flow control valves to construct multiple resistance fields. A flow control valve is a device that can precisely regulate the gas flow rate and direction, including various types such as butterfly valves, ball valves, and diaphragm valves, and adjusts the opening degree according to the control signal. The airflow channel is the path through which the waste gas flows in the absorption tower, and a specific flow pattern is formed through the baffle plate and the packing layer. The multiple resistance field refers to the spatial distribution of different flow resistances set in different areas of the absorption tower, which is achieved by changing the packing density, baffle plate angle, etc., with the aim of controlling the gas flow rate and residence time in different areas. According to the time-removal rate correspondence relationship, the multiple resistance field is precisely adjusted to form an airflow channel with an increasing residence time. Precise adjustment is a process of finely tuning the opening degree of the flow control valve and the baffle plate angle based on real-time monitoring data to ensure that the actual residence time in each area is consistent with the theoretical requirement. The airflow channel with an increasing residence time means the design of the flow path where the residence time of the waste gas in the absorption tower gradually increases along the flow direction, enabling the easily degradable components to be processed in the short residence time area in the front section and the difficult-to-degrade components to fully react in the long residence time area in the rear section.

[0127] The waste gas entering the airflow channel with an increasing residence time is processed step by step according to the degradation difficulty to construct a hierarchical absorption sequence. Step-by-step processing refers to the process in which components with different degradation difficulties in the waste gas are selectively removed in the treatment areas that match their kinetic characteristics, forming a gradient treatment effect. The hierarchical absorption sequence is a sequence of multiple treatment areas with different functions formed along the height direction of the absorption tower. Each area is optimized for specific components, including the directional regulation of parameters such as the type, concentration, pH value, and temperature of the reactant, to achieve targeted treatment of different components.

[0128] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for treating malodorous and organic waste gases with enhanced absorption by pretreatment, characterized in that, The method for treating malodor and organic waste gas by pretreatment and enhanced absorption comprises: S101, collecting exhaust gas composition data through sensors, monitoring the exhaust gas components entering the pretreatment tower in real time, and obtaining exhaust gas component characteristic parameters, step S101 includes: quantitatively detecting hydrogen sulfide, ammonia, methyl mercaptan, dimethyl sulfide, and dimethyl disulfide in the exhaust gas through multiple groups of gas component detectors to obtain exhaust gas component concentration data; continuously sampling the temperature, humidity, pressure, and flow rate of the exhaust gas by the airflow parameter acquisition unit to form exhaust gas physical property data; classifying and marking the exhaust gas source based on the emission source type identification program, and generating exhaust gas source characteristic values in combination with the industry characteristic library; performing time series analysis on the exhaust gas component concentration data according to the time series data processing algorithm to generate a component fluctuation trend map; performing correlation analysis on the exhaust gas component concentration data and the exhaust gas physical property data through data clustering processing to construct a multi-dimensional exhaust gas characteristic correlation matrix; performing data fusion processing on the exhaust gas component concentration data, the exhaust gas physical property data, the exhaust gas source characteristic values, and the component fluctuation trend map to comprehensively form the exhaust gas component characteristic parameters; S102, calculating the pretreatment agent formula using a convolutional neural network model according to the exhaust gas component characteristic parameters, and generating a pretreatment agent dosing instruction; S103, according to the pretreatment agent adding instruction, adding the pretreatment agent into the pretreatment tower, removing the easy-to-treat components in the exhaust gas, and performing interface conditioning on the difficult-to-treat components to obtain conditioned exhaust gas; S104, monitoring the absorption liquid parameters through the sensors in the enhanced absorption tower, collecting the pH value and concentration data of the absorption liquid, and forming the absorption liquid characteristic index, step S104 includes: regularly sampling the absorption liquid through the liquid sampling sensor group installed at different heights of the enhanced absorption tower, and obtaining the original absorption liquid sample at each sampling point; using the pH microelectrode sensor to detect the pH value of the original absorption liquid sample in real time, generating a continuous time series of pH data, and constructing the pH value spatial distribution record in the tower; using the conductivity probe to measure the total ion concentration in the original absorption liquid sample, recording the change trend of the conductivity of the absorption liquid, and obtaining the absorption capacity attenuation index; quantitatively analyzing the dissolved organic matter content in the original absorption liquid sample based on ultraviolet absorption spectrum analysis, determining the organic matter concentration value, and forming an organic load data set; integrating and calculating the pH value spatial distribution record, the absorption capacity attenuation index and the organic load data set through the data association analysis processor to generate the absorption liquid state evaluation parameter; comparing and analyzing the absorption liquid state evaluation parameter with the historical absorption efficiency data, and calculating the value characterizing the current absorption liquid performance through the weighted average algorithm to form the absorption liquid characteristic index; S105, based on the absorption liquid characteristic index and the characteristics of the exhaust gas after conditioning, using a fluid dynamics model to adjust the gas-liquid distribution in the enhanced absorption tower to construct an optimal absorption environment; S106, introducing the conditioned exhaust gas into the optimal absorption environment, removing the odor and organic matter in the exhaust gas by enhanced absorption, and obtaining exhaust gas that meets the standards.

2. The pretreatment-enhanced absorption method for treating malodorous and organic waste gases according to claim 1, characterized in that, Step S102 includes: Extract waste gas treatment case data from the historical processing database, match and compare it with the waste gas composition characteristic parameters to form a waste gas treatment similarity matrix; Process the waste gas composition characteristic parameters through a three-layer convolution structure. Among them, the first layer of convolution uses a 5×5 convolution kernel to extract the basic characteristics of waste gas components, the second layer of convolution uses a 3×3 convolution kernel to capture the interaction characteristics between components, and the third layer of convolution uses a 1×1 convolution kernel to integrate the characteristic information. After being processed by the ReLU activation function, a waste gas component characteristic map is generated; Send the waste gas component characteristic map into the fully connected layer network. Through the linear transformation of the weight matrix and the bias vector, combined with the Softmax function, the reactivity of waste gas components is quantified, and the waste gas component reactivity index is calculated; Construct a reagent-component interaction matrix based on the waste gas component reactivity index. Separate the component sensitivity parameter and the reagent reaction ability parameter through the matrix decomposition algorithm, and select the reagent candidate set from oxidants, acid-base neutralizing agents, complexing agents, and surfactants accordingly; Input the reagent candidate set into the decision-making layer of the convolutional neural network. After dimensionality reduction through pooling operations, perform residual connection with the historical reaction efficiency data to generate a reagent ratio prediction tensor, and then optimize the ratio parameters through backpropagation to form a reagent ratio efficiency map; Extract the optimal ratio point from the reagent ratio efficiency map according to the gradient descent method, perform constraint optimization in combination with the cost coefficient matrix, determine the optimal reagent formula, and convert the optimal reagent formula into a time series control instruction to generate the pretreatment reagent dosing instruction; 3. The pretreatment-enhanced absorption method for treating malodorous odors and organic waste gases according to claim 1, wherein, Step S103 includes: Add the pretreatment reagent to different height positions in the pretreatment tower according to the concentration gradient through a multi-point distribution injection system to form a reagent concentration stratified structure; The tangential mixing of the waste gas and the pretreatment reagent is carried out by a swirl intensification device to generate a waste gas-reagent turbulent contact zone, so that hydrogen sulfide and ammonia in the waste gas react with the reagent quickly through redox reaction to generate primary reaction products; Use the temperature gradient control unit in the tower to adjust the waste gas-reagent reaction temperature, and promote the catalytic degradation of methanethiol and dimethyl sulfide through the temperature field gradient distribution to generate a polarity-enhanced intermediate; Based on the interfacial activity regulator, adjust the surface tension of the primary reaction product and the polarity-enhanced intermediate, reduce the gas-liquid interface transfer resistance of the difficult-to-treat components, and form a microemulsion phase; Send the microemulsion phase into the molecular polarity regulation area, change the molecular polarity of the difficult-to-treat components through the action of the polarity field, increase their water solubility, and generate a polarity-enhanced waste gas stream; Perform gas-solid separation on the polarity-enhanced waste gas stream to remove the particulate matter and aerosol generated during the reaction process to obtain the conditioned waste gas; 4. The method for treating malodorous and organic waste gas with enhanced absorption by pretreatment according to claim 1, characterized in that, Step S105 includes: Refined parameter extraction of the absorption liquid characteristic index through a multiphase flow analyzer to generate a liquid phase flow characteristic map; Calculate the gas flow performance according to the characteristics of the conditioned waste gas to form gas phase diffusion mode data; Input the liquid phase flow characteristic map and the gas phase diffusion mode data into the computational fluid dynamics analysis system to calculate the gas-liquid contact probability distribution; According to the gas-liquid contact probability distribution, adjust the spraying angle and flow rate distribution of the spraying device in the enhanced absorption tower to form a gradient distribution absorption area; Through the turbulence intensity optimizer, adjust the boundary layer perturbation of the gradient distribution absorption area to improve the gas-liquid exchange efficiency and generate a multi-dimensional interactive absorption pattern; Conduct dynamic detection and calibration on the multi-dimensional interactive absorption pattern to construct the optimal absorption environment.

5. The pretreatment-enhanced absorption method for treating malodorous and organic waste gases according to claim 1, characterized in that, Step S106 includes: Uniformly introduce the conditioned waste gas into the optimal absorption environment through a porous distributor to form a microbubble flow state; Use an ion activation device to enhance the polarity of the absorbent in the optimal absorption environment to generate a highly reactive absorption zone; According to the microbubble flow state and the highly reactive absorption zone, adjust the temperature gradient in the absorption tower to form a temperature-controlled absorption field; Collect the reaction rate data of the waste gas and the absorbent from the temperature-controlled absorption field to establish a reaction kinetics parameter library; According to the reaction kinetics parameter library, dynamically regulate the residence time in the absorption tower through an intelligent feedback mechanism to construct a hierarchical absorption sequence; Pass the purified gas generated by the hierarchical absorption sequence through a catalytic conversion unit for final degradation to obtain the up-to-standard discharge gas.

6. The pretreatment enhanced absorption method for treating malodorous odors and organic waste gases according to claim 5, characterized in that, The adjusting the temperature gradient in the absorption tower according to the microbubble flow state and the highly reactive absorption zone to form a temperature-controlled absorption field includes: Collect cross-sectional images of the microbubble flow state through a high-speed camera and extract bubble size distribution data from the cross-sectional images; Use an optical detector to scan the polarity field intensity of the highly reactive absorption zone to generate a reaction activity distribution map; Conduct spatial matching analysis on the bubble size distribution data and the reaction activity distribution map, and calculate the gas-liquid contact area coefficient of each spatial point; According to the gas-liquid contact area coefficient, solve the temperature requirements of each area of the absorption tower through the heat conduction equation to obtain a temperature control target matrix; According to the temperature control target matrix, adjust the heating power distribution of the temperature control device on the absorption tower wall to establish a longitudinal temperature gradient; Use a temperature sensor array to monitor and calibrate the longitudinal temperature gradient in real time to form the temperature-controlled absorption field.

7. The pretreatment-enhanced absorption method for treating malodorous odors and organic waste gases according to claim 5, characterized in that, The dynamically regulating the residence time in the absorption tower according to the reaction kinetics parameter library through an intelligent feedback mechanism to construct a hierarchical absorption sequence includes: Extract the key reaction rate constants from the reaction kinetics parameter library to generate a multi-component degradation rate spectrum; Conduct a three-dimensional configuration evaluation of the organic molecules in the multi-component degradation rate spectrum through reaction steric hindrance analysis to form a table of easy and difficult degradation classification; Based on the table of easy and difficult degradation classification, calculate the theoretical residence time required for different organic components to establish a time-removal rate correspondence; Use flow control valves to manage the airflow channels inside the absorption tower in zones to construct a multi-resistance field; According to the time-removal rate correspondence, precisely adjust the multi-resistance field to form an airflow channel with an increasing residence time; Process the waste gas entering the airflow channel with an increasing residence time step by step according to the degradation difficulty to construct the hierarchical absorption sequence.

Images