Dynamic management method and system for high and low concentration organic waste gas treatment

Abstract

The present invention discloses a dynamic management method and system for high and low concentration organic waste gas treatment, and relates to the technical field of waste gas management. The system comprises a waste gas monitoring module, a waste gas pretreatment module, a dynamic treatment module and an effect monitoring module, wherein the dynamic treatment module serves as the core of the dynamic management system. The technical key points are as follows: through the combination of an intelligent algorithm optimization unit and a model prediction control unit, the system can automatically determine the type of waste gas and select an appropriate treatment method, thereby improving the intelligence of the system; according to the singleness or complexity of the waste gas components, different units are triggered respectively, thereby realizing targeted optimization treatment; through model prediction control, intelligent algorithm optimization and online adjustment, real-time dynamic management of high and low concentration organic waste gas treatment is realized, thereby ensuring that the treatment system can operate efficiently and stably under different working conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of waste gas management, and in particular to a dynamic management method and system for treating high and low concentration organic waste gases. Background Art

[0002] Waste gas management is a comprehensive task, which refers to the monitoring, treatment and control of waste gas generated in industrial production, transportation, daily life, etc. through appropriate technologies and procedures, so as to reduce or prevent the harm of waste gas emissions to the environment and human health. It usually includes the following contents: waste gas emission monitoring, technical application and treatment methods, emission control and compliance management.

[0003] The existing application publication number is CN116609487A, and the technical solution pointed out in the document named "An Intelligent Management System for Waste Gas Monitoring" includes: the system discloses a concentration acquisition module, a monitoring marking module, and a major maintenance selection module. By setting the monitoring marking module, the waste gas emission points can be graded and marked according to the collected waste gas emission concentration. Through the classification, not only can the waste gas purification status of different waste gas emission points be intuitively understood, but the corresponding intervention range can be adjusted according to the classification situation, and maintenance personnel can be arranged to repair the waste gas emission points within the intervention range. By setting the major maintenance selection module, appropriate maintenance personnel can be selected according to the maintenance records to repair the waste gas emission points. On the basis of the rationality of the maintenance task arrangement of the maintenance personnel, it is guaranteed that the waste gas emission points can be efficiently maintained. However, the above-mentioned design scheme cannot solve the dynamic adjustment of the parameters of the waste gas treatment equipment.

[0004] In combination with the above documents and existing technologies, at present, in the treatment or management of organic waste gas at factory emission points, it is usually necessary to detect the concentration of the corresponding waste gas to ensure that the waste gas concentration at the time of emission reaches the preset standard, and configure corresponding sensors at the emission point for real-time monitoring. Only when the waste gas concentration exceeds the standard is it monitored that the corresponding treatment is carried out. However, the timing of treatment is too late at this time and it cannot be predicted in advance, resulting in low treatment efficiency. At the same time, there is also a lack of effective mechanism in the traditional setting to realize the intelligent adjustment of equipment parameters to adapt to the waste gas treatment needs under different working conditions. Summary of the invention

[0005] 1. Technical issues to be solved

[0006] In view of the shortcomings of the prior art, the present invention provides a dynamic management method and system for the treatment of high and low concentration organic waste gases. Through model predictive control, intelligent algorithm optimization and online adjustment, real-time dynamic management of high and low concentration organic waste gas treatment is achieved, ensuring that the treatment system can operate efficiently and stably under different working conditions, thereby solving the problems raised in the background technology.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] Dynamic management system for high and low concentration organic waste gas treatment, including:

[0010] The exhaust gas monitoring module obtains the exhaust gas data of the emission point, including the exhaust gas composition and exhaust gas concentration, and outputs the exhaust gas data through wireless communication technology;

[0011] An exhaust gas pretreatment module pre-treats the exhaust gas discharged from the emission point, and the pre-treatment means include chemical adsorption, condensation and oxidation;

[0012] The dynamic processing module includes a model prediction control unit and an intelligent algorithm optimization unit; receives the output exhaust gas data, and determines whether the exhaust gas data is single-component exhaust gas through a built rule engine;

[0013] Under the condition that it is determined to be a single-component exhaust gas, the model prediction control unit is triggered to obtain the concentration and timestamp of the exhaust gas corresponding to the component emitted at the emission point, and whether there is a periodic fluctuation is determined by Fourier analysis. If so, a time series prediction model is established to predict the emission concentration of the exhaust gas corresponding to the component in the future; if not, a multivariate linear regression model is established based on historical data to predict the emission concentration of the exhaust gas corresponding to the component in the future; the predicted emission concentration of the exhaust gas corresponding to the component is extracted, and the predicted concentration value is compared with the preset standard threshold value. If the predicted concentration value exceeds the standard threshold value, the optimization treatment strategy is executed, otherwise, no response action is taken;

[0014] Among them, the multiple linear regression model is applicable to the following situations:

[0015] There is no obvious cyclical change in the data.

[0016] There is a linear relationship between the variables, which can be modeled and predicted by a multiple linear regression model.

[0017] Assume that the noise in the data is random and does not change regularly over time.

[0018] Time series forecasting models are suitable for the following situations:

[0019] The data has clear periodicity, such as patterns that repeat daily, weekly, or yearly.

[0020] There may be trends (long-term changes upward or downward) or seasonality (cyclical, short-term changes) in the data.

[0021] Time series models can capture these cyclical changes and trends, so they are more suitable in this case. Therefore, when choosing a model, you need to analyze the characteristics and trends of the data first. If the data shows cyclical changes, such as seasonal fluctuations in sales data every year, then time series models (such as ARIMA, seasonal ARIMA, or neural network-based models) may be more suitable for prediction than multiple linear regression models. If the data does not have obvious cyclical changes, but only linear relationships between various variables, then multiple linear regression models may be a better choice;

[0022] When it is determined that the exhaust gas is not a single-component exhaust gas, the intelligent algorithm optimization unit is triggered;

[0023] The effect monitoring module, after executing the optimized treatment strategy, if the concentration of the corresponding exhaust gas component at the actual prediction time still exceeds the standard threshold, the secondary online adjustment strategy will be executed.

[0024] Furthermore, the historical data includes the concentration of the exhaust gas corresponding to the components emitted from the emission point, the production cycle and the production parameters, and the production parameters include temperature and pressure.

[0025] Furthermore, the time series model includes any one of ARIMA, seasonal ARIMA, and a neural network based model.

[0026] Furthermore, the process of establishing a multiple linear regression model is as follows:

[0027] Collect historical data as variables and establish a multivariate linear regression model. The concentration of the corresponding component waste gas is used as the dependent variable, and the production cycle and production parameters are used as independent variables. The coefficients of the model will reflect the influence of the independent variables on the concentration of the corresponding component waste gas. Use the established regression model and input the independent variable values ​​at future time points to predict the concentration of the corresponding component waste gas at future time points.

[0028] Furthermore, the optimization processing strategy implemented is:

[0029] According to the concentration value predicted in the future time period T and its difference with the standard threshold, after dimensionless processing, the estimated adsorbent dosage increment U is generated, and before that, the adsorbent dosage in the activated carbon adsorption tower is increased according to the estimated adsorbent dosage increment U. The formula for calculating the estimated adsorbent dosage increment U is as follows:

[0030]

[0031] Where G is the error factor, and the value range of G is 0 to 1; i = 1, 2, ..., n, n is a positive integer, and n = T, nd i represents the predicted concentration value at the corresponding time i, nd avgIt represents the mean of the predicted concentration values ​​at each time i within the period T, Bz represents the standard threshold, and the weight coefficient is: 0≤W1≤1, 0≤W2≤1, and W1+W2=1.

[0032] Furthermore, when the intelligent algorithm optimization unit is triggered, the intelligent algorithm is used to optimize the processing equipment parameters, and the intelligent algorithm used includes any one of a genetic algorithm and a fuzzy logic control algorithm.

[0033] Further, the process of optimizing the processing equipment parameters is as follows:

[0034] Parameter selection: determine the parameters that need to be optimized, including reaction temperature; Optimization target: set optimization targets, including minimizing energy consumption; Algorithm selection and implementation: select a suitable optimization algorithm, i.e. genetic algorithm, write the algorithm program and debug it; Optimization execution: run the optimization algorithm in real time and dynamically adjust equipment parameters according to monitoring data.

[0035] Furthermore, the secondary online adjustment strategy implemented is to increase the temperature of the oxidation reactor to cope with the high concentration of the corresponding exhaust gas, and the temperature difference each time is increased by S degrees until the concentration of the corresponding exhaust gas component does not exceed the standard threshold.

[0036] Furthermore, in the secondary online adjustment strategy, the value range of S is 2-5.

[0037] The dynamic management method for high and low concentration organic waste gas treatment includes the following steps:

[0038] S1. Obtain exhaust gas data from the emission point, including exhaust gas composition and exhaust gas concentration, and output the exhaust gas data through wireless communication technology;

[0039] S2. Pre-treating the waste gas discharged from the emission point, and the pre-treatment means include chemical adsorption, condensation and oxidation;

[0040] S3, receiving the output exhaust gas data, and determining whether the exhaust gas data is single-component exhaust gas through the established rule engine;

[0041] Under the condition that it is determined to be a single-component exhaust gas, the model prediction control unit is triggered to obtain the concentration and timestamp of the exhaust gas corresponding to the component emitted at the emission point, and whether there is a periodic fluctuation is determined by Fourier analysis. If so, a time series prediction model is established to predict the emission concentration of the exhaust gas corresponding to the component in the future; if not, a multivariate linear regression model is established based on historical data to predict the emission concentration of the exhaust gas corresponding to the component in the future; the predicted emission concentration of the exhaust gas corresponding to the component is extracted, and the predicted concentration value is compared with the preset standard threshold value. If the predicted concentration value exceeds the standard threshold value, the optimization treatment strategy is executed, otherwise, no response action is taken;

[0042] When it is determined that the exhaust gas is not a single-component exhaust gas, the intelligent algorithm optimization unit is triggered;

[0043] S4. After executing the optimized treatment strategy, if the concentration of the corresponding exhaust gas component at the actual prediction time still exceeds the standard threshold, a secondary online adjustment strategy is executed.

[0044] (III) Beneficial effects

[0045] The present invention provides a dynamic management method and system for high and low concentration organic waste gas treatment, which has the following beneficial effects:

[0046] 1. Through the combination of intelligent algorithm optimization unit and model prediction control unit, the system can automatically determine the type of exhaust gas and select the appropriate treatment method, which improves the intelligence of the system; according to the singleness or complexity of the exhaust gas components, different units are triggered respectively to achieve targeted optimization treatment; when the predicted exhaust gas concentration value exceeds the standard threshold, the system can immediately take optimization treatment measures to reduce the risk of environmental pollution and improve environmental protection efficiency;

[0047] 2. Based on the predicted concentration value in the future time period T and the difference between it and the standard threshold, the estimated increment of adsorbent dosage is calculated, and the adsorbent dosage can be finely adjusted according to specific working conditions and needs; through the intelligent algorithm optimization unit, the processing equipment parameters are adjusted dynamically in real time by using intelligent algorithms such as genetic algorithms, which effectively improves the processing efficiency and reduces energy consumption;

[0048] 3. Through model predictive control, intelligent algorithm optimization and online adjustment, real-time dynamic management of high and low concentration organic waste gas treatment is achieved, ensuring that the treatment system can operate efficiently and stably under different working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic diagram of the module structure of the dynamic management system for high and low concentration organic waste gas treatment in the present invention;

[0050] Figure 2 It is a flowchart of the overall steps of the dynamic management method for treating high and low concentration organic waste gases in the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Example 1: Please refer to Figure 1This embodiment provides a dynamic management system for high and low concentration organic waste gas treatment, the system includes a waste gas monitoring module, a waste gas pretreatment module, a dynamic treatment module and an effect monitoring module, wherein the dynamic treatment module is the core of the dynamic management system, and is intended to dynamically adjust the treatment method according to the waste gas characteristics monitored in real time, and the dynamic treatment module includes a model prediction control unit and an intelligent algorithm optimization unit;

[0053] The exhaust gas monitoring module obtains the exhaust gas data of the emission point, including the exhaust gas composition and exhaust gas concentration, and outputs the exhaust gas data through wireless communication technology;

[0054] The emission point is usually the emission point configured in the chemical plant, and the sensor technology is used at the emission point to obtain the exhaust gas composition and exhaust gas concentration;

[0055] Sensor technologies include infrared absorption sensors and mass spectrometry;

[0056] Infrared absorption sensor: The principle is to use the absorption characteristics of organic substances such as benzene to infrared rays of specific wavelengths, and measure the degree of infrared light absorption to determine the concentration of benzene; Advantages: high sensitivity, fast response speed, and continuous monitoring; Mass spectrometry: The principle is to ionize the exhaust gas sample, analyze it according to the mass-to-charge ratio (m / z), and detect the concentration of each component; Advantages: high sensitivity, high resolution, and can accurately detect the composition of complex mixed gases;

[0057] Wireless communication technology includes any of 4G, 5G and WiFi;

[0058] The exhaust gas pretreatment module pre-treats the exhaust gas discharged from the emission point, and the pre-treatment means include chemical adsorption, condensation and oxidation. After the preliminary pre-treatment, the exhaust gas can be treated more efficiently in the future;

[0059] It should be noted that pretreatment is the process of treating exhaust gas according to a predetermined pattern or quantity;

[0060] Among them, chemical adsorption is activated carbon adsorption. Activated carbon is a commonly used adsorbent that can effectively remove organic substances in exhaust gas, such as benzene and formaldehyde. The working principle is that through the large surface area and microporous structure of activated carbon, organic molecules are adsorbed on its surface, thereby reducing their concentration in the exhaust gas.

[0061] Condensation technology: by lowering the exhaust gas temperature, some easily condensable organic substances are converted from gas to liquid, which is convenient for subsequent separation and treatment. Condensation is usually used in combination with adsorption or other treatment technologies to enhance the exhaust gas treatment effect;

[0062] Oxidation, also known as catalytic oxidation: the use of catalysts (such as metal oxides) to oxidize organic substances into more easily handled or harmless substances. The oxidation reaction is usually carried out at higher temperatures and helps to decompose complex organic compounds;

[0063] Example: Removal of benzene from exhaust gas by adsorption using activated carbon;

[0064] First, install the activated carbon adsorption equipment:

[0065] Equipment location: Install the activated carbon adsorption equipment near the exhaust gas outlet of the chemical plant; Adsorbent selection: Select activated carbon suitable for treating benzene, and ensure that its adsorption performance and life meet the treatment requirements; Process control: Set the flow rate and adsorption time to ensure that the benzene in the exhaust gas can be effectively adsorbed;

[0066] The working principle is:

[0067] Adsorption process: When the exhaust gas passes through the activated carbon adsorption bed, the benzene molecules are physically or chemically adsorbed on the surface or in the micropores of the activated carbon; Adsorption efficiency: The high specific surface area and good adsorption characteristics of activated carbon ensure the efficient adsorption of organic matter such as benzene; Periodic operation: According to the characteristics and output of the exhaust gas, the activated carbon is replaced or regenerated regularly to maintain the adsorption efficiency and treatment effect.

[0068] Treated waste gas:

[0069] Treatment effect: After activated carbon adsorption, the concentration of benzene in the exhaust gas is significantly reduced, meeting environmental emission standards; Subsequent treatment: The treated exhaust gas can be further condensed or oxidized to meet higher emission standards or reuse requirements.

[0070] in conclusion:

[0071] The waste gas pretreatment module can effectively reduce the concentration of organic matter in the waste gas through technologies such as activated carbon adsorption, providing clean and effective pre-treatment preparation for subsequent waste gas treatment processes. Choosing appropriate pretreatment technology and equipment is crucial for chemical companies to achieve environmental protection requirements and energy conservation and emission reduction goals.

[0072] A dynamic processing module, which includes a model prediction control unit and an intelligent algorithm optimization unit;

[0073] Receive the output exhaust gas data, and determine whether the exhaust gas data is single-component exhaust gas through the established rule engine. If it is determined to be single-component exhaust gas, trigger the model prediction control unit;

[0074] When it is determined that the exhaust gas is not a single-component exhaust gas, the intelligent algorithm optimization unit is triggered;

[0075] Among them, when the model prediction control unit is triggered, the concentration and timestamp of the exhaust gas corresponding to the emission component at the emission point are obtained, and Fourier analysis is used to determine whether there is a periodic fluctuation, which helps to detect the periodic components in the corresponding data;

[0076] If it exists, it means that the emission concentration of the corresponding waste gas component fluctuates with the production cycle. By establishing a time series prediction model, the emission concentration of the corresponding waste gas component in the future can be predicted;

[0077] It should be noted that the established time series forecasting model adopts the ARIMA model;

[0078] If it does not exist, the emission concentration of the corresponding exhaust gas components in the future is predicted by establishing a multivariate linear regression model based on historical data;

[0079] The predicted emission concentration of the corresponding component exhaust gas is extracted, and the predicted concentration value is compared with the preset standard threshold. If the predicted concentration value exceeds the standard threshold, the optimization processing strategy is executed. If the predicted concentration value does not exceed the standard threshold, the system does not respond.

[0080] The historical data includes the concentration of the corresponding component waste gas emitted at the emission point, the production cycle and the production parameters. The production parameters include temperature and pressure. The temperature and pressure are the values ​​detected by the corresponding sensors installed at the emission point during the production process.

[0081] The process of building a multiple linear regression model is as follows:

[0082] Collect historical data as variables, establish a multivariate linear regression model, take the concentration of the corresponding component waste gas as the dependent variable, and the production cycle and production parameters as independent variables. The coefficient of the model (i.e. the weight of each variable) will reflect their influence on the concentration of the corresponding component waste gas. Use the established regression model and input the independent variable value at a future time point to predict the concentration of the corresponding component waste gas at a future time point.

[0083] Example:

[0084] Suppose that through multiple linear regression analysis, it is found that benzene concentration has a certain relationship with temperature and production batches; now, the chemical plant predicts that the temperature will rise in the next week and there will be two production batches; according to the established regression model, the expected change trend of benzene concentration after each production batch can be predicted. The multiple linear regression model can provide an estimate of the future change in benzene concentration, assuming a linear relationship between the variables;

[0085] It should be noted that when conducting data analysis and prediction, it is very important to choose an appropriate model based on the characteristics of the data;

[0086] 1. The multiple linear regression model is applicable to the following situations:

[0087] There is no obvious cyclical change in the data.

[0088] There is a linear relationship between the variables, which can be modeled and predicted by a multiple linear regression model.

[0089] Assume that the noise in the data is random and does not change regularly over time.

[0090] 2. The time series forecasting model is suitable for the following situations:

[0091] The data has clear periodicity, such as patterns that repeat daily, weekly, or yearly.

[0092] There may be trends (long-term changes upward or downward) or seasonality (cyclical, short-term changes) in the data.

[0093] Time series models can capture these cyclical changes and trends, so they are more suitable in this case. Therefore, when choosing a model, you need to analyze the characteristics and trends of the data first. If the data shows cyclical changes, such as seasonal fluctuations in sales data every year, then time series models (such as ARIMA, seasonal ARIMA, or neural network-based models) may be more suitable for prediction than multiple linear regression models. If the data does not have obvious cyclical changes, but only linear relationships between various variables, then multiple linear regression models may be a better choice.

[0094] Specifically, through the combination of intelligent algorithm optimization unit and model prediction control unit, the system can automatically determine the type of exhaust gas and select the appropriate treatment method, thereby improving the intelligence of the system; different units are triggered according to the simplicity or complexity of the exhaust gas components to achieve targeted optimization treatment; when the predicted exhaust gas concentration value exceeds the standard threshold, the system can immediately take optimization treatment measures to reduce the risk of environmental pollution and improve environmental protection efficiency.

[0095] The optimization processing strategy implemented is:

[0096] Based on the predicted concentration value in the future time period T and its difference with the standard threshold, after dimensionless processing, the estimated adsorption dose increment U is generated, based on the following formula:

[0097]

[0098] Wherein, the future time period T represents the time period of the forward prediction period T, with the detection of the predicted concentration value exceeding the standard threshold as the initial time node, G is the error factor, the specific value of which can be adjusted and set by the user, or generated by the analysis function fitting, and the value range of G is 0 to 1; i = 1, 2, ..., n, n is a positive integer, and n = T, nd i represents the predicted concentration value at the corresponding time i, nd avg It represents the mean value of the predicted concentration value at each time i in the period T, Bz represents the standard threshold value, and its specific value can be set according to actual needs. The weight coefficient is: 0≤W1≤1, 0≤W2≤1, and W1+W2=1;

[0099] Prior to this, the adsorbent dosage in the activated carbon adsorption tower is increased according to the estimated adsorbent dosage increment U, and the "prior to this" refers to the actual operation;

[0100] When the intelligent algorithm optimization unit is triggered, the intelligent algorithm is used to optimize the processing equipment parameters;

[0101] Among them, the intelligent algorithms used include any one of the genetic algorithm and the fuzzy logic control algorithm. The genetic algorithm is used in this scheme to ensure that the exhaust gas treatment process can achieve the best effect under various working conditions;

[0102] The process of optimizing the processing equipment parameters is as follows:

[0103] Parameter selection: determine the parameters that need to be optimized, such as adsorbent dosage, reaction temperature, flow rate, etc.; Optimization objectives: set optimization objectives, such as maximizing treatment efficiency, minimizing energy consumption, ensuring emissions meet standards, etc.; Algorithm selection and implementation: select a suitable optimization algorithm, such as genetic algorithm, particle swarm optimization algorithm (PSO), fuzzy logic control, etc., write the algorithm program and debug it; Optimization execution: run the optimization algorithm in real time and dynamically adjust equipment parameters according to monitoring data.

[0104] Example:

[0105] In the same chemical plant, the exhaust gas contains not only benzene, but also toluene and xylene. The genetic algorithm is used to optimize the operating parameters of the adsorption tower so that the total organic matter removal efficiency is optimal under different concentrations and component ratios. The genetic algorithm finds the optimal adsorbent type and dosage combination, as well as the optimal reaction temperature through multiple iterative calculations.

[0106] Specifically, based on the concentration value predicted in the future time period T and the difference between it and the standard threshold, the estimated increase in adsorbent dosage is calculated, and the adsorbent dosage can be finely adjusted according to specific working conditions and needs. Through the intelligent algorithm optimization unit, intelligent algorithms such as genetic algorithms are used to dynamically adjust the processing equipment parameters in real time, effectively improving the processing efficiency and reducing energy consumption.

[0107] Effect monitoring module, under the condition of executing the optimized treatment strategy, if the concentration of the corresponding exhaust gas component at the actual prediction time still exceeds the standard threshold, then execute the secondary online adjustment strategy to increase the temperature of the oxidation reactor to deal with the high concentration of the corresponding exhaust gas, and the temperature difference each time is increased by S degrees until the concentration of the exhaust gas component does not exceed the standard threshold;

[0108] The value range of S is 2 to 5, and the value usually adopted in this scheme is 2.

[0109] Specifically, through model predictive control, intelligent algorithm optimization and online adjustment, real-time dynamic management of high and low concentration organic waste gas treatment is achieved, ensuring that the treatment system can operate efficiently and stably under different operating conditions.

[0110] Example 2: Please refer to Figure 2 Based on Example 1, this embodiment also provides a dynamic management method for high and low concentration organic waste gas treatment, including the following specific steps:

[0111] S1. Obtain exhaust gas data from the emission point, including exhaust gas composition and exhaust gas concentration, and output the exhaust gas data through wireless communication technology;

[0112] S2. Pre-treating the waste gas discharged from the emission point, and the pre-treatment means include chemical adsorption, condensation and oxidation;

[0113] S3, receiving the output exhaust gas data, and determining whether the exhaust gas data is single-component exhaust gas through the established rule engine;

[0114] Under the condition that it is determined to be a single-component exhaust gas, the model prediction control unit is triggered to obtain the concentration and timestamp of the exhaust gas corresponding to the component emitted at the emission point, and whether there is a periodic fluctuation is determined by Fourier analysis. If so, a time series prediction model is established to predict the emission concentration of the exhaust gas corresponding to the component in the future; if not, a multivariate linear regression model is established based on historical data to predict the emission concentration of the exhaust gas corresponding to the component in the future; the predicted emission concentration of the exhaust gas corresponding to the component is extracted, and the predicted concentration value is compared with the preset standard threshold value. If the predicted concentration value exceeds the standard threshold value, the optimization treatment strategy is executed, otherwise, no response action is taken;

[0115] When it is determined that the exhaust gas is not a single-component exhaust gas, the intelligent algorithm optimization unit is triggered;

[0116] S4. After executing the optimized treatment strategy, if the concentration of the corresponding exhaust gas component at the actual prediction time still exceeds the standard threshold, a secondary online adjustment strategy is executed.

[0117] In the application, the several formulas involved are all calculated by removing the dimensions and taking their numerical values. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The formula is set by technical personnel in this field according to actual conditions.

[0118] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0120] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. Dynamic management system for high and low concentration organic waste gas treatment, characterized by: include: The exhaust gas monitoring module obtains the exhaust gas data of the emission point, including the exhaust gas composition and exhaust gas concentration, and outputs the exhaust gas data through wireless communication technology; An exhaust gas pretreatment module pre-treats the exhaust gas discharged from the emission point, and the pre-treatment means include chemical adsorption, condensation and oxidation; The dynamic processing module includes a model prediction control unit and an intelligent algorithm optimization unit; receives the output exhaust gas data, and determines whether the exhaust gas data is single-component exhaust gas through a built rule engine; Under the condition that it is determined to be a single-component exhaust gas, the model prediction control unit is triggered to obtain the concentration and timestamp of the exhaust gas corresponding to the component emitted at the emission point, and whether there is a periodic fluctuation is determined by Fourier analysis. If so, a time series prediction model is established to predict the emission concentration of the exhaust gas corresponding to the component in the future; if not, a multivariate linear regression model is established based on historical data to predict the emission concentration of the exhaust gas corresponding to the component in the future; the predicted emission concentration of the exhaust gas corresponding to the component is extracted, and the predicted concentration value is compared with the preset standard threshold value. If the predicted concentration value exceeds the standard threshold value, the optimization treatment strategy is executed, otherwise, no response action is taken; The historical data includes the concentration of the corresponding component waste gas emitted from the emission point, the production cycle and the production parameters, and the production parameters include temperature and pressure; the time series model includes any one of ARIMA, seasonal ARIMA and neural network-based models; the process of establishing the multivariate linear regression model is as follows: Collect historical data as variables, establish a multivariate linear regression model, take the concentration of the corresponding component waste gas as the dependent variable, and the production cycle and production parameters as independent variables. The coefficient of the model will reflect the influence of the independent variable on the concentration of the corresponding component waste gas. Use the established regression model and input the independent variable value at a future time point to predict the concentration of the corresponding component waste gas at a future time point. The optimization processing strategy implemented is: According to the concentration value predicted in the future time period T and its difference with the standard threshold, after dimensionless processing, the estimated adsorbent dosage increment U is generated, and before that, the adsorbent dosage in the activated carbon adsorption tower is increased according to the estimated adsorbent dosage increment U. The formula for calculating the estimated adsorbent dosage increment U is as follows: Where G is the error factor, and the value range of G is 0 to 1; i = 1, 2, ..., n, n is a positive integer, and n = T, nd i represents the predicted concentration value at the corresponding time i, nd avg represents the mean value of the predicted concentration value at each time i in the period T, Bz represents the standard threshold, and the weight coefficient is: 0≤W1≤1, 0≤W2≤1, and W1+W2=1; When it is determined that the exhaust gas is not a single-component exhaust gas, the intelligent algorithm optimization unit is triggered; When the intelligent algorithm optimization unit is triggered, the intelligent algorithm is used to optimize the processing equipment parameters, and the intelligent algorithm used includes any one of a genetic algorithm and a fuzzy logic control algorithm; The process of optimizing the processing equipment parameters is as follows: Parameter selection: determine the parameters that need to be optimized, including reaction temperature; Optimization target: set the optimization target, including minimizing energy consumption; Algorithm selection and implementation: select a suitable optimization algorithm, i.e. genetic algorithm, write the algorithm program and debug it; Optimization execution: run the optimization algorithm in real time and dynamically adjust the equipment parameters according to the monitoring data; The effect monitoring module, after executing the optimized treatment strategy, if the concentration of the corresponding exhaust gas component at the actual prediction time still exceeds the standard threshold, the secondary online adjustment strategy will be executed.

2. The dynamic management system for high and low concentration organic waste gas treatment according to claim 1 is characterized by: The secondary online adjustment strategy implemented is to increase the temperature of the oxidation reactor to cope with the high concentration of the corresponding exhaust gas, with the temperature difference increased by S degrees each time until the concentration of the corresponding exhaust gas component does not exceed the standard threshold.

3. The dynamic management system for high and low concentration organic waste gas treatment according to claim 2 is characterized in that: In the secondary online adjustment strategy, the value range of S is 2 to 5.

4. A dynamic management method for high and low concentration organic waste gas treatment, using any system described in claims 1 to 3, characterized in that: The steps include: S1. Obtain exhaust gas data from the emission point, including exhaust gas composition and exhaust gas concentration, and output the exhaust gas data through wireless communication technology; S2. Pre-treating the waste gas discharged from the emission point, and the pre-treatment means include chemical adsorption, condensation and oxidation; S3, receiving the output exhaust gas data, and determining whether the exhaust gas data is single-component exhaust gas through the established rule engine; Under the condition that it is determined to be a single-component exhaust gas, the model prediction control unit is triggered to obtain the concentration and timestamp of the exhaust gas corresponding to the component emitted at the emission point, and whether there is a periodic fluctuation is determined by Fourier analysis. If so, a time series prediction model is established to predict the emission concentration of the exhaust gas corresponding to the component in the future; if not, a multivariate linear regression model is established based on historical data to predict the emission concentration of the exhaust gas corresponding to the component in the future; the predicted emission concentration of the exhaust gas corresponding to the component is extracted, and the predicted concentration value is compared with the preset standard threshold value. If the predicted concentration value exceeds the standard threshold value, the optimization treatment strategy is executed, otherwise, no response action is taken; The historical data includes the concentration of the corresponding component waste gas emitted from the emission point, the production cycle and the production parameters, and the production parameters include temperature and pressure; the time series model includes any one of ARIMA, seasonal ARIMA and neural network-based models; the process of establishing the multivariate linear regression model is as follows: Collect historical data as variables, establish a multivariate linear regression model, take the concentration of the corresponding component waste gas as the dependent variable, and the production cycle and production parameters as independent variables. The coefficient of the model will reflect the influence of the independent variable on the concentration of the corresponding component waste gas. Use the established regression model and input the independent variable value at a future time point to predict the concentration of the corresponding component waste gas at a future time point. The optimization processing strategy implemented is: According to the concentration value predicted in the future time period T and its difference with the standard threshold, after dimensionless processing, the estimated adsorbent dosage increment U is generated, and before that, the adsorbent dosage in the activated carbon adsorption tower is increased according to the estimated adsorbent dosage increment U. The formula for calculating the estimated adsorbent dosage increment U is as follows: Where G is the error factor, and the value range of G is 0 to 1; i = 1, 2, ..., n, n is a positive integer, and n = T, nd i represents the predicted concentration value at the corresponding time i, nd avg represents the mean value of the predicted concentration value at each time i in the period T, Bz represents the standard threshold, and the weight coefficient is: 0≤W1≤1, 0≤W2≤1, and W1+W2=1; When it is determined that the exhaust gas is not a single-component exhaust gas, the intelligent algorithm optimization unit is triggered; When the intelligent algorithm optimization unit is triggered, the intelligent algorithm is used to optimize the processing equipment parameters, and the intelligent algorithm used includes any one of a genetic algorithm and a fuzzy logic control algorithm; The process of optimizing the processing equipment parameters is as follows: Parameter selection: determine the parameters that need to be optimized, including reaction temperature; Optimization target: set the optimization target, including minimizing energy consumption; Algorithm selection and implementation: select a suitable optimization algorithm, i.e. genetic algorithm, write the algorithm program and debug it; Optimization execution: run the optimization algorithm in real time and dynamically adjust the equipment parameters according to the monitoring data; S4. After executing the optimized treatment strategy, if the concentration of the corresponding exhaust gas component at the actual prediction time still exceeds the standard threshold, a secondary online adjustment strategy is executed.

Images