Resin adsorption waste gas treatment evaluation optimization method based on multi-source data fusion

Abstract

The invention discloses a resin adsorption waste gas treatment evaluation optimization method based on multi-source data fusion, and particularly relates to the field of waste gas treatment.The method comprises the steps of S1, collection area determination, S2, waste gas monitoring sensor deployment, S3, resin data collection, S4, resin data preprocessing, S5, resin treatment effect analysis, S6, numerical treatment effect judgment and S7, real-time optimization. In order to solve the problems that a traditional waste gas treatment monitoring mode is single, evaluation is not accurate, and dynamic optimization cannot be achieved, a first-level collection area and a second-level collection area are divided, multiple sensors are deployed in a targeted mode to collect multi-source data, the multi-source data are preprocessed, and the multi-source data are analyzed and analyzed. The preprocessed data are input into a mathematical model to evaluate the waste gas treatment effect, and when an evaluation index exceeds a threshold value, the system performs early warning and automatically adjusts operation parameters of the adsorption tower, so that multi-source data fusion monitoring, accurate evaluation and intelligent optimization are realized, the treatment cost is reduced, and the treatment effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of waste gas treatment, and more specifically, to an evaluation and optimization method for resin adsorption waste gas treatment based on multi-source data fusion. Background Technique

[0002] In the process of industrial production, the environmental pollution problem caused by waste gas emissions is becoming increasingly severe. As an efficient waste gas treatment technology, the resin adsorption method is widely used in many industries such as chemical industry, electronics, and painting. It realizes waste gas purification by means of the adsorption characteristics of the resin for harmful substances in the waste gas, ensuring environmental quality. However, with the continuous innovation of industrial technology, the production scale is constantly expanding, and the production process is becoming more and more complex. The waste gas emissions show new characteristics such as complex and variable components and unstable emissions.

[0003] Currently, there are many limitations in the traditional monitoring and evaluation methods for the resin adsorption waste gas treatment process. For example, manual inspection not only has low efficiency, but also is greatly affected by the subjective factors of the inspectors, making it difficult to monitor in real time and comprehensively. At the same time, the monitoring method that only relies on a small number of specific sensors can only obtain limited parameter information and cannot reflect the operation status of the waste gas treatment system in all aspects. This makes it difficult for enterprises to accurately grasp key information such as the change in the adsorption performance of the resin and the fluctuation of the waste gas treatment effect, and they cannot discover potential problems in time and take effective measures, resulting in poor waste gas treatment effect and high treatment cost, which seriously restricts the sustainable development and environmental protection compliance of enterprises. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an evaluation and optimization method for resin adsorption waste gas treatment based on multi-source data fusion, through the following solutions, to solve the problems raised in the above background technique.

[0005] To achieve the above object, the present invention provides the following technical solution: An evaluation and optimization method for resin adsorption waste gas treatment based on multi-source data fusion, including, S1: Determination of the collection area: According to the distribution of waste gas emission sources, equipment operation conditions, and environmental conditions in the workshop, the adsorption tower area near the waste gas discharge port of the reaction kettle, the area near the adsorption tower switching valve that frequently switches work, and the edge adsorption tower area that is greatly affected by the workshop ventilation are divided into the first-level monitoring areas; other adsorption tower areas far from the emission source and with stable environment are divided into the second-level monitoring areas;

[0006] S2: Deployment of waste gas monitoring sensors: Install waste gas monitoring sensors based on the areas divided in step S1;

[0007] S3: Resin data collection: The gas composition sensor collects data every 3 minutes, and the temperature sensor, pressure sensor, humidity sensor, and flow sensor collect data every second. Before each adsorption starts and after regeneration ends, the resin performance sensor is activated to collect data to obtain resin performance data;

[0008] The resin performance data includes waste gas concentration, temperature, pressure, humidity, flow rate, adsorption capacity, and mechanical strength;

[0009] S4: Resin data preprocessing: Process the resin performance data according to the established preprocessing method, and then calculate the processed resin performance data to obtain resin performance preprocessed data;

[0010] S5: Analysis of resin treatment effect: Input the resin performance preprocessed data obtained in step S4 into the mathematical model of the comprehensive evaluation index of waste gas treatment effect to obtain the comprehensive evaluation index of waste gas treatment effect;

[0011] S6: Judgment of numerical processing effect: Compare the calculated evaluation index E with the preset threshold TE = 0.5. If E≥0.5, it is judged that there is a risk of decline in the waste gas treatment effect, and the system issues a warning; if E<0.5, it is considered that the system is operating normally, but the data changes are continuously monitored

[0012] S7: Real-time optimization: When the system detects that E≥0.5, it immediately notifies the workshop management personnel by text message and email, and issues an audible and visual alarm on site. At the same time, the system automatically adjusts the operating parameters of the adsorption tower according to the data analysis results.

[0013] Preferably, the installation method of the waste gas monitoring sensor is as follows: In the secondary monitoring area, an electrochemical sensor for detecting the VOCs concentration, a thermocouple temperature sensor, a pressure sensor, and a humidity sensor are installed every 8 meters, and a vortex flow sensor is installed every 15 meters. In the primary monitoring area, a GC-MS gas composition sensor, a thermal resistance temperature sensor, a high-precision pressure sensor, and a humidity sensor are installed every 3 meters, and a thermal mass flow sensor is installed every 8 meters. At the inlet and outlet of the adsorption tower and at key positions inside the resin layer, a resin adsorption capacity sensor based on the principle of spectral analysis and a stress-strain type resin mechanical strength sensor are installed.

[0014] Preferably, the preprocessing method refers to using the median filtering algorithm to smooth the collected resin performance data, filling in the missing values using the Lagrange interpolation method, removing the outliers, and normalizing the data;

[0015] The method for obtaining the resin performance preprocessed data is as follows:

[0016] Rate of change of concentration (RC ) Let \(C_i\) be the concentration value of a certain type of volatile organic compound (VOCs) at the \(i\)-th time point, with a time interval of \(\Delta t\). The calculation formula for the concentration change rate \(RC\) within the time period \([i, i + n]\) is For example, if the concentration data of benzene is collected every 1 minute, \(\Delta t = 1\) minute. Within 10 minutes (\(n = 10\)), the initial concentration \(C\) of benzene i = 50 ppm, and the concentration \(C\) after 10 minutes i+n = 60 ppm. Then the concentration change rate \(R\) of benzene C = \(\frac{60 - 50}{10}= 1\) ppm / minute;

[0017] Temperature change slope \((dT / dt)\): Taking time \(t\) as the abscissa and temperature \(T\) as the ordinate to plot a curve, and calculating the slope of the curve to obtain the temperature change slope. For example, within a certain 5 minutes, the temperature rises from \(x\) °C to \(y\) °C, then the temperature change slope is \(\frac{y - x}{5}\) °C / minute;

[0018] Standard deviation of pressure fluctuation \((\sigma\) P ): Calculating the standard deviation of the pressure data over a period of time to reflect the fluctuation of pressure. Assuming that the sampling frequency is 1 time per second within 1 hour, a total of 3600 data points, and the average value of the pressure data is \(P\) avg = 101325 Pa. By calculation, the standard deviation of pressure fluctuation \(\sigma\) P = 100 Pa, indicating the degree of dispersion of pressure fluctuation during this period;

[0019] Adsorption capacity decay rate \((R\) Q ): Let \(Q\) i be the adsorption capacity of the resin at the \(i\)-th detection. The calculation formula for the adsorption capacity decay rate \(R\) within the time period \([i, i + m]\) is Q For example, at the first detection, the adsorption capacity \(Q\) of the resin i = 500 mg / g. After 5 adsorption - regeneration cycles (\(m = 5\)), the adsorption capacity becomes \(Q\) i+m = 450 mg / g. Then the adsorption capacity decay rate \(R\) Q = \(\frac{500 - 450}{5\times500}= 0.02\);

[0020] Mechanical strength change rate \((R\) M ): Let \(M\) i be the mechanical strength value of the resin at the \(i\)-th time point. The calculation formula for the mechanical strength change rate \(R\) within the time period \([i, i + n]\) is M If the mechanical strength of the resin is detected every 1 day (\(\Delta t = 1\) day), within 5 days (\(n = 5\)), the initial mechanical strength \(M\) i = 50 MPa, and the mechanical strength \(M\) after 5 days i+n ​​= 48 MPa, then the mechanical strength change rate R M = 5×148 - 50 = -0.4 MPa / day;

[0021] Humidity change rate (R H ): Humidity data (expressed as relative humidity RH (%)) was obtained during data collection. Let H i be the humidity value at the i-th time point, with a time interval of Δt. The humidity change rate R H in the time period [i, i + n] is calculated as follows

[0022] It should be further noted that humidity changes may affect the adsorption effect of the resin on certain components in the waste gas. The humidity change rate can reflect the humidity fluctuation situation. For example, if humidity data is collected every 10 minutes (Δt = 10 minutes), within 30 minutes (n = 3), the initial humidity H i = 50%, and the humidity H i+n after 30 minutes = 55%, then the humidity change rate

[0023] Flow rate change rate (R F ): Waste gas flow rate data is available during data collection, with the unit of m 3 / h or L / min. Let F i be the flow rate value at the i-th time point, with a time interval of Δt. The flow rate change rate R F in the time period [i, i + n] is calculated as follows Flow rate changes will affect the contact time between the resin and the waste gas and the adsorption efficiency. For example, if flow rate data is collected every 5 minutes (Δt = 5 minutes), within 20 minutes (n = 4), the initial flow rate F i = 100 m3 / h, and the flow rate F i+n after 20 minutes = 110 m 3 / h, then the flow rate change rate R F = (110 - 100) / 4×5 = 0.5 m 3 / h / min;

[0024] Pressure change rate (R P ): For the collected pressure data with the unit of Pa, let P i be the pressure value at the i-th time point, with a time interval of Δt. The pressure change rate R P in the time period [i, i + n] is calculated as follows Pressure changes can reflect the stability of the gas flow in the adsorption tower, etc. Assuming that pressure data is collected every 1 minute (Δt = 1 minute), within 10 minutes (n = 10), the initial pressure P i = 101000 Pa, and the pressure P after 10 minutesi+n =101200Pa, then the pressure change rate R P =(101200-101000) / 10×1=20Pa / minute, which is consistent with the pressure fluctuation standard deviation σ calculated previously P Reflects the pressure change characteristics from different angles. The pressure change rate reflects the linear change trend of pressure over a period of time, while the standard deviation reflects the degree of dispersion of pressure data.

[0025] Resin mechanical strength fluctuation characteristics: In addition to the mechanical strength change rate, calculate the variance σ of the resin mechanical strength data M 2 To describe the fluctuation of mechanical strength, the larger the variance, the greater the fluctuation of mechanical strength, and the worse the stability of the resin structure may be. Assume that within a period of time, such as 10 tests, the mechanical strength data of the resin are M1, M2, ..., M10 respectively. First calculate the average value Then the variance σ M 2 The calculation formula is: This characteristic and the mechanical strength change rate together reflect the change in the mechanical strength of the resin. The change rate reflects the trend, and the variance reflects the degree of fluctuation.

[0026] Preferably, the mathematical model of the comprehensive evaluation index of the exhaust gas treatment effect is specifically:

[0027]

[0028] Among them, E is the comprehensive evaluation index of exhaust gas treatment effect, R C is the concentration change rate, which reflects the speed of change of the pollutant concentration in the exhaust gas. It is of great significance for judging the stability of the exhaust gas source and the adsorption effect. dT / dt is the temperature change slope, which reflects the temperature change trend inside the system. Abnormal temperature changes may affect the resin adsorption performance and chemical reaction process. P is the standard deviation of pressure fluctuation, which measures the pressure fluctuation. Excessive pressure fluctuation may indicate problems such as uneven airflow distribution inside the adsorption tower and equipment leakage. H R is the humidity change rate, which reflects the change of humidity over time. Humidity changes will affect the resin's adsorption capacity for waste gas components. F R is the flow rate change rate. The fluctuation of flow rate will change the contact time between resin and exhaust gas and the adsorption efficiency. This indicator is used to evaluate the impact of flow stability on the treatment effect. P R is the pressure change rate, which reflects the pressure change from the perspective of linear change trend, complements the pressure fluctuation standard deviation, and more comprehensively reflects the impact of pressure on the treatment process. Q R is the adsorption capacity attenuation rate, which is directly related to the adsorption performance and service life of the resin and is a key indicator for evaluating the effectiveness of the resin. Mis the mechanical strength change rate, reflecting the change in the structural stability of the resin during use. A decrease in mechanical strength may lead to problems such as resin fragmentation and loss, σ M 2 is the mechanical strength fluctuation characteristic of the resin, reflecting the degree of mechanical strength fluctuation. Together with the mechanical strength change rate, it reflects the stability of the resin's mechanical properties. α1, α2, α3, α4, α5, α6, β1, β2, β3, and θ are model coefficients, and α1 + α2 + α3 + α4 + α5 + α6 + β1 + β2 + β3 + θ = 1.

[0029] Preferably, α1, α2, α3, α4, α5, α6, β1, β2, β3, and θ are determined through regression analysis of a large amount of historical data, machine learning algorithm training, and combined with actual engineering experience. They reflect the relative importance and influence degree of different parameters on the evaluation of waste gas treatment effect. For example, in some chemical waste gas treatment scenarios, through data analysis and empirical judgment, it is found that the change in waste gas concentration has a greater impact on the treatment effect, so the value of α1 may be relatively large; while in some adsorption processes sensitive to temperature, the value of α2 is more important. In actual application, the coefficients need to be adjusted and optimized according to specific production processes, waste gas components, equipment characteristics and other factors to ensure that the model can accurately evaluate the waste gas treatment effect.

[0030] Preferably, the operations for the system to automatically adjust the operating parameters of the adsorption tower include, but are not limited to, increasing the adsorption time and adjusting the waste gas flow distribution to optimize the waste gas treatment effect. If the problem is relatively serious, such as excessive attenuation of the resin adsorption capacity, the system will arrange a resin regeneration or replacement plan to ensure the stable operation of the waste gas treatment system.

[0031] The technical effects and advantages of the present invention:

[0032] 1. By dividing the primary and secondary monitoring areas and deploying various types of sensors such as GC-MS gas component sensors and thermal resistance temperature sensors in a targeted manner, the present invention can comprehensively collect multi-source data such as waste gas concentration, temperature, pressure, humidity, flow rate, and resin adsorption capacity and mechanical strength. Moreover, the sensors set different installation densities and collection frequencies according to the regional risk level, with encrypted monitoring in the primary monitoring area and regular monitoring in the secondary monitoring area, ensuring real-time and accurate grasp of the system operation status and making up for the deficiencies of traditional monitoring methods;

[0033] 2. The present invention performs preprocessing such as median filtering and Lagrange interpolation on the collected multi-source data, extracts key features such as the concentration change rate and the temperature change slope, and then substitutes these features into a comprehensive evaluation index mathematical model for waste gas treatment effects carefully constructed. The model coefficients are determined through regression analysis of a large amount of historical data, training of machine learning algorithms, and combined with actual engineering experience, enabling the model to accurately evaluate the waste gas treatment effects, discover potential risks in advance, and provide a scientific basis for timely measures.

[0034] 3. The present invention judges the waste gas treatment effects through the evaluation index E and notifies the workshop management personnel in real time via text messages and emails, while giving an on-site audible and visual alarm. The present invention automatically adjusts the operating parameters of the adsorption tower according to the data analysis results, such as increasing the adsorption time and adjusting the waste gas flow distribution. If serious problems such as excessive attenuation of the resin adsorption capacity occur, the system will also arrange for resin regeneration or replacement plans to ensure the stable operation of the waste gas treatment system, effectively reduce the treatment cost, and improve the economic and environmental benefits of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0036] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] As shown in the Figure 1 accompanying drawings, a resin adsorption waste gas treatment evaluation and optimization method based on multi-source data fusion includes:

[0038] S1: Determination of the collection area: According to the distribution of waste gas emission sources, equipment operation conditions, and environmental conditions in the workshop, the adsorption tower area near the waste gas discharge port of the reaction kettle, the area near the adsorption tower switching valve that frequently switches jobs, and the edge adsorption tower area that is greatly affected by the workshop ventilation are divided into the first-level monitoring areas; other adsorption tower areas far from the emission sources and with stable environments are divided into the second-level monitoring areas.

[0039] S2: Deployment of waste gas monitoring sensors: Install waste gas monitoring sensors based on the areas divided in step S1.

[0040] Specifically, the installation method of the exhaust gas monitoring sensor is as follows: In the secondary monitoring area, an electrochemical sensor for detecting the concentration of VOCs, a thermocouple temperature sensor, a pressure sensor, and a humidity sensor are installed every 8 meters, and a vortex flow sensor is installed every 15 meters. In the primary monitoring area, a GC-MS gas composition sensor, a thermal resistance temperature sensor, a high-precision pressure sensor, and a humidity sensor are installed every 3 meters, and a thermal mass flow sensor is installed every 8 meters. At the inlet and outlet of the adsorption tower and at key positions inside the resin layer, a resin adsorption capacity sensor and a stress-strain type resin mechanical strength sensor based on the principle of spectral analysis are installed;

[0041] S3: Resin data acquisition: The gas composition sensor collects data every 3 minutes, and the temperature sensor, pressure sensor, humidity sensor, and flow sensor collect data every second. Before each adsorption starts and after regeneration ends, the resin performance sensor is activated to collect data to obtain resin performance data.

[0042] Specifically, the resin performance data includes exhaust gas concentration, temperature, pressure, humidity, flow rate, adsorption capacity, and mechanical strength;

[0043] S4: Resin data preprocessing: Process the resin performance data according to the established preprocessing method, and then calculate the processed resin performance data to obtain resin performance preprocessing data;

[0044] The preprocessing method refers to using the median filtering algorithm to smooth the collected resin performance data, filling the missing values using the Lagrange interpolation method, removing the outliers, and normalizing the data;

[0045] The method for obtaining the resin performance preprocessing data is as follows:

[0046] Concentration change rate (R C ): Let Ci be the concentration value of a certain type of volatile organic compound VOCs at the i-th time point, and the time interval is Δt. The concentration change rate RC calculation formula within the time period [i, i + n] is For example, if the concentration data of benzene is collected every 1 minute, Δt = 1 minute, within 10 minutes (n = 10), the initial concentration C i of benzene = 50 ppm, and the concentration C i+n after 10 minutes = 60 ppm, then the concentration change rate R C of benzene = 10×160 - 50 = 1 ppm / minute;

[0047] Temperature change slope (dT / dt): With time t as the abscissa and temperature T as the ordinate, plot a curve and calculate the slope of the curve to obtain the temperature change slope. For example, within a certain 5 minutes, if the temperature rises from ℃ to ℃, then the temperature change slope is ℃ / minute;

[0048] Standard deviation of pressure fluctuation (σ P ): Calculate the standard deviation of the pressure data over a period of time to reflect the pressure fluctuation. Assume that the acquisition frequency is once per second within 1 hour, with a total of 3600 data points, and the average value of the pressure data is P avg = 101325 Pa. By calculation, the standard deviation of pressure fluctuation σ P = 100 Pa, indicating the degree of dispersion of the pressure fluctuation during this period;

[0049] Adsorption capacity decay rate (R Q ): Let Q i be the adsorption capacity of the resin at the i-th detection. The adsorption capacity decay rate R Q within the time period [i, i + m] is calculated as follows For example, at the first detection, the adsorption capacity of the resin Q i = 500 mg / g. After 5 adsorption-regeneration cycles (m = 5), the adsorption capacity becomes Q i+m = 450 mg / g. Then the adsorption capacity decay rate R Q = 5×500500 - 450 = 0.02;

[0050] Mechanical strength change rate (R M ): Let M i be the mechanical strength value of the resin at the i-th time point. The mechanical strength change rate R M within the time period [i, i + n] is calculated as follows If the mechanical strength of the resin is detected once every 1 day (Δt = 1 day), within 5 days (n = 5), the initial mechanical strength M i = 50 MPa, and after 5 days, the mechanical strength M i+n = 48 MPa. Then the mechanical strength change rate R M = 5×148 - 50 = -0.4 MPa / day;

[0051] Humidity change rate (R H ): Humidity data (expressed as relative humidity RH (%)) is obtained during data acquisition. Let H i be the humidity value at the i-th time point, with a time interval of Δt. The humidity change rate R H within the time period [i, i + n] is calculated as follows

[0052] It should be further noted that humidity changes may affect the adsorption effect of the resin on certain components in the waste gas. The humidity change rate can reflect the humidity fluctuation. For example, if humidity data is collected every 10 minutes (Δt = 10 minutes), within 30 minutes (n = 3), the initial humidity H i = 50%, and the humidity H i+n after 30 minutes = 55%, then the humidity change rate

[0053] Flow rate change rate (R F ): There is waste gas flow rate data in the data collection, with the unit of m 3 / h or L / min. Let F i be the flow rate value at the i-th time point, and the time interval is Δt. The flow rate change rate R F within the time period [i, i + n] is calculated as Flow rate changes will affect the contact time and adsorption efficiency between the resin and the waste gas. For example, if flow rate data is collected every 5 minutes (Δt = 5 minutes), within 20 minutes (n = 4), the initial flow rate F i = 100 m3 / h, and the flow rate F i+n after 20 minutes = 110 m 3 / h, then the flow rate change rate R F = (110 - 100) / 4×5 = 0.5 m 3 / h / min;

[0054] Pressure change rate (R P ): For the collected pressure data with the unit of Pa, let P i be the pressure value at the i-th time point, and the time interval is Δt. The pressure change rate R P within the time period [i, i + n] is calculated as Pressure changes can reflect the stability of the gas flow in the adsorption tower, etc. Assume that pressure data is collected every 1 minute (Δt = 1 minute), within 10 minutes (n = 10), the initial pressure P i = 101000 Pa, and the pressure P i+n after 10 minutes = 101200 Pa, then the pressure change rate R P = (101200 - 101000) / 10×1 = 20 Pa / min, which reflects the pressure change characteristics from different angles compared with the previously calculated standard deviation σ P of the pressure fluctuation. The pressure change rate reflects the linear change trend of the pressure over a period of time, while the standard deviation reflects the dispersion degree of the pressure data;

[0055] Fluctuation characteristics of the mechanical strength of the resin: In addition to the mechanical strength change rate, calculate the variance σ M 2To describe the fluctuation of mechanical strength, the greater the variance, the greater the fluctuation of mechanical strength, and the worse the resin structure stability may be. Suppose that within a period of time, such as 10 detections, the resin mechanical strength data are M1, M2, …, M10. First, calculate the average value. Then the variance σ M 2 The calculation formula is: This characteristic and the mechanical strength change rate jointly reflect the change of resin mechanical strength. The change rate reflects the trend, and the variance reflects the fluctuation degree.

[0056] S5: Analysis of resin treatment effect: Input the resin performance pretreatment data obtained in step S4 into the mathematical model of the comprehensive evaluation index of waste gas treatment effect to obtain the comprehensive evaluation index of waste gas treatment effect;

[0057] It should be further explained that the mathematical model of the comprehensive evaluation index of waste gas treatment effect is specifically:

[0058]

[0059] Among them, E is the comprehensive evaluation index of waste gas treatment effect, R C is the concentration change rate, which reflects the change speed of pollutant concentration in waste gas and is of great significance for judging the stability of waste gas generation source and adsorption effect. dT / dt is the temperature change slope, which reflects the change trend of the internal temperature of the system. Abnormal temperature change may affect the resin adsorption performance and chemical reaction process. σ P is the standard deviation of pressure fluctuation, which measures the pressure fluctuation. Excessive pressure fluctuation may imply problems such as uneven air flow distribution inside the adsorption tower and equipment leakage. R H is the humidity change rate, which reflects the change of humidity over time. Humidity change will affect the adsorption capacity of resin for waste gas components. R F is the flow rate change rate. The fluctuation of flow rate will change the contact time between resin and waste gas and the adsorption efficiency. This index is used to evaluate the influence of flow rate stability on treatment effect. R P is the pressure change rate, which reflects the pressure change from the perspective of linear change trend and complements the standard deviation of pressure fluctuation to more comprehensively reflect the influence of pressure on the treatment process. R Q is the adsorption capacity attenuation rate, which is directly related to the adsorption performance and service life of resin and is a key index for evaluating the effectiveness of resin. R M is the mechanical strength change rate, which reflects the change of resin structure stability during use. The decrease of mechanical strength may lead to problems such as resin fragmentation and loss. σ M 2It is the fluctuation characteristic of the mechanical strength of the resin, reflecting the degree of fluctuation of the mechanical strength, and jointly reflecting the stability of the mechanical properties of the resin with the mechanical strength change rate. α1, α2, α3, α4, α5, α6, β1, β2, β3, and θ are model coefficients, and α1 + α2 + α3 + α4 + α5 + α6 + β1 + β2 + β3 + θ = 1.

[0060] It should be further noted that α1, α2, α3, α4, α5, α6, β1, β2, β3, and θ are determined by performing regression analysis on a large amount of historical data, training machine learning algorithms, and combining actual engineering experience. They reflect the relative importance and influence degree of different parameters on the evaluation of the waste gas treatment effect. For example, in some chemical waste gas treatment scenarios, through data analysis and empirical judgment, it is found that the change in waste gas concentration has a greater impact on the treatment effect, then the value of α1 may be relatively large; while in some adsorption processes sensitive to temperature, the value of α2 is more important. In actual application, the coefficients need to be adjusted and optimized according to specific production processes, waste gas components, equipment characteristics, etc. to ensure that the model can accurately evaluate the waste gas treatment effect.

[0061] S6: Judgment of numerical processing effect: Compare the calculated evaluation index E with the pre-set threshold TE = 0.5. If E ≥ 0.5, it is judged that there is a risk of decline in the waste gas treatment effect, and the system issues a warning; if E < 0.5, it is considered that the system is operating normally, but the change of the continuously monitored data is continued.

[0062] S7: Real-time optimization: When the system detects that E ≥ 0.5, it immediately notifies the workshop management personnel by text message and email, and issues an audible and visual alarm on site. At the same time, the system automatically adjusts the operating parameters of the adsorption tower according to the data analysis results.

[0063] It should be further noted that the operations of the system automatically adjusting the operating parameters of the adsorption tower include but are not limited to increasing the adsorption time and adjusting the waste gas flow distribution to optimize the waste gas treatment effect. If the problem is relatively serious, such as the resin adsorption capacity decays too much, the system will arrange a resin regeneration or replacement plan to ensure the stable operation of the waste gas treatment system.

[0064] Secondly: In the accompanying drawings of the disclosed embodiments of the present invention, only the structures related to the disclosed embodiments of the present disclosure are involved. Other structures can refer to the usual designs. Without conflict, the same embodiment and different embodiments of the present invention can be combined with each other;

[0065] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An evaluation and optimization method for resin adsorption waste gas treatment based on multi-source data fusion, characterized in that, include: S1: Determination of collection area: Based on the distribution of exhaust gas emission sources in the workshop, equipment operation status and environmental conditions, the adsorption tower area close to the exhaust gas outlet of the reactor, the area near the switching valve of the adsorption tower where work is frequently switched, and the edge adsorption tower area which is greatly affected by workshop ventilation are divided into the first-level monitoring area; Other adsorption tower areas that are far away from emission sources and have stable environments are classified as secondary monitoring areas; S2: Exhaust gas monitoring sensor deployment: Install exhaust gas monitoring sensors based on the areas divided in step S1; S3: Resin data collection: The gas composition sensor collects data every 3 minutes, and the temperature sensor, pressure sensor, humidity sensor, and flow sensor collect data once a second. Before each adsorption and regeneration, the resin performance sensor is started to collect data to obtain resin performance data. The resin performance data include exhaust gas concentration, temperature, pressure, humidity, flow rate, adsorption capacity, and mechanical strength; S4: Resin data preprocessing: processing the resin performance data according to a predetermined preprocessing method, and then calculating the processed resin performance data to obtain resin performance preprocessing data; S5: Resin treatment effect analysis: The resin performance pretreatment data obtained in step S4 is input into the mathematical model of the comprehensive evaluation index of the exhaust gas treatment effect to obtain the comprehensive evaluation index of the exhaust gas treatment effect; S6: Determine the numerical treatment effect: Compare the calculated evaluation index E with the pre-set threshold value TE = 0.

5. If E ≥ 0.5, it is determined that there is a risk of deterioration in the exhaust gas treatment effect, and the system issues an early warning. If E < 0.5, the system is considered to be operating normally, but data changes are continuously monitored. S7: Real-time optimization: When the system detects E≥0.5, it immediately notifies the workshop management personnel via SMS or email, and issues an audible and visual alarm on site. At the same time, the system automatically adjusts the operating parameters of the adsorption tower based on the data analysis results.

2. The resin adsorption waste gas treatment evaluation and optimization method based on multi-source data fusion according to claim 1, wherein: The installation method of the exhaust gas monitoring sensor is as follows: in the secondary monitoring area, an electrochemical sensor for detecting VOCs concentration, a thermocouple temperature sensor, a pressure sensor and a humidity sensor are installed every 8 meters, and a vortex flow sensor is installed every 15 meters; in the primary monitoring area, a GC-MS gas composition sensor, a thermistor temperature sensor, a high-precision pressure sensor and a humidity sensor are installed every 3 meters, and a thermal mass flow sensor is installed every 8 meters; resin adsorption capacity sensors based on spectral analysis principles and stress-strain resin mechanical strength sensors are installed at the inlet and outlet of the adsorption tower and at key positions inside the resin layer.

3. The resin adsorption waste gas treatment evaluation and optimization method based on multi-source data fusion according to claim 1, characterized in that: The preprocessing method refers to using a median filter algorithm to smooth the collected resin performance data, using a Lagrange interpolation method to fill in missing values, eliminating outliers, and normalizing the data.

4. A method for evaluating and optimizing the treatment of resin-adsorbed waste gas based on multi-source data fusion according to claim 1, characterized in that: The method for obtaining the resin performance pretreatment data is as follows: Concentration change rate (R C ): Let Ci be the concentration value of a certain type of volatile organic compound VOCs at the i-th time point, the time interval is Δt, and the concentration change rate RC in the time period [i, i+n] is calculated as follows: Temperature change slope (dT / dt): Draw a curve with time t as the horizontal axis and temperature T as the vertical axis, and calculate the slope of the curve to get the temperature change slope; Standard deviation of pressure fluctuation (σ P ): Calculate the standard deviation of pressure data over a period of time to reflect the fluctuation of pressure; Adsorption capacity decay rate (R Q ): Let Q i is the adsorption capacity of the resin at the time of the i-th test, and the adsorption capacity decay rate R in the time period [i, i+m] Q The calculation formula is Mechanical strength change rate (R M ): Let M i is the mechanical strength value of the resin at the i-th time point, and the mechanical strength change rate R in the time period [i, i+n] M The calculation formula is Humidity change rate (R H ):Humidity data (expressed as relative humidity RH (%)) was obtained during data collection. Let H i is the humidity value at the i-th time point, the time interval is Δt, and the humidity change rate R in the time period [i,i+n] H The calculation formula is Flow rate change rate (R F ): There is exhaust gas flow rate data in data acquisition, with the unit of m 3 / h or L / min. Let F i be the flow rate value at the i-th time point, with the time interval of Δt. The flow rate change rate R F in the time period [i, i + n] is calculated as Rate of pressure change (R P ): For the pressure data collected with the unit of Pa, let P i be the pressure value at the i-th time point, the time interval is Δt, and the rate of pressure change R P in the time period [i, i + n] is calculated by the formula Resin mechanical strength fluctuation characteristics: In addition to the mechanical strength change rate, calculate the variance σ of the resin mechanical strength data M 2 To describe the fluctuation of mechanical strength, the larger the variance, the greater the fluctuation of mechanical strength, and the worse the stability of the resin structure may be. Assume that within a period of time, such as 10 tests, the mechanical strength data of the resin are M1, M2, ..., M10 respectively. First calculate the average value Then the variance σ M 2 The calculation formula is:

5. The resin adsorption waste gas treatment evaluation and optimization method based on multi-source data fusion according to claim 1, characterized in that: The mathematical model of the comprehensive evaluation index of the exhaust gas treatment effect is specifically: Among them, E is the comprehensive evaluation index of the waste gas treatment effect, R C is the concentration change rate, dT / dt is the temperature change slope, σ P is the standard deviation of pressure fluctuation, R H is the humidity change rate, R F is the flow rate change rate, R P is the pressure change rate, R Q is the adsorption capacity attenuation rate, R M is the mechanical strength change rate, σ M 2 is the mechanical strength fluctuation characteristic of the resin, α1, α2, α3, α4, α5, α6, β1, β2, β3, θ are model coefficients, and α1 + α2 + α3 + α4 + α5 + α6 + β1 + β2 + β3 + θ = 1.

Images