Intelligent control and data analysis method and system for gas purification in adsorption tank

Abstract

The application provides an intelligent control and data analysis method and system for gas purification in an adsorption tank. The multi-dimensional environmental data in the adsorption tank is collected in real time to obtain a multi-dimensional environmental data set. Based on the multi-dimensional environmental data set, the time series analysis and machine learning algorithm are used to predict the concentration change trend of the gas components and the performance degradation of the adsorption material, and the abnormal detection algorithm is applied to identify potential abnormal working conditions to obtain a comprehensive prediction result. According to the comprehensive prediction result, a multi-objective optimization algorithm and a fuzzy logic control theory are used to generate an optimized operation parameter setting. The optimized operation parameter setting is applied to the actual operation of the adsorption tank, the implementation effect is monitored, and the actual operation effect is fed back to the prediction model to generate a more accurate prediction model. The technical scheme provided by the application can improve the data analysis effect of gas purification in the adsorption tank.

Description

Technical Field

[0001] This application relates to the field of intelligent control and data analysis technology, and in particular to intelligent control and data analysis methods and systems for gas purification in adsorption tanks. Background Technology

[0002] Adsorption tanks are widely used in chemical, environmental protection, and energy fields to remove harmful components from gases or recover valuable gases. In these applications, the operating status of the adsorption tank directly affects the gas treatment effect and the overall system performance. To ensure the efficient, safe, and stable operation of the adsorption tank, it is necessary to collect and analyze multi-dimensional environmental data (such as gas component concentration, temperature, pressure, humidity, and flow rate) in real time. This data not only helps predict the concentration change trend of gas components and the performance degradation of the adsorption material, but also allows for the timely identification of potential abnormal operating conditions, enabling corresponding adjustments and optimizations.

[0003] Currently, most adsorption tank systems possess basic data acquisition capabilities, enabling real-time monitoring of parameters such as gas component concentration, temperature, pressure, humidity, and flow rate. Existing data analysis methods primarily rely on traditional statistical analysis and simple predictive models, such as linear regression and moving averages. While these methods offer a degree of prediction and analysis, their accuracy and robustness are limited. Existing anomaly detection methods are typically based on threshold judgments and simple rule engines, capable of identifying some obvious anomalies, but their accuracy and timeliness need improvement for complex and variable operating conditions. Existing parameter optimization methods often employ single-objective optimization strategies, such as optimal control theory, but these methods often struggle to consider multiple optimization objectives simultaneously, resulting in incomplete and ineffective optimization results.

[0004] Traditional statistical analysis and simple prediction models struggle to capture the inherent patterns and dynamic changes in complex, multidimensional environmental data, resulting in low prediction accuracy. Existing anomaly detection methods primarily rely on fixed thresholds and rules, which fail to identify potential anomalies accurately and promptly under complex and changing operating conditions, leading to false alarms or missed alarms. Existing parameter optimization methods often employ single-objective optimization strategies, making it difficult to simultaneously consider multiple optimization objectives (such as minimizing the prediction error of gas component concentration change trends, minimizing system operating costs, minimizing the frequency of abnormal operating condition warnings, and maximizing system stability), resulting in incomplete and ineffective optimization results. Most existing solutions lack effective closed-loop feedback mechanisms, failing to continuously optimize and adjust the prediction model based on actual operating performance, resulting in poor system adaptability and robustness. Summary of the Invention

[0005] This application provides an intelligent control and data analysis method and system for gas purification in an adsorption tank, which solves the problem of poor data analysis effect for adsorption tanks in the prior art.

[0006] In a first aspect, embodiments of this application provide an intelligent control and data analysis method for gas purification in an adsorption tank, including:

[0007] Real-time acquisition of multidimensional environmental data within the adsorption tank yields a multidimensional environmental dataset, which includes gas component concentration, temperature, pressure, humidity, and flow rate.

[0008] Based on the multidimensional environmental dataset, time series analysis and machine learning algorithms are used to predict the concentration change trend of gas components and the performance degradation of adsorption materials. At the same time, anomaly detection algorithms are applied to identify potential abnormal operating conditions and obtain comprehensive prediction results.

[0009] Based on the comprehensive prediction results, a multi-objective optimization algorithm and fuzzy logic control theory are used to intelligently adjust the operating parameters of the adsorption tank and generate optimized operating parameter settings.

[0010] The optimized operating parameters are applied to the actual operation of the adsorption tank, its implementation effect is monitored, and the actual operating effect is fed back to the prediction model to generate a more accurate prediction model.

[0011] Secondly, embodiments of this application provide an intelligent control and data analysis method for gas purification in an adsorption tank, including:

[0012] The acquisition module is used to acquire multidimensional environmental data in the adsorption tank in real time to obtain a multidimensional environmental dataset, which includes gas component concentration, temperature, pressure, humidity and flow rate.

[0013] The processing module is used to predict the concentration change trend of gas components and the performance degradation of adsorption materials based on the multidimensional environmental dataset using time series analysis and machine learning algorithms. At the same time, it applies anomaly detection algorithms to identify potential abnormal operating conditions and obtain comprehensive prediction results.

[0014] The generation module is used to intelligently adjust the operating parameters of the adsorption tank based on the comprehensive prediction results, using a multi-objective optimization algorithm and fuzzy logic control theory, and generate optimized operating parameter settings.

[0015] The monitoring module is used to apply the optimized operating parameter settings to the actual operation of the adsorption tank, monitor its implementation effect, and feed the actual operating effect back to the prediction model to generate a more accurate prediction model.

[0016] Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are to be invoked and executed by the processing component to realize an intelligent control and data analysis method for gas purification in an adsorption tank as described in the first aspect.

[0017] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements an intelligent control and data analysis method for gas purification in an adsorption tank as described in the first aspect.

[0018] In this embodiment, multidimensional environmental data within the adsorption tank is collected in real time to obtain a multidimensional environmental dataset, including gas component concentration, temperature, pressure, humidity, and flow rate. Based on this multidimensional environmental dataset, time series analysis and machine learning algorithms are used to predict the concentration change trend of gas components and the performance degradation of the adsorption material. Simultaneously, an anomaly detection algorithm is applied to identify potential abnormal operating conditions, resulting in a comprehensive prediction result. According to the comprehensive prediction result, a multi-objective optimization algorithm and fuzzy logic control theory are used to intelligently adjust the operating parameters of the adsorption tank, generating optimized operating parameter settings. These optimized operating parameter settings are applied to the actual operation of the adsorption tank, monitoring their implementation effect, and feeding the actual operating effect back to the prediction model to generate a more accurate prediction model. By collecting real-time data on gas composition concentration, temperature, pressure, humidity, and flow rate within the adsorption tank, the comprehensiveness and accuracy of the data are ensured, providing a reliable foundation for subsequent prediction and optimization. Time series analysis and machine learning algorithms are used to predict the concentration change trend of gas components and the performance degradation of the adsorption material, improving the accuracy and reliability of the predictions. Anomaly detection algorithms are applied to identify potential abnormal operating conditions, enabling timely detection and handling of anomalies and preventing equipment failures or production accidents caused by abnormal conditions. A multi-objective optimization algorithm is employed, comprehensively considering multiple optimization objectives (such as minimizing the prediction error of gas composition concentration change trends, minimizing system operating costs, etc.). The system prioritizes minimizing abnormal operating condition warnings and maximizing system stability, generating preliminary optimization schemes to ensure optimal operation of the adsorption tank under various constraints. Utilizing fuzzy logic control theory, the system intelligently adjusts its operating parameters to enhance adaptability and robustness, ensuring efficient and stable operation even under complex and changing conditions. The optimized operating parameters are applied to the actual operation of the adsorption tank, with real-time monitoring of their effectiveness to ensure the optimization scheme's validity. The actual operating results are fed back to the predictive model, and through continuous learning and optimization, a more accurate predictive model is generated, forming a closed-loop feedback mechanism for continuous system improvement and optimization. Real-time monitoring and intelligent adjustment significantly improve the operating efficiency of the adsorption tank and the lifespan of the adsorption material; optimized operating parameters reduce system operating costs and improve economic efficiency; timely identification and handling of abnormal operating conditions reduce equipment failures and production accidents, improving system safety and reliability; the closed-loop feedback mechanism continuously optimizes the predictive model, improving prediction accuracy and reliability; and the combination of multi-objective optimization and fuzzy logic control enables intelligent management of the adsorption tank, enhancing the system's adaptability and robustness. The above measures can comprehensively improve the operating performance and management level of the adsorption tank, achieving efficient, economical, safe and reliable operation.

[0019] The system performs missing value imputation, outlier handling, and data standardization on multidimensional environmental data to generate a high-quality dataset. Time series analysis is used to model the concentration change trends of gas components, generating gas concentration change trend predictions. Machine learning algorithms are employed to predict the performance degradation of adsorption materials based on the gas concentration change trend predictions, generating adsorption material performance degradation predictions. Anomaly detection algorithms are implemented to identify potential abnormal conditions in the system, generating abnormal operating condition warnings. Finally, a comprehensive analysis is performed, combining the high-quality dataset and various prediction results, to generate a comprehensive prediction result. Missing value imputation ensures data integrity and avoids prediction bias caused by missing data; outlier handling improves data reliability and reduces prediction errors caused by outliers; data standardization ensures data consistency and comparability, improving model training effectiveness; time series analysis is used to model the concentration change trends of gas components, improving prediction accuracy and reliability; machine learning algorithms are used to predict the performance degradation of adsorption materials, capturing complex nonlinear relationships and improving prediction accuracy; anomaly detection algorithms are implemented to promptly identify potential anomalies in the system, providing early warnings and preventing equipment failures or production accidents caused by abnormal operating conditions; comprehensive analysis of gas component concentration change trends, adsorption material performance degradation, and abnormal operating condition warnings generates comprehensive prediction results, providing a scientific basis for system optimization; based on the comprehensive prediction results, multi-objective optimization algorithms and fuzzy logic control theory are used to intelligently adjust the operating parameters of the adsorption tank, improving system operating efficiency and stability; actual operating results are fed back to the prediction model, and through continuous learning and optimization, a more accurate prediction model is generated, forming a closed-loop feedback mechanism to achieve continuous system improvement and optimization.

[0020] By utilizing comprehensive prediction results, the optimization objectives are identified and generated. Based on these objectives, a multi-objective optimization algorithm is selected to construct an optimization model for the adsorption tank's operating parameters, generating a preliminary optimization scheme. Using fuzzy logic control theory, by defining fuzzy rules and membership functions, the preliminary optimization scheme is transformed into more easily executable operating instructions, generating fuzzy logic control instructions. Based on these instructions, the operating parameters of the adsorption tank are intelligently adjusted, generating optimized operating parameter settings. Through comprehensive prediction results, the specific optimization objectives are clearly defined, ensuring the targeted and effective nature of the optimization work. The multi-objective optimization algorithm comprehensively considers multiple optimization objectives, generating a more comprehensive and accurate optimization scheme. Fuzzy logic control theory transforms complex optimization results into easily understood and executed operating instructions, improving operational convenience and accuracy. Intelligent adjustment of the adsorption tank's operating parameters improves system efficiency and stability, extending the lifespan of the adsorption material. Through continuous feedback and adjustment, a closed-loop optimization mechanism is formed, ensuring the system remains in optimal operating condition under different operating conditions.

[0021] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating an intelligent control and data analysis method for gas purification in an adsorption tank, provided as an embodiment of this application;

[0024] Figure 2 This is a schematic diagram of the structure of an intelligent control and data analysis system for gas purification in an adsorption tank, provided in an embodiment of this application.

[0025] Figure 3 This is a schematic diagram of the structure of a computing device provided in an embodiment of this application. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0027] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0028] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0029] Figure 1 This application provides a flowchart of an intelligent control and data analysis method for gas purification in an adsorption tank, as shown in the embodiments of this application. Figure 1 As shown, the method includes:

[0030] 101. Real-time acquisition of multidimensional environmental data within the adsorption tank to obtain a multidimensional environmental dataset, wherein the multidimensional environmental data includes gas component concentration, temperature, pressure, humidity, and flow rate;

[0031] Acquire real-time environmental data inside the adsorption tank for subsequent analysis and processing; collect multi-dimensional environmental data inside the adsorption tank in real time, including but not limited to gas component concentration, temperature, pressure, humidity, and flow rate; and use sensors and other monitoring devices to transmit this data to the central processing system.

[0032] 102. Based on the multidimensional environmental dataset, time series analysis and machine learning algorithms are used to predict the concentration change trend of gas components and the performance degradation of adsorption materials. At the same time, anomaly detection algorithms are applied to identify potential abnormal operating conditions and obtain comprehensive prediction results.

[0033] By analyzing the collected data, the system predicts the concentration trends of gas components and the performance degradation of adsorbent materials, and identifies potential abnormal operating conditions. Time series analysis is used to model the gas component concentration trends over time, generating a predictive model. Machine learning algorithms are applied to predict the performance degradation of adsorbent materials under different conditions based on historical and current data. Anomaly detection algorithms are used to identify potential abnormal conditions in the system and generate early warnings for abnormal operating conditions. The system incorporates data preprocessing, time series modeling, machine learning model training, and anomaly detection algorithms.

[0034] Optionally, step 102, which involves using time series analysis and machine learning algorithms to predict the concentration change trend of gas components and the performance degradation of adsorbent materials based on the multidimensional environmental dataset, and simultaneously applying anomaly detection algorithms to identify potential abnormal operating conditions, to obtain a comprehensive prediction result, includes: using data preprocessing techniques to impute missing values, handle outliers, and standardize data in the time series data of the multidimensional environmental dataset to obtain a high-quality dataset; using time series analysis techniques to model the concentration change trend of gas components over time to generate a gas concentration change trend prediction; using machine learning algorithms to predict the performance degradation of adsorbent materials under different conditions based on the gas concentration change trend prediction to obtain an adsorbent material performance degradation prediction; implementing anomaly detection algorithms based on multidimensional environmental data under normal operating conditions to identify and process possible abnormal conditions in the system and generate an abnormal operating condition warning; and using the high-quality dataset and the gas concentration change trend prediction to comprehensively analyze the gas component concentration change trend, the performance degradation of the adsorbent material, and the abnormal operating condition warning to generate a comprehensive prediction result.

[0035] Suppose we have an adsorption tank for purifying harmful gases in industrial waste gas; sensors are used to collect data on gas composition concentration, temperature, pressure, humidity, and flow rate in real time within the adsorption tank; data is recorded every minute and stored in a database; missing values ​​are imputed in the dataset using values ​​from the previous time step; outliers are identified and handled using the Z-score method, marking data points exceeding 3 standard deviations as outliers and removing them; the data is standardized using Min-Max, transforming all data to the range of 0-1; an LSTM model is selected to model the gas composition concentration data; exploratory analysis of the data reveals significant periodic changes in gas composition concentration; an LSTM model is built to predict the gas composition concentration for the next 10 minutes using data from the past hour; model parameters are adjusted to optimize prediction accuracy, ultimately generating a prediction of gas concentration change trends; a feature set for predicting the performance degradation of the adsorption material is constructed, including predicted values ​​of gas composition concentration change trends, temperature, and flow rate. Pressure, humidity, and flow rate are considered. A random forest algorithm is selected to learn the performance degradation of the adsorbent material under different conditions. Cross-validation is used to evaluate model performance, and the model is optimized by adjusting the number and depth of trees. The optimized model is applied to predict the performance degradation of the adsorbent material under new or variable conditions, generating a performance degradation prediction. A baseline model is generated using multi-dimensional environmental data under normal operating conditions. The IsolationForest algorithm is selected to detect anomalies in real-time collected multi-dimensional environmental data. Domain knowledge is combined to analyze the detected abnormal data points to determine whether they represent potential faults or abnormal operating conditions in the system. When abnormal operating conditions are detected, early warning information is issued in a timely manner, generating an abnormal operating condition warning. High-quality datasets and gas concentration change trend predictions are used to comprehensively analyze gas component concentration change trends, adsorbent material performance degradation, and abnormal operating condition warnings. A comprehensive prediction result is generated to provide a basis for subsequent intelligent adjustment and optimization.

[0036] Through the above steps, the system can effectively predict and manage the gas purification process in the adsorption tank, improving purification efficiency and system stability.

[0037] Optionally, based on the high-quality dataset, the step of modeling the trend of gas component concentration over time using time series analysis techniques to generate a gas concentration change trend prediction includes:

[0038] Using statistical analysis methods, exploratory analysis is performed on the gas component concentration data in the high-quality dataset to identify the time dependence and periodicity characteristics of the data, obtaining data feature analysis results. Based on the data feature analysis results, a suitable time series analysis model is selected to construct a prediction model for gas component concentration, generating a preliminary prediction model. Using the high-quality dataset, the preliminary prediction model is trained, and the prediction accuracy is optimized by adjusting the model parameters until a predetermined accuracy standard is reached, obtaining an optimized prediction model. Based on the optimized prediction model, future gas component concentration change trends are predicted, generating a gas concentration change trend prediction. Using actual measurements, the gas concentration change trend prediction is validated to evaluate the accuracy and reliability of the prediction model, and the model is adjusted and optimized to generate the final gas concentration change trend prediction.

[0039] Suppose we have an adsorption tank for purifying sulfur dioxide (SO2) from industrial waste gas; sensors are used to collect real-time data on the concentration of SO2, temperature, pressure, humidity, and flow rate within the adsorption tank; data is recorded every minute and stored in a database; missing values ​​are imputed using values ​​from the previous time step; outliers are identified and handled using the Z-score method, marking data points exceeding 3 standard deviations as outliers and removing them; Min-Max standardization is performed on the data, transforming all data to the range of 0-1; exploratory analysis of the SO2 concentration data is conducted using autocorrelation plots and moving average plots; it is found that the SO2 concentration data exhibits a clear diurnal periodicity, with daily concentration changes showing a certain regularity; time series plots and box plots are used to further confirm the data characteristics; based on the data characteristic analysis results, Proph is selected. The Prophet model is used because it can handle data with periodic characteristics well. A preliminary Prophet model is built using a high-quality dataset. This preliminary model is then trained using the high-quality dataset. Hyperparameters of the model, such as the length and intensity of seasonal cycles, are adjusted through cross-validation until the model reaches a predetermined accuracy standard (e.g., RMSE less than 0.1 ppm), generating an optimized prediction model. The optimized Prophet model is then used to predict future SO2 concentrations. A prediction of SO2 concentration trends over the next 24 hours is generated. Actual measurements are collected and compared with the prediction results. MAE and RMSE are calculated to evaluate the model's accuracy and reliability. Based on the evaluation results, model parameters are further adjusted, such as by adding more seasonal factors. Finally, a final prediction of SO2 concentration trends is generated.

[0040] Through the above steps, the system can effectively predict the trend of gas component concentration changes in the adsorption tank, providing a reliable basis for subsequent intelligent adjustment and optimization.

[0041] Optionally, the step of predicting the performance degradation of the adsorbent material under different conditions based on the gas concentration change trend prediction and using a machine learning algorithm to obtain the adsorbent material performance degradation prediction includes: constructing a feature set for the adsorbent material performance degradation prediction model using the gas concentration change trend prediction and multi-dimensional environmental data to obtain a complete feature set; selecting a suitable machine learning algorithm based on the complete feature set and historical data to learn the performance degradation of the adsorbent material under different conditions to generate a preliminary performance degradation prediction model; evaluating the preliminary performance degradation prediction model using cross-validation technology, and optimizing the model's prediction performance by adjusting model parameters and structure until satisfactory accuracy and stability are achieved to obtain an optimized performance degradation prediction model; predicting the performance degradation of the adsorbent material under new or variable conditions based on the optimized performance degradation prediction model to generate an adsorbent material performance degradation prediction; and comparing and verifying the adsorbent material performance degradation prediction with actual monitoring data to evaluate the applicability and reliability of the prediction model and generate the final adsorbent material performance degradation prediction.

[0042] Suppose we have an adsorption tank for purifying sulfur dioxide (SO2) in industrial waste gas, and we want to predict the performance degradation of the adsorption material. Sensors are used to collect real-time data on gas component concentration (SO2), temperature, pressure, humidity, and flow rate within the adsorption tank. Data is recorded every minute and stored in a database. Missing values ​​are imputed using values ​​from the previous time step. The Z-score method is used to identify and handle outliers, marking and removing data points exceeding 3 standard deviations. Min-Max normalization is applied to transform all data to the range of 0-1. The predicted gas concentration trend (e.g., SO2 concentration trend over the next 24 hours) is combined with multidimensional environmental data (temperature, pressure, humidity, and flow rate) to construct a feature set. A complete feature set is generated, including the predicted gas concentration trend, temperature, pressure, humidity, and flow rate. A random forest algorithm is selected as the machine learning model. The process involves: training an initial performance degradation prediction model using historical data and a complete feature set; evaluating the performance of the initial model using cross-validation; optimizing the model's predictive performance by adjusting model parameters (such as the number of trees in a random forest and its maximum depth) and structure; generating an optimized performance degradation prediction model until the model achieves satisfactory accuracy and stability (e.g., RMSE less than 0.05); using the optimized model to predict the performance of adsorbent materials under new or variable conditions; generating performance degradation predictions for adsorbent materials under different conditions, such as predicting performance degradation over the next 24 hours; collecting actual monitoring data and comparing it with the prediction results; calculating MAE and RMSE to evaluate the model's applicability and reliability; further adjusting model parameters based on the evaluation results, such as adding more features or using different machine learning algorithms; and generating the final performance degradation prediction for the adsorbent materials.

[0043] Through the above steps, the system can effectively predict the performance degradation of adsorption materials under different conditions, providing a scientific basis for the optimized operation and maintenance of adsorption tanks.

[0044] This application considers that the formula is used to generate a performance degradation prediction model for adsorbent materials under different conditions. Specifically, it is a comprehensive function trained using historical data and a complete feature set. The model combines support vector machines, additional feature functions, and regularization terms to improve the accuracy and robustness of predictions.

[0045] Optionally, the step of using historical data and selecting a suitable machine learning algorithm based on the complete feature set to learn the performance degradation of the adsorbent material under different conditions and generate a preliminary performance degradation prediction model includes: using historical data and the complete feature set, selecting a suitable machine learning algorithm, and training to obtain model parameters.

[0046]

[0047] Where f(x) is a comprehensive function used to generate a performance degradation prediction model for adsorbent materials under different conditions, x=[C t [T,P,H,F] is the input vector, containing the predicted value C of the gas component concentration change trend. t Temperature T, pressure P, humidity H, and flow rate F; N is the number of support vectors; α i It is a Lagrange multiplier; y i It is the label of the i-th support vector; w i (z) is the adaptive weight of the i-th support vector; It is the dynamic exponent term of the i-th support vector; K(z) i (x) is the kernel function; b is the bias term; M is the number of extra characteristic functions; β j These are the weights of the additional feature functions; g j (x) is the extra characteristic function; L is the number of regularization terms; γ k It is the weight of the regularization term; h k (x) is the regularization term; i represents the index of the support vector, from 1 to N; j represents the index of the extra feature function, from 1 to M; k represents the index of the regularization term, from 1 to L;

[0048] For the new input vector x = [C t ,T,P,H,F], calculate the adaptive weight w for each support vector. i (x) and dynamic exponential term and kernel function K(x) i ,x):

[0049]

[0050] Where β is the weight adjustment parameter. λ is the average of all support vector labels, ρ is the distance adjustment parameter, σ is the region distance adjustment parameter, and μ is the similarity adjustment parameter. i It is the mean of the i-th support vector, dist(x, x) i ) is the input vector x and the support vector x. i The distance, similarity(x,x) i ) is the input vector x and the support vector x. i Similarity;

[0051]

[0052] Where α is the exponential adjustment parameter for the mean; γ is the exponential adjustment parameter for the standard deviation; δ is the exponential adjustment parameter for the variance; η is the exponential adjustment parameter for skewness; φ is the exponential adjustment parameter for kurtosis; and ψ is the exponential adjustment parameter for mean(x). i std(x) is the mean of the i-th support vector; i ) is the standard deviation of the i-th support vector; var(x) i ) is the variance of the i-th support vector; skew(x) i ) is the skewness of the i-th support vector; kurt(x) i ) is the kurtosis of the i-th support vector; entropy(x) i ) is the entropy of the i-th support vector;

[0053] K(x i ,x)=exp(-γ||x i -x|| 2 )

[0054] Where γ is the parameter of the kernel function;

[0055] Calculate the additional characteristic function g j (x) and regularization term h k (x):

[0056] g j (x)=exp(α j ·mean(x)+γ j std(x)+δ j ·var(x)+η j ·skew(x)+φ j

[0057] ·kurt(x)+ψ j ·entropy(x))

[0058] Where, α j It is the exponential adjustment parameter of the mean; γ j It is the exponentially adjusted parameter of the standard deviation; δ j It is the exponential adjustment parameter for variance; η j It is the exponential adjustment parameter for skewness; φ j It is the exponential adjustment parameter for kurtosis; ψ j It is the exponential adjustment parameter for entropy;

[0059] h k (x)=exp(-λ k ||x-μ k || 2 )

[0060] Where, λ kIt is the regularization parameter; μ k It is the center of the regularization term;

[0061] The model's predictive performance is evaluated using validation set data. Based on the evaluation results, the model's parameter settings are adjusted, and repeated testing and tuning are performed to generate a preliminary performance degradation prediction model.

[0062] Suppose there is an adsorption tank used to purify sulfur dioxide (SO2) from industrial waste gas, and we want to predict the performance degradation of the adsorption material:

[0063] The historical dataset contains 1000 samples, each including a predicted value C for the trend of gas component concentration changes. t Temperature T, pressure P, humidity H, and flow velocity F; Example of input vector x: x = [C t [T,P,H,F]=[0.5,25,1013,50,2]; Support Vector Machine (SVM) is selected as the machine learning algorithm; The model is trained using historical data and a complete feature set to obtain the model parameters; Assume the first support vector x1=[0.6,26,1014,51,2.1], label y1=1, mean μ1=[0.6,26,1014,51,2.1]; Parameters are set to β=0.1, y - =0.5, λ=0.01, ρ=0.01, σ=0.01; Calculate the distance. The similarity is calculated as similarity(x,x1) = 0.95; the adaptive weight w1(x) is calculated as follows: Assume the statistical characteristics of the first support vector are mean(x1) = 0.6, standard deviation std(x1) = 0.1, variance var(x1) = 0.01, skewness(x1) = 0.05, kurtosis(x1) = 3.0, and entropy(x1) = 0.5. Set the parameters to α = 0.1, γ = 0.1, δ = 0.1, η = 0.1, φ = 0.1, and ψ = 0.1. Calculate the dynamic exponential term c. ∧ 1(x) is c ∧ 1(x) = exp(0.1·0.6 + 0.1·0.1 + 0.1·0.01 + 0.1·0.05 + 0.1·3.0 + 0.1·0.5) ≈ 1.65; Assuming the kernel function parameter γ = 0.1, calculate the kernel function K(x1,x) as follows:

[0064] K(x1,x)=exp(-0.1·2.236 2Assume the statistical characteristics of the first extra feature function are mean(x) = 0.5, standard deviation std(x) = 0.1, variance var(x) = 0.01, skewness(x) = 0.05, kurtiness(x) = 3.0, entropy(x) = 0.5, and parameters are set as α1 = 0.1, γ1 = 0.1, δ1 = 0.1, η1 = 0.1, φ1 = 0.1, ψ1 = 0.1; calculate the extra feature function g1(x) as g1(x) = exp(0.1·0.5 + 0.1·0.1 + 0.1·0.01 + 0.1·0.05 + 0.1·3.0 + 0.1·0.5) ≈ 1.65; assume the regularization parameter λ k =0.1, the center μ of the regularization term k =[0.5,25,1013,50,2], calculate the regularization term h1(x) as h1(x)=exp(-0.1·0)=1.0; assume the number of support vectors N=1, the Lagrange multiplier α1=1, the bias term b=0, the number of extra feature functions M=1, the number of regularization terms L=1, the weight of extra feature function β1=1, and the weight of regularization term γ1=1; calculate the comprehensive function f(x) as f(x)=1·1·0.88·1.65·0.65+0+1·1.65+1·1.0≈3.47;

[0065] Through the above steps, predicted values ​​of performance degradation of the adsorbent material under specific conditions are obtained. These predicted values ​​can be used to guide the operation and maintenance of the adsorption tank, improving the system's efficiency and stability.

[0066] Optionally, the step of implementing an anomaly detection algorithm based on multidimensional environmental data under normal operating conditions to identify and process potential abnormal conditions in the system and generate an abnormal operating condition warning includes: using multidimensional environmental data under normal operating conditions to perform statistical analysis on the distribution characteristics of gas composition parameters to obtain a baseline model; based on the baseline model, selecting a suitable anomaly detection algorithm to perform anomaly detection processing on the real-time collected multidimensional environmental data to generate abnormal data points; using the abnormal data points and combining domain knowledge to further analyze and process these abnormal data points to determine whether they represent potential faults or abnormal operating conditions in the system, and generating a preliminary abnormal operating condition report; and based on the preliminary abnormal operating condition report, formulating corresponding early warning strategies, and issuing early warning information in a timely manner when an abnormal operating condition is detected, generating an abnormal operating condition warning.

[0067] Suppose we have an adsorption tank for purifying sulfur dioxide (SO2) from industrial waste gas and want to identify potential abnormal operating conditions in the system. We collect multidimensional environmental data under normal operating conditions, including SO2 concentration, temperature, pressure, humidity, and flow rate. Data is recorded every minute and stored in a database. We perform statistical analysis on the multidimensional environmental data under normal operating conditions, calculating the mean, standard deviation, maximum, and minimum values ​​of each parameter. A baseline model is generated with the following parameters: mean SO2 concentration of 0.5 ppm, mean temperature of 25℃, mean pressure of 1013 hPa, mean humidity of 50%, mean flow rate of 2 m / s, and the standard deviation of each parameter. We select IsolationForest as the anomaly detection algorithm. Using the baseline model and IsolationForest, we perform anomaly detection on the real-time collected multidimensional environmental data. An example of real-time collected data is x = [0.8, 30, 1015, 60, 3]. We use IsolationForest to perform anomaly detection on the real-time data, generating abnormal data points and anomaly counts. Data point x1 = [0.8, 30, 1015, 60, 3], and anomalous data point 2 x2 = [0.7, 32, 1016, 65, 3.5]. Based on domain knowledge, further analysis is performed on the anomalous data points. It is determined whether these anomalous data points represent potential faults or abnormal operating conditions in the system. A preliminary anomalous operating condition report is generated. Anomalous operating condition 1 is a significant increase in SO2 concentration, which may indicate adsorption material failure; anomalous operating condition 2 is a simultaneous increase in temperature and humidity, which may indicate equipment overheating or environmental changes. Based on the preliminary anomalous operating condition report, a plan is formulated. The corresponding early warning strategy is as follows: Set the SO2 concentration threshold to 0.7 ppm; a Level 1 warning is issued when this threshold is exceeded. Set the temperature threshold to 30℃; a Level 2 warning is issued when this threshold is exceeded. Define alarm levels, such as a Level 1 warning notifying the operator and a Level 2 warning notifying the engineer. When abnormal operating conditions are detected, the system automatically issues warning information to notify relevant personnel to take necessary measures. Abnormal operating condition warnings are generated as follows: Level 1 warning: SO2 concentration exceeds 0.7 ppm, suggesting inspection of the adsorption material; Level 2 warning: temperature exceeds 30℃, suggesting inspection of the equipment cooling system.

[0068] Through the above steps, the system can effectively identify and warn of potential abnormal operating conditions, ensuring the normal operation and safety of the adsorption tank.

[0069] 103. Based on the comprehensive prediction results, a multi-objective optimization algorithm and fuzzy logic control theory are used to intelligently adjust the operating parameters of the adsorption tank to generate optimized operating parameter settings.

[0070] Based on the comprehensive prediction results, the operating parameters of the adsorption tank are optimized to improve gas purification efficiency and system stability; the optimization objective is determined, a suitable multi-objective optimization algorithm is selected, an optimization model is constructed, and a preliminary optimization scheme is generated; using fuzzy logic control theory, the optimization results are converted into specific operation instructions to generate fuzzy logic control instructions; the optimization objective is defined, a multi-objective optimization algorithm is selected, an optimization model is constructed, and a preliminary optimization scheme is generated; fuzzy rules and membership functions are defined to convert the optimization results into specific operation instructions.

[0071] Optionally, step 103, based on the comprehensive prediction results, employs a multi-objective optimization algorithm and fuzzy logic control theory to intelligently adjust the operating parameters of the adsorption tank, generating optimized operating parameter settings, including:

[0072] Using the comprehensive prediction results, the target to be optimized is determined to obtain the optimization target; based on the optimization target, a multi-objective optimization algorithm is selected to construct an optimization model for the adsorption tank operating parameters, resulting in a preliminary optimization scheme; using fuzzy logic control theory, the preliminary optimization scheme is processed, and by defining fuzzy rules and membership functions, the optimization results are converted into more easily executable operating instructions, generating fuzzy logic control instructions; according to the fuzzy logic control instructions, the operating parameters of the adsorption tank are intelligently adjusted to generate optimized operating parameter settings.

[0073] Suppose there is an adsorption tank used to purify sulfur dioxide from industrial waste gas, and the goal is to improve gas purification efficiency and system stability by optimizing operating parameters. Analyze comprehensive prediction results to determine the optimization objectives. For example, if the prediction shows that the SO2 concentration will increase within the next 24 hours, the optimization objective is to improve gas purification efficiency; simultaneously, if the prediction shows that the adsorption material's performance degrades rapidly, the optimization objective also includes extending the adsorption material's lifespan. NSGA-II is selected as the multi-objective optimization algorithm. An optimization model is constructed, defining the relationship between operating parameters and optimization objectives. The optimization model can be expressed as MaximizeE(T,P,F) and... Maximize L(T,P,F), where E(T,P,F) represents the gas purification efficiency and L(T,P,F) represents the adsorption material lifetime. The optimization model is solved using NSGA-II to generate a preliminary optimization scheme: temperature T = 26℃, pressure P = 1014 hPa, and flow rate F = 2.1 m / s. Fuzzy rules are defined to describe the fuzzy relationship between the operating parameters and the optimization objective: Rule 1: If the temperature is high, decrease the flow rate; Rule 2: If the pressure is low, increase the temperature. Membership functions are defined to describe the degree to which an input variable belongs to a certain fuzzy set: the membership functions for temperature are low temperature (0-20℃), medium temperature (20-30℃), and high temperature (30-40℃); the membership functions for pressure are low pressure (1000-1010 hPa) and medium pressure (1010-1020 hPa). The initial optimization scheme is converted into specific operating instructions using a fuzzy logic controller. For example, if the current temperature is 26℃ (medium temperature), pressure is 1014hPa (medium pressure), and flow rate is 2.1m / s (medium speed), according to fuzzy rules, rule 1 is that the temperature is too high, so the flow rate should be reduced to 2.0m / s; rule 2 is that the pressure is moderate, so the temperature should remain unchanged. The fuzzy logic control instruction is then generated: adjust the flow rate to 2.0m / s. Based on the fuzzy logic control instruction, the operating parameters of the adsorption tank are adjusted. The optimized operating parameters are then generated: temperature T = 26℃, pressure P = 1014hPa, and flow rate F = 2.1m / s.

[0074] Through the above steps, the system can intelligently adjust the operating parameters of the adsorption tank based on comprehensive prediction results, thereby improving gas purification efficiency and system stability. This optimization method not only considers a single objective but also takes into account multiple optimization objectives, ensuring that the system operates in its optimal state.

[0075] This application considers that the formula is used to construct a multi-objective optimization model for the operating parameters of the adsorption tank. Multiple optimization objectives are integrated through a comprehensive objective function, including minimizing the prediction error of gas component concentration change trends, minimizing system operating costs, minimizing the frequency of abnormal operating condition warnings, and maximizing system stability. Each objective function has its specific expression, which are combined into a comprehensive objective function through a weighted sum.

[0076] Optionally, based on the optimization objective, a suitable multi-objective optimization algorithm is selected to construct an optimization model for the operating parameters of the adsorption tank, thereby obtaining a preliminary optimization scheme, including:

[0077] Multiple optimization objectives are integrated by combining the objective function f(x):

[0078]

[0079] Where x = [x1, x2, ..., x n [] is the operation parameter vector, containing the operation parameters for adsorbing gnats; m is the number of objective functions; ω i The weight of the i-th objective satisfies f i (x) is the i-th objective function; f i min f is the minimum value of the i-th objective function; i max It is the maximum value of the i-th objective function; γ i (x) is the nonlinear adjustment parameter of the i-th objective function, which depends on the operation parameter vector; λ i It is a distance adjustment parameter; It is the optimal operation parameter vector of the i-th objective function; It is the operation parameter vector x and the optimal operation parameter vector The distance; i represents the index of the weight, from 1 to m;

[0080] Minimize the prediction error of gas component concentration change trend.

[0081]

[0082] Minimize system operating cost

[0083]

[0084] Minimize the frequency of abnormal operating condition warnings to

[0085]

[0086] Maximizing system stability is

[0087]

[0088] Where T is the number of time steps; These are predicted values ​​for the concentration of gas components; This is the actual value of the gas component concentration; c i It is the cost coefficient of the i-th operating parameter; d ij It is the interaction cost coefficient between operating parameters; I(anomaly) t ) is the abnormal operating condition indication function at time t. It is 1 if there is an abnormality, and 0 otherwise. This is a predicted value for system pressure; This represents the actual value of the system pressure; i represents the index of the cost coefficient, from 1 to n; t represents the index of the time step, from 1 to T;

[0089] The values ​​of each objective function are calculated using the above formulas, and these values ​​are substituted into the comprehensive objective function f(x) to optimize the operation parameter vector x, thus obtaining the optimization results. Based on the operation parameter vector x obtained from the optimization results, a preliminary optimization scheme is obtained.

[0090] Suppose there is an adsorption tank used to purify sulfur dioxide from industrial waste gas, and the goal is to improve gas purification efficiency and system stability by optimizing operating parameters; the operating parameter vector is x = [T, P, F], where T is temperature, P is pressure, and F is flow rate; the objective functions are f1(x) to minimize the prediction error of gas component concentration change trend; f2(x) to minimize system operating cost; f3(x) to minimize the abnormal operating condition warning frequency; and f4(x) to maximize system stability; the weights are ω1 = 0.3, ω2 = 0.2, ω3 = 0.2, ω4 = 0.3; the nonlinear adjustment parameters are γ1(x) = 1, γ2(x) = 1, γ3(x) = 1, γ4(x) = 1; the distance adjustment parameters are λ1 = 0.1, λ2 = 0.1, λ3 = 0.1, λ4 = 0.1; and the optimal operating parameter vector is given. Assuming the current operating parameters x = [26, 1014, 2.1]; the prediction error for the gas component concentration change trend should be minimized as follows:

[0091] Assuming T = 10, the predicted value and the actual value are respectively Calculated Minimize system operating costs Assume c1 = 0.1, c2 = 0.1, c3 = 0.1, d 12 =0.01, d 13 =0.01, d 23 =0.01; calculated, f2(x) = 0.1·26 2 +0.1·1014 2+0.1·2.1 2 +0.01·26·1014+0.01·26·2.1+0.01·1014·2.1≈103000; Minimize the frequency of abnormal operating condition warnings. Assuming T = 10, the abnormal operating condition indication functions are I(anomaly) t = [0,0,1,0,0,1,0,0,0,0]; calculate f3(x) = log(1+0) + log(1+0) + log(1+1) + log(1+0) + log(1+0) + log(1+1) + log(1+0) + log(1+0); maximize system stability. Assuming T = 10, the predicted value and the actual value are respectively Calculated

[0092]

[0093] Assume f1 min =0,f1 max =1,f2 min =0,f2 max =100000, f3 min =0,f3 max =2,f4 min =-1,f4 max =0, calculate distance for

[0094]

[0095] Calculate the comprehensive objective function value.

[0096]

[0097] f(x)=0.3·0·exp(-0.147)+0.2·1.03·exp(-0.147)+0.2·0.693·exp(-0.147)+0.3·1·exp(-0.147)

[0098] f(x)=0+0.2·1.03·0.863+0.2·0.693·0.863+0.3·1·0.863

[0099] f(x)=0+0.180+0.121+0.259≈0.560

[0100] Through the above steps, we obtained the comprehensive objective function value f(x)≈0.560 for the operating parameter vector x=[26,1014,2.1]. This value can be used to guide the adjustment of the operating parameters of the adsorption tank to achieve multi-objective optimization.

[0101] Optionally, the preliminary optimization scheme is processed using fuzzy logic control theory. By defining fuzzy rules and membership functions, the optimization result is converted into more easily executable operation instructions, generating fuzzy logic control instructions, including:

[0102] Using the preliminary optimization scheme, the optimization results of the adsorption tank operating parameters are analyzed and processed to determine the key parameters that need to be converted into operating instructions, resulting in a list of key parameters. Based on the list of key parameters, fuzzy rules are defined to describe the nonlinear relationships and interactions between different parameters, generating a set of fuzzy rules. Membership functions are defined to convert specific numerical values ​​into membership degrees in fuzzy sets. Each set corresponds to a membership function, representing the degree to which different temperature values ​​belong to each set, generating a set of membership functions. Using the set of fuzzy rules and the set of membership functions, the specific numerical values ​​in the optimization results are converted into fuzzy logic control instructions, generating fuzzy logic control instructions.

[0103] Suppose there is an adsorption tank used to purify sulfur dioxide in industrial waste gas, and we want to improve gas purification efficiency and system stability by optimizing operating parameters;

[0104] The preliminary optimization scheme was analyzed to identify the key parameters that need to be converted into operational instructions. The preliminary optimization scheme suggests the following operational parameters: temperature T = 26℃, pressure P = 1014 hPa, and flow velocity F = 2.1 m / s. The key parameter list includes temperature T, pressure P, and flow velocity F. Based on the key parameter list, fuzzy rules were defined to describe the nonlinear relationships and interactions between different parameters. Rule 1 states that if the temperature is high, the flow velocity should be reduced; rule 2 states that if the pressure is low, the temperature should be increased; and rule 3 states that if the flow velocity is high, the pressure should be reduced. The resulting fuzzy rule set is as follows: Rule 1 states that if the temperature is high, the flow velocity should be low. Rule 2 states that lower pressure corresponds to higher temperature, and rule 3 states that higher flow rate corresponds to lower pressure. Membership functions are defined to convert specific numerical values ​​into membership degrees within fuzzy sets. The membership functions for temperature (TT) are as follows: low temperature (0-20℃) is a triangular membership function, medium temperature (20-30℃) is a trapezoidal membership function, and high temperature (30-40℃) is a triangular membership function. The membership functions for pressure (PP) are as follows: low pressure (1000-1010 hPa) is a triangular membership function, medium pressure (1010-1020 hPa) is a trapezoidal membership function, and high pressure (1020-1030 hPa) is a triangular membership function. Membership functions; the membership functions for flow velocity FF are as follows: low speed (0-1.5 m / s) is a triangular membership function, medium speed (1.5-2.5 m / s) is a trapezoidal membership function, and high speed (2.5-3.5 m / s) is a triangular membership function. Using fuzzy rule sets and membership function sets, the specific values ​​in the optimization results are converted into fuzzy logic control instructions. The current operating parameters are temperature T = 26℃, pressure P = 1014 hPa, and flow velocity F = 2.1 m / s. According to the membership functions, the specific values ​​are converted into membership degrees in the fuzzy set. Temperature T = 26℃ belongs to medium temperature, and pressure P = 1014 hPa... The pressure is medium, and the flow velocity F = 2.1 m / s is medium speed. Applying the following fuzzy rules: Rule 1 states that a higher temperature corresponds to a lower flow velocity; however, since T = 26℃ is medium temperature, this rule is not triggered. Rule 2 states that a lower pressure corresponds to a higher temperature; however, since P = 1014 hPa is medium pressure, this rule is not triggered. Rule 3 states that a higher flow velocity corresponds to a lower pressure; however, since F = 2.1 m / s is medium speed, this rule is not triggered. Combining these rules, the fuzzy logic control instruction is generated: keep the current operating parameters unchanged, i.e., T = 26℃, pressure P = 1014 hPa, and flow velocity F = 2.1 m / s.

[0105] Through the above steps, the system can generate specific operating instructions based on the preliminary optimization scheme using fuzzy logic control theory, ensuring that the adsorption tank operates in an optimal state. This method can not only handle complex nonlinear relationships but also flexibly respond to different operating conditions, improving the system's stability and efficiency.

[0106] 104. Apply the optimized operating parameter settings to the actual operation of the adsorption tank, monitor its implementation effect, and feed the actual operation effect back to the prediction model to generate a more accurate prediction model.

[0107] The optimized operating parameters are applied to actual operation, the implementation effect is monitored, and the actual operating effect is fed back to the prediction model for further optimization. The generated optimized operating parameter settings are applied to the actual operation of the adsorption tank. The operating status of the adsorption tank is monitored in real time, and actual operating data is collected. The actual operating effect is fed back to the prediction model to evaluate the accuracy and reliability of the model, and the model is adjusted and optimized based on the feedback results. The optimized operating parameters are implemented, the system operating status is monitored in real time, actual operating data is collected, and the data is fed back to the prediction model for adjustment and optimization.

[0108] Figure 2 This application provides a schematic diagram of the structure of an intelligent control and data analysis system for gas purification in an adsorption tank, as shown in the embodiment of the present application. Figure 2 As shown, the device includes:

[0109] The acquisition module 21 is used to acquire multidimensional environmental data in the adsorption tank in real time to obtain a multidimensional environmental dataset, which includes gas component concentration, temperature, pressure, humidity and flow rate.

[0110] Processing module 22 is used to predict the concentration change trend of gas components and the performance degradation of adsorption materials based on the multidimensional environmental dataset using time series analysis and machine learning algorithms, and at the same time apply anomaly detection algorithm to identify potential abnormal operating conditions and obtain comprehensive prediction results.

[0111] The generation module 23 is used to intelligently adjust the operating parameters of the adsorption tank based on the comprehensive prediction results, using a multi-objective optimization algorithm and fuzzy logic control theory, and generate optimized operating parameter settings.

[0112] The monitoring module 24 is used to apply the optimized operating parameter settings to the actual operation of the adsorption tank, monitor its implementation effect, and feed the actual operation effect back to the prediction model to generate a more accurate prediction model.

[0113] Figure 2 The aforementioned intelligent control and data analysis system for gas purification in an adsorption tank can perform... Figure 1The implementation principle and technical effects of the intelligent control and data analysis method for gas purification in an adsorption tank as described in the above embodiment will not be repeated here. The specific operation methods of each module and unit in the intelligent control and data analysis system for gas purification in an adsorption tank in the above embodiment have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0114] In one possible design, Figure 2 The intelligent control and data analysis system for gas purification in an adsorption tank, as shown in the embodiment, can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0115] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0116] The processing component 32 is used to: collect multi-dimensional environmental data in the adsorption tank in real time to obtain a multi-dimensional environmental dataset, which includes gas component concentration, temperature, pressure, humidity, and flow rate; based on the multi-dimensional environmental dataset, use time series analysis and machine learning algorithms to predict the concentration change trend of gas components and the performance degradation of adsorption materials, and simultaneously apply anomaly detection algorithms to identify potential abnormal operating conditions to obtain a comprehensive prediction result; according to the comprehensive prediction result, use a multi-objective optimization algorithm and fuzzy logic control theory to intelligently adjust the operating parameters of the adsorption tank to generate optimized operating parameter settings; apply the optimized operating parameter settings to the actual operation of the adsorption tank, monitor its implementation effect, and feed the actual operating effect back to the prediction model to generate a more accurate prediction model.

[0117] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0118] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0119] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0120] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0121] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0122] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0123] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates an intelligent control and data analysis method for gas purification in an adsorption tank.

[0124] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0125] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0126] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0127] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions 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 this application.

Claims

1. A method for intelligent control and data analysis of gas purification in an adsorption tank, characterized in that, include: Real-time acquisition of multidimensional environmental data within the adsorption tank yields a multidimensional environmental dataset, which includes gas component concentration, temperature, pressure, humidity, and flow rate. Based on the multidimensional environmental dataset, time series analysis and machine learning algorithms are used to predict the concentration change trend of gas components and the performance degradation of adsorption materials. At the same time, anomaly detection algorithms are applied to identify potential abnormal operating conditions and obtain comprehensive prediction results. Based on the comprehensive prediction results, a multi-objective optimization algorithm and fuzzy logic control theory are used to intelligently adjust the operating parameters of the adsorption tank and generate optimized operating parameter settings. The optimized operating parameters are applied to the actual operation of the adsorption tank, its implementation effect is monitored, and the actual operating effect is fed back to the prediction model of gas component concentration after multiple optimizations to generate a more accurate prediction model. Based on the comprehensive prediction results, a multi-objective optimization algorithm and fuzzy logic control theory are used to intelligently adjust the operating parameters of the adsorption tank, generating optimized operating parameter settings, including: Using the comprehensive prediction results, the target that needs to be optimized is determined to obtain the optimized target; Based on the aforementioned optimization objectives, a multi-objective optimization algorithm is selected to construct an optimization model for the operating parameters of the adsorption tank, thereby obtaining a preliminary optimization scheme. Using fuzzy logic control theory, the preliminary optimization scheme is processed. By defining fuzzy rules and membership functions, the optimization results are converted into more easily executable operation instructions, generating fuzzy logic control instructions. Based on the fuzzy logic control instructions, the operating parameters of the adsorption tank are intelligently adjusted to generate optimized operating parameter settings. The step of selecting a multi-objective optimization algorithm based on the optimization objective to construct an optimization model for the adsorption tank operating parameters and obtain a preliminary optimization scheme includes: By synthesizing the objective function To integrate multiple optimization objectives: in, It is an operational parameter vector containing the operational parameters for adsorbing snakes. It refers to the number of operation parameters; It is the number of objective functions; It is the first The weights of each objective satisfy... ; It is the first One objective function; It is the first The minimum value of an objective function; It is the first The maximum value of each objective function; It is the first The nonlinear adjustment parameters of each objective function depend on the operation parameter vector; It is a distance adjustment parameter; It is the first The optimal operation parameter vector of each objective function; It is an operation parameter vector With the optimal operation parameter vector The distance; Minimize the prediction error of gas component concentration change trends: Minimize the operating cost of the adsorption tank system: Minimize the frequency of abnormal operating condition warnings: Maximizing the stability of the adsorption tank system: in, It is the number of time steps; These are predicted values ​​for the concentration of gas components; This is the actual value of the gas component concentration; It is the first Cost coefficients for each operating parameter; It is the interaction cost coefficient between operating parameters; anomaly It is the first The abnormal operating condition indication function at any given time is 1 if an abnormality exists, and 0 otherwise. Indicates the first Are there any abnormal operating conditions at any given time? It is a predicted value of the adsorption tank system pressure; This is the actual value of the adsorption tank system pressure; The values ​​of each objective function are calculated using the above formulas, and these values ​​are then substituted into the comprehensive objective function. In this process, a multi-objective optimization algorithm is used to optimize the operation parameter vector. Optimize to obtain the optimization result; obtain the operation parameter vector based on the optimization result. A preliminary optimization plan was obtained.

2. The method according to claim 1, characterized in that, Based on the multidimensional environmental dataset, time series analysis and machine learning algorithms are used to predict the concentration trends of gas components and the performance degradation of adsorbent materials. Simultaneously, anomaly detection algorithms are applied to identify potential abnormal operating conditions, resulting in a comprehensive prediction outcome, including: By using data preprocessing techniques, missing value imputation, outlier handling, and data standardization are performed on the time series data in the multidimensional environmental dataset to obtain a high-quality dataset. Based on the high-quality dataset, time series analysis techniques are used to model the trend of gas component concentration over time, generating a prediction of gas concentration change trend. Based on the predicted trend of gas concentration change, a machine learning algorithm is used to predict the performance degradation of the adsorption material under different conditions, thereby obtaining the performance degradation prediction of the adsorption material. Based on multidimensional environmental data under normal operating conditions, an anomaly detection algorithm is implemented to identify and process possible abnormal conditions in the adsorption tank system and generate early warnings of abnormal operating conditions. Using the high-quality dataset and the predicted gas concentration change trend, a comprehensive analysis is performed on the gas component concentration change trend, the performance degradation of the adsorption material, and the early warning of abnormal operating conditions to generate a comprehensive prediction result.

3. The method according to claim 2, characterized in that, Based on the high-quality dataset, time series analysis techniques are used to model the concentration variation trend of gas components over time, generating a gas concentration variation trend prediction, including: Using statistical analysis methods, exploratory analysis is performed on the gas component concentration data in the high-quality dataset to identify the time dependence and periodicity of the data, and to obtain the data feature analysis results. Based on the data feature analysis results, a time series analysis model is selected to construct a prediction model for gas component concentration, generating a preliminary prediction model. Using the high-quality dataset, the preliminary prediction model is trained, and the prediction accuracy is optimized by adjusting the model parameters until the predetermined accuracy standard is reached, thus obtaining the optimized prediction model. Based on the optimized prediction model, the future trend of gas component concentration change is predicted to generate a gas concentration change trend prediction. The gas concentration change trend prediction is verified using actual measured values. The accuracy and reliability of the optimized prediction model are evaluated. The optimized prediction model is then adjusted and optimized to generate the final gas concentration change trend prediction.

4. The method according to claim 2, characterized in that, The step of predicting the performance degradation of the adsorbent material under different conditions based on the gas concentration change trend, and using a machine learning algorithm, to obtain the adsorbent material performance degradation prediction, includes: Using the gas concentration change trend prediction and combined with multidimensional environmental data, the feature set of the adsorption material performance degradation prediction model is constructed and processed to obtain a complete feature set. Based on the complete feature set, historical data is used to select a machine learning algorithm to learn the performance degradation of the adsorbent material under different conditions and generate a preliminary performance degradation prediction model. Cross-validation technology is used to evaluate the preliminary performance degradation prediction model. By adjusting the model parameters and structure of the preliminary performance degradation prediction model, the prediction performance of the preliminary performance degradation prediction model is optimized until the preset accuracy and stability are achieved, thus obtaining the optimized performance degradation prediction model. Based on the optimized performance degradation prediction model, the performance degradation of the adsorption material under new or variable conditions is predicted to generate an adsorption material performance degradation prediction. The predicted performance degradation of the adsorbent material is compared and verified with actual monitoring data to evaluate the applicability and reliability of the optimized performance degradation prediction model and generate the final predicted performance degradation of the adsorbent material. Specifically, based on the complete feature set, historical data is used to select a machine learning algorithm to learn the performance degradation of the adsorbent material under different conditions, generating a preliminary performance degradation prediction model, including: Using historical data and a complete feature set, a machine learning algorithm is selected and trained to obtain model parameters: in, It is a comprehensive function used to generate a performance degradation prediction model for adsorbent materials under different conditions. It is an input vector containing predicted values ​​of gas component concentration change trends. ,temperature ,pressure ,humidity and flow rate ; It is the number of support vectors; They are Lagrange multipliers; It is the first Labels of the support vectors; It is the first Adaptive weights for each support vector; It is the first The dynamic exponential term of each support vector; It is a kernel function; It is a bias term; It is the number of additional characteristic functions; These are the weights of the additional feature functions; It is an additional characteristic function; It is the number of regularization terms; It is the weight of the regularization term; is the regularization term; i represents the index of the support vector, from 1 to N; j represents the index of the extra feature function, from 1 to M; k represents the index of the regularization term, from 1 to L. For the new input vector Calculate the adaptive weights for each support vector. and dynamic index term and kernel function : in, It is a weight adjustment parameter. It is the average of all support vector labels. It's a distance adjustment parameter. This is a regional distance adjustment parameter. It is a similarity adjustment parameter. It is the first The mean of the support vectors, It is the input vector With support vectors distance, It is the input vector With support vectors Similarity; in, It is the exponential adjustment parameter of the mean; It is the exponentially adjusted parameter of the standard deviation; It is the exponential adjustment parameter for variance; It is the exponential adjustment parameter for skewness; It is the exponential adjustment parameter for kurtosis; It is the exponential adjustment parameter of entropy; mean It is the first The mean of the support vectors; It is the first The standard deviation of each support vector; It is the first The variance of the support vectors; It is the first The skewness of each support vector; It is the first kurtosis of support vectors; entropy It is the first The entropy of each support vector; in, These are the parameters of the kernel function; Calculate additional characteristic functions and regularization term : in, It is the exponential adjustment parameter of the mean; It is the exponentially adjusted parameter of the standard deviation; It is the exponential adjustment parameter for variance; It is the exponential adjustment parameter for skewness; It is the exponential adjustment parameter for kurtosis; It is the exponential adjustment parameter for entropy; in, It is a regularization parameter; It is the center of the regularization term; The model's predictive performance is evaluated using validation set data. Based on the evaluation results, the model's parameter settings are adjusted, and repeated testing and tuning are performed to generate a preliminary performance degradation prediction model.

5. The method according to claim 2, characterized in that, Based on multi-dimensional environmental data under normal operating conditions, an anomaly detection algorithm is implemented to identify and process potential abnormal conditions in the adsorption tank system, generating abnormal operating condition warnings, including: By using multidimensional environmental data under normal operating conditions, the distribution characteristics of gas composition parameters are statistically analyzed to obtain a baseline model; Based on the baseline model, an anomaly detection algorithm is selected to perform anomaly detection processing on the real-time collected multidimensional environmental data and generate anomaly data points. Further analysis and processing of the abnormal data points are performed to determine whether they represent potential faults or abnormal operating conditions in the adsorption tank system, and a preliminary abnormal operating condition report is generated. Based on the preliminary abnormal operating condition report, a corresponding early warning strategy is formulated. When an abnormal operating condition is detected, an early warning message is issued in a timely manner, generating an abnormal operating condition warning.

6. The method according to claim 1, characterized in that, The preliminary optimization scheme is processed using fuzzy logic control theory. By defining fuzzy rules and membership functions, the optimization result is converted into more easily executable operation instructions, generating fuzzy logic control instructions, including: Using the aforementioned preliminary optimization scheme, the optimization results of the adsorption tank operating parameters are analyzed and processed to determine the key parameters that need to be converted into operating instructions, thus obtaining a list of key parameters. Based on the list of key parameters, fuzzy rules are defined to describe the nonlinear relationships and interactions between different key parameters, and a set of fuzzy rules is generated. Define a membership function to convert specific numerical values ​​into membership degrees in a fuzzy rule set. Each fuzzy rule set corresponds to a membership function, which represents the degree to which different temperature values ​​belong to each fuzzy rule set, and generates a membership function set. Using the set of fuzzy rules and the set of membership functions, the specific numerical values ​​in the optimization results are converted into fuzzy logic control instructions, thereby generating fuzzy logic control instructions.

7. An intelligent control and data analysis system for gas purification in an adsorption tank, used to execute the intelligent control and data analysis method for gas purification in an adsorption tank as described in any one of claims 1 to 6, characterized in that, include: The acquisition module is used to acquire multidimensional environmental data in the adsorption tank in real time to obtain a multidimensional environmental dataset, which includes gas component concentration, temperature, pressure, humidity and flow rate. The processing module is used to predict the concentration change trend of gas components and the performance degradation of adsorption materials based on the multidimensional environmental dataset using time series analysis and machine learning algorithms. At the same time, it applies anomaly detection algorithms to identify potential abnormal operating conditions and obtain comprehensive prediction results. The generation module is used to intelligently adjust the operating parameters of the adsorption tank based on the comprehensive prediction results, using a multi-objective optimization algorithm and fuzzy logic control theory, and generate optimized operating parameter settings. The monitoring module is used to apply the optimized operating parameter settings to the actual operation of the adsorption tank, monitor its implementation effect, and feed the actual operating effect back to the prediction model to generate a more accurate prediction model.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to realize the intelligent control and data analysis method for gas purification in an adsorption tank as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements an intelligent control and data analysis method for gas purification in an adsorption tank as described in any one of claims 1 to 6.

Images