A precise ammonia injection control method and system applicable to denitrification of coke oven flue gas

Abstract

The present invention discloses a precise ammonia injection control method and system applicable to denitrification of coke oven flue gas. The method includes: collecting historical data and real-time data of coke oven flue gas flow rate and NOx concentration, using a time series prediction algorithm to model the collected historical data of flue gas flow rate and NOx concentration, predicting the flue gas flow rate and NOx emission values in a future period of time, calculating the corresponding ammonia injection amount based on the predicted flue gas flow rate and NOx emission values by the model, the system monitors the actual NOx emission value in real time, compares it with the predicted value, if there is an obvious deviation, then corrects the prediction model in the next cycle and transmits the calculated ammonia injection amount data to the denitrification master control system, and the denitrification master control system automatically adjusts the ammonia supply valve, and if the denitrification precise ammonia injection control system fails, an alarm is immediately triggered and it automatically switches to the manual control mode. The present invention realizes the accurate control and stability of the total ammonia regulation of the denitrification system.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial pollution control, and particularly to a precise ammonia injection control method and system applicable to denitrification of coke oven flue gas. Background Art

[0002] During the production process of a coking plant, the coke oven flue gas contains a high concentration of nitrogen oxides (NOx), which causes serious pollution to the environment.

[0003] Currently, the medium and low temperature SCR denitrification process is generally adopted to purify NOx in the flue gas. Since the NOx value in the coke oven flue gas fluctuates greatly and changes periodically, the addition amount of ammonia cannot be automatically controlled by the PID proportional regulation method. Currently, manual adjustment is generally carried out according to the hourly average emission monitoring data, lacking the prediction of future emission trends, resulting in unstable denitrification effects and low ammonia utilization rate.

[0004] Therefore, there is an urgent need for a denitrification control system that can predict NOx emissions based on historical data and accurately adjust the ammonia injection amount to improve the denitrification effect and reduce the ammonia escape rate. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] Therefore, the purpose of the present invention is to provide a precise ammonia injection system and its control method applicable to denitrification of coke oven flue gas, so as to achieve accurate control of the total ammonia regulation of the denitrification system, and at the same time improve the timeliness, accuracy, reliability, and stability of the ammonia injection system for denitrification.

[0007] To solve the above technical problems, according to one aspect of the present invention, the following technical solutions are provided:

[0008] A precise ammonia injection control method applicable to denitrification of coke oven flue gas, comprising the following steps:

[0009] S1. Collect historical data and real-time data of the coke oven flue gas flow rate and NOx concentration, and store them in the on-site database;

[0010] S2. Use the time series prediction algorithm to model the historical data of the flue gas flow rate and NOx concentration collected, and predict the NOx emission value within a future period of time;

[0011] S3. Calculate the corresponding ammonia injection amount based on the NOx emission value predicted by the model. The system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is an obvious deviation, the prediction model will be corrected in the next cycle.

[0012] S4. Transmit the calculated ammonia injection amount data to the denitration master control system, which automatically adjusts the ammonia injection valve. If the denitration precise ammonia injection system fails, an alarm will be triggered immediately and it will automatically switch to the manual control mode.

[0013] As a preferred solution of a precise ammonia injection control method for coke oven flue gas denitration according to the present invention, in step S1, the NOx data is stored in the DCS system, and the data is collected through the modbus TCP protocol.

[0014] As a preferred solution of a precise ammonia injection control method for coke oven flue gas denitration according to the present invention, in step S2, the time series prediction algorithm is used to model the historical data of flue gas flow and NOx concentration collected, and the specific steps for predicting the NOx emission value in the next period are as follows:

[0015] Clean the NOx time series training data, remove outliers, and fill in missing values to ensure the accuracy and integrity of the data.

[0016] Resample the data according to the determined time period. Divide the data into a training set and a test set for model training and verification.

[0017] Use the TPR time series prediction algorithm to train the data, and verify the error of the model prediction on the test set. The error uses the MASE mean absolute percentage error, and the calculation formula is as follows:

[0018]

[0019] Where Ft is the predicted value at time t, At is the actual value at time t, n is the number of prediction time points, and N is the total number of actual values.

[0020] According to the evaluation results, use all the training data for training to form the final model, and fix it as the model for predicting the actual total NOx emissions.

[0021] As a preferred solution of a precise ammonia injection control method for coke oven flue gas denitration according to the present invention, in step S3, the system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is an obvious deviation, the specific steps for correcting the prediction model in the next cycle are as follows:

[0022] Compare the predicted values of flue gas flow and NOx concentration with the actual operating values, first correct the total ammonia injection amount, and then perform a secondary correction on the total ammonia injection amount according to the hourly average values of NOx and ammonia slip actually discharged at the CEMS outlet.

[0023] During the actual operation process, compare the predicted values of flue gas flow and NOx concentration with the actual operating values, perform a primary correction on the total ammonia supply amount according to the difference, and then perform a secondary correction on the total ammonia supply amount according to the hourly average values of NOx and ammonia slip actually discharged at the CEMS outlet to meet the hourly average environmental protection emission index.

[0024] As a preferred solution of the precise ammonia injection control method for coke oven flue gas denitrification according to the present invention, the ammonia injection amount is calculated in real time, and the upper and lower limits of the ammonia injection amount are set according to the on-site operation conditions. The primary correction is to compare the difference between the predicted value and the actual measured value. If the absolute value of the difference is greater than a certain limit value, the data input for the next prediction is adjusted. The secondary correction is only performed on the second half of each hour, and the correction amplitude is calculated with reference to the over-standard emission amplitude of the first half.

[0025] A precise ammonia injection control system for coke oven flue gas denitrification, which is used to implement the precise ammonia injection control method for coke oven flue gas denitrification according to any one of claims 1-5, is characterized in that it includes:

[0026] A data collection module, configured to: collect historical data and real-time data of coke oven flue gas flow and NOx concentration, and store them in the on-site database;

[0027] A model training and prediction module, configured to: use a time series prediction algorithm to model the historical data of flue gas flow and NOx concentration collected, and predict the NOx emission value in the next period of time;

[0028] A calculation and correction module, configured to: calculate the corresponding ammonia injection amount based on the NOx emission value predicted by the model, the system monitors the actual NOx emission value in real time, compares it with the predicted value, and if there is an obvious deviation, correct the prediction model in the next cycle;

[0029] A control module, configured to: transmit the calculated ammonia injection amount data to the main control system of denitrification, and the main control system of denitrification automatically adjusts the ammonia injection valve, and if a failure occurs in the precise ammonia injection system for denitrification, immediately trigger an alarm and automatically switch to the manual control mode.

[0030] As a preferred solution of the precise ammonia injection control system for coke oven flue gas denitrification according to the present invention, it further includes a kinetic catalytic reaction platform module, which is configured to: simulate according to the actual catalyst performance of the target denitrification device, combine the experimental data of the catalyst manufacturer and the operation data of the coking plant, and synthesize a multi-dimensional continuous reaction model platform through a neural network intelligent algorithm. This is beneficial to optimizing the operation of the precise ammonia injection prediction control system through this reaction model platform before production operation, and has a guiding role for actual production;

[0031] Among them, the data includes flue gas flow rate, denitrification reaction temperature, catalyst attenuation parameter, NOx value at the inlet of denitrification, NOx value at the outlet of denitrification, total ammonia supply, and ammonia escape amount in emissions.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: By combining the time series prediction algorithm, the present invention can predict the trend of NOx at the inlet of denitrification in advance, accurately adjust the ammonia injection amount, improve the denitrification efficiency and the utilization rate of ammonia. The kinetic catalytic reaction model platform is used for simulation operation to optimize the precise ammonia injection control system and improve its operation stability. The introduction of the correction logic enhances the response ability of the system to actual working condition changes, and the upper and lower limits of the ammonia injection amount ensure the safe operation of the system. At the same time, the design of automatic fault alarm and switching to the manual mode improves the reliability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below with reference to the drawings and specific embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts. Among them:

[0034] Figure 1 It is a flowchart of a precise ammonia injection control method for coke oven flue gas denitrification according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the drawings.

[0036] Secondly, the present invention is described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention in detail, for the sake of convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally out of the general proportion, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention here. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0037] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0038] The present invention provides a precise ammonia injection system and its control method for coke oven flue gas denitration, which can achieve accurate control of the total ammonia regulation of the denitration system, and improve the timeliness, accuracy, reliability, and stability of the ammonia injection system for denitration.

[0039] Figure 1 The following shows a flowchart of a precise ammonia injection control method for coke oven flue gas denitration according to the present invention. Please refer to Figure 1 A precise ammonia injection control method for coke oven flue gas denitration in this embodiment is as follows:

[0040] S1. Collect historical and real-time data of coke oven flue gas flow rate and NOx concentration, and store them in the on-site database. Among them, the NOx data is stored in the control system, and the data is collected through the modbus TCP protocol.

[0041] S2. Use the time series prediction algorithm to model the collected historical data of flue gas flow rate and NOx concentration, and predict the NOx emission value in the future period. Specifically, it includes:

[0042] Clean the NOx time series training data, remove outliers, and fill in missing values to ensure the accuracy and integrity of the data;

[0043] Resample the data according to a determined time period. Divide the data into a training set and a test set for model training and verification;

[0044] Use the TPR time series prediction algorithm to train the data, and verify the prediction error of the model on the test set. The error uses the MASE (Mean Absolute Scaled Error), and the calculation formula is as follows:

[0045]

[0046] Among them, Ft is the predicted value at time t, At is the actual value at time t, n is the number of prediction time points, and N is the total number of actual values;

[0047] According to the evaluation results, use all the training data for training to form a final model, and fix it as the model for actual NOx emission total prediction.

[0048] S3. Based on the NOx emission value predicted by the model, calculate the corresponding ammonia injection amount. The system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is an obvious deviation, the prediction model will be corrected in the next cycle. Specifically, compare the predicted values of flue gas flow and NOx concentration with the actual operating values, and make a primary correction to the total ammonia supply according to the difference. Then, based on the NOx actually emitted at the CEMS outlet and the hourly average value of ammonia slip, make a secondary correction to the total ammonia supply. More specifically: the ammonia injection amount is calculated in real time, and the upper and lower limits of the ammonia injection amount are set according to the on-site operating conditions. The primary correction is to compare the difference between the predicted value and the actual measured value. If the absolute value of the difference is greater than a certain limit, adjust the data input for the next prediction. The secondary correction is only carried out for the second half of each hour, and the correction amplitude is calculated with reference to the over-standard emission amplitude in the first half.

[0049] S4. Transmit the calculated ammonia injection amount data to the denitration main control system. The denitration main control system automatically adjusts the ammonia supply valve. And if the denitration precise ammonia injection system fails, an alarm will be triggered immediately and it will automatically switch to the manual control mode.

[0050] To implement the above method steps, the present invention also provides a precise ammonia injection control system applicable to coke oven flue gas denitration. The precise ammonia injection control system applicable to coke oven flue gas denitration specifically includes a data collection module, a model training and prediction module, a calculation and correction module, and a control module.

[0051] Among them, the data collection module is configured to: collect the historical data and real-time data of coke oven flue gas flow and NOx concentration, and store them in the on-site database;

[0052] The model training and prediction module is configured to: use the time series prediction algorithm to model the historical data of flue gas flow and NOx concentration collected, and predict the NOx emission value in the next period of time;

[0053] The calculation and correction module is configured to: based on the NOx emission value predicted by the model, calculate the corresponding ammonia injection amount. The system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is an obvious deviation, the prediction model will be corrected in the next cycle;

[0054] The control module is configured to: transmit the calculated ammonia injection amount data to the denitration main control system. The denitration main control system automatically adjusts the ammonia supply valve. And if the denitration precise ammonia injection system fails, an alarm will be triggered immediately and it will automatically switch to the manual control mode.

[0055] Further, it further includes a kinetic catalytic reaction platform module, which is configured to: simulate according to the actual catalyst performance of the target denitration device, combine the experimental data of the catalyst manufacturer and the operation data of the coking plant, and synthesize a multi-dimensional continuous reaction model through a neural network intelligent algorithm, where the data includes flue gas flow rate, denitration reaction temperature, catalyst attenuation parameter, NOx value at the denitration inlet, NOx value at the denitration outlet, total ammonia supply, and ammonia escape in emissions.

[0056] Although the present invention has been described above with reference to the embodiments, various improvements can be made to it and its components can be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed in the present invention can be combined with each other in any way, and the exhaustive description of these combinations is omitted in this specification only for the consideration of saving space and resources. Therefore, the present invention is not limited to the specific embodiments disclosed in the text, but includes all technical solutions falling within the scope of the claims.

Claims

1. A precise ammonia injection control method suitable for denitrification of coke oven flue gas, characterized in that: The steps include: S1. Collect historical and real-time data of coke oven flue gas flow and NOx concentration, and store them in the on-site database; S2. Use time series prediction algorithm to model the collected historical data of flue gas flow and NOx concentration to predict the NOx emission value in the future; S3. Based on the NOx emission value predicted by the model, the corresponding ammonia injection amount is calculated. The system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is a significant deviation, the prediction model will be corrected in the next cycle. S4. The calculated ammonia injection amount data is transmitted to the denitration main control system, which adjusts the ammonia supply valve. If the precise ammonia injection system fails, an alarm is triggered immediately and the system automatically switches to manual control mode. In step S3, the system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is a significant deviation, the specific steps for correcting the prediction model in the next cycle are as follows: During operation, the predicted values ​​of flue gas flow and NOx concentration are compared with the actual operating values, and the total ammonia supply is corrected once according to the difference. Then, the total ammonia supply is corrected twice according to the hourly average values ​​of NOx and ammonia escape actually emitted from the CEMS outlet. The ammonia injection amount is calculated in real time, and the upper and lower limits of the ammonia injection amount are set according to the on-site operating conditions. The first correction is made by comparing the difference between the predicted value and the actual measured value. If the absolute value of the difference is greater than a certain limit, the data input for the next prediction is adjusted. The second correction is only made for the second half of each hour, and the correction range is calculated based on the excess emission range in the first half.

2. The precise ammonia injection control method for coke oven flue gas denitrification according to claim 1, characterized in that: In step S1, the NOx data is stored in the DCS system, and the data is collected through the modbus TCP protocol.

3. The precise ammonia injection control method for coke oven flue gas denitrification according to claim 1, characterized in that: In step S2, the time series prediction algorithm is used to model the collected historical data of flue gas flow and NOx concentration. The specific steps for predicting the NOx emission value in the future are as follows: Clean the NOx time series training data, remove outliers, fill in missing values, and ensure the accuracy and completeness of the data; Resample the data according to a certain time period and divide the data into training set and test set for model training and verification; The TPR time series prediction algorithm is used to train the data, and the error of the model prediction is verified on the test set. The error uses the MASE mean absolute ratio error, and the calculation formula is as follows: Where Ft is the predicted value at time t, At is the actual value at time t, n is the number of predicted time points, and N is the total number of actual values; According to the evaluation results, all training data are used for training to form the final model, which is fixed as the model for actual total NOx emission prediction.

4. A precise ammonia injection control system suitable for coke oven flue gas denitrification, to realize the precise ammonia injection control method suitable for coke oven flue gas denitrification as claimed in any one of claims 1 to 3, characterized in that: include: The data collection module is configured to: collect historical data and real-time data of coke oven flue gas flow and NOx concentration, and store them in a field database; The model training prediction module is configured to: use the time series prediction algorithm to model the collected flue gas flow and NOx concentration historical data to predict the NOx emission value in the future; The calculation and correction module is configured to: calculate the corresponding ammonia injection amount based on the NOx emission value predicted by the model, and the system monitors the actual NOx emission value in real time and compares it with the predicted value. If there is a significant deviation, the prediction model will be corrected in the next cycle; The control module is configured to: transmit the calculated ammonia injection amount data to the denitrification main control system, and the denitrification main control system adjusts the ammonia supply valve. If the denitrification main control system fails, an alarm is immediately triggered and the system automatically switches to manual control mode.

5. The precise ammonia injection control system suitable for coke oven flue gas denitrification according to claim 4, characterized in that: It also includes a kinetic catalytic reaction platform module, which is configured to: simulate the actual catalyst performance of the target denitration device, combine the experimental data of the catalyst manufacturer and the operation data of the coking plant, and fit it into a multi-dimensional continuous reaction model platform through a neural network intelligent algorithm; Among them, the reaction model platform data includes flue gas flow, denitrification reaction temperature, catalyst attenuation parameters, denitrification inlet NOx value, denitrification outlet NOx value, total ammonia supply and ammonia escape emission.

Images