Dynamic correction method for deviation of wet desulphurization slurry pH value prediction model

Abstract

The invention discloses a dynamic deviation correction method for a wet desulphurization slurry pH value prediction model. According to the method, a model verification and model prediction parallel operation mode is adopted, and desulfurization performance calculation model dynamic deviation calculation and slurry pH value model prediction for correcting input parameters are carried out respectively. The model verification adopts a desulfurization performance calculation model to calculate the dynamic deviation of the desulfurization performance calculation model in real time; a slurry pH value calculation model is adopted for model prediction, SO2 data input into the model are dynamically corrected, and the control target of the slurry pH value is predicted in real time. According to the method, the desulfurization performance calculation model deviation is calculated in real time, and the input data of the slurry pH value prediction model is corrected, so that the problems of low precision and short timeliness of a wet desulfurization intelligent control static model can be effectively solved, and the wet desulfurization intelligent control level is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of environmental protection and relates to a dynamic deviation correction method for intelligent control of wet desulfurization. Background Art

[0002] The intelligent control system for wet flue gas desulfurization of limestone and gypsum uses the method of learning historical data modeling to predict the control target of the circulating slurry pH value in real time, which is used as the basis for adjusting the supply of limestone slurry to realize the model predictive control of the system. In the wet flue gas desulfurization industrial device system, due to various external factors such as the dynamic deviation characteristics of the measuring instrument itself, instrument maintenance and calibration, and the deviation of the desulfurization system slurry characteristics from the prediction model, the model calculation and the measured data deviate. In response to the above problems, the conventional treatment method is to statically correct the model or regenerate a new model. However, this technical means has the significant disadvantages of low accuracy and short timeliness. External disturbances of the system state and measuring instruments or deviations from the model establishment conditions will cause the model to be unusable. Based on this, the industry has been exploring the use of dynamic models to solve this problem. The dynamic model has high requirements for data cleaning and filtering, and any disturbance of the measuring instrument will be learned by the model, so the applicability of the dynamic model is limited.

[0003] Since the pH value of the slurry has a significant impact on the SO2 concentration at the desulfurization outlet, the direct deviation correction of the pH value of the slurry calculated by the model will lead to a large fluctuation of the SO2 concentration at the outlet of the intelligent control. Therefore, the reasonable deviation correction of the desulfurization slurry pH prediction model has a decisive influence on the operation effect of the intelligent system.

[0004] Based on the above situation, the present invention proposes a method for dynamically correcting the deviation of the wet desulfurization slurry pH value prediction model to solve the problems of low accuracy and short timeliness of the static model of wet desulfurization intelligent control and improve the level of wet desulfurization intelligent control. Summary of the invention

[0005] The technical problem to be solved by the present invention is the low accuracy and short timeliness of the static model of wet flue gas desulfurization intelligent control.

[0006] In order to achieve the above-mentioned purpose, the present invention proposes a method for dynamically correcting the deviation of the prediction model of pH value of wet desulfurization slurry. The method adopts a mode of parallel operation of model verification and model prediction to respectively calculate the dynamic deviation of the desulfurization performance calculation model and correct the slurry pH value model prediction of the input parameters.

[0007] The model verification method is to use the desulfurization performance calculation model to calculate the deviation between the outlet SO2 concentration of the desulfurization system and the measured value in real time, which is recorded as the dynamic deviation of the desulfurization performance calculation model. The flue gas volume data input to the desulfurization performance calculation model is the current real-time operation data of the desulfurization system. The unit of the dynamic deviation of the desulfurization performance calculation model is ppm. The dynamic deviation limit of the model verification is set. When the dynamic deviation between the outlet SO2 concentration of the desulfurization system and the measured value output by the desulfurization performance calculation model is greater than the set deviation limit, it is prompted that the desulfurization performance calculation model needs to be updated.

[0008] The model prediction method is to use the slurry pH value calculation model to dynamically correct the SO2 data input to the model and predict the control target of the slurry pH value in real time. The flue gas volume data input to the slurry pH value calculation model is the flue gas volume predicted according to the unit load instruction. The dynamic correction method of the model input SO2 data is to calculate the difference between the SO2 concentration setting value at the outlet of the desulfurization system and the dynamic deviation of the SO2 concentration model in real time. The SO2 concentration setting value at the outlet of the desulfurization system is in ppm.

[0009] In the calculation process of the slurry pH value calculation model, the pH value range is controlled to be ±0.2 of the current slurry pH measured value. The slurry pH value calculation model is the reverse process of the desulfurization performance calculation model. DETAILED DESCRIPTION

[0010] The present invention relates to a method for dynamically correcting the deviation of a wet desulfurization slurry pH value prediction model. The method adopts a mode in which model verification and model prediction are run in parallel, and respectively performs dynamic deviation calculation of a desulfurization performance calculation model and slurry pH value model prediction of corrected input parameters. The specific steps are as follows:

[0011] 1. Calculate the dynamic deviation of the desulfurization performance calculation model. In the model verification mode, the desulfurization performance calculation model is used to calculate the dynamic deviation in real time. The model verification method is to use the desulfurization performance calculation model to calculate in real time the deviation between the SO2 concentration at the outlet of the desulfurization system and the measured value, that is, the dynamic deviation of the desulfurization performance calculation model. The flue gas volume data input to the desulfurization performance calculation model is the current real-time operation data of the desulfurization system. The unit of the dynamic deviation of the desulfurization performance calculation model is ppm. Set the dynamic deviation limit of the model verification. When the dynamic deviation between the SO2 concentration at the outlet of the desulfurization system output by the desulfurization performance calculation model and the measured value is greater than the set deviation limit, it is prompted that the desulfurization performance calculation model needs to be updated.

[0012] 2. Real-time deviation correction and model prediction. In the model prediction mode, the slurry pH value calculation model is used to dynamically correct the SO2 data input to the model, and the control target of the slurry pH value is predicted in real time. The flue gas volume data input to the slurry pH value calculation model is the flue gas volume predicted according to the unit load instruction, and the input SO2 concentration data is the difference between the SO2 concentration setting value at the outlet of the intelligent desulfurization system converted to the value under the actual oxygen concentration conditions and the dynamic deviation of the SO2 concentration. During the calculation process of the slurry pH value calculation model, the control pH value range is the current slurry measured pH ± 0.2.

[0013] 3. During the operation of the desulfurization system, the dynamic deviation of the desulfurization performance calculation model in step 1 and the real-time correction of the deviation in step 2 are performed simultaneously, and the predicted value of the control target of the slurry pH value is output in real time.

Claims

1. A method for dynamically correcting the deviation of a wet flue gas desulfurization slurry pH prediction model, characterized in that: The model verification and model prediction are operated in parallel to perform real-time dynamic deviation calculation of the desulfurization performance calculation model and slurry pH value model prediction of the corrected input parameters. During the model verification process, a model verification dynamic deviation limit is set. When the dynamic deviation is greater than the set deviation limit, it is prompted that the desulfurization performance calculation model needs to be updated.

2. The model verification as described in claim 1, wherein the method is to use a desulfurization performance calculation model to calculate in real time the deviation of the SO2 concentration at the outlet of the desulfurization system and the actual measured value, which is recorded as the dynamic deviation of the desulfurization performance calculation model; during the model verification process, the flue gas volume data input to the desulfurization performance calculation model is the current real-time operation data of the desulfurization system.

3. The dynamic deviation of the desulfurization performance calculation model as claimed in claim 2, wherein the unit is ppm.

4. The model prediction as described in claim 1, wherein the method is to use a slurry pH value calculation model, dynamically correct the model input SO2 data, and predict the control target of the slurry pH value in real time.

5. The slurry pH value calculation model as described in claim 4, wherein the flue gas volume data input is the flue gas volume predicted according to the unit load instruction.

6. The correction input parameter as claimed in claim 1, wherein the method is to calculate in real time the difference between the set value of the SO2 concentration at the outlet of the desulfurization system and the dynamic deviation of the SO2 concentration model.