Sulfur dioxide concentration prediction and model optimization method, device, equipment and medium

Abstract

The invention discloses a sulfur dioxide concentration prediction and model optimization method, device, equipment and medium, and the sulfur dioxide concentration prediction model optimization method comprises the steps: obtaining original desulfurization sample data from historical desulfurization record data, the original desulfurization sample data comprises original influence concentration factors at all moments in a preset time period and corresponding original sulfur dioxide concentrations, and determining response lag moments of the original influence concentration factors at the starting moment according to correlation coefficients of the original influence concentration factors at the starting moment and the original sulfur dioxide concentrations at all moments in the preset time period; replacing the original influence concentration factor at the response lag moment with the original influence concentration factor at the starting moment, and replacing the original influence concentration factor at each moment after the starting moment with the original influence concentration factor at each moment after the response lag moment; and creating target desulfurization sample data according to the replacement result to train the initial desulfurization prediction model to obtain a target desulfurization prediction model.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer technology, and in particular to a method, device, equipment and medium for predicting sulfur dioxide concentration and optimizing a model. Background Art

[0002] With the acceleration of industrialization and the improvement of environmental protection awareness, the monitoring and control of air pollutant emissions has become a global focus. 2 ) is one of the common air pollutants, and the accurate prediction of its concentration is of great significance for formulating effective emission reduction measures, optimizing industrial production processes and protecting human health.

[0003] However, the existing sulfur dioxide concentration prediction model only considers the processing of data noise, outliers and other problems in the training data set, but does not consider the time delay between the input data set and the output data set, resulting in insufficient accuracy in the sulfur dioxide concentration prediction.

[0004] Therefore, it is urgent to propose a new method to solve the above problems. Summary of the invention

[0005] The present invention provides a sulfur dioxide concentration prediction and model optimization method, device, equipment and medium, which can improve the accuracy of sulfur dioxide concentration prediction.

[0006] In a first aspect, an embodiment of the present invention provides a method for optimizing a sulfur dioxide concentration prediction model, comprising:

[0007] Obtaining desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain original desulfurization sample data, wherein the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculating the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment;

[0008] Determining the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time;

[0009] The original influencing concentration factor at the response lag moment is replaced by the original influencing concentration factor at the starting moment, and the original influencing concentration factor at each moment after the response lag moment is replaced by the original influencing concentration factor at each moment after the starting moment;

[0010] Target desulfurization sample data is created according to the replacement result, and the target desulfurization sample data is used to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0011] An embodiment of the present invention provides a method for optimizing a sulfur dioxide concentration prediction model, comprising: acquiring desulfurization record data of a preset time period from historical desulfurization record data of a wet desulfurization system to obtain original desulfurization sample data, the original desulfurization sample data including original influencing concentration factors and corresponding original sulfur dioxide concentrations at each moment within the preset time period, and calculating the correlation coefficient between the original influencing concentration factors at the starting moment within the preset time period and the original sulfur dioxide concentrations at each moment; determining the response lag time of the original influencing concentration factors at the starting moment according to the correlation coefficient between the original influencing concentration factors at the starting moment and the original sulfur dioxide concentrations at each moment; replacing the original influencing concentration factors at the response lag moment with the original influencing concentration factors at the starting moment, and replacing the original influencing concentration factors at each moment after the response lag moment with the original influencing concentration factors at each moment after the starting moment; creating target desulfurization sample data according to the replacement result, and using the target desulfurization sample data to train the initial desulfurization prediction model to obtain a target desulfurization prediction model. The above technical solution first obtains the desulfurization record data of the preset period from the historical desulfurization record data of the wet desulfurization system to obtain the original desulfurization sample data, and calculates the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time within the preset period, which provides a data basis for determining the response lag time of the original influencing concentration factor at the starting time. Then, according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time, the response lag time of the original influencing concentration factor at the starting time is determined. Then, the original influencing concentration factor at the response lag time is replaced with the original influencing concentration factor at the starting time, and the original influencing concentration factor at each time after the response lag time is replaced with the original influencing concentration factor at each time after the starting time, which provides a data basis for creating the target desulfurization sample data later. Finally, the target desulfurization sample data is created according to the replacement result, and the interference factors caused by time delay in the original data are removed, so that the target desulfurization sample data can more accurately reflect the relationship between the influencing concentration factor and the sulfur dioxide concentration. The target desulfurization sample data is then used to train the initial desulfurization prediction model, thereby obtaining a target desulfurization prediction model with higher prediction accuracy, thereby improving the accuracy of predicting sulfur dioxide concentration. Therefore, the present application can solve the problem that the existing sulfur dioxide concentration prediction model only considers the processing of data noise, outliers and other issues in the training data set, but does not consider the time delay between the input data set and the output data set, resulting in insufficient accuracy in sulfur dioxide concentration prediction.

[0012] In a second aspect, an embodiment of the present invention further provides a method for predicting sulfur dioxide concentration, comprising:

[0013] Obtain the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment;

[0014] The real-time concentration influencing factors are input into a target desulfurization prediction model obtained by using the sulfur dioxide concentration prediction model optimization method described in any one of the first aspects to obtain the predicted sulfur dioxide concentration at the current moment.

[0015] In a third aspect, an embodiment of the present invention further provides a sulfur dioxide concentration prediction model optimization device, the device comprising:

[0016] An acquisition module is used to acquire desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain original desulfurization sample data, wherein the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculate the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment;

[0017] A determination module, used to determine the response lag time of the original influencing concentration factor at the starting moment according to the correlation coefficient between the original influencing concentration factor at the starting moment and the original sulfur dioxide concentration at each moment;

[0018] A replacement module, used to replace the original influencing concentration factors at the response lag moment with the original influencing concentration factors at the starting moment, and to replace the original influencing concentration factors at each moment after the response lag moment with the original influencing concentration factors at each moment after the starting moment;

[0019] The training module is used to create target desulfurization sample data according to the replacement result, and use the target desulfurization sample data to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0020] In a fourth aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the sulfur dioxide concentration prediction model optimization method as described in any one of the first aspects when executing the program, or implements the sulfur dioxide concentration prediction method as described in the second aspect when executing the program.

[0021] In a fifth aspect, an embodiment of the present invention further provides a storage medium comprising computer executable instructions, on which a computer program is stored, characterized in that when the program is executed by a processor, the sulfur dioxide concentration prediction model optimization method as described in any one of the first aspects is implemented, or when the program is executed by a processor, the sulfur dioxide concentration prediction method as described in the second aspect is implemented.

[0022] It should be noted that the above-mentioned computer instructions may be stored in whole or in part on a computer-readable storage medium. The computer-readable storage medium may be packaged together with the processor of the sulfur dioxide concentration prediction model optimization device, or may be packaged separately from the processor of the sulfur dioxide concentration prediction model optimization device, and this application does not limit this.

[0023] In this application, the name of the above-mentioned sulfur dioxide concentration prediction model optimization device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they belong to the scope of the claims of this application and their equivalent technologies.

[0024] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A flowchart of a sulfur dioxide concentration prediction model optimization method provided by an embodiment of the present invention;

[0027] Figure 2 A flowchart of another sulfur dioxide concentration prediction model optimization method provided by an embodiment of the present invention;

[0028] Figure 3 A schematic diagram of the structure of a method for predicting sulfur dioxide concentration provided by an embodiment of the present invention;

[0029] Figure 4 A schematic diagram of the structure of a sulfur dioxide concentration prediction model optimization device provided by an embodiment of the present invention;

[0030] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0032] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0033] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0034] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0035] It should be mentioned before discussing exemplary embodiments in more detail that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe various operations (or steps) as sequential processes, many operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of various operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the accompanying drawings. The process can correspond to methods, functions, procedures, subroutines, subprograms, etc. In addition, the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict.

[0036] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0037] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0038] Figure 1A flowchart of a sulfur dioxide concentration prediction model optimization method provided by an embodiment of the present invention. This embodiment can be applied to situations where the prediction accuracy of the sulfur dioxide concentration prediction model needs to be improved. The method can be executed by a sulfur dioxide concentration prediction model optimization device, which can be implemented in software and / or hardware. Exemplarily, the device can be a computer or a server. Figure 1 The sulfur dioxide concentration prediction model optimization method of this embodiment specifically includes the following steps:

[0039] Step 110, obtaining desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system, obtaining original desulfurization sample data, and calculating the correlation coefficient between the original influencing concentration factor at the starting time of the preset period and the original sulfur dioxide concentration at each time.

[0040] Specifically, a wet flue gas desulfurization system refers to a system that uses a liquid absorbent (such as an alkaline solution) to remove sulfides such as sulfur dioxide from flue gas. Historical desulfurization record data refers to data records generated during the operation of a wet flue gas desulfurization system over a period of time (such as one year). A preset period refers to a specific period of time set in advance according to actual conditions or needs, and is used to obtain original desulfurization sample data, for example: the preset period is from 16:00 on October 1, 2024 to 17:00 on October 1, 2024. Original desulfurization sample data refers to data within a preset period of time selected from historical desulfurization record data, including the original influencing concentration factors at each moment in the preset period of time and the corresponding original sulfur dioxide concentration. The original influencing concentration factors refer to factors that affect the sulfur dioxide concentration in a wet flue gas desulfurization system, for example: the original influencing concentration factors may be original flue gas flow, limestone slurry density, unit load, main steam pressure and / or coal consumption, etc. The original sulfur dioxide concentration refers to the actual measured sulfur dioxide concentration in the flue gas discharged from industrial production equipment (such as thermal power generation units) after desulfurization. The correlation coefficient refers to an indicator used to measure the correlation between the original influencing concentration factors and the original sulfur dioxide concentration.

[0041] In the specific implementation, first, a database query statement can be used to filter out all desulfurization record data that meet the preset time period from the database storing the historical desulfurization record data of the wet desulfurization system to obtain the original desulfurization sample data. Then, according to the actual situation or application requirements, a suitable correlation coefficient calculation method can be selected to calculate the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at each time. For example, the correlation coefficient calculation method can be the maximum information coefficient, Pearson correlation coefficient or Spearman rank correlation coefficient. Then, the selected correlation coefficient calculation method is used to calculate the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at each time.

[0042] Exemplarily, if the original influencing concentration factor at time t1 within the preset time period is x1, the original influencing concentration factor at time t2 is x2, the original influencing concentration factor at time t3 is x3, the original sulfur dioxide concentration at time t1 within the preset time period is y1, the original sulfur dioxide concentration at time t2 within the preset time period is y2, and the original sulfur dioxide concentration at time t3 within the preset time period is y3, the correlation coefficient calculation method is the Pearson correlation coefficient, and time t1 is the starting time, then the mean value of the original influencing concentration factor in the preset time period A=(x1+x2+x3) / 3, the mean value of the original sulfur dioxide concentration in the preset time period B=(y1+y2+y3) / 3, the correlation coefficient between the original influencing concentration factor at time t1 and the original sulfur dioxide concentration at time t1 is The correlation coefficient between the original influencing concentration factor at time t1 and the original sulfur dioxide concentration at time t2 is The correlation coefficient between the original influencing concentration factor at time t1 and the original sulfur dioxide concentration at time t3 is

[0043] In this embodiment, by obtaining the original desulfurization sample data and calculating the correlation coefficient between the original influencing concentration factor at the starting time within the preset time period and the original sulfur dioxide concentration at each time, a data basis is provided for subsequently determining the response lag time of the original influencing concentration factor at the starting time.

[0044] Step 120: Determine the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time.

[0045] Specifically, in a wet flue gas desulfurization system, changes in various factors affecting concentration (such as flue gas temperature, flue gas flow, etc.) are not immediately reflected in changes in sulfur dioxide concentration, but there is a certain time delay. Therefore, the response lag time refers to the time corresponding to when the original influencing concentration factors at the starting time have an impact on the sulfur dioxide concentration.

[0046] In a specific implementation, after obtaining the correlation coefficients between the original influencing concentration factors at the starting time within a preset time period and the original sulfur dioxide concentrations at various times, these correlation coefficients can be sorted. Then, the moment of the original sulfur dioxide concentration corresponding to the maximum correlation coefficient is found from the sorted correlation coefficients. The moment corresponding to this maximum correlation coefficient means that the linear relationship between the original influencing concentration factors at the starting time and the sulfur dioxide concentration is the closest at this moment, that is, the original influencing concentration factors at the starting time have the most significant effect on the sulfur dioxide concentration. Therefore, the moment of the original sulfur dioxide concentration corresponding to this maximum correlation coefficient can be determined as the response lag moment of the original influencing concentration factors at the starting time.

[0047] Step 130: Replace the original influencing concentration factors at the response lag time with the original influencing concentration factors at the start time, and replace the original influencing concentration factors at each time after the response lag time with the original influencing concentration factors at each time after the start time.

[0048] In the specific implementation, after determining the response lag time of the original influencing concentration factor at the starting time, the replacement operation of the original influencing concentration factor is performed. Specifically, the original influencing concentration factor at the response lag time is first replaced with the original influencing concentration factor at the starting time. For example, if the original influencing concentration factor at the response lag time is A1 and the original influencing concentration factor at the starting time is B1, then A1 is replaced with B1. Then, for each moment after the response lag time, the original influencing concentration factor at each moment after the response lag time is replaced with the original influencing concentration factor at each moment after the starting time.

[0049] For example, if the original desulfurization sample data is as shown in Table 1, and the response lag time is T3, the original influencing concentration factors at the response lag time are replaced with the original influencing concentration factors at the starting time, and the original influencing concentration factors at each time after the response lag time are replaced with the original influencing concentration factors at each time after the starting time, as shown in Table 2.

[0050] Table 1 Example table of original desulfurization sample data 1

[0051] time Original influence concentration factor Original sulfur dioxide concentration T1 X1 Y1 T2 X2 Y2 T3 X3 Y3 T4 X4 Y4 T5 X5 Y5 T6 X6 Y6

[0052] Table 2 Example of desulfurization sample data after replacement

[0053]

[0054]

[0055] In this embodiment, by operating the original influencing concentration factors at the response lag moment and each moment thereafter, a data basis is provided for subsequently creating target desulfurization sample data.

[0056] Step 140: Create target desulfurization sample data according to the replacement result, and use the target desulfurization sample data to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0057] Specifically, the target desulfurization sample data refers to a data set used to train the desulfurization prediction model determined according to the replacement result. The initial desulfurization prediction model refers to the desulfurization prediction model before training. The target desulfurization prediction model is a model with higher sulfur dioxide concentration prediction accuracy obtained after training the initial desulfurization prediction model using the target desulfurization sample data.

[0058] In a specific implementation, the target desulfurization sample data can be created based on the original influencing concentration factors replaced at each moment and the corresponding original sulfur dioxide concentration. Then the target desulfurization sample data is used as training data to train the initial desulfurization prediction model. This process usually includes two steps: forward propagation and back propagation. Forward propagation is to obtain the prediction result through the model of the input data, and then calculate the loss between the prediction result and the true target. Back propagation is to update the parameters of the model according to the loss function to reduce the gap between the prediction result and the true target. The initial desulfurization prediction model can be CornerNet, FCOS, RetinaNet, YolV7, etc., and the embodiment of the present invention does not limit this. Finally, the model can be optimized based on the back propagation algorithm until the loss function converges, thereby obtaining the target desulfurization prediction model. For example: the gradient descent method can be selected to determine hyperparameters such as the learning rate in order to optimize the model.

[0059] Exemplarily, if the replacement result is as shown in the aforementioned Table 2, the target desulfurization sample data can be created based on the original influencing concentration factors replaced at each moment and the corresponding original sulfur dioxide concentration. The target desulfurization sample data is as shown in Table 3.

[0060] Table 3 Example table of target desulfurization sample data 1

[0061] time Original influence concentration factor Original sulfur dioxide concentration T3 X1 Y3 T4 X2 Y4 T5 X3 Y5 T6 X4 Y6

[0062] In this embodiment, target desulfurization sample data is created based on the replacement result, and the interference factors caused by time delay in the original data are removed, so that the target desulfurization sample data can more accurately reflect the relationship between the factors affecting the concentration and the sulfur dioxide concentration. The target desulfurization sample data is then used to train the initial desulfurization prediction model, thereby obtaining a target desulfurization prediction model with higher prediction accuracy, thereby improving the accuracy of predicting sulfur dioxide concentration.

[0063] In the embodiment of the present invention, the desulfurization record data of the preset period is first obtained from the historical desulfurization record data of the wet desulfurization system to obtain the original desulfurization sample data, and the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time in the preset period is calculated, which provides a data basis for determining the response lag time of the original influencing concentration factor at the starting time. Then, according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time, the response lag time of the original influencing concentration factor at the starting time is determined. Then, the original influencing concentration factor at the response lag time is replaced with the original influencing concentration factor at the starting time, and the original influencing concentration factor at each time after the response lag time is replaced with the original influencing concentration factor at each time after the starting time, which provides a data basis for creating the target desulfurization sample data. Finally, the target desulfurization sample data is created according to the replacement result, and the interference factors caused by time delay in the original data are removed, so that the target desulfurization sample data can more accurately reflect the relationship between the influencing concentration factor and the sulfur dioxide concentration. The target desulfurization sample data is then used to train the initial desulfurization prediction model, thereby obtaining a target desulfurization prediction model with higher prediction accuracy, thereby improving the accuracy of predicting sulfur dioxide concentration. Therefore, the present application can solve the problem that the existing sulfur dioxide concentration prediction model only considers the processing of data noise, outliers and other issues in the training data set, but does not consider the time delay between the input data set and the output data set, resulting in insufficient accuracy in sulfur dioxide concentration prediction.

[0064] Figure 2 A flowchart of another sulfur dioxide concentration prediction model optimization method provided by an embodiment of the present invention is a specific embodiment based on the above embodiment. In this embodiment, the method may also include:

[0065] Step 201: Obtain desulfurization record data of a preset period from historical desulfurization record data of a wet desulfurization system, and calculate the correlation coefficient between the original concentration influencing factor at the starting time of the preset period and the original sulfur dioxide concentration at each time.

[0066] Step 202: Determine whether the maximum correlation coefficient is greater than a first correlation threshold.

[0067] If it is greater, execute step 204; if it is not greater, execute step 203.

[0068] Specifically, the first correlation threshold refers to a correlation coefficient threshold set according to actual conditions or requirements for determining the original desulfurization sample data. For example, the first correlation threshold may be 0.5.

[0069] In a specific implementation, after calculating the correlation coefficients between the original influencing concentration factors at the start time of the preset time period and the original sulfur dioxide concentration at each time, these correlation coefficients can be sorted first. Then, the maximum correlation coefficient is found from the sorted correlation coefficients, and then it is determined whether the maximum correlation coefficient is greater than the first correlation threshold. If it is greater than, it means that there is a close correlation between the original influencing concentration factors and the original sulfur dioxide concentration in the current preset time period. At this time, the desulfurization record data of the preset time period can be determined as the original desulfurization sample data. If it is not greater than, it means that in the current preset time period, the correlation between the original influencing concentration factors and the original sulfur dioxide concentration is weak and cannot meet the requirements as the original desulfurization sample data. At this time, the preset time period can be updated according to the preset time update method to obtain the updated preset time period.

[0070] It should be noted that if there are multiple original factors affecting concentration, the maximum correlation coefficient corresponding to each original factor affecting concentration must be determined to be greater than the first correlation threshold. When the maximum correlation coefficient corresponding to each original factor affecting concentration is greater than the first correlation threshold, the desulfurization record data of the preset period is determined to be the original desulfurization sample data.

[0071] In this embodiment, by determining whether the maximum correlation coefficient is greater than the first correlation threshold, the quality of the determined original desulfurization sample data can be improved.

[0072] Step 203: Update the preset time period according to the preset time update method to obtain an updated preset time period.

[0073] Specifically, the preset time updating method refers to a method of updating the preset time period determined according to actual conditions or needs.

[0074] In a specific implementation, after determining that the maximum correlation coefficient is not greater than the first correlation threshold, the preset time period can be updated according to the preset time update method to obtain an updated preset time period. For example, the preset time update method can be to add 1 hour to the preset time period. The preset time period is from 11:00 on January 1, 2024 to 12:00 on January 1, 2024. The updated preset time period is from 12:00 on January 1, 2024 to 13:00 on January 1, 2024.

[0075] After executing step 203 , the preset period may be replaced based on the updated preset period, and the process returns to executing step 201 .

[0076] In this embodiment, by updating the preset time period, it is helpful to find data with a high correlation between the original concentration influencing factors and the original sulfur dioxide concentration, thereby helping to determine the original desulfurization sample data and improving the quality of the original desulfurization sample data.

[0077] Step 204: Determine the desulfurization record data for a preset period of time as the original desulfurization sample data.

[0078] In a specific implementation, after determining that the maximum correlation coefficient is greater than the first correlation threshold, the desulfurization record data of the preset time period can be directly determined as the original desulfurization sample data.

[0079] In addition, after obtaining the original desulfurization sample data, the outlier processing method can be used to process the original desulfurization sample number. For example, the outlier processing method can be a box plot method, Laida criterion, Z-score method, local outlier factor method or Hampel filter. For example, if the outlier processing method is a Hampel filter, the median and median absolute deviation of the original sulfur dioxide concentration are first calculated, and then the difference between the original sulfur dioxide concentration and the median at each moment is compared to see if it exceeds the preset threshold (such as 4 times the median absolute deviation). Then, the original sulfur dioxide concentration at each moment that exceeds the preset threshold is replaced with the median of the original sulfur dioxide concentration, and the rest of the data remains unchanged. And the original influencing concentration factors are processed according to the same process.

[0080] Step 205: for the original sulfur dioxide concentration at the current moment, determine the mutual information between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment according to the occurrence probability of the original sulfur dioxide concentration at the current moment, the occurrence probability of the original influencing concentration factor at the starting moment of the preset time period, and the probability of the simultaneous occurrence of the original sulfur dioxide concentration at the current moment and the original influencing concentration factor at the starting moment of the preset time period.

[0081] Specifically, mutual information is an important concept in information theory, which is used to measure the correlation or mutual dependence between two random variables. In this embodiment, mutual information is used to measure the correlation between the original influencing concentration factor at the start time of the preset time period and the original sulfur dioxide concentration at the current time.

[0082] In a specific implementation, for the original sulfur dioxide concentration at the current moment, the probability of occurrence of the original sulfur dioxide concentration at the current moment can be obtained by calculating the ratio of the number of factors in the original desulfurization sample data that are equal to the original sulfur dioxide concentration at the current moment to the total number of factors in the original desulfurization sample data. Then, the probability of occurrence of the original factors influencing the concentration at the starting moment of the preset period is obtained by calculating the ratio of the number of factors in the original desulfurization sample data that are equal to the original factors influencing the concentration at the starting moment of the preset period to the total number of factors in the original desulfurization sample data. Then, the probability of occurrence of the original factors influencing the concentration at the starting moment of the preset period is obtained by calculating the ratio of the number of factors in the original desulfurization sample data that are equal to the original factors influencing the concentration at the starting moment of the preset period and the original sulfur dioxide concentration at the current moment to the total number of factors influencing the concentration in the original desulfurization sample data. Afterwards, the mutual information between the original influencing concentration factor at the starting time of the preset period and the original sulfur dioxide concentration at the current time can be determined based on the occurrence probability of the original sulfur dioxide concentration at the current time, the occurrence probability of the original influencing concentration factor at the starting time of the preset period, and the probability of the original sulfur dioxide concentration at the current time and the original influencing concentration factor at the starting time of the preset period appearing at the same time. The specific calculation formula is as follows:

[0083]

[0084] Where I represents the mutual information between the original influencing concentration factor at the starting time of the preset period and the original sulfur dioxide concentration at the current time, P 1 represents the probability of occurrence of the original sulfur dioxide concentration at the current moment, P 2 Represents the probability of occurrence of the original influencing concentration factor at the starting time within the preset period, P 3 It indicates the probability of the original sulfur dioxide concentration at the current moment and the original influencing concentration factor at the starting moment of the preset time period occurring at the same time.

[0085] In this embodiment, by determining the mutual information between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment, a data basis is provided for subsequently determining the correlation coefficient between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment.

[0086] Step 206, calculate the ratio of the mutual information of the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at the current time to the preset normalization parameter, and obtain the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at the current time.

[0087] Specifically, the preset normalization parameter refers to a parameter that is preset according to actual conditions or requirements and is used to normalize the mutual information so that the calculated correlation coefficients are comparable.

[0088] In a specific implementation, the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at the current time can be obtained by calculating the ratio of the mutual information between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at the current time and the preset normalization parameter. In practical applications, the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at the current time can be determined as = the mutual information between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at the current time / the preset normalization parameter.

[0089] In this embodiment, by obtaining the correlation coefficient between the original influencing concentration factor at the starting time within the preset time period and the original sulfur dioxide concentration at the current time, a data basis is provided for subsequently determining the response lag time of the original influencing concentration factor at the starting time.

[0090] Step 207: Determine the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time.

[0091] Furthermore, step 207 may specifically include: determining the time of the original sulfur dioxide concentration corresponding to the maximum correlation coefficient as the response lag time of the original influencing concentration factor at the starting time.

[0092] In a specific implementation, after obtaining the correlation coefficient between the original influencing concentration factor at the starting moment and the original sulfur dioxide concentration at each moment, all the calculated correlation coefficients can be compared to determine the maximum value, that is, the maximum correlation coefficient. Then, the moment of the original sulfur dioxide concentration corresponding to the maximum correlation coefficient can be determined as the response lag moment of the original influencing concentration factor at the starting moment.

[0093] In this embodiment, the moment of the original sulfur dioxide concentration corresponding to the maximum correlation coefficient is determined as the response lag moment of the original influencing concentration factor at the starting moment, which helps to reduce the error caused by time lag and helps improve the accuracy and reliability of the target desulfurization sample data.

[0094] Further, step 207 may specifically include: determining a preset influencing concentration factor from each original influencing concentration factor at the starting time, and determining a response lag time of the preset influencing concentration factor at the starting time.

[0095] Specifically, the preset influencing concentration factor is the influencing concentration factor whose maximum correlation coefficient with the original sulfur dioxide concentration at each moment is greater than the second correlation threshold, and the original influencing concentration factors at the starting moment include at least two. The second correlation threshold is a correlation coefficient threshold for determining the preset influencing concentration factor set according to actual conditions or needs, for example, the second correlation threshold may be 0.8.

[0096] In a specific implementation, the largest of the correlation coefficients between the original influencing concentration factors at the starting time and the original sulfur dioxide concentration at each time is greater than the second correlation threshold, and is determined as the preset influencing concentration factor. Then, the moment corresponding to the largest correlation coefficient in each preset influencing concentration factor is determined as the response lag moment of the preset influencing concentration factor at the starting time.

[0097] Exemplarily, if the preset influencing concentration factors are X and V, the original sulfur dioxide concentration is Y, the time corresponding to the maximum of the correlation coefficients between X at the starting time and Y at each time is T3, and the time corresponding to the maximum of the correlation coefficients between V at the starting time and Y at each time is T5, then the response lag time of X at the starting time is T3, and the response lag time of V at the starting time is T5.

[0098] In this embodiment, through the above steps, the accuracy and reliability of the target desulfurization sample data can be improved.

[0099] Step 208: Replace the original influencing concentration factors at the response lag time with the original influencing concentration factors at the start time, and replace the original influencing concentration factors at each time after the response lag time with the original influencing concentration factors at each time after the start time.

[0100] Step 209: Create target desulfurization sample data according to the replacement result, and use the target desulfurization sample data to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0101] Further, target desulfurization sample data is created according to the replacement result, including: determining a candidate moment from the response lag moment and each moment after the response lag moment, and creating target desulfurization sample data according to the original sulfur dioxide concentration at the candidate moment and the corresponding current influencing concentration factor.

[0102] Specifically, the candidate time is the time at which all corresponding original factors affecting the concentration are replaced.

[0103] In the specific implementation, starting from the response lag moment, each subsequent moment is checked in turn. For each checked moment, it is determined whether the moment is the moment when all the corresponding original influencing concentration factors are replaced. If so, the moment is determined as a candidate moment. Then the original sulfur dioxide concentration at the candidate moment and the corresponding current influencing concentration factor are combined to form the target desulfurization sample data.

[0104] Exemplarily, if the original desulfurization sample data 2 is as shown in Table 4, wherein U and P are the original influencing concentration factors, Y is the original sulfur dioxide concentration, the starting time is T1, the response lag time of U at the starting time is T2, and the response lag time of P at the starting time is T4, then the result after replacement is as shown in Table 5, and the target desulfurization sample data 2 is as shown in Table 6.

[0105] Table 4 Example table of original desulfurization sample data 2

[0106] time U P Y T1 U1 P1 Y1 T2 U2 P2 Y2 T3 U3 P3 Y3 T4 U4 P4 Y4 T5 U5 P5 Y5 T6 U6 P6 Y6

[0107] Table 5 Example table of desulfurization sample data 2 after replacement

[0108] time U P Y T1 U1 P1 Y1 T2 U1 P2 Y2 T3 U2 P3 Y3 T4 U3 P1 Y4 T5 U4 P2 Y5 T6 U5 P3 Y6

[0109] Table 6 Example table of target desulfurization sample data 2

[0110] time U P Y T4 U3 P1 Y4 T5 U4 P2 Y5 T6 U5 P3 Y6

[0111] In this embodiment, through the above steps, the interference factors caused by time delay in the original data are removed, ensuring that the target desulfurization sample data can more accurately reflect the relationship between each concentration influencing factor and sulfur dioxide concentration.

[0112] In the embodiment of the present invention, the desulfurization record data of the preset period is first obtained from the historical desulfurization record data of the wet desulfurization system, and the correlation coefficient between the original influencing concentration factor at the starting time of the preset period and the original sulfur dioxide concentration at each time is calculated. Then, it is determined whether the maximum correlation coefficient is greater than the first correlation threshold value. If it is not greater than, it means that in the current preset period, the correlation between the original influencing concentration factor and the original sulfur dioxide concentration is weak and cannot meet the requirements of being the original desulfurization sample data. At this time, the preset period can be updated according to the preset time update method to obtain the updated preset period, and return to execute the steps of obtaining the desulfurization record data of the preset period from the historical desulfurization record data of the wet desulfurization system, and calculating the correlation coefficient between the original influencing concentration factor at the starting time of the preset period and the original sulfur dioxide concentration at each time. If it is greater than, it means that there is a close correlation between the original influencing concentration factor and the original sulfur dioxide concentration in the current preset time period. At this time, the desulfurization record data of the preset time period can be determined as the original desulfurization sample data. Then, for the original sulfur dioxide concentration at the current moment, according to the probability of occurrence of the original sulfur dioxide concentration at the current moment, the probability of occurrence of the original influencing concentration factor at the starting moment of the preset time period, and the probability of the simultaneous occurrence of the original sulfur dioxide concentration at the current moment and the original influencing concentration factor at the starting moment of the preset time period, the mutual information between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment is determined, which provides a data basis for determining the correlation coefficient between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment. Then, the ratio of the mutual information between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment and the preset normalization parameter is calculated to obtain the correlation coefficient between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment, which provides a data basis for determining the response lag time of the original influencing concentration factor at the starting moment. Then, according to the correlation coefficient between the original influencing concentration factor at the starting moment and the original sulfur dioxide concentration at each moment, the response lag time of the original influencing concentration factor at the starting moment is determined. The original influencing concentration factor at the response lag moment is replaced with the original influencing concentration factor at the starting moment, and the original influencing concentration factor at each moment after the response lag moment is replaced with the original influencing concentration factor at each moment after the starting moment, which provides a data basis for the subsequent creation of target desulfurization sample data. Finally, the target desulfurization sample data is created based on the replacement result, and the interference factors caused by time delay in the original data are removed, so that the target desulfurization sample data can more accurately reflect the relationship between the influencing concentration factor and the sulfur dioxide concentration. The target desulfurization sample data is then used to train the initial desulfurization prediction model, thereby obtaining a target desulfurization prediction model with higher prediction accuracy, thereby improving the accuracy of predicting sulfur dioxide concentration.Therefore, the present application can solve the problem that the existing sulfur dioxide concentration prediction model only considers the processing of data noise, outliers and other issues in the training data set, but does not consider the time delay between the input data set and the output data set, resulting in insufficient accuracy of sulfur dioxide concentration prediction.

[0113] Figure 3 The present invention provides a structural diagram of a method for predicting sulfur dioxide concentration. The present invention can be applied to situations where sulfur dioxide concentration needs to be predicted. Figure 3 As shown,

[0114] Step 310: Obtain the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment.

[0115] Specifically, the current moment refers to the current time point. The real-time influencing concentration factor refers to the value of the influencing concentration factor corresponding to the current time point.

[0116] In a specific implementation, the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment can be obtained through a data acquisition system (such as a distributed control system of a power plant).

[0117] In this embodiment, by obtaining the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment, a data basis is provided for subsequently determining the predicted sulfur dioxide concentration at the current moment.

[0118] Step 320: input the real-time concentration influencing factors into the target desulfurization prediction model obtained by the sulfur dioxide concentration prediction model optimization method, and process to obtain the predicted sulfur dioxide concentration at the current moment.

[0119] Specifically, the predicted sulfur dioxide concentration refers to the predicted value of the sulfur dioxide emission concentration at the current moment obtained by the target desulfurization prediction model. The sulfur dioxide concentration prediction model optimization method may be the sulfur dioxide concentration prediction model optimization method provided by any embodiment of the present invention.

[0120] In a specific implementation, after obtaining the real-time factors affecting concentration, the real-time factors affecting concentration can be input into the target desulfurization prediction model obtained by the sulfur dioxide concentration prediction model optimization method. The target desulfurization prediction model will generate the predicted sulfur dioxide concentration at the current moment based on the input real-time factors affecting concentration.

[0121] In this embodiment, by inputting the real-time factors affecting the concentration into the target desulfurization prediction model obtained by the sulfur dioxide concentration prediction model optimization method, a more accurate predicted sulfur dioxide concentration at the current moment can be obtained.

[0122] The sulfur dioxide concentration prediction method provided by the embodiment of the present invention includes: obtaining the real-time concentration influencing factors of the wet flue gas desulfurization system at the current moment; inputting the real-time concentration influencing factors into the target desulfurization prediction model obtained by the sulfur dioxide concentration prediction model optimization method for processing, and obtaining the predicted sulfur dioxide concentration at the current moment. The above technical scheme first obtains the real-time concentration influencing factors of the wet flue gas desulfurization system at the current moment, which provides a data basis for subsequently determining the predicted sulfur dioxide concentration at the current moment. Then, the real-time concentration influencing factors are input into the target desulfurization prediction model obtained by the sulfur dioxide concentration prediction model optimization method for processing, and a predicted sulfur dioxide concentration at the current moment with higher accuracy is obtained.

[0123] Figure 4 A schematic structural diagram of a sulfur dioxide concentration prediction model optimization device provided in an embodiment of the present invention. The device and the sulfur dioxide concentration prediction model optimization method of the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiment of the sulfur dioxide concentration prediction model optimization device, reference can be made to the embodiment of the above-mentioned sulfur dioxide concentration prediction model optimization method.

[0124] like Figure 4 As shown, the device comprises:

[0125] The acquisition module 410 is used to acquire the desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system, obtain the original desulfurization sample data, the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculate the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment;

[0126] A determination module 420, configured to determine a response lag time of the original influencing concentration factor at the starting time according to a correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time;

[0127] A replacement module 430 is used to replace the original influencing concentration factor at the response lag moment with the original influencing concentration factor at the start moment, and replace the original influencing concentration factor at each moment after the response lag moment with the original influencing concentration factor at each moment after the start moment;

[0128] The training module 440 is used to create target desulfurization sample data according to the replacement result, and use the target desulfurization sample data to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0129] On the basis of the above embodiment, the acquisition module 410 acquires the desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain the original desulfurization sample data, including:

[0130] Acquire the desulfurization record data of the preset period from the historical desulfurization record data of the wet desulfurization system, and calculate the correlation coefficient between the original concentration influencing factor at the starting time of the preset period and the original sulfur dioxide concentration at each time;

[0131] If the maximum correlation coefficient is greater than the first correlation threshold, determining the desulfurization record data of the preset time period as the original desulfurization sample data;

[0132] If the maximum correlation coefficient is not greater than the first correlation threshold, the preset time period is updated according to the preset time update method to obtain the updated preset time period, and the step of returning to obtain the desulfurization record data of the preset time period from the historical desulfurization record data of the wet desulfurization system, and calculating the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at each time is executed.

[0133] Based on the above embodiment, the acquisition module 410 calculates the correlation coefficient between the original concentration influencing factor at the start time of the preset time period and the original sulfur dioxide concentration at each time, including:

[0134] For the original sulfur dioxide concentration at the current moment, determine the mutual information between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment according to the occurrence probability of the original sulfur dioxide concentration at the current moment, the occurrence probability of the original influencing concentration factor at the starting moment of the preset time period, and the probability of the original sulfur dioxide concentration at the current moment and the original influencing concentration factor at the starting moment of the preset time period appearing at the same time;

[0135] The ratio of the mutual information to a preset normalization parameter is calculated to obtain a correlation coefficient between the original concentration influencing factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment.

[0136] Based on the above embodiment, the training module 440 creates target desulfurization sample data according to the replacement result, including:

[0137] Determine a candidate time from the response lag time and each time after the response lag time, the candidate time being a time at which all corresponding original concentration-influencing factors are replaced;

[0138] The target desulfurization sample data is created according to the original sulfur dioxide concentration at the candidate time and the corresponding current influencing concentration factor.

[0139] Based on the above embodiment, the determination module 420 is specifically used for:

[0140] The time of the original sulfur dioxide concentration corresponding to the maximum correlation coefficient is determined as the response lag time of the original influencing concentration factor at the starting time.

[0141] Based on the above embodiment, the original factors affecting the concentration at the starting time include at least two, and the determination module 420 is specifically used for:

[0142] Determine a preset influencing concentration factor from the original influencing concentration factors at the starting moment, wherein the preset influencing concentration factor is an influencing concentration factor whose maximum correlation coefficient with the original sulfur dioxide concentration at each moment is greater than a second correlation threshold;

[0143] Determine the response lag time of the preset concentration influencing factor at the starting time.

[0144] The sulfur dioxide concentration prediction model optimization device provided in the embodiment of the present invention can execute the sulfur dioxide concentration prediction model optimization method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0145] It is worth noting that in the embodiment of the above-mentioned sulfur dioxide concentration prediction model optimization device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0146] Figure 5 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 5 A block diagram of an exemplary electronic device 5 suitable for implementing embodiments of the present invention is shown. Figure 5 The electronic device 5 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0147] like Figure 5 As shown, the electronic device 5 is in the form of a general purpose computing electronic device. The components of the electronic device 5 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 connecting different system components (including the system memory 28 and the processing unit 16).

[0148] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0149] The electronic device 5 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 5, including volatile and non-volatile media, removable and non-removable media.

[0150] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 5 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0151] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0152] The electronic device 5 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the electronic device 5, and / or any device that enables the electronic device 5 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the electronic device 5 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 20. Figure 5 As shown, the network adapter 20 communicates with other modules of the electronic device 5 via the bus 18. It should be understood that although Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with the electronic device 5, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0153] The processing unit 16 executes various functional applications and page displays by running programs stored in the system memory 28, such as implementing the sulfur dioxide concentration prediction model optimization method or sulfur dioxide concentration prediction method provided in the embodiment of the present invention.

[0154] The sulfur dioxide concentration prediction model optimization method includes:

[0155] Obtaining desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain original desulfurization sample data, wherein the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculating the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment;

[0156] Determining the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time;

[0157] The original influencing concentration factor at the response lag moment is replaced by the original influencing concentration factor at the starting moment, and the original influencing concentration factor at each moment after the response lag moment is replaced by the original influencing concentration factor at each moment after the starting moment;

[0158] Target desulfurization sample data is created according to the replacement result, and the target desulfurization sample data is used to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0159] The sulfur dioxide concentration prediction method includes:

[0160] Obtain the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment;

[0161] The real-time concentration influencing factors are input into the target desulfurization prediction model obtained by using the aforementioned sulfur dioxide concentration prediction model optimization method to obtain the predicted sulfur dioxide concentration at the current moment.

[0162] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the sulfur dioxide concentration prediction model optimization method or the sulfur dioxide concentration prediction method provided by any embodiment of the present invention.

[0163] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, for example, the sulfur dioxide concentration prediction model optimization method or the sulfur dioxide concentration prediction method provided in the embodiment of the present invention is implemented.

[0164] The sulfur dioxide concentration prediction model optimization method includes:

[0165] Obtaining desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain original desulfurization sample data, wherein the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculating the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment;

[0166] Determining the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time;

[0167] The original influencing concentration factor at the response lag moment is replaced by the original influencing concentration factor at the starting moment, and the original influencing concentration factor at each moment after the response lag moment is replaced by the original influencing concentration factor at each moment after the starting moment;

[0168] Target desulfurization sample data is created according to the replacement result, and the target desulfurization sample data is used to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

[0169] The sulfur dioxide concentration prediction method includes:

[0170] Obtain the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment;

[0171] The real-time concentration influencing factors are input into the target desulfurization prediction model obtained by using the aforementioned sulfur dioxide concentration prediction model optimization method to obtain the predicted sulfur dioxide concentration at the current moment.

[0172] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0173] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0174] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0175] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0176] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0177] In addition, the acquisition, storage, use, and processing of data in the technical solution of the present invention are in compliance with the relevant provisions of national laws and regulations.

[0178] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A sulfur dioxide concentration prediction model optimization method, characterized in that: include: Obtaining desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain original desulfurization sample data, wherein the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculating the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment; Determining the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time; The original influencing concentration factor at the response lag moment is replaced by the original influencing concentration factor at the starting moment, and the original influencing concentration factor at each moment after the response lag moment is replaced by the original influencing concentration factor at each moment after the starting moment; Target desulfurization sample data is created according to the replacement result, and the target desulfurization sample data is used to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

2. The sulfur dioxide concentration prediction model optimization method according to claim 1, characterized in that: The desulfurization record data of the preset period is obtained from the historical desulfurization record data of the wet desulfurization system to obtain the original desulfurization sample data, including: Acquire the desulfurization record data of the preset period from the historical desulfurization record data of the wet desulfurization system, and calculate the correlation coefficient between the original concentration influencing factor at the starting time of the preset period and the original sulfur dioxide concentration at each time; If the maximum correlation coefficient is greater than the first correlation threshold, determining the desulfurization record data of the preset time period as the original desulfurization sample data; If the maximum correlation coefficient is not greater than the first correlation threshold, the preset time period is updated according to the preset time update method to obtain the updated preset time period, and the step of returning to obtain the desulfurization record data of the preset time period from the historical desulfurization record data of the wet desulfurization system, and calculating the correlation coefficient between the original influencing concentration factor at the starting time of the preset time period and the original sulfur dioxide concentration at each time is executed.

3. The sulfur dioxide concentration prediction model optimization method according to claim 1, characterized in that: Calculating the correlation coefficient between the original concentration influencing factor at the start time of the preset time period and the original sulfur dioxide concentration at each time, including: For the original sulfur dioxide concentration at the current moment, determine the mutual information between the original influencing concentration factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment according to the occurrence probability of the original sulfur dioxide concentration at the current moment, the occurrence probability of the original influencing concentration factor at the starting moment of the preset time period, and the probability of the original sulfur dioxide concentration at the current moment and the original influencing concentration factor at the starting moment of the preset time period appearing at the same time; The ratio of the mutual information to a preset normalization parameter is calculated to obtain a correlation coefficient between the original concentration influencing factor at the starting moment of the preset time period and the original sulfur dioxide concentration at the current moment.

4. The sulfur dioxide concentration prediction model optimization method according to claim 1, characterized in that: Create target desulfurization sample data based on the replacement results, including: Determine a candidate time from the response lag time and each time after the response lag time, the candidate time being a time at which all corresponding original concentration-influencing factors are replaced; The target desulfurization sample data is created according to the original sulfur dioxide concentration at the candidate time and the corresponding current influencing concentration factor.

5. The sulfur dioxide concentration prediction model optimization method according to claim 1, characterized in that: Determining the response lag time of the original influencing concentration factor at the starting time according to the correlation coefficient between the original influencing concentration factor at the starting time and the original sulfur dioxide concentration at each time, comprising: The time of the original sulfur dioxide concentration corresponding to the maximum correlation coefficient is determined as the response lag time of the original influencing concentration factor at the starting time.

6. The sulfur dioxide concentration prediction model optimization method according to claim 1, characterized in that: The original influencing concentration factors at the starting time include at least two, and determining the response lag time of the original influencing concentration factors at the starting time according to the correlation coefficient between the original influencing concentration factors at the starting time and the original sulfur dioxide concentrations at each time includes: Determine a preset influencing concentration factor from the original influencing concentration factors at the starting moment, wherein the preset influencing concentration factor is an influencing concentration factor whose maximum correlation coefficient with the original sulfur dioxide concentration at each moment is greater than a second correlation threshold; Determine the response lag time of the preset concentration influencing factor at the starting time.

7. A method for predicting sulfur dioxide concentration, characterized in that: include: Obtain the real-time influencing concentration factors of the wet flue gas desulfurization system at the current moment; The real-time influencing concentration factors are input into the target desulfurization prediction model obtained by the sulfur dioxide concentration prediction model optimization method described in any one of claims 1 to 6, so as to obtain the predicted sulfur dioxide concentration at the current moment.

8. A sulfur dioxide concentration prediction model optimization device, characterized in that: include: An acquisition module is used to acquire desulfurization record data of a preset period from the historical desulfurization record data of the wet desulfurization system to obtain original desulfurization sample data, wherein the original desulfurization sample data includes the original influencing concentration factors and the corresponding original sulfur dioxide concentrations at each moment in the preset period, and calculate the correlation coefficient between the original influencing concentration factors at the starting moment in the preset period and the original sulfur dioxide concentrations at each moment; A determination module, used to determine the response lag time of the original influencing concentration factor at the starting moment according to the correlation coefficient between the original influencing concentration factor at the starting moment and the original sulfur dioxide concentration at each moment; A replacement module, used to replace the original influencing concentration factors at the response lag moment with the original influencing concentration factors at the starting moment, and to replace the original influencing concentration factors at each moment after the response lag moment with the original influencing concentration factors at each moment after the starting moment; The training module is used to create target desulfurization sample data according to the replacement result, and use the target desulfurization sample data to train the initial desulfurization prediction model to obtain the target desulfurization prediction model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method for optimizing the sulfur dioxide concentration prediction model according to any one of claims 1 to 6 is implemented; or, when the processor executes the program, the method for predicting sulfur dioxide concentration according to claim 7 is implemented.

10. A storage medium containing computer executable instructions, on which a computer program is stored, characterized in that: When the program is executed by a processor, the method for optimizing the sulfur dioxide concentration prediction model according to any one of claims 1 to 6 is implemented; or, when the program is executed by a processor, the method for predicting sulfur dioxide concentration according to claim 7 is implemented.

Images