CN115345376A - Method and device for predicting oxygen content of boiler flue gas - Google Patents
Method and device for predicting oxygen content of boiler flue gas Download PDFInfo
- Publication number
- CN115345376A CN115345376A CN202211035640.7A CN202211035640A CN115345376A CN 115345376 A CN115345376 A CN 115345376A CN 202211035640 A CN202211035640 A CN 202211035640A CN 115345376 A CN115345376 A CN 115345376A
- Authority
- CN
- China
- Prior art keywords
- oxygen content
- flue gas
- parameter time
- data
- series data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The disclosure relates to the technical field of computers, and provides a method and a device for predicting oxygen content of boiler flue gas. The method comprises the following steps: acquiring historical sample operation parameter time-series data and a corresponding smoke oxygen content label; preprocessing historical sample operation parameter time series data; clustering the preprocessed historical sample operation parameter time sequence data to obtain historical operation parameter time sequence data of different clustering clusters; and training the initial extreme gradient lifting model by adopting the historical operating parameter time sequence data of different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model, and predicting the boiler flue gas oxygen content by adopting the flue gas oxygen content prediction model.
Description
Technical Field
The disclosure relates to the technical field of computers, in particular to a method and a device for predicting oxygen content of boiler flue gas.
Background
Thermal efficiency is an important measure for gas boilers. The heat efficiency of energy equipment such as a boiler can be improved by carrying out closed-loop control on the operation of the boiler according to the oxygen content of the smoke of the boiler.
A zirconia measuring instrument is arranged at a flue gas outlet of the boiler, so that the oxygen content of the flue gas of the boiler can be measured. However, the cost of using the zirconia measuring instrument to perform the flue gas oxygen content test is high, which results in that the small gas-fired boiler in some application scenes abandons the installation of the zirconia measuring instrument for saving the cost, so that the flue gas oxygen content of the boiler cannot be obtained, and the thermal efficiency of the boiler cannot be improved.
Disclosure of Invention
In view of this, the embodiments of the present disclosure provide a method and an apparatus for predicting an oxygen content of boiler flue gas, an electronic device, and a computer-readable storage medium, so as to solve the problem in the prior art that the oxygen content of boiler flue gas cannot be known.
In a first aspect of the embodiments of the present disclosure, a method for predicting oxygen content of boiler flue gas is provided, including: acquiring historical sample operation parameter time-series data and a corresponding smoke oxygen content label; preprocessing historical sample operation parameter time series data; clustering the preprocessed historical sample operation parameter time sequence data to obtain historical operation parameter time sequence data of different clusters; training an initial extreme gradient lifting model by adopting historical operating parameter time series data of different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model, and predicting the flue gas oxygen content of the boiler by adopting the flue gas oxygen content prediction model.
In a second aspect of the embodiments of the present disclosure, a device for predicting oxygen content in flue gas of a boiler is provided, including: the acquisition module is used for acquiring historical sample operation parameter time series data and a corresponding smoke oxygen content label; the preprocessing module is used for preprocessing the historical sample operation parameter time series data; the clustering module is used for clustering the preprocessed historical sample operation parameter time sequence data to obtain historical operation parameter time sequence data of different clustering clusters; and the training module is used for training the initial extreme gradient lifting model by adopting the historical operating parameter time series data of different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model so as to predict the boiler flue gas oxygen content by adopting the flue gas oxygen content prediction model.
In a third aspect of the embodiments of the present disclosure, an electronic device is provided, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.
In a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program, which when executed by a processor, implements the steps of the above-mentioned method.
Compared with the prior art, the embodiment of the disclosure has the following beneficial effects: through clustering historical sample operation parameter time series data and training an extreme gradient lifting model based on the historical operation parameter time series data obtained through processing, the flue gas oxygen content can be accurately predicted by the trained flue gas oxygen content prediction model, closed-loop control can be performed on the boiler based on the flue gas oxygen content, and the thermal efficiency of the boiler is improved.
Drawings
To more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings needed for the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without inventive efforts.
FIG. 1 is a schematic flow chart of a method for predicting oxygen content in boiler flue gas according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart of an application method of a flue gas oxygen content prediction model provided by the embodiment of the disclosure;
FIG. 3 is a flow chart of a pre-processing process of historical sample operating parameter time series data provided by an embodiment of the disclosure;
FIG. 4 is a schematic structural diagram of a device for predicting oxygen content in flue gas of a boiler according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the disclosed embodiments. However, it will be apparent to one skilled in the art that the present disclosure may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present disclosure with unnecessary detail.
The method and the device for predicting the oxygen content of the boiler flue gas according to the embodiment of the disclosure will be described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for predicting oxygen content in boiler flue gas according to an embodiment of the present disclosure. The method provided by the embodiment of the present disclosure can be executed by any electronic device with computer processing capability, such as a terminal or a server. As shown in fig. 1, the method for predicting the oxygen content of the boiler flue gas comprises the following steps:
and step S101, acquiring historical sample operation parameter time series data and a corresponding smoke oxygen content label.
Specifically, the historical sample operating parameter time-series data and the corresponding smoke oxygen content label form a training data set for training the smoke oxygen content model. The training data set may be split into a training set and a test set. When splitting the training data set, a splitting point may be selected to split the training data set into a training set and a test set, and the splitting point may be a time point. For example, if the training data included in the training data set is a historical sample operation parameter time sequence within the time length of one year and a corresponding smoke oxygen content label, the last day of the 8 th month can be set as a splitting point, the training data of the first 8 months are constructed into a training set after splitting, and the training data of the last 4 months are constructed into a test set.
And step S102, preprocessing the historical sample operation parameter time series data.
Specifically, bad data in the historical sample operation parameter time sequence can be identified and eliminated in the preprocessing process, and normalization processing is performed to normalize the training data.
And step S103, clustering the preprocessed historical sample operation parameter time series data to obtain historical operation parameter time series data of different clusters.
Specifically, historical sample operation parameter time series data are clustered, so that training data groups in a cluster form can be obtained, and each training data group belongs to the same cluster.
And S104, training an initial extreme gradient lifting model by adopting the historical operating parameter time series data of different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model, and predicting the boiler flue gas oxygen content by adopting the flue gas oxygen content prediction model.
Specifically, the extreme gradient lifting model may be trained according to the cluster of the historical operating parameter time series data. For example, a first sub-training set may be constructed based on the historical operating parameter time-series data belonging to the first set cluster and the smoke oxygen content label corresponding thereto, a smoke oxygen content prediction model is trained on the first sub-training set, and then the trained smoke oxygen content prediction model is evaluated and the model is adjusted by using the test set. Further, a second sub-training set can be constructed based on the historical operating parameter time-series data belonging to a second set cluster and the corresponding smoke oxygen content label, the adjusted smoke oxygen content prediction model is trained on the first sub-training set, the test set is adopted to evaluate the trained smoke oxygen content prediction model, and the model is adjusted again. The above training and testing process is repeated until the initial pair of extreme gradient boosting models converges. The first set cluster can be one, two or more clusters, and the second set cluster can be one, two or more clusters.
According to the technical scheme of the embodiment of the disclosure, clustering operation is performed on historical sample operation parameter time-series data based on an optimized clustering algorithm, an extreme gradient boost model is trained based on the historical operation parameter time-series data obtained by the clustering operation, after an xgboost (extreme gradient boost) model is obtained, the oxygen content of flue gas can be predicted by using the extreme gradient boost model, and compared with a method for measuring the oxygen content of flue gas of a boiler by installing a zirconia measuring instrument, the cost for predicting the oxygen content of flue gas can be reduced, and a better flue gas oxygen content prediction effect is achieved.
As shown in fig. 2, the application method of the flue gas oxygen content prediction model comprises the following steps:
step S201, acquiring time series data of current operation parameters of the boiler equipment.
Specifically, the current operating parameter time-series data includes time-series data of an operating parameter during operation of the boiler acquired in real time. In the technical scheme of the embodiment of the disclosure, the oxygen content of the flue gas of the boiler can be predicted according to the time series data of the current operation parameters.
And S202, calling a flue gas oxygen content prediction model according to the current operation parameter time series data.
And S203, obtaining the prediction data of the oxygen content of the flue gas of the boiler equipment by using the flue gas oxygen content prediction model.
Specifically, when the time-series data of the operation parameters of the historical samples of the boiler and the corresponding tags of the oxygen content of the flue gas are acquired, the time-series data of the operation parameters of the historical samples can be generated, and the oxygen content of the flue gas of the corresponding boiler can be detected by using a zirconia measuring instrument, but not limited to this. And training the initial extreme gradient lifting model by taking the detected oxygen content of the flue gas as a training label of the corresponding time series data of the operation parameters of the historical samples, so as to obtain the flue gas oxygen content prediction model capable of predicting the oxygen content of the flue gas.
As shown in fig. 3, in step S102, the process of preprocessing the time-series data of the historical sample operation parameters includes the following steps:
in step S301, bad data identification is performed on the time-series data of the historical sample operation parameters.
Step S302, bad data correction is carried out on the historical sample operation parameter time series data after bad data identification.
Step S303, the historical sample operation parameter time series data after bad data correction is processed with normalization processing.
In step S301, when bad data identification is performed on the historical sample operating parameter time-series data, a data point with abnormal oxygen content in flue gas in the historical sample operating parameter time-series data may be screened out based on the three-sigma principle.
Specifically, bad data can be discriminated according to the following formulas (1) and (2).
Wherein x is n,i Is the oxygen content of the smoke at the ith moment of the nth day,
is the average value of the oxygen content of the smoke at the ith moment from the 1 st day to the Nth day,
the variance of the oxygen content of the smoke at the time i from day 1 to day N, epsilon, is a set threshold, typically 1 to 1.5. In the embodiment of the disclosure, after the mean value and the variance of the oxygen content of the flue gas at the ith time from the 1 st day to the nth day are calculated by the formula (1), bad data discrimination can be realized by using the formula (2) based on the three-sigma (3 sigma) principle. If the flue gas oxygen content, the mean value and the variance data do not meet the formula (2), the current flue gas oxygen content data can be determined to be normal data and reserved. If the flue gas oxygen content data meets the formula (2), the current flue gas oxygen content data can be determined to be bad data, and the bad data needs to be corrected.
In step S302, when bad data correction is performed on the historical sample operating parameter time-series data, the historical sample operating parameter time-series data after bad data correction is performed on the data point with the abnormal oxygen content of the flue gas may be obtained based on the weighted sum of the average values of the data points with the abnormal oxygen content of the flue gas and the flue gas oxygen content at the same time on at least one similar day before and after the data point with the abnormal oxygen content of the flue gas.
Specifically, the bad data correction may be performed according to the following formula (3).
Wherein the content of the first and second substances,
is a corrected value of the oxygen content of the flue gas at the ith time of the nth day, x n±1,i Is x n,i The oxygen content value of the smoke at the ith time of the previous and the next similar days. Alpha is alpha 1 、β 1 And gamma 1 Is a weighted value.
In step S303, the historical sample operation parameter time-series data are normalized, and a normalized value of the flue gas oxygen content time-series data may be determined based on the maximum value and the minimum value of the flue gas oxygen content of the historical sample operation parameter time-series data at each time of each day. Specifically, the difference between the smoke oxygen content value at each time in any day in the training data set and the minimum value at each time in the day and the extremely poor ratio between the difference and the smoke oxygen content value in the day can be obtained, and the ratio is used as the normalization value corresponding to the historical sample operation parameter time series data at each time in the day.
Specifically, the normalization processing may be performed according to the following formula (4).
Wherein x is ij Is the element in the ith row and the jth column in the oxygen content matrix of the flue gas,
is the element in the ith row and the jth column in the normalized flue gas oxygen content matrix,
respectively the minimum value and the maximum value of the oxygen content of the smoke at each moment of the ith day,
for the extreme difference of the oxygen content of the flue gas on the i day, i =1,2, …, n, j =1,2, …,24, n is a natural number.
In step S103, determining an initial clustering center of the historical sample operation parameter time sequence data according to fitness corresponding to the characteristics of the historical sample operation parameter time sequence data and a set maximum genetic algebra by using a genetic algorithm; based on the initial clustering center, grouping the historical sample operation parameter time sequence data into a plurality of clustering clusters by using a clustering algorithm to obtain grouped historical operation parameter time sequence data.
Specifically, training data are clustered, so that the data of the same cluster have more similar data characteristics, and a genetic algorithm is adopted for optimization during clustering of the training data, so that a clustering center of the training data can be acquired more accurately.
In step S104, the initial extreme gradient boost model is trained using the root mean square error and the relative error rate as loss functions until the initial extreme gradient boost model converges.
Specifically, in the training process, test data concentrated in the test are input into the current extreme gradient lifting model, and the output value of the current extreme gradient lifting model obtained is the predicted value of the oxygen content of the flue gas. The test index of the prediction result can select rmse (root mean square error) and re (equivalent error) as evaluation indexes.
Wherein, the calculation formula of rmse is shown as the following formula (5):
wherein, y i Is the true value of the time-series data,
is a time series data prediction value.
Wherein, the calculation formula of re is shown as the following formula (6):
wherein, y i Is the true value of the time-series data,
is a time series data prediction value.
For example, the sum of rmse and re can be used as a loss function, and the training and testing process of the initial extreme gradient lifting model is repeated until the loss function converges to obtain final flue gas oxygen content prediction data.
All the above optional technical solutions may be combined arbitrarily to form optional embodiments of the present application, and are not described herein again.
According to the method for predicting the oxygen content of the boiler flue gas, the historical sample operation parameter time series data are subjected to clustering processing, and the extreme gradient lifting model is trained on the basis of the processed historical operation parameter time series data, so that the trained flue gas oxygen content prediction model can predict the oxygen content of the flue gas more accurately, the boiler can be subjected to closed-loop control on the basis of the flue gas oxygen content, and the thermal efficiency of the boiler is improved.
The following are embodiments of the disclosed apparatus that may be used to perform embodiments of the disclosed methods. The following description of the prediction device of the oxygen content of the boiler flue gas and the above description of the prediction method of the oxygen content of the boiler flue gas can be referred to with each other. For details not disclosed in the embodiments of the apparatus of the present disclosure, refer to the embodiments of the method of the present disclosure.
Fig. 4 is a schematic diagram of a device for predicting oxygen content in boiler flue gas according to an embodiment of the disclosure. As shown in fig. 4, the apparatus for predicting the oxygen content of the flue gas of the boiler comprises:
and the acquisition module can be used for acquiring the historical sample operation parameter time series data and the corresponding smoke oxygen content label.
Specifically, the historical sample operating parameter time-series data and the corresponding smoke oxygen content label form a training data set for training the smoke oxygen content model. The training data set may be split into a training set and a test set. When splitting the training data set, a splitting point may be selected to split the training data set into a training set and a test set, and the splitting point may be a time point.
And the preprocessing module can be used for preprocessing the historical sample operation parameter time sequence data.
Specifically, bad data in the time sequence of the historical sample operation parameters can be identified and eliminated in the preprocessing process, and normalization processing is performed to normalize the training data.
And the clustering module can be used for clustering the preprocessed historical sample operation parameter time sequence data to obtain historical operation parameter time sequence data of different clustering clusters.
Specifically, historical sample operation parameter time series data are clustered, so that training data groups in a cluster form can be obtained, and each training data group belongs to the same cluster.
And the training module can be used for training the initial extreme gradient lifting model by adopting the historical operating parameter time sequence data of different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model so as to predict the boiler flue gas oxygen content by adopting the flue gas oxygen content prediction model.
Specifically, the extreme gradient lifting model may be trained according to the cluster of the historical operating parameter time series data. For example, a first sub-training set may be constructed based on the historical operating parameter time-series data belonging to the first set cluster and the smoke oxygen content label corresponding thereto, a smoke oxygen content prediction model is trained on the first sub-training set, and then the trained smoke oxygen content prediction model is evaluated and the model is adjusted by using the test set. Further, a second sub-training set can be constructed based on the historical operating parameter time series data belonging to the second set cluster and the corresponding flue gas oxygen content label, the adjusted flue gas oxygen content prediction model is trained on the first sub-training set, and then the test set is adopted to evaluate the trained flue gas oxygen content prediction model and adjust the model again. The above training and testing process is repeated until the initial pair of extreme gradient boosting models converges. The first set cluster can be one, two or more clusters, and the second set cluster can be one, two or more clusters.
According to the technical scheme of the embodiment of the disclosure, clustering operation is carried out on historical sample operation parameter time series data based on an optimized clustering algorithm, an extreme gradient lifting model is trained based on the historical operation parameter time series data obtained by the clustering operation, after an xgboost model is obtained, the oxygen content of flue gas can be predicted by using the extreme gradient lifting model, and compared with the method of installing a zirconia measuring instrument to measure the oxygen content of flue gas of a boiler, the method can reduce the cost of predicting the oxygen content of flue gas and has a better flue gas oxygen content predicting effect.
In this disclosure, the apparatus for predicting the oxygen content of the boiler flue gas further includes an application module for applying the above flue gas oxygen content prediction model to predict the oxygen content of the boiler flue gas, where the application module includes:
and the acquisition sub-module can be used for acquiring the time series data of the current operating parameters of the boiler equipment.
Specifically, the current operating parameter time-series data includes time-series data of an operating parameter during operation of the boiler acquired in real time. In the technical scheme of the embodiment of the disclosure, the oxygen content of the flue gas of the boiler can be predicted according to the time series data of the current operation parameters.
And the calling submodule is used for calling a flue gas oxygen content prediction model according to the current operation parameter time series data.
And the prediction submodule is used for obtaining the prediction data of the oxygen content of the flue gas of the boiler equipment by utilizing the flue gas oxygen content prediction model.
Specifically, when the time-series data of the operation parameters of the historical samples of the boiler and the corresponding tags of the oxygen content of the flue gas are acquired, the time-series data of the operation parameters of the historical samples can be generated, and the oxygen content of the flue gas of the corresponding boiler can be detected by using a zirconia measuring instrument, but not limited to this. And training the initial extreme gradient lifting model by taking the detected oxygen content of the flue gas as a training label of the corresponding time series data of the operation parameters of the historical samples, so as to obtain the flue gas oxygen content prediction model capable of predicting the oxygen content of the flue gas.
In the embodiment of the disclosure, the preprocessing module may be further configured to perform bad data identification, bad data correction, and normalization processing on the time-series data of the historical sample operation parameters.
Specifically, when the preprocessing module identifies bad data of the time series data of the operation parameters of the historical samples, a data point with abnormal oxygen content of the flue gas in the time series data of the operation parameters of the historical samples can be screened out based on the three sigma principle.
Specifically, bad data can be discriminated according to the following formulas (1) and (2).
Wherein x is n,i Is the oxygen content of the smoke at the ith moment of the nth day,
is the average value of the oxygen content of the smoke at the ith moment from the 1 st day to the Nth day,
the variance of the oxygen content of the smoke at the time i from day 1 to day N, epsilon, is a set threshold, typically 1 to 1.5. In the embodiment of the disclosure, after the mean value and the variance of the oxygen content of the flue gas at the ith time from the 1 st day to the nth day are calculated by the formula (1), bad data discrimination can be realized by using the formula (2) based on the three-sigma (3 sigma) principle. If the flue gas oxygen content, the average value and the variance data do not meet the formula (2), the current flue gas oxygen content data can be determined to be normal data and reserved. If the flue gas oxygen content data meets the formula (2), the current flue gas oxygen content data can be determined to be bad data, and the bad data needs to be corrected.
When the preprocessing module corrects the bad data of the time-series data of the operation parameters of the historical samples, the time-series data of the operation parameters of the historical samples after correcting the bad data of the data points of the abnormal oxygen content of the flue gas can be obtained based on the weighted sum of the average values of the data points of the abnormal oxygen content of the flue gas and the flue gas oxygen content at the same time on at least one same day before and after the data points of the abnormal oxygen content of the flue gas.
Specifically, the bad data correction may be performed according to the following formula (3).
Wherein the content of the first and second substances,
is a corrected value, x, of the oxygen content of the flue gas at the ith moment of the nth day n±1,i Is x n,i The oxygen content value of the smoke at the ith time of the previous and the next similar days. Alpha is alpha 1 、β 1 And gamma 1 Is a weighted value.
When the preprocessing module normalizes the historical sample operation parameter time series data, the normalization value of the flue gas oxygen content time series data can be determined based on the maximum value and the minimum value of the flue gas oxygen content of the historical sample operation parameter time series data at each time of each day. Specifically, the difference between the smoke oxygen content value at each time in any day in the training data set and the minimum value at each time in the day and the extremely poor ratio between the difference and the smoke oxygen content value in the day can be obtained, and the ratio is used as the normalization value corresponding to the historical sample operation parameter time series data at each time in the day.
Specifically, the normalization processing may be performed according to the following formula (4).
Wherein x is ij Is the element in the ith row and the jth column in the oxygen content matrix of the flue gas,
is the element of the ith row and the jth column in the normalized oxygen content matrix of the flue gas,
respectively the minimum value and the maximum value of the oxygen content of the smoke at each moment of the ith day,
for the extreme difference of the oxygen content of the flue gas on the i day, i =1,2, …, n, j =1,2, …,24, n is a natural number.
The clustering module can also determine an initial clustering center of the historical sample operation parameter time sequence data according to fitness corresponding to the characteristics of the historical sample operation parameter time sequence data and a set maximum genetic algebra by using a genetic algorithm; and based on the initial clustering center, grouping the historical sample operation parameter time sequence data into a plurality of clustering clusters by using a clustering algorithm to obtain grouped historical operation parameter time sequence data.
Specifically, training data are clustered, so that the data of the same cluster have more similar data characteristics, and a genetic algorithm is adopted for optimization when the training data are clustered, so that the clustering center of the training data can be acquired more accurately.
The training module may train the initial extreme gradient boost model using the root mean square error and the relative error rate as a loss function until the initial extreme gradient boost model converges.
Specifically, in the training process, test data concentrated in the test is input into the current extreme gradient lifting model, and the output value of the current extreme gradient lifting model obtained is the predicted value of the oxygen content of the flue gas. The test index of the prediction result can select rmse (root mean square error) and re (equivalent error) as evaluation indexes.
Wherein, the calculation formula of rmse is shown as the following formula (5):
wherein, y i Is the true value of the time-series data,
is a time series data prediction value.
Wherein, the calculation formula of re is shown as the following formula (6):
wherein, y i Is the true value of the time-series data,
is a time series data prediction value.
Since each functional module of the apparatus for predicting oxygen content in boiler flue gas in the exemplary embodiment of the present disclosure corresponds to the steps of the exemplary embodiment of the method for predicting oxygen content in boiler flue gas, please refer to the embodiment of the method for predicting oxygen content in boiler flue gas in the present disclosure for details that are not disclosed in the embodiment of the apparatus of the present disclosure.
According to the prediction device of the boiler flue gas oxygen content of the embodiment of the disclosure, the historical sample operation parameter time series data are clustered, and the extreme gradient lifting model is trained based on the processed historical operation parameter time series data, so that the flue gas oxygen content can be accurately predicted by the trained flue gas oxygen content prediction model, the boiler can be subjected to closed-loop control based on the flue gas oxygen content, and the thermal efficiency of the boiler is improved.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present disclosure.
Fig. 5 is a schematic diagram of an electronic device 5 provided in an embodiment of the present disclosure. As shown in fig. 5, the electronic apparatus 5 of this embodiment includes: a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and operable on the processor 501. The steps in the various method embodiments described above are implemented when the processor 501 executes the computer program 503. Alternatively, the processor 501 implements the functions of the respective modules in the above-described respective apparatus embodiments when executing the computer program 503.
The electronic device 5 may be a desktop computer, a notebook, a palm computer, a cloud server, or other electronic devices. The electronic device 5 may include, but is not limited to, a processor 501 and a memory 502. Those skilled in the art will appreciate that fig. 5 is merely an example of the electronic device 5, and does not constitute a limitation of the electronic device 5, and may include more or less components than those shown, or different components.
The Processor 501 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like.
The storage 502 may be an internal storage unit of the electronic device 5, for example, a hard disk or a memory of the electronic device 5. The memory 502 may also be an external storage device of the electronic device 5, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the electronic device 5. The memory 502 may also include both internal and external storage units of the electronic device 5. The memory 502 is used for storing computer programs and other programs and data required by the electronic device.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules, so as to perform all or part of the functions described above. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit.
The integrated module, if implemented in the form of a software functional unit and sold or used as a separate product, may be stored in a computer readable storage medium. Based on such understanding, the present disclosure may implement all or part of the flow of the method in the above embodiments, and may also be implemented by a computer program to instruct related hardware, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the computer program may implement the steps of the above methods and embodiments. The computer program may comprise computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, recording medium, U.S. disk, removable hard disk, magnetic diskette, optical disk, computer Memory, read-Only Memory (ROM), random Access Memory (RAM), electrical carrier wave signal, telecommunications signal, software distribution medium, etc. It should be noted that the computer readable medium may contain suitable additions or additions that may be required in accordance with legislative and patent practices within the jurisdiction, for example, in some jurisdictions, computer readable media may not include electrical carrier signals or telecommunications signals in accordance with legislative and patent practices.
The above examples are only intended to illustrate the technical solutions of the present disclosure, not to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present disclosure, and are intended to be included within the scope of the present disclosure.
Claims (10)
1. A prediction method for the oxygen content of boiler flue gas is characterized by comprising the following steps:
acquiring historical sample operation parameter time-series data and a corresponding smoke oxygen content label;
preprocessing the historical sample operation parameter time-series data;
clustering the preprocessed historical sample operation parameter time sequence data to obtain historical operation parameter time sequence data of different clusters;
training an initial extreme gradient lifting model by adopting the historical operating parameter time-series data of the different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model, and predicting the boiler flue gas oxygen content by adopting the flue gas oxygen content prediction model.
2. The method of claim 1, wherein after obtaining the flue gas oxygen content prediction model, the method further comprises:
acquiring time series data of current operating parameters of boiler equipment;
calling the flue gas oxygen content prediction model according to the current operation parameter time series data;
and obtaining the prediction data of the oxygen content of the flue gas of the boiler equipment by using the flue gas oxygen content prediction model.
3. The method of claim 1, wherein the pre-processing the historical sample run parameter time series data comprises:
and carrying out bad data identification, bad data correction and normalization processing on the historical sample operation parameter time series data.
4. The method according to claim 1, wherein the clustering the preprocessed historical sample operating parameter time series data to obtain the historical operating parameter time series data of different clusters comprises:
determining an initial clustering center of the historical sample operation parameter time sequence data according to fitness corresponding to the characteristics of the historical sample operation parameter time sequence data and a set maximum genetic algebra by using a genetic algorithm;
and grouping the historical sample operation parameter time sequence data into a plurality of clustering clusters by using a clustering algorithm based on the initial clustering center to obtain the grouped historical operation parameter time sequence data.
5. The method of claim 3, wherein performing bad data identification and bad data correction on the historical sample operating parameter time series data comprises:
screening out data points with abnormal oxygen content of the flue gas in the historical sample operation parameter time series data based on a three-sigma principle;
and obtaining the historical sample operation parameter time series data after correcting the data points with abnormal smoke oxygen content according to the weighted sum of the smoke oxygen content of the data points with abnormal smoke oxygen content and the average value of the data points with abnormal smoke oxygen content at the same time on at least one similar day before and after the data points with abnormal smoke oxygen content.
6. The method of claim 3, wherein normalizing the historical sample run parameter time series data comprises:
and determining a normalized value of the time-series data of the oxygen content of the smoke based on the maximum value and the minimum value of the oxygen content of the smoke of the time-series data of the historical sample operation parameters at each moment of each day.
7. The method of claim 1, wherein the training an initial extreme gradient boost model using the historical operating parameter time series data of the different clusters until the initial extreme gradient boost model converges comprises:
training the initial extreme gradient boost model by using the root mean square error and the relative error rate as loss functions until the initial extreme gradient boost model converges.
8. An apparatus for predicting oxygen content in boiler flue gas, the apparatus comprising:
the acquisition module is used for acquiring historical sample operation parameter time series data and a corresponding smoke oxygen content label;
the preprocessing module is used for preprocessing the historical sample operation parameter time series data;
the clustering module is used for clustering the preprocessed historical sample operation parameter time sequence data to obtain historical operation parameter time sequence data of different clustering clusters;
and the training module is used for training an initial extreme gradient lifting model by adopting the historical operating parameter time sequence data of the different clustering clusters until the initial extreme gradient lifting model converges to obtain a flue gas oxygen content prediction model, so that the flue gas oxygen content prediction model is adopted to predict the boiler flue gas oxygen content.
9. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of a method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211035640.7A CN115345376A (en) | 2022-08-26 | 2022-08-26 | Method and device for predicting oxygen content of boiler flue gas |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211035640.7A CN115345376A (en) | 2022-08-26 | 2022-08-26 | Method and device for predicting oxygen content of boiler flue gas |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115345376A true CN115345376A (en) | 2022-11-15 |
Family
ID=83954834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211035640.7A Pending CN115345376A (en) | 2022-08-26 | 2022-08-26 | Method and device for predicting oxygen content of boiler flue gas |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115345376A (en) |
-
2022
- 2022-08-26 CN CN202211035640.7A patent/CN115345376A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113239314A (en) | Method, device, terminal and computer-readable storage medium for carbon emission prediction | |
| CN112465138A (en) | Model distillation method, device, storage medium and equipment | |
| CN111461445B (en) | Short-term wind speed prediction method and device, computer equipment and storage medium | |
| CN113743673B (en) | Power load prediction method during typhoon | |
| CN110991761B (en) | Heat supply load prediction method and device | |
| CN112417734B (en) | Wind speed correction method and device based on geographic information of wind farm | |
| CN113837596A (en) | Fault determination method and device, electronic equipment and storage medium | |
| CN108399505B (en) | Distributed photovoltaic power supply access capacity planning method and terminal equipment | |
| CN114912720A (en) | Memory network-based power load prediction method, device, terminal and storage medium | |
| CN115016966A (en) | Measurement automation system fault prediction method and device based on Transformer and storage medium | |
| CN112257958A (en) | Power saturation load prediction method and device | |
| CN101206727A (en) | Data processing apparatus, data processing method data processing program and computer readable medium | |
| CN116706884A (en) | Photovoltaic power generation amount prediction method, device, terminal and storage medium | |
| CN115345376A (en) | Method and device for predicting oxygen content of boiler flue gas | |
| CN115935822A (en) | Method, system, device and medium for predicting long-term degradation trend of computer parameters | |
| CN113705929B (en) | Spring festival holiday load prediction method based on load characteristic curve and typical characteristic value fusion | |
| CN116432908A (en) | Near-zero carbon park carbon accounting method, device, terminal and storage medium | |
| CN111563078B (en) | Data quality detection method and device based on time sequence data and storage device | |
| CN112906757A (en) | Model training method, optical power prediction method, device, equipment and storage medium | |
| CN111861259A (en) | Load modeling method, system and storage medium considering time sequence | |
| CN112101806A (en) | Harmonic responsibility evaluation method, terminal equipment and storage medium | |
| CN113449258B (en) | Quality evaluation method and device for intelligent electric meter, terminal equipment and storage medium | |
| CN112580700B (en) | Data correction method, system and storage medium of electric power Internet of things meter | |
| CN113435058B (en) | Data dimension reduction method, system, terminal and medium for distribution network self-healing test model | |
| CN116187299B (en) | Scientific and technological project text data verification and evaluation method, system and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |