A flue gas denitrification device abnormality monitoring method and system

Abstract

The present invention discloses a method and system for monitoring abnormalities of a flue gas denitrification device, and the method includes data collection, data preprocessing, establishing a flue gas denitrification device operation monitoring model, and monitoring abnormalities of a flue gas denitrification device. The present invention belongs to the field of device operation monitoring, and specifically refers to a method and system for monitoring abnormalities of a flue gas denitrification device. This scheme introduces Lévy noise to describe sudden disturbances and Gaussian noise to describe continuous disturbances, thereby enhancing the adaptability to complex disturbances in actual working conditions, thereby improving the accuracy of the final abnormal monitoring of the flue gas denitrification device; based on dynamic gain, the ability to quickly adjust large deviations is improved, while suppressing excessive adjustment of small deviations; based on nonlinear terms, feedback adjustment is accurately performed for different abnormal degrees, and time lags and cumulative effects are fully considered to better handle complex dynamic environments; and by dynamically adjusting the convergence time, a higher monitoring sensitivity is obtained, thereby improving the abnormal monitoring effect of the flue gas denitrification device.

Description

Technical Field

[0001] The present invention relates to the technical field of device operation monitoring, and in particular to a method and system for monitoring abnormality of a flue gas denitration device. Background Art

[0002] The abnormal monitoring method of the flue gas denitrification device is a systematic means for real-time collection, analysis, processing and monitoring of various operating parameters generated by the flue gas denitrification device during operation, so as to detect whether the device is in normal operation, mild abnormal state or serious abnormal state. Its goal is to ensure the stability and operation efficiency of the denitrification device by monitoring the operation data and detecting abnormalities. However, the general abnormal monitoring method of the flue gas denitrification device has insufficient adaptability to complex disturbances in actual working conditions, which leads to the problem of poor accuracy of abnormal monitoring of the flue gas denitrification device; the general abnormal monitoring method of the flue gas denitrification device has the problem of being unable to accurately reflect the impact of changes in working conditions, poor sensitivity to abnormal deviations, and insufficient consideration of the dynamic changes of deviations, which leads to the problem of poor abnormal monitoring effect of the flue gas denitrification device. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a method and system for monitoring abnormalities of a flue gas denitrification device. In view of the fact that the general method for monitoring abnormalities of a flue gas denitrification device is not adaptable enough to complex disturbances in actual working conditions, which leads to poor accuracy of abnormal monitoring of the flue gas denitrification device, the present invention introduces Lévy noise to describe sudden disturbances and Gaussian noise to describe continuous disturbances, so as to enhance the adaptability to complex disturbances in actual working conditions, thereby improving the accuracy of abnormal monitoring of the flue gas denitrification device; in view of the fact that the general method for monitoring abnormalities of a flue gas denitrification device is unable to accurately reflect the influence of changes in working conditions, has poor sensitivity to abnormal deviations, and does not consider the dynamic changes of deviations sufficiently, thereby leading to poor abnormal monitoring effect of the flue gas denitrification device, the present invention improves the ability to quickly adjust large deviations based on dynamic gain, while suppressing excessive adjustment of small deviations, thereby enhancing the ability to respond to abnormal deviations; based on nonlinear terms, feedback adjustment is accurately performed for different abnormal degrees, time lags and cumulative effects are fully considered, and complex dynamic environments are better handled; and by dynamically adjusting the convergence time, a higher monitoring sensitivity is obtained, thereby improving the abnormal monitoring effect of the flue gas denitrification device.

[0004] The technical solution adopted by the present invention is as follows: The present invention provides a method for monitoring abnormality of a flue gas denitrification device, the method comprising the following steps:

[0005] Step S1: data collection;

[0006] Step S2: data preprocessing;

[0007] Step S3: Establishing a flue gas denitrification device operation monitoring model;

[0008] Step S4: abnormality monitoring of flue gas denitrification device.

[0009] Furthermore, in step S1, the data collection is to collect historical flue gas denitrification device operation data; the historical flue gas denitrification device operation data includes flue gas characteristic data, system operation parameters, environmental parameters, operation time and device operation status; the device operation status is used as a data label; the device operation status includes normal operation, mild abnormal operation and severe abnormal operation.

[0010] Further, in step S2, the data preprocessing is to clean the collected data, convert the data, standardize the data and correct the disturbance; the data cleaning is to process the missing values ​​and duplicate values; the data conversion is to convert the data into a vector form; the standardization is based on the maximum and minimum normalization method with data standardization; the disturbance correction is due to the randomness of the ambient temperature and pressure in the actual flue gas denitrification operation, so the ambient pressure and ambient temperature are introduced with sudden disturbance correction and continuous disturbance correction, which is expressed as: ; ;in, and is the standardized historical environmental pressure and temperature; and are the ambient pressure and ambient temperature after introducing disturbance correction; is the Lévy intensity coefficient; It is the increment of the Lévy process, which is used to describe the mutation characteristics in the real environment; is the Gaussian noise intensity; N(0,1) is the standard normal distribution, which is used to simulate continuous disturbances.

[0011] Further, in step S3, the flue gas denitrification device operation monitoring model is established by processing the pre-processed historical flue gas denitrification device operation data, which specifically includes the following steps:

[0012] Step S31: construct a dynamic sub-model of flue gas denitrification; use the dynamic model of random cellular neural network, combined with nonlinear excitation function and feedback control strategy, to describe the evolution of various parameters of flue gas denitrification device over time, take the typical value of the operating parameter as the ideal operating data, and the deviation between the ideal operating data and the actual operating data as the monitoring parameter deviation; the dynamic sub-model of flue gas denitrification is expressed as follows:

[0013] ; In the formula, and are the deviations of the ith and jth monitoring parameter variables, respectively; t is the running time; L is the total number of monitoring parameter variables; is the rate of change of deviation; is the attenuation coefficient; , , and is the coupling weight, which indicates the mutual influence strength between parameters; F(·) is the nonlinear excitation function, and the tanh function is selected; is the time lag delay; is a feedback controller; is Lévy noise; , γ and It is a dynamic adjustment parameter used to simulate the impact of changes in operating conditions;

[0014] Step S32: design a feedback control strategy; based on the abnormal characteristics of flue gas denitrification, design a nonlinear feedback controller; the formula used is as follows:

[0015] ; ;

[0016] In the formula, is the linear feedback gain; , and is a nonlinear control parameter used to adjust the weights of different nonlinear feedback terms; and is the gain coefficient; p and q are the nonlinear suppression degrees for mild anomalies and severe anomalies, respectively, where p∈(0,1), q>1; is the gain adjustment factor;

[0017] Step S33: synchronization criterion: Based on the synchronization criterion under Lévy noise, the dynamic sub-model of flue gas denitrification is ensured to achieve synchronous switching between normal and abnormal operating states within a fixed time. The formula used is as follows:

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] in, It is the threshold value of the synchronous stability criterion, which is used to ensure that the flue gas denitrification dynamic sub-model works stably under the actual flue gas denitrification conditions; is the noise intensity parameter; is the correction coefficient for the effect of time delay on synchronization; is the controller gain for severe anomalies; and are coupling weight parameters, which respectively represent the dynamic coupling relationship of the j-th monitoring parameter to the ith monitoring parameter and the dynamic coupling relationship of the ith monitoring parameter to the j-th monitoring parameter;

[0024] Step S34: Calculate the convergence time; Calculate the convergence time for the flue gas denitrification dynamic monitoring model to reach synchronization , the formula used is as follows:

[0025] ;

[0026] ;

[0027] in, is the control correction amount; is the adjustment factor;

[0028] Step S35: Run monitoring; a maximum convergence time T is preset. , the device is judged to be operating normally; otherwise, if there is , then the device operation status is judged to be slightly abnormal, otherwise the device operation status is judged to be seriously abnormal; is the deviation threshold;

[0029] Step S36: loss function design; construct deviation convergence loss , expressed as: ; Build synchronization performance loss , expressed as:

[0030] ; Construct classification accuracy loss , expressed as: ; Obtain the loss function of the flue gas denitrification device operation monitoring model , expressed as ; Where N is the total number of samples; C1 is the total category; is the true label of the vth sample; is the predicted label; c is the category index; the parameters of the flue gas denitrification device operation monitoring model are updated using the gradient descent algorithm;

[0031] Step S37: model determination; the pre-processed historical flue gas denitrification device operation data is divided into a training set and a test set; when the flue gas denitrification device operation monitoring model converges with the loss of the training set, the flue gas denitrification device operation monitoring model training is completed; a prediction threshold is set in advance, when the prediction accuracy of the trained flue gas denitrification device operation monitoring model for the test set is higher than the prediction threshold, the flue gas denitrification device operation monitoring model is established, otherwise the initial flue gas denitrification device operation monitoring model parameters are optimized based on the particle swarm search algorithm, and the data set is re-divided for training.

[0032] Furthermore, in step S4, the abnormal monitoring of the flue gas denitrification device is based on the established flue gas denitrification device operation monitoring model, and the flue gas characteristic data, system operation parameters, environmental parameters and operation time are collected in real time, and are input into the flue gas denitrification device operation monitoring model after preprocessing. When the device operation status output by the model is seriously abnormal, early warning processing is performed.

[0033] The present invention provides a flue gas denitration device abnormality monitoring system, comprising a data acquisition module, a data preprocessing module, a flue gas denitration device operation monitoring model establishment module and a flue gas denitration device abnormality monitoring module;

[0034] The data acquisition module collects historical flue gas denitrification device operation data and sends the data to the data preprocessing module;

[0035] The data preprocessing module performs data cleaning, data conversion, standardization processing and disturbance correction on the collected data; and sends the data to the flue gas denitrification device operation monitoring model establishment module;

[0036] The flue gas denitrification device operation monitoring model establishment module builds a dynamic monitoring model, combines nonlinear feedback control strategy, synchronization criteria and convergence time calculation, models, controls and optimizes the deviation of the flue gas denitrification device operation state, so as to achieve accurate monitoring of the device operation state; and sends the data to the flue gas denitrification device abnormality monitoring module;

[0037] The flue gas denitrification device abnormality monitoring module realizes the operation monitoring of the flue gas denitrification device based on the real-time collected data based on the established flue gas denitrification device operation monitoring model.

[0038] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0039] (1) In view of the problem that the general abnormal monitoring method of flue gas denitrification equipment is not adaptable enough to the complex disturbances in actual working conditions, which leads to poor accuracy of abnormal monitoring of flue gas denitrification equipment, this scheme introduces Lévy noise to describe sudden disturbances and Gaussian noise to describe continuous disturbances, so as to enhance the adaptability to complex disturbances in actual working conditions and thus improve the accuracy of abnormal monitoring of flue gas denitrification equipment.

[0040] (2) In view of the fact that the general abnormal monitoring method of flue gas denitrification equipment cannot accurately reflect the impact of changes in operating conditions, has poor sensitivity to abnormal deviations, and does not take into account the dynamic changes of deviations, which leads to poor abnormal monitoring effect of flue gas denitrification equipment, this scheme improves the ability to quickly adjust large deviations based on dynamic gain, while suppressing excessive adjustment of small deviations and enhancing the response ability to abnormal deviations; based on nonlinear terms, it accurately performs feedback adjustment for different abnormal degrees, comprehensively considers time lag and cumulative effects, and better handles complex dynamic environments; and by dynamically adjusting the convergence time, it obtains higher monitoring sensitivity and improves the abnormal monitoring effect of flue gas denitrification equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a flow chart of a method for monitoring abnormality of a flue gas denitration device provided by the present invention;

[0042] Figure 2 A schematic diagram of a flue gas denitration device abnormality monitoring system provided by the present invention;

[0043] Figure 3 It is a schematic diagram of the process of step S3.

[0044] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0046] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0047] Example 1, see Figure 1 The present invention provides a method for monitoring abnormality of a flue gas denitration device, the method comprising the following steps:

[0048] Step S1: data collection, collecting historical flue gas denitrification device operation data;

[0049] Step S2: Data preprocessing, data cleaning, data conversion, standardization and disturbance correction of the collected data;

[0050] Step S3: Establishing a flue gas denitrification device operation monitoring model, by constructing a dynamic monitoring model, combining nonlinear feedback control strategy, synchronization criterion and convergence time calculation, modeling, controlling and optimizing the deviation of the flue gas denitrification device operation state, so as to achieve accurate monitoring of the device operation state;

[0051] Step S4: abnormality monitoring of the flue gas denitrification device, based on the established flue gas denitrification device operation monitoring model, the operation monitoring of the flue gas denitrification device is realized based on the real-time collected data.

[0052] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the historical flue gas denitrification device operation data includes flue gas characteristic data, system operation parameters, environmental parameters, operation time and device operation status; the device operation status is used as a data label; the flue gas characteristic data includes flue gas inlet temperature, flue gas outlet temperature, nitrogen oxide concentration, ammonia escape concentration and oxygen content; the system operation parameters include SCR system pressure, fan flow, ammonia injection control valve opening and catalyst activity; the environmental parameters include ambient temperature and ambient pressure; the device operation status includes normal operation, mild abnormal operation and severe abnormal operation.

[0053] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S2, data cleaning is to process missing values ​​and duplicate values; data conversion is to convert data into vector form; standardization is based on the maximum and minimum normalization method with data standardization; disturbance correction is due to the randomness of ambient temperature and pressure in actual flue gas denitrification operation, so sudden disturbance correction and continuous disturbance correction are introduced to ambient pressure and ambient temperature, which are expressed as: ; ;in, and is the standardized historical environmental pressure and temperature; and are the ambient pressure and ambient temperature after introducing disturbance correction; is the Lévy intensity coefficient; It is the increment of the Lévy process, which is used to describe the mutation characteristics in the real environment; is the Gaussian noise intensity; N(0,1) is the standard normal distribution, which is used to simulate continuous disturbances.

[0054] By performing the above operations, the general flue gas denitrification device abnormal monitoring method has insufficient adaptability to complex disturbances in actual working conditions, which leads to poor accuracy of abnormal monitoring of flue gas denitrification devices. This scheme introduces Lévy noise to describe sudden disturbances and Gaussian noise to describe continuous disturbances, thereby enhancing the adaptability to complex disturbances in actual working conditions and improving the accuracy of abnormal monitoring of flue gas denitrification devices.

[0055] Example 4, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S3, the flue gas denitrification device operation monitoring model is established by processing the pre-processed historical flue gas denitrification device operation data, which specifically includes the following steps:

[0056] Step S31: construct a dynamic sub-model of flue gas denitrification; use the dynamic model of random cellular neural network, combined with nonlinear excitation function and feedback control strategy, to describe the evolution of various parameters of flue gas denitrification device over time, take the typical value of the operating parameter as the ideal operating data, and the deviation between the ideal operating data and the actual operating data as the monitoring parameter deviation; the dynamic sub-model of flue gas denitrification is expressed as follows:

[0057] ; In the formula, and are the deviations of the ith and jth monitoring parameter variables, respectively; t is the running time; L is the total number of monitoring parameter variables; is the rate of change of deviation; is the attenuation coefficient; , , and is the coupling weight, which indicates the mutual influence strength between parameters; F(·) is the nonlinear excitation function, and the tanh function is selected; is the time lag delay; is a feedback controller; is Lévy noise; , γ and It is a dynamic adjustment parameter used to simulate the impact of changes in operating conditions;

[0058] Step S32: design a feedback control strategy; based on the abnormal characteristics of flue gas denitrification, design a nonlinear feedback controller; the formula used is as follows:

[0059] ; ;

[0060] In the formula, is the linear feedback gain; , and is a nonlinear control parameter used to adjust the weights of different nonlinear feedback terms; and is the gain coefficient; p and q are the nonlinear suppression degrees for mild anomalies and severe anomalies, respectively, where p∈(0,1), q>1; is the gain adjustment factor;

[0061] Step S33: synchronization criterion: Based on the synchronization criterion under Lévy noise, the dynamic sub-model of flue gas denitrification is ensured to achieve synchronous switching between normal and abnormal operating states within a fixed time. The formula used is as follows:

[0062] ;

[0063] ;

[0064] ;

[0065] ;

[0066] ;

[0067] in, It is the threshold of the synchronous stability criterion, which is used to ensure that the dynamic sub-model of flue gas denitrification works stably under the actual flue gas denitrification conditions; is the noise intensity parameter; is the correction coefficient for the effect of time delay on synchronization; is the controller gain for severe anomalies; and are coupling weight parameters, which respectively represent the dynamic coupling relationship of the j-th monitoring parameter to the ith monitoring parameter and the dynamic coupling relationship of the ith monitoring parameter to the j-th monitoring parameter;

[0068] Step S34: Calculate the convergence time; Calculate the convergence time for the flue gas denitrification dynamic monitoring model to reach synchronization , the formula used is as follows:

[0069] ;

[0070] ;

[0071] in, is the control correction amount; is the adjustment factor;

[0072] Step S35: Run monitoring; a maximum convergence time T is preset. , the device is judged to be operating normally; otherwise, if there is , then the device operation status is judged to be slightly abnormal, otherwise the device operation status is judged to be seriously abnormal; is the deviation threshold;

[0073] Step S36: loss function design; construct deviation convergence loss , expressed as: ; Build synchronization performance loss , expressed as:

[0074] ; Construct classification accuracy loss , expressed as: ; Obtain the loss function of the flue gas denitrification device operation monitoring model , expressed as ; Where N is the total number of samples; C1 is the total category; is the true label of the vth sample; is the predicted label; c is the category index; the parameters of the flue gas denitrification device operation monitoring model are updated using the gradient descent algorithm;

[0075] Step S37: model determination; the pre-processed historical flue gas denitrification device operation data is divided into a training set and a test set; when the flue gas denitrification device operation monitoring model converges with the loss of the training set, the flue gas denitrification device operation monitoring model training is completed; a prediction threshold is set in advance, when the prediction accuracy of the trained flue gas denitrification device operation monitoring model for the test set is higher than the prediction threshold, the flue gas denitrification device operation monitoring model is established, otherwise the initial flue gas denitrification device operation monitoring model parameters are optimized based on the particle swarm search algorithm, and the data set is re-divided for training.

[0076] By performing the above operations, the general abnormal monitoring method of flue gas denitrification device cannot accurately reflect the impact of changes in operating conditions, has poor sensitivity to abnormal deviations, and does not consider the dynamic changes of deviations, which leads to poor abnormal monitoring effect of flue gas denitrification device. This scheme improves the ability to quickly adjust large deviations based on dynamic gain, while suppressing excessive adjustment of small deviations and enhancing the response ability to abnormal deviations; accurately performs feedback adjustment for different degrees of abnormality based on nonlinear terms, comprehensively considers time lag and cumulative effects, and better handles complex dynamic environments; and obtains higher monitoring sensitivity by dynamically adjusting the convergence time, thereby improving the abnormal monitoring effect of flue gas denitrification device.

[0077] Example 5, see Figure 1This embodiment is based on the above embodiment. In step S4, the abnormal monitoring of the flue gas denitrification device is based on the established flue gas denitrification device operation monitoring model. The flue gas characteristic data, system operation parameters, environmental parameters and operation time are collected in real time, and are input into the flue gas denitrification device operation monitoring model after preprocessing. When the device operation status output by the model is seriously abnormal, early warning processing is performed.

[0078] Example 6, see Figure 2 , this embodiment is based on the above embodiment, and a flue gas denitrification device abnormality monitoring system provided by the present invention includes a data acquisition module, a data preprocessing module, a flue gas denitrification device operation monitoring model establishment module and a flue gas denitrification device abnormality monitoring module;

[0079] The data acquisition module collects historical flue gas denitrification device operation data and sends the data to the data preprocessing module;

[0080] The data preprocessing module performs data cleaning, data conversion, standardization processing and disturbance correction on the collected data; and sends the data to the flue gas denitrification device operation monitoring model establishment module;

[0081] The flue gas denitrification device operation monitoring model establishment module builds a dynamic monitoring model, combines nonlinear feedback control strategy, synchronization criteria and convergence time calculation, models, controls and optimizes the deviation of the flue gas denitrification device operation state, so as to achieve accurate monitoring of the device operation state; and sends the data to the flue gas denitrification device abnormality monitoring module;

[0082] The flue gas denitrification device abnormality monitoring module realizes the operation monitoring of the flue gas denitrification device based on the real-time collected data based on the established flue gas denitrification device operation monitoring model.

[0083] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0084] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0085] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A method for monitoring abnormality of a flue gas denitrification device, characterized in that: The method comprises the following steps: Step S1: data collection, collecting historical flue gas denitrification device operation data; Step S2: Data preprocessing, data cleaning, data conversion, standardization and disturbance correction of the collected data; Step S3: Establishing a flue gas denitrification device operation monitoring model; Step S4: abnormality monitoring of the flue gas denitrification device, realizing operation monitoring of the flue gas denitrification device based on the real-time collected data based on the established flue gas denitrification device operation monitoring model; Step S3 includes step S31: constructing a flue gas denitrification dynamic sub-model; using a dynamic model of a random cell neural network, combined with a nonlinear excitation function and a feedback control strategy, describing the evolution of various parameters of the flue gas denitrification device over time, taking the typical value of the operating parameter as the ideal operating data, and the deviation between the ideal operating data and the actual operating data as the monitoring parameter deviation; the flue gas denitrification dynamic sub-model is expressed as follows: ； In the formula, and are the deviations of the ith and jth monitoring parameter variables, respectively; t is the running time; L is the total number of monitoring parameter variables; is the rate of change of deviation; is the attenuation coefficient; , , and is the coupling weight, which indicates the mutual influence strength between parameters; F(·) is the nonlinear excitation function, and the tanh function is selected; is the time lag delay; is a feedback controller; is Lévy noise; , γ and It is a dynamic adjustment parameter used to simulate the impact of changes in operating conditions.

2. The method for monitoring abnormality of a flue gas denitration device according to claim 1, characterized in that: In step S3, the flue gas denitrification device operation monitoring model is established by processing the pre-processed historical flue gas denitrification device operation data, which specifically includes the following steps: Step S31: constructing a flue gas denitrification dynamic sub-model; Step S32: design a feedback control strategy; based on the abnormal characteristics of flue gas denitrification, design a nonlinear feedback controller; the formula used is as follows: ； ； In the formula, is the linear feedback gain; , and is a nonlinear control parameter used to adjust the weights of different nonlinear feedback terms; and is the gain coefficient; p and q are the nonlinear suppression degrees for mild anomalies and severe anomalies, respectively, where p∈(0,1), q>1; is the gain adjustment factor; Step S33: synchronization criterion: Based on the synchronization criterion under Lévy noise, the dynamic sub-model of flue gas denitrification is ensured to achieve synchronous switching between normal and abnormal operating states within a fixed time. The formula used is as follows: ； ； ； ； ； in, It is the threshold value of the synchronous stability criterion, which is used to ensure that the flue gas denitrification dynamic sub-model works stably under the actual flue gas denitrification conditions; is the noise intensity parameter; is the correction coefficient for the effect of time delay on synchronization; is the controller gain for severe anomalies; and are coupling weight parameters, which respectively represent the dynamic coupling relationship of the j-th monitoring parameter to the ith monitoring parameter and the dynamic coupling relationship of the ith monitoring parameter to the j-th monitoring parameter; Step S34: Calculate the convergence time; Calculate the convergence time for the flue gas denitrification dynamic monitoring model to reach synchronization , the formula used is as follows: ； ； in, is the control correction amount; is the adjustment factor; Step S35: Run monitoring; a maximum convergence time T is preset. , the device is judged to be operating normally; otherwise, if there is , then the device operation status is judged to be slightly abnormal, otherwise the device operation status is judged to be seriously abnormal; is the deviation threshold; Step S36: loss function design; construct deviation convergence loss , expressed as: ; Build synchronization performance loss , expressed as: ； Constructing classification accuracy loss , expressed as: ; Obtain the loss function of the flue gas denitrification device operation monitoring model , expressed as ; Where N is the total number of samples; C1 is the total category; is the true label of the vth sample; is the predicted label; c is the category index; the parameters of the flue gas denitrification device operation monitoring model are updated using the gradient descent algorithm; Step S37: model determination; the pre-processed historical flue gas denitrification device operation data is divided into a training set and a test set; when the flue gas denitrification device operation monitoring model converges with the loss of the training set, the flue gas denitrification device operation monitoring model training is completed; a prediction threshold is set in advance, when the prediction accuracy of the trained flue gas denitrification device operation monitoring model for the test set is higher than the prediction threshold, the flue gas denitrification device operation monitoring model is established, otherwise the initial flue gas denitrification device operation monitoring model parameters are optimized based on the particle swarm search algorithm, and the data set is re-divided for training.

3. The method for monitoring abnormality of a flue gas denitration device according to claim 2, characterized in that: In step S2, the disturbance correction is due to the randomness of the ambient temperature and pressure in the actual flue gas denitrification operation, so sudden disturbance correction and continuous disturbance correction are introduced to the ambient pressure and ambient temperature, which are expressed as: ; ;in, and is the standardized historical environmental pressure and temperature; and are the ambient pressure and ambient temperature after introducing disturbance correction; is the Lévy intensity coefficient; It is the increment of the Lévy process, which is used to describe the mutation characteristics in the real environment; is the Gaussian noise intensity; N(0,1) is the standard normal distribution used to simulate continuous disturbances.

4. The method for monitoring abnormality of a flue gas denitration device according to claim 1, characterized in that: In step S1, the historical flue gas denitrification device operation data includes flue gas characteristic data, system operation parameters, environmental parameters, operation time and device operation status; the device operation status is used as a data label; the device operation status includes normal operation, slightly abnormal operation and seriously abnormal operation.

5. The method for monitoring abnormality of a flue gas denitration device according to claim 1, characterized in that: In step S4, the abnormal monitoring of the flue gas denitrification device is based on the established flue gas denitrification device operation monitoring model, and the flue gas characteristic data, system operation parameters, environmental parameters and operation time are collected in real time, and input into the flue gas denitrification device operation monitoring model after preprocessing. When the device operation status output by the model is seriously abnormal, early warning processing is performed.

6. The method for monitoring abnormality of a flue gas denitration device according to claim 1, characterized in that: In step S2, the data cleaning is to process missing values ​​and duplicate values; the data conversion is to convert the data into a vector form; and the standardization processing is based on the maximum and minimum normalization method and data standardization.

7. A flue gas denitration device abnormality monitoring system, used to implement a flue gas denitration device abnormality monitoring method as described in any one of claims 1 to 6, characterized in that: It includes a data acquisition module, a data preprocessing module, a flue gas denitrification device operation monitoring model establishment module and a flue gas denitrification device abnormality monitoring module; The data acquisition module collects historical flue gas denitrification device operation data and sends the data to the data preprocessing module; The data preprocessing module performs data cleaning, data conversion, standardization processing and disturbance correction on the collected data; and sends the data to the flue gas denitrification device operation monitoring model establishment module; The flue gas denitrification device operation monitoring model establishment module models, controls and optimizes the deviation of the flue gas denitrification device operation state by constructing a dynamic monitoring model, combining nonlinear feedback control strategy, synchronization criteria and convergence time calculation, so as to realize accurate monitoring of the device operation state; And send the data to the abnormal monitoring module of the flue gas denitrification device; The flue gas denitrification device abnormality monitoring module realizes the operation monitoring of the flue gas denitrification device based on the real-time collected data based on the established flue gas denitrification device operation monitoring model.

Images