CN116060201B - Deflagration monitoring abnormality positioning and identifying method and system for coal mill of thermal power station - Google Patents
Deflagration monitoring abnormality positioning and identifying method and system for coal mill of thermal power station Download PDFInfo
- Publication number
- CN116060201B CN116060201B CN202310212801.3A CN202310212801A CN116060201B CN 116060201 B CN116060201 B CN 116060201B CN 202310212801 A CN202310212801 A CN 202310212801A CN 116060201 B CN116060201 B CN 116060201B
- Authority
- CN
- China
- Prior art keywords
- monitoring
- association
- data set
- sensor
- 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.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C25/00—Control arrangements specially adapted for crushing or disintegrating
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention discloses a deflagration monitoring abnormality positioning and identifying method and system for a coal mill of a thermal power station, and relates to the technical field of data processing, wherein the method comprises the following steps: information acquisition is carried out on a pulverizing system of a target thermal power station, so as to obtain a basic information set; acquiring historical monitoring information of a monitoring sensor according to an industrial control system of a target thermal power station; performing association locking on the historical sensor data set according to the basic information set; constructing a deflagration anomaly monitoring model; acquiring real-time data of a monitoring sensor in a preset time window to obtain a real-time monitoring data set; inputting the deflagration anomaly monitoring model to obtain anomaly monitoring results; and generating abnormal reminding information, and carrying out positioning identification on the deflagration monitoring abnormality. The invention solves the technical problems of long deflagration monitoring feedback period and low anomaly identification accuracy of the coal mill of the thermal power station in the prior art, and achieves the technical effects of carrying out association locking on sensor data and improving anomaly identification efficiency and intelligent degree.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a method and a system for positioning and identifying deflagration monitoring abnormality of a coal mill of a thermal power station.
Background
In recent years, as the amount of electric energy required is increasing, energy problems are becoming urgent. Although non-fossil energy is being greatly developed in the demand of low carbon and environmental protection, thermal power still occupies most of the market for electric power. Therefore, research on the production condition of the thermal power station has very important significance for ensuring the power supply.
At present, in the thermal power generation production process, the danger coefficient of the deflagration phenomenon of the coal mill in a plurality of abnormal production conditions is higher, once an internal deflagration accident occurs, the damage of powder making system equipment is caused by light weight, and the casualties can be caused when the damage is too serious. The coal mill has excessive impact factors of deflagration, and the occurring process is relatively short, and the abnormality recognition is mainly carried out by utilizing a monitoring sensor.
However, since the number of items to be monitored is excessive in the actual thermal power generation process, the abnormal data of the monitoring sensor cannot be timely obtained by monitoring personnel, so that the abnormal feedback time is long, the abnormal period is long, and serious economic loss and personnel danger are caused. For various types of monitoring data, an abnormality identification method capable of rapidly identifying and responding to abnormal data, feeding back reasons of the abnormality and rapidly positioning the monitored abnormality is lacking currently. In the prior art, the technical problems of long deflagration monitoring feedback period and low anomaly identification accuracy of the coal mill of the thermal power station exist.
Disclosure of Invention
The application provides a deflagration monitoring anomaly positioning identification method and system for a coal mill in a thermal power station, which are used for solving the technical problems of long deflagration monitoring feedback period and low anomaly identification accuracy of the coal mill in the thermal power station in the prior art.
In view of the above problems, the application provides a method and a system for positioning and identifying abnormal deflagration monitoring of a coal mill in a thermal power station.
In a first aspect of the application, a method for positioning and identifying deflagration monitoring abnormality of a coal mill in a thermal power station is provided, and the method comprises the following steps:
information acquisition is carried out on a pulverizing system of a target thermal power station, so as to obtain a basic information set;
acquiring historical monitoring information of a monitoring sensor according to an industrial control system of the target thermal power station, and acquiring a historical sensor data set, wherein the historical sensor data set is provided with double identifications, and the double identifications comprise sensor position identifications and sensor type identifications;
performing association locking on the historical sensor data set according to the basic information set to obtain an association locking result;
constructing a deflagration anomaly monitoring model according to the basic information set and the associated locking result;
acquiring real-time data of a monitoring sensor in a preset time window to obtain a real-time monitoring data set;
Inputting the real-time monitoring data set into the deflagration anomaly monitoring model to obtain anomaly monitoring results;
and generating abnormal reminding information according to the abnormal monitoring result, and carrying out positioning identification on the deflagration monitoring abnormality.
In a second aspect of the present application, there is provided a deflagration monitoring anomaly locating and identifying system for a coal pulverizer in a thermal power plant, the system comprising:
the basic information acquisition module is used for acquiring information of a pulverizing system of the target thermal power station to obtain a basic information set;
the historical data set obtaining module is used for obtaining historical monitoring information of a monitoring sensor according to an industrial control system of the target thermal power station to obtain a historical sensor data set, wherein the historical sensor data set is provided with dual identifications, and the dual identifications comprise sensor position identifications and sensor type identifications;
the association locking module is used for carrying out association locking on the historical sensor data set according to the basic information set to obtain an association locking result;
the monitoring model construction module is used for constructing a deflagration abnormal monitoring model according to the basic information set and the association locking result;
The monitoring data set acquisition module is used for acquiring real-time data of the monitoring sensor in a preset time window to obtain a real-time monitoring data set;
the monitoring result obtaining module is used for inputting the real-time monitoring data set into the deflagration abnormal monitoring model to obtain an abnormal monitoring result;
and the abnormality reminding module is used for generating abnormality reminding information according to the abnormality monitoring result and carrying out positioning identification on the deflagration monitoring abnormality.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
according to the embodiment of the application, the basic information set capable of reflecting the basic production condition of the target thermal power station is obtained through collecting the information of the pulverizing system of the target thermal power station, then the industrial control system of the target thermal power station is in communication connection, the history monitoring information of the monitoring sensor is obtained, the history sensor data set is obtained, double identification is carried out on the history sensor data set, the double identification comprises the sensor position identification and the sensor type identification, then the history sensor data set is subjected to association locking according to the basic information set, an association locking result is obtained, further, a deflagration anomaly monitoring model for intelligently monitoring anomalies is constructed through utilizing the basic information set and the association locking result, then real-time data collection is carried out on the monitoring sensor in a preset time window, a real-time monitoring data set is obtained, the anomaly monitoring result is obtained through inputting the real-time monitoring data set into the deflagration anomaly monitoring model, the anomaly reminding information is generated according to the anomaly monitoring result, and the positioning and identification are carried out on the monitoring anomalies. The method has the advantages that high-quality and high-efficiency analysis is carried out on the data of the deflagration monitoring sensor of the coal mill in the thermal power station, the data processing efficiency is improved, and the abnormal conditions are fed back quickly.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a method for positioning and identifying abnormal deflagration monitoring of a coal mill in a thermal power station according to an embodiment of the present application;
fig. 2 is a schematic flow chart of performing association locking on a historical sensor data set in a method for positioning and identifying abnormal deflagration monitoring of a coal mill in a thermal power station according to an embodiment of the present application;
fig. 3 is a schematic flow chart of constructing a deflagration anomaly monitoring model in the deflagration monitoring anomaly positioning and identifying method of the coal mill in the thermal power station according to the embodiment of the present application;
fig. 4 is a schematic structural diagram of a deflagration monitoring abnormality positioning and identifying system of a coal mill in a thermal power station according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a basic information obtaining module 11, a historical data set obtaining module 12, an association locking module 13, a monitoring model constructing module 14, a monitoring data set obtaining module 15, a monitoring result obtaining module 16 and an abnormality reminding module 17.
Detailed Description
The utility model provides a deflagration monitoring anomaly location identification method for a coal mill in a thermal power station, which is used for solving the technical problems of long deflagration monitoring feedback period and low anomaly identification accuracy rate of the coal mill in the thermal power station in the prior art.
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It will be apparent that the described embodiments are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, are intended to be within the scope of the present application.
It should be noted that the terms "comprises" and "comprising," along with any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or server that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed or inherent to such process, method, article, or apparatus, but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
As shown in fig. 1, the application provides a method for positioning and identifying abnormal deflagration monitoring of a coal mill in a thermal power station, wherein the method comprises the following steps:
step S100: information acquisition is carried out on a pulverizing system of a target thermal power station, so as to obtain a basic information set;
specifically, the target thermal power station is any thermal power station for analyzing and monitoring deflagration abnormality of the coal mill. The coal pulverizing system is an important auxiliary system of a coal-fired unit of the thermal power station, and a hearth is conveyed by grinding qualified coal powder, so that combustion of a boiler is guaranteed, and the coal pulverizing system mainly comprises a raw coal bin, a coal feeder, a coal mill, a sealing fan, a primary fan, an auxiliary system and the like. The basic information set is a data set obtained after collecting the basic condition of a pulverizing system of the target thermal power station, and comprises pulverizing types (such as a direct-fired pulverizing system or an intermediate warehouse pulverizing system), the distribution position and the quantity of coal mills, the air quantity of a fan, the distribution position of the fan and the like. Basic information of a pulverizing system of the target thermal power station is acquired, so that the working environment of the coal mill is mastered, and the technical effects of providing analysis background and analysis data for subsequent analysis abnormality are achieved.
Step S200: acquiring historical monitoring information of a monitoring sensor according to an industrial control system of the target thermal power station, and acquiring a historical sensor data set, wherein the historical sensor data set is provided with double identifications, and the double identifications comprise sensor position identifications and sensor type identifications;
specifically, the industrial control system is a system for carrying out real-time management and real-time monitoring on working conditions in the production process by the target thermal power station. And extracting monitoring data of a detection sensor in the industrial control system to obtain the historical sensor data set. Wherein the historical sensor dataset reflects a production status of the target thermal power station in a production process. The dual identification is to identify the extracted historical sensor data according to the sensor position and the type of the sensor, so that clear distinguishing identification is provided for the data abnormality of the subsequent sensor. The sensor position identification is used for identifying the position of the sensor in the pulverizing system of the target thermal power station, and in order to ensure the accuracy of sensor data, the sensor is generally installed near a monitored object, so that the position of the sensor is used for identification, and the abnormal data is rapidly positioned for subsequent bedding. The sensor type identification refers to identification information of a sensor according to the purpose of the sensor, and the sensor type comprises a temperature sensor, a wind speed sensor, a wind pressure sensor, a coal dust concentration sensor and the like. Through double identification of the historical sensor data set, the technical effect of providing basis for multidimensional analysis and division of data is achieved.
Step S300: performing association locking on the historical sensor data set according to the basic information set to obtain an association locking result;
further, as shown in fig. 2, the performing association locking on the historical sensor data set according to the basic information set, step S300 in the embodiment of the present application further includes:
step S310: extracting coal mill distribution information based on the basic information set to obtain a coal mill distribution data set;
step S320: based on the sensor position identification, clustering the historical sensor data sets by combining the coal mill distribution data sets to obtain a sensor clustering result;
step S330: and performing association locking on the historical sensor data set according to the sensor clustering result to obtain the association locking result.
Further, the step S330 of the embodiment of the present application further includes:
step S331: performing dimension division on the sensor clustering result according to the sensor type identifier to obtain a dimension division result;
step S332: similar association is carried out on the dimension division results according to the sensor position identifiers, and similar association results are obtained;
Step S333: carrying out heterogeneous association on the dimension division result according to the basic information set to obtain a heterogeneous association result;
step S334: and performing association locking on the historical sensor data set according to the similar association results and the heterogeneous association results to obtain the association locking results.
Further, the step S334 of the embodiment of the present application further includes:
step S335: extracting the historical sensor data according to the similar association results and the heterogeneous association results to obtain an association data set;
step S336: obtaining the data change trend of each associated sensor according to the associated data set, and obtaining a plurality of trend change results;
step S337: evaluating the association degrees of the similar association results and the heterogeneous association results according to the trend change results to obtain an association degree grade;
step S338: and carrying out relevance identification on the relevance locking result by utilizing the relevance grade.
Specifically, the performing association locking on the historical sensor dataset according to the basic information set refers to performing association on data associated with the same type of deflagration reasons in the historical sensor dataset according to coal mill information in the basic information set, and locking connection relations among the data, so that rapid response to deflagration anomalies is performed, and the situation that failure reasons are checked one by one according to the anomaly data after deflagration phenomena occur, and time and resource waste are caused is avoided. The association locking result reflects the association relation between the data, and the association of the related sensor data is carried out on the reasons generated by the deflagration, so that the technical effect of improving the anomaly identification efficiency is achieved.
Specifically, extracting coal mill distribution information from the basic information set, grasping the distribution condition of the coal mills in the target thermal power station, and obtaining the coal mill distribution data after summarizing. The distribution data of the coal mills refer to the distribution quantity and the distribution positions of the coal mills in the thermal power station. The sensor clustering result is obtained by performing cluster analysis on a historical sensor data set and dividing sensor data for deflagration monitoring on the same coal mill into one class.
Specifically, the number of coal mills and the positions of the coal mills in the coal mill distribution data set are used as clustering labels, the historical sensor data set is input into the root node of the sensor position clustering tree, and the positions of the coal mills are used for setting leaf nodes, wherein the number of the leaf nodes is determined by the number of the coal mills. And inputting the historical sensor data set into the sensor position clustering tree, and comparing the sensor position identification with the coal mill position in the leaf node by taking the sensor position identification as a division identification, so that the historical sensor data in the same position is divided into the same sub-node and is stored. And taking the sensor data in each child node as a class, thereby obtaining the sensor clustering result.
Furthermore, the dimension division result is obtained by further subdividing the sensor clustering result according to the sensor type identification, namely the monitoring purpose of the sensor on the coal mill is a division standard. The similar association result is obtained when the monitoring targets are consistent according to the sensor position identification, namely, whether the monitoring targets of the sensor are consistent or not is determined according to the monitoring positions of the sensors. The heterogeneous association result is a result of determining association relations among sensors of different monitoring targets in the dimension division result according to the working state of the coal mill in the basic information set, and comprises the step of identifying the heterogeneous association result according to whether the data growth amplitude of the monitoring sensors is consistent.
Illustratively, the temperature sensor is used for collecting the temperature in the coal grinding process of the thermal power station in real time. The temperature sensor is arranged at the outlet of the coal mill and in the working space of the coal mill, so that the outlet temperature and the working temperature of the coal mill are collected in real time. According to the working principle of the coal mill, when the coal mill stops grinding, the outlet temperature of the coal mill is adapted to the ambient temperature in order to avoid the deflagration phenomenon of the coal mill. At this time, there is a similar correlation between the outlet temperature sensor data at the outlet and the ambient temperature sensor data. The data have a locking relation, and when the ambient temperature is lower than 15 ℃, the outlet temperature data of the coal mill should be within 60-65 ℃; when the ambient temperature is 15-30 ℃, the outlet temperature data of the coal mill is 57-60 ℃; when the ambient temperature is higher than 30 ℃, the outlet temperature data of the coal mill is within 54-58 ℃, so that spontaneous combustion of coal powder can be effectively avoided. When the ambient temperature is detected to be higher than 30 ℃, the outlet temperature data of the coal mill is further determined according to the similar incidence relation between the outlet temperature sensor data at the outlet and the ambient temperature sensor data, so that the abnormal deflagration condition of the coal mill is rapidly positioned.
Specifically, the historical sensor data are respectively extracted according to the similar association results and the heterogeneous association results to obtain the association data set. Wherein the associated data sets include homogeneous associated data sets and heterogeneous associated data sets. The similar association data set is obtained by extracting data according to the sensors in the similar association results. The heterogeneous association data set is obtained by extracting data according to the sensors in the heterogeneous association result.
Specifically, according to the similar associated data set and the heterogeneous associated data set, a data change trend, namely, the change degree of the ascending trend or the change degree of the descending trend of the data in unit time is carried out, so that a plurality of trend change results are obtained. Wherein the plurality of trend change results are in one-to-one correspondence with the sensors. And further, respectively carrying out trend consistency analysis on the similar association results and the heterogeneous association results according to the plurality of trend change results, namely judging the consistency degree of the trend change degree. The higher the corresponding degree of correlation level, the higher the degree of agreement. And carrying out relevance identification on the relevance locking results according to the relevance grade, so that for the case that the same sensor has a plurality of relevance locking results, sorting is carried out according to the relevance grade corresponding to each relevance locking result, and monitoring objects corresponding to the sensors with high relevance grade are firstly identified. Therefore, the technical effects of grading the association locking result, establishing a clear and hierarchical association structure and improving the abnormality identification efficiency are achieved.
Further, step S338 in the embodiment of the present application further includes:
step S3381: extracting a historical detonation record of the target thermal power station recorded in the industrial control system to obtain a historical detonation record data set;
step S3382: performing frequency statistics on the association locking result according to the historical deflagration record data set to obtain a frequency data set;
step S3383: randomly selecting one frequency data from the frequency data set without replacement as a primary dividing node;
step S3384: randomly selecting one frequency data from the frequency data set without replacement again to serve as a secondary dividing node;
step S3385: continuously constructing a multi-level partition node for obtaining a data cleaning unit;
step S3386: marking a plurality of final dividing nodes of the multi-stage dividing nodes according to the association locking result corresponding to the frequency data set to obtain the constructed data cleaning unit;
step S3387: inputting the frequency data set into the data cleaning unit, and determining the frequency data lower than a preset dividing node as an obsolete frequency data set;
step S3388: and screening and optimizing the association locking result according to the elimination frequency data set.
Specifically, the historical deflagration record data set is obtained by summarizing production record data of a target thermal power station when deflagration occurs in a historical manner, and comprises deflagration time, deflagration related equipment, deflagration reasons, deflagration data and the like. Illustratively, the coal mill D is started up and operated at 12 days 2016, 6 and the coal mill E is switched and shut down. The expansion joint of the hot air pipeline of the No. D coal mill is broken, the negative pressure of the hearth is greatly changed, the temperature of the coal mill after grinding is rapidly increased from 80 ℃ to 220 ℃, and the air pressure of an outlet is rapidly increased from 3.1kpa to 7.9kpa. And analyzing the deflagration record data to obtain the association relation between the temperature after grinding and the wind pressure of the outlet, wherein the temperature after grinding and the wind pressure of the outlet are rapidly increased due to the rupture of the expansion joint.
Specifically, the frequency data set is obtained by counting the frequency of occurrence of each associated locking result in the deflagration influence in the historical deflagration record data set. The preset occurrence frequency is set according to the production condition of the target thermal power station, and is set by the staff at will, without limitation. The obsolete frequency data sets are data sets that do not perform associative locking. When the frequency in the frequency data set is smaller than the preset occurrence frequency, the occurrence frequency is too low, which belongs to the accidental phenomenon, and the fault analysis is not considered. And then, eliminating the associated locking result object corresponding to the elimination frequency data set from the associated locking result, so as to optimize the associated locking result. The technical effects of improving the reliability of the associated locking result, reducing the operation data and improving the efficiency are achieved.
Specifically, the first-level dividing node is a node for dividing the data set by randomly selecting one frequency data from the frequency data set without replacement, and taking the frequency data as the basis of the node data division. The secondary partition node is a node for partitioning the data set by randomly selecting one frequency data from the frequency data set without replacement and taking the frequency data as the basis of the secondary node data partitioning. The multi-stage dividing node is a node for dividing data in multiple stages. The data cleaning unit is a functional unit for performing data screening on the frequency data set to obtain frequency data with low occurrence frequency, and comprises multiple levels of dividing nodes. And obtaining an association locking result corresponding to the frequency data by utilizing the frequency data corresponding to the multi-stage dividing nodes, and marking a plurality of final dividing nodes of the multi-stage dividing nodes, namely, each final dividing node corresponds to one association locking result. The preset dividing nodes are dividing nodes constructed according to preset data occurrence frequency, and if the preset dividing nodes are lower than the corresponding frequency of the nodes, the preset dividing nodes indicate that the corresponding frequency of occurrence of the associated locking result is too low, and the corresponding historical deflagration records are accidental and have no referential property. Therefore, the technical effects of cleaning the data of the frequency data set, ensuring the reliability of the data and improving the operation efficiency of data processing are achieved by screening the data set with lower frequency.
Step S400: constructing a deflagration anomaly monitoring model according to the basic information set and the associated locking result;
further, as shown in fig. 3, the constructing a knock anomaly monitoring model according to the basic information set and the associated locking result, step S400 of the embodiment of the present application further includes:
step S410: building a BP feedforward network structure, which comprises an input layer, an implicit layer and an output layer, wherein the implicit layer comprises an operation mode identification layer and an association result matching layer;
step S420: extracting information from the basic information data set, and establishing an operation mode training set, wherein the operation mode training set comprises historical operation mode data and corresponding mode type identification information;
step S430: training convergence is carried out on the operation mode identification layer by utilizing the operation mode training set until a preset training stopping requirement is reached;
step S440: according to the historical operation mode data, determining the corresponding relation between each operation mode data type and the associated locking result, and constructing an associated result matching training set;
step S450: and training and converging the association result matching layer by using the association result matching training set until reaching a preset training stopping requirement, and obtaining the deflagration anomaly monitoring model.
Specifically, the deflagration anomaly monitoring model is a functional model for intelligently identifying deflagration conditions of coal mills in the target thermal power station. The operation mode identification layer is a functional layer for identifying the operation state of the coal mill. The association result matching layer is a functional layer for efficiently matching the running state of the coal mill with the corresponding association locking result.
Specifically, the operation data of each coal mill in the basic information data set is extracted by taking the BP feedforward network structure as a construction basis, and the operation mode data is used as the operation mode training set. The operation mode training set is a data set for training the operation mode identification layer and comprises the historical operation mode data and corresponding mode type identification information, wherein the mode type identification information is used for identifying the operation state of the coal mill, and the operation state of the coal mill comprises shutdown, startup, operation work and the like. And training the operation mode identification layer by the operation mode training set until the output result of the operation mode identification layer reaches convergence, thereby meeting the preset training stopping requirement. The preset training stopping requirement means that the training result of the operation mode identification layer is verified according to the mode type identification information, and the accuracy rate is required to meet the requirement.
Specifically, the operation mode data type obtained in the operation model identification layer is input according to the history operation mode data, namely, history mode type information is output, a history association locking result is obtained according to the history mode type information, and an association relation between the operation state of the coal mill and the association locking result is determined, wherein the association relation refers to a mapping relation between the history mode type information and the history association locking result. And matching the historical operation mode data type with the historical association locking result and the association relation as the association result with a training set. The association result matching training set is used for training the association result matching layer until the output result of the operation mode identification layer reaches convergence, and meets the preset training stopping requirement.
Specifically, the input data of the operation mode identification layer is operation mode data, and the output data is mode type information. And the input data of the association result matching layer is mode type information, and the output data is an association locking result. At this time, the input layer, the hidden layer and the output layer are connected in series to obtain the deflagration anomaly monitoring model. Therefore, the intelligent determination of the relation between the deflagration abnormal data of the coal mill according to the association relation between the state of the coal mill and the sensor data is achieved, and the technical effect of supporting the subsequent rapid recognition feedback of the sensor abnormality is achieved.
Step S500: acquiring real-time data of a monitoring sensor in a preset time window to obtain a real-time monitoring data set;
further, the step S500 of the embodiment of the present application further includes:
step S510: collecting monitoring sensor data in the preset time window in real time to obtain an initial real-time monitoring data set;
step S520: acquiring monitoring sensor data in a last preset time window to obtain a comparison monitoring data set;
step S530: and carrying out data comparison on the comparison monitoring data set and the initial real-time monitoring data set to obtain the real-time monitoring data set.
Specifically, the preset time window is a time period for performing sensor data analysis, and is set by the staff at his own discretion, which is not limited herein. The real-time monitoring data set is obtained by collecting monitoring data of each sensor in a preset time window. The initial real-time monitoring data set is a data set which is directly acquired by monitoring data of each sensor in a preset time window and is not processed. The comparison monitoring data set is a data set obtained by data acquisition of the monitoring sensor in the last preset time window. And comparing the monitoring data set with the initial real-time monitoring data set to obtain data with abnormal data change, and screening the data to obtain the real-time monitoring data set. Therefore, the technical effect of screening the monitoring data and reducing the monitoring data quantity is achieved.
Step S600: inputting the real-time monitoring data set into the deflagration anomaly monitoring model to obtain anomaly monitoring results;
step S700: and generating abnormal reminding information according to the abnormal monitoring result, and carrying out positioning identification on the deflagration monitoring abnormality.
Specifically, the real-time monitoring data set is input into the deflagration anomaly detection model, anomaly analysis is carried out on the real-time monitoring data set according to the running state of the real-time coal mill, a corresponding association locking result is obtained, and the output association locking result is used as the anomaly detection result. And further, carrying out sensor analysis on the association locking result in the abnormality detection result to obtain an association sensor, and using the position of the association sensor as abnormality reminding information so as to locate and identify the explosion monitoring abnormality. The abnormal reminding information is information for reminding abnormal conditions of the sensor. Therefore, the technical effects of rapidly positioning the abnormal deflagration condition of the coal mill of the thermal power station, determining the abnormal sensor and directly positioning the abnormal position are achieved.
In summary, the embodiments of the present application have at least the following technical effects:
according to the embodiment of the application, the basic condition of the pulverizing system of the target thermal power station is acquired, the distribution condition and the running state of the coal mill are obtained, then the historical data of the monitoring sensor in the thermal power station are analyzed, double identification is carried out on the historical data for the convenience of analysis, the analysis is carried out from two angles of position and type, furthermore, the historical data are associated and locked according to the distribution position of the coal mill, the correlation degree between the historical data is analyzed, in order to improve the efficiency of abnormal identification, the abnormal data are subjected to locking identification of associated data by constructing a deflagration abnormal monitoring model, then the monitoring sensor is subjected to real-time data acquisition within a preset time period, the real-time acquired result is input into the abnormal monitoring model, the abnormal monitoring result is obtained, and then abnormal reminding information is generated according to the abnormal monitoring result, and positioning identification is carried out on deflagration monitoring abnormality. The method and the device have the advantages that the efficiency of abnormality identification is improved, the abnormal situation is rapidly identified, and the corresponding associated sensor is matched, so that the technical effect of abnormality processing quality is improved.
Example two
Based on the same inventive concept as the method for identifying the abnormal location of the deflagration monitoring of the coal mill in the thermal power station in the previous embodiment, as shown in fig. 4, the present application provides a system for identifying the abnormal location of the deflagration monitoring of the coal mill in the thermal power station, and the embodiments of the system and the method in the embodiments of the present application are based on the same inventive concept. Wherein the system comprises:
the basic information acquisition module 11 is used for acquiring information of a pulverizing system of a target thermal power station to obtain a basic information set;
a historical data set obtaining module 12, where the historical data set obtaining module 12 is configured to obtain historical monitoring information of a monitoring sensor according to an industrial control system of the target thermal power station, and obtain a historical sensor data set, where the historical sensor data set has a dual identifier, and the dual identifier includes a sensor position identifier and a sensor type identifier;
the association locking module 13 is used for performing association locking on the historical sensor data set according to the basic information set to obtain an association locking result;
a monitoring model construction module 14, wherein the monitoring model construction module 14 is used for constructing a deflagration abnormal monitoring model according to the basic information set and the association locking result;
The monitoring data set obtaining module 15, wherein the monitoring data set obtaining module 15 is used for collecting real-time data of the monitoring sensor in a preset time window to obtain a real-time monitoring data set;
the monitoring result obtaining module 16, where the monitoring result obtaining module 16 is configured to input the real-time monitoring dataset into the deflagration anomaly monitoring model to obtain an anomaly monitoring result;
the abnormality reminding module 17 is used for generating abnormality reminding information according to the abnormality monitoring result and carrying out positioning identification on the explosion monitoring abnormality.
Further, the system further comprises:
the distribution data set obtaining unit is used for extracting the distribution information of the coal mill based on the basic information set to obtain a distribution data set of the coal mill;
the clustering result obtaining unit is used for clustering the historical sensor data set by combining the coal mill distribution data set based on the sensor position identification to obtain a sensor clustering result;
and the association locking result obtaining unit is used for carrying out association locking on the historical sensor data set according to the sensor clustering result to obtain the association locking result.
Further, the system further comprises:
the dimension dividing unit is used for dimension dividing the sensor clustering result according to the sensor type identifier to obtain a dimension dividing result;
the similar association unit is used for carrying out similar association on the dimension division result according to the sensor position identification to obtain a similar association result;
the heterogeneous association unit is used for carrying out heterogeneous association on the dimension division result according to the basic information set to obtain a heterogeneous association result;
and the locking result obtaining unit is used for carrying out association locking on the historical sensor data set according to the similar association results and the heterogeneous association results to obtain the association locking result.
Further, the system further comprises:
the detonation record data set obtaining unit is used for extracting the historical detonation record of the target thermal power station recorded in the industrial control system to obtain a historical detonation record data set;
the frequency statistics unit is used for carrying out frequency statistics on the association locking result according to the historical deflagration record data set to obtain a frequency data set;
The primary dividing unit is used for randomly selecting one frequency data from the frequency data set without being put back and is used as a primary dividing node;
the secondary dividing unit is used for randomly selecting one frequency data from the frequency data set again without replacement and is used as a secondary dividing node;
the multi-stage dividing unit is used for continuously constructing multi-stage dividing nodes for obtaining the data cleaning unit;
the data cleaning unit is used for marking a plurality of final dividing nodes of the multi-stage dividing nodes according to the association locking result corresponding to the frequency data set to obtain the constructed data cleaning unit;
the elimination data set determining unit is used for inputting the frequency data set into the data cleaning unit and determining frequency data lower than a preset dividing node as an elimination frequency data set;
and the screening optimization unit is used for screening and optimizing the association locking result according to the elimination frequency data set.
Further, the system further comprises:
the association data set obtaining unit is used for extracting the historical sensor data according to the similar association results and the heterogeneous association results to obtain an association data set;
The trend change result obtaining unit is used for obtaining the data change trend of each associated sensor according to the associated data set to obtain a plurality of trend change results;
the relevance grade obtaining unit is used for evaluating relevance of the similar relevance results and the heterogeneous relevance results according to the trend change results to obtain relevance grade;
and the association degree identification unit is used for carrying out association degree identification on the association locking result by utilizing the association degree grade.
Further, the system further comprises:
the network structure building unit is used for building a BP feedforward network structure and comprises an input layer, an implicit layer and an output layer, wherein the implicit layer comprises an operation mode identification layer and an association result matching layer;
the training set establishing unit is used for extracting information from the basic information data set and establishing an operation mode training set, wherein the operation mode training set comprises historical operation mode data and corresponding mode type identification information;
the training convergence unit is used for training and converging the operation mode identification layer by using the operation mode training set until a preset training stopping requirement is reached;
The matching training set construction unit is used for determining the corresponding relation between each operation mode data type and the associated locking result according to the historical operation mode data and constructing an associated result matching training set;
and the anomaly monitoring model obtaining unit is used for training and converging the association result matching layer by using the association result matching training set until reaching a preset training stopping requirement to obtain the deflagration anomaly monitoring model.
Further, the system further comprises:
the initial monitoring data set obtaining unit is used for collecting monitoring sensor data in the preset time window in real time to obtain an initial real-time monitoring data set;
the comparison data set obtaining unit is used for obtaining the data of the monitoring sensor in the last preset time window to obtain a comparison monitoring data set;
the real-time monitoring data set obtaining unit is used for comparing the comparison monitoring data set with the initial real-time monitoring data set to obtain the real-time monitoring data set.
It should be noted that the sequence of the embodiments of the present application is merely for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing description has been directed to specific embodiments of this specification. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing description of the preferred embodiments of the present application is not intended to limit the invention to the particular embodiments of the present application, but to limit the scope of the invention to the particular embodiments of the present application.
The specification and drawings are merely exemplary of the application and are to be regarded as covering any and all modifications, variations, combinations, or equivalents that are within the scope of the application. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (7)
1. The method for positioning and identifying the deflagration monitoring abnormality of the coal mill of the thermal power station is characterized by comprising the following steps:
information acquisition is carried out on a pulverizing system of a target thermal power station, so as to obtain a basic information set;
acquiring historical monitoring information of a monitoring sensor according to an industrial control system of the target thermal power station, and acquiring a historical sensor data set, wherein the historical sensor data set is provided with double identifications, and the double identifications comprise sensor position identifications and sensor type identifications;
performing association locking on the historical sensor data set according to the basic information set to obtain an association locking result;
constructing a deflagration anomaly monitoring model according to the basic information set and the associated locking result, wherein the deflagration anomaly monitoring model comprises the following steps: building a BP feedforward network structure, which comprises an input layer, an implicit layer and an output layer, wherein the implicit layer comprises an operation mode identification layer and an association result matching layer; extracting information from the basic information data set, and establishing an operation mode training set, wherein the operation mode training set comprises historical operation mode data and corresponding mode type identification information; training convergence is carried out on the operation mode identification layer by utilizing the operation mode training set until a preset training stopping requirement is reached; according to the historical operation mode data, determining the corresponding relation between each operation mode data type and the associated locking result, and constructing an associated result matching training set; training convergence is carried out on the association result matching layer by utilizing the association result matching training set until a preset training stopping requirement is reached, so as to obtain the deflagration anomaly monitoring model;
Acquiring real-time data of a monitoring sensor in a preset time window to obtain a real-time monitoring data set;
inputting the real-time monitoring data set into the deflagration anomaly monitoring model to obtain anomaly monitoring results;
and generating abnormal reminding information according to the abnormal monitoring result, and carrying out positioning identification on the deflagration monitoring abnormality.
2. The method of claim 1, wherein the association locking of the historical sensor dataset from the base information set, the method further comprising:
extracting coal mill distribution information based on the basic information set to obtain a coal mill distribution data set;
based on the sensor position identification, clustering the historical sensor data sets by combining the coal mill distribution data sets to obtain a sensor clustering result;
and performing association locking on the historical sensor data set according to the sensor clustering result to obtain the association locking result.
3. The method of claim 2, wherein the performing an association lock on the historical sensor dataset according to the sensor cluster result results to obtain the association lock result, the method further comprising:
Performing dimension division on the sensor clustering result according to the sensor type identifier to obtain a dimension division result;
similar association is carried out on the dimension division results according to the sensor position identifiers, and similar association results are obtained;
carrying out heterogeneous association on the dimension division result according to the basic information set to obtain a heterogeneous association result;
and performing association locking on the historical sensor data set according to the similar association results and the heterogeneous association results to obtain the association locking results.
4. The method of claim 1, wherein the method further comprises:
extracting a historical detonation record of the target thermal power station recorded in the industrial control system to obtain a historical detonation record data set;
performing frequency statistics on the association locking result according to the historical deflagration record data set to obtain a frequency data set;
randomly selecting one frequency data from the frequency data set without replacement as a primary dividing node;
randomly selecting one frequency data from the frequency data set without replacement again to serve as a secondary dividing node;
continuously constructing a multi-level partition node for obtaining a data cleaning unit;
Marking a plurality of final dividing nodes of the multi-stage dividing nodes according to the association locking result corresponding to the frequency data set to obtain the constructed data cleaning unit;
inputting the frequency data set into the data cleaning unit, and determining the frequency data lower than a preset dividing node as an obsolete frequency data set;
and screening and optimizing the association locking result according to the elimination frequency data set.
5. The method of claim 3, wherein the historical sensor dataset is relatedly locked based on the homogeneous and heterogeneous association results, the method further comprising:
extracting the historical sensor data according to the similar association results and the heterogeneous association results to obtain an association data set;
obtaining the data change trend of each associated sensor according to the associated data set, and obtaining a plurality of trend change results;
evaluating the association degrees of the similar association results and the heterogeneous association results according to the trend change results to obtain an association degree grade;
and carrying out relevance identification on the relevance locking result by utilizing the relevance grade.
6. The method of claim 1, wherein the real-time data acquisition is performed on the monitoring sensor within a preset time window to obtain a real-time monitoring data set, and the method further comprises:
collecting monitoring sensor data in the preset time window in real time to obtain an initial real-time monitoring data set;
acquiring monitoring sensor data in a last preset time window to obtain a comparison monitoring data set;
and carrying out data comparison on the comparison monitoring data set and the initial real-time monitoring data set to obtain the real-time monitoring data set.
7. A deflagration monitoring anomaly locating and identifying system for a coal mill of a thermal power station, the system comprising:
the basic information acquisition module is used for acquiring information of a pulverizing system of the target thermal power station to obtain a basic information set;
the historical data set obtaining module is used for obtaining historical monitoring information of a monitoring sensor according to an industrial control system of the target thermal power station to obtain a historical sensor data set, wherein the historical sensor data set is provided with dual identifications, and the dual identifications comprise sensor position identifications and sensor type identifications;
The association locking module is used for carrying out association locking on the historical sensor data set according to the basic information set to obtain an association locking result;
the monitoring model construction module is used for constructing a deflagration abnormal monitoring model according to the basic information set and the association locking result;
the network structure building unit is used for building a BP feedforward network structure and comprises an input layer, an implicit layer and an output layer, wherein the implicit layer comprises an operation mode identification layer and an association result matching layer;
the training set establishing unit is used for extracting information from the basic information data set and establishing an operation mode training set, wherein the operation mode training set comprises historical operation mode data and corresponding mode type identification information;
the training convergence unit is used for training and converging the operation mode identification layer by using the operation mode training set until a preset training stopping requirement is reached;
the matching training set construction unit is used for determining the corresponding relation between each operation mode data type and the associated locking result according to the historical operation mode data and constructing an associated result matching training set;
The abnormal monitoring model obtaining unit is used for training and converging the association result matching layer by using the association result matching training set until reaching a preset training stopping requirement to obtain the deflagration abnormal monitoring model;
the monitoring data set acquisition module is used for acquiring real-time data of the monitoring sensor in a preset time window to obtain a real-time monitoring data set;
the monitoring result obtaining module is used for inputting the real-time monitoring data set into the deflagration abnormal monitoring model to obtain an abnormal monitoring result;
and the abnormality reminding module is used for generating abnormality reminding information according to the abnormality monitoring result and carrying out positioning identification on the deflagration monitoring abnormality.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310212801.3A CN116060201B (en) | 2023-03-08 | 2023-03-08 | Deflagration monitoring abnormality positioning and identifying method and system for coal mill of thermal power station |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310212801.3A CN116060201B (en) | 2023-03-08 | 2023-03-08 | Deflagration monitoring abnormality positioning and identifying method and system for coal mill of thermal power station |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116060201A CN116060201A (en) | 2023-05-05 |
CN116060201B true CN116060201B (en) | 2023-06-23 |
Family
ID=86169927
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310212801.3A Active CN116060201B (en) | 2023-03-08 | 2023-03-08 | Deflagration monitoring abnormality positioning and identifying method and system for coal mill of thermal power station |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116060201B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116383769B (en) * | 2023-05-19 | 2023-08-15 | 陕西科诺特斯科技技术有限公司 | Sensor-based gas data detection method and system |
CN116859843B (en) * | 2023-07-05 | 2024-01-09 | 安徽如柒信息科技有限公司 | Production equipment fault monitoring method and system based on industrial big data |
CN116582574B (en) * | 2023-07-13 | 2023-09-26 | 深圳市聚亚科技有限公司 | Atmospheric monitoring system based on Internet of things |
CN116578845B (en) * | 2023-07-14 | 2023-11-07 | 杭州小策科技有限公司 | Risk identification method and system for batch identification data learning |
CN117095361B (en) * | 2023-10-18 | 2023-12-22 | 南通飞海电子科技有限公司 | Antifouling cleaning method and system for digital cash register |
CN117387699B (en) * | 2023-12-11 | 2024-04-02 | 浙江万胜智能科技股份有限公司 | Data monitoring method and system for identifying circuit breaker abnormality |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR900002655B1 (en) * | 1983-08-01 | 1990-04-21 | 더 뱁콕 앤드 윌콕스 컴퍼니 | Safety system for coal pulverizers |
CN103909006B (en) * | 2013-01-08 | 2016-12-28 | 宝山钢铁股份有限公司 | Detonation detection method and detecting system thereof in coal pulverizer outlet pulverized coal channel |
CN112863116B (en) * | 2020-12-03 | 2023-04-28 | 华能国际电力股份有限公司玉环电厂 | Intelligent diagnosis and alarm display method for deflagration of coal pulverizing system of thermal power plant |
CN113000192B (en) * | 2021-02-23 | 2022-07-12 | 南方电网电力科技股份有限公司 | Method and system for quickly warning deflagration of coal mill |
CN113588308A (en) * | 2021-07-22 | 2021-11-02 | 上海电力大学 | Coal mill detonation fault early warning and diagnosis method based on LSTM and Stacking fusion algorithm |
CN115081926B (en) * | 2022-07-14 | 2022-11-11 | 石家庄良村热电有限公司 | Operation safety early warning method and system suitable for intelligent power plant |
-
2023
- 2023-03-08 CN CN202310212801.3A patent/CN116060201B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN116060201A (en) | 2023-05-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116060201B (en) | Deflagration monitoring abnormality positioning and identifying method and system for coal mill of thermal power station | |
CN108154291B (en) | Power supply plant performance index evaluation method and system based on strategy library | |
CN106709662B (en) | Power equipment operation condition division method | |
CN111539553B (en) | Wind turbine generator fault early warning method based on SVR algorithm and off-peak degree | |
CN108133316B (en) | Detection method for familial defects of power equipment of power supply plant | |
CN107516174B (en) | Generation method of power plant technical supervision strategy library | |
Kabir et al. | Isolation forest based anomaly detection and fault localization for solar PV system | |
CN114856941A (en) | Offshore wind power plant and unit fault diagnosis operation and maintenance system and diagnosis operation and maintenance method thereof | |
CN108710689A (en) | Wind power plant meteorological data management system and method | |
CN112001511A (en) | Equipment reliability and dynamic risk evaluation method, system and equipment based on data mining | |
CN111596643A (en) | Visual dynamic energy consumption diagnosis, analysis and pre-control system based on big data | |
CN103218535B (en) | The system of selection of electric power communication device detection scheme and device | |
CN116186624A (en) | Boiler assessment method and system based on artificial intelligence | |
CN117151684A (en) | Wind power fan data analysis early warning method, system, device and readable storage medium | |
CN116467658A (en) | Equipment fault tracing method based on Markov chain | |
CN115563885A (en) | Energy conversion system for energy storage battery of wind power plant | |
KR20160091942A (en) | Method, system, and computer program product for analyzing production and/or process-engineering processes and/or process steps in a plant | |
CN109469919A (en) | A kind of stifled grey monitoring method of power station air preheater based on weight cluster | |
CN115309729A (en) | Thermal power plant multi-type data allocation method and system based on ETL technology | |
CN114157026A (en) | Scheduling method for offshore wind power operation and maintenance | |
Zhang et al. | A data-driven photovoltaic string current mismatch fault diagnosis method based on IV curve | |
CN113946977A (en) | Application method for early warning of fan variable pitch fault based on decision tree algorithm | |
CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network | |
CN114070198B (en) | Fault diagnosis method and device for distributed photovoltaic power generation system and electronic equipment | |
Jiao et al. | Photovoltaic Power Abnormal Data Cleaning Based on Variance Change Point and Correlation Analysis |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |