Intelligent monitoring and early warning method and system for boiler induced draft fan

By building the operation monitoring channel and abnormal traceability of boiler induced fan, the problem of imperfect early warning mechanism in the existing technology is solved, and the accurate early warning and alarm processing of boiler induced fan is achieved, and the monitoring of equipment in complex environments is improved. Maintenance efficiency.

CN119068641AInactive Publication Date: 2024-12-03NANTONG WANDILAI ELECTROMECHANICAL CO LTD

Patent Information

Application Number
CN202411554273.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-04
Publication Date
2024-12-03
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing boiler induced fan monitoring technology is incomplete in complex and changing industrial environments, making it difficult to achieve accurate early warning.

Method used

The operating parameters of the boiler induced fan are detected in real time through the data sensing unit, and the operation control unit is used to select and construct the monitoring channel, generate an early warning signal, and transmit the abnormal data to the remote control terminal for alarm processing through abnormal traceability.

Benefits of technology

It realizes accurate early warning of boiler induced fans, improves monitoring and alarm efficiency in complex environments, and ensures the safe and stable operation of the equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119068641A_ABST
    Figure CN119068641A_ABST
Patent Text Reader

Abstract

The invention discloses an intelligent monitoring and early warning method and system for a boiler induced-draft fan, and relates to the technical field of data processing.The intelligent monitoring and early warning method comprises the steps that the boiler induced-draft fan is started, real-time detection is conducted through a data sensing unit, and a real-time operation parameter set of the boiler induced-draft fan is obtained; acquiring a real-time operation parameter set through an operation control unit to obtain a plurality of operation signals, and performing feature selection based on the plurality of operation signals to determine a plurality of operation features; constructing an operation monitoring channel of the boiler induced draft fan based on the plurality of operation characteristics; the boiler induced draft fan is monitored through the operation monitoring channel, and a monitoring result is synchronized to the operation early warning unit to generate an early warning signal; and according to the early warning signal, performing abnormity tracing, determining a plurality of abnormal data sources, and transmitting the plurality of abnormal data sources to a remote control terminal for alarm processing. The boiler induced draft fan early warning method solves the problem that an early warning mechanism is still imperfect when a complex and changeable industrial environment is dealt with in the prior art, and achieves the technical effect of accurate early warning of the boiler induced draft fan.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an intelligent monitoring and early warning method and system for a boiler induced draft fan. Background Art

[0002] With the continuous improvement of industrial automation, the monitoring and maintenance methods of boiler induced draft fans, as an important part of industrial boiler systems, are also gradually upgraded. The existing boiler induced draft fan monitoring technology has achieved a certain degree of automation, including monitoring parameters such as pressure, temperature, flow and vibration through sensors, and using control systems for data collection and processing. However, these existing technologies still have the problem of imperfect early warning mechanisms when dealing with complex and changing industrial environments. Summary of the invention

[0003] The present application provides an intelligent monitoring and early warning method and system for a boiler induced draft fan, which is used to solve the technical problem that the existing technology still has an imperfect early warning mechanism when dealing with a complex and changeable industrial environment.

[0004] In view of the above problems, the present application provides an intelligent monitoring and early warning method and system for a boiler induced draft fan.

[0005] In a first aspect of the present application, an intelligent monitoring and early warning method for a boiler induced draft fan is provided, the method comprising: Start the boiler induced draft fan, perform real-time detection through the data sensing unit, and obtain the real-time operating parameter set of the boiler induced draft fan; collect the real-time operating parameter set through the operation control unit to obtain multiple operating signals, perform feature selection based on the multiple operating signals, and determine multiple operating features; construct an operation monitoring channel for the boiler induced draft fan based on the multiple operating features; monitor the boiler induced draft fan using the operation monitoring channel, and synchronize the monitoring results to the operation early warning unit to generate an early warning signal; trace the abnormality of the boiler induced draft fan according to the early warning signal, determine multiple abnormal data sources, and transmit the multiple abnormal data sources to the remote control terminal for alarm processing.

[0006] The second aspect of the present application provides an intelligent monitoring and early warning system for a boiler induced draft fan, the system comprising: An operation parameter acquisition module, wherein the operation parameter acquisition module starts the boiler induced draft fan, performs real-time detection through a data sensing unit, and obtains a real-time operation parameter set of the boiler induced draft fan; a feature selection module, wherein the feature selection module collects the real-time operation parameter set through an operation control unit, obtains multiple operation signals, performs feature selection based on the multiple operation signals, and determines multiple operation features; an operation monitoring channel construction module, wherein the operation monitoring channel construction module constructs an operation monitoring channel of the boiler induced draft fan based on the multiple operation features; an early warning signal generation module, wherein the early warning signal generation module monitors the boiler induced draft fan using the operation monitoring channel, and synchronizes the monitoring results to the operation early warning unit to generate an early warning signal; an alarm module, wherein the alarm module traces the abnormality of the boiler induced draft fan according to the early warning signal, determines multiple abnormal data sources, and transmits the multiple abnormal data sources to the remote control terminal for alarm processing.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application starts the boiler induced draft fan, performs real-time detection through a data sensing unit, and obtains a real-time operating parameter set of the boiler induced draft fan; collects the real-time operating parameter set through an operation control unit, obtains multiple operating signals, performs feature selection based on multiple operating signals, and determines multiple operating features; constructs an operation monitoring channel for the boiler induced draft fan based on multiple operating features; monitors the boiler induced draft fan using the operation monitoring channel, and synchronizes the monitoring results to the operation early warning unit to generate an early warning signal; traces the abnormality of the boiler induced draft fan according to the early warning signal, determines multiple abnormal data sources, and transmits the multiple abnormal data sources to the remote control terminal for alarm processing. The present invention solves the problem that the existing technology still has an imperfect early warning mechanism when dealing with a complex and changeable industrial environment. By constructing an operation monitoring channel, monitoring the induced draft fan in real time and generating an early warning signal, tracing the abnormality according to the early warning signal, determining and transmitting abnormal data to the remote control terminal for alarm processing, the technical effect of accurate early warning of the boiler induced draft fan is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1 A schematic flow chart of an intelligent monitoring and early warning method for a boiler induced draft fan provided in an embodiment of the present application; Figure 2 A schematic diagram of the structure of an intelligent monitoring and early warning system for a boiler induced draft fan provided in an embodiment of the present application.

[0010] Explanation of the accompanying drawings: operating parameter acquisition module 11, feature selection module 12, operating monitoring channel construction module 13, early warning signal generation module 14, alarm module 15. DETAILED DESCRIPTION

[0011] The present application provides an intelligent monitoring and early warning method and system for boiler induced draft fans, aiming to solve the problem that the existing technology still has an imperfect early warning mechanism when dealing with complex and changeable industrial environments. By constructing an operation monitoring channel, the induced draft fan is monitored in real time and an early warning signal is generated. The abnormality is traced according to the early warning signal, and the abnormal data is determined and transmitted to the remote control terminal for alarm processing, thereby achieving the technical effect of accurate early warning of the boiler induced draft fan.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides an intelligent monitoring and early warning method for a boiler induced draft fan, the method comprising: Step S100: Start the boiler induced draft fan, perform real-time detection through a data sensing unit, and obtain a real-time operating parameter set of the boiler induced draft fan.

[0015] In the embodiment of the present application, the boiler induced draft fan is started and real-time detection is performed through a data sensing unit. The data sensing unit is a device that can continuously obtain the operating status of the equipment. It collects data during the operation in real time through sensors installed in various parts of the induced draft fan. These data include but are not limited to operating parameters such as pressure, temperature, flow rate and vibration. Through real-time detection, a real-time operating parameter set of the boiler induced draft fan is obtained.

[0016] Step S200: collecting the real-time operation parameter set through an operation control unit to obtain a plurality of operation signals, performing feature selection based on the plurality of operation signals, and determining a plurality of operation features.

[0017] In an embodiment of the present application, the operation control unit is an intelligent control system responsible for managing and processing real-time data from the data sensing unit. This unit is capable of collecting and processing the operation parameter set in a refined manner to obtain multiple operation signals. The operation signals include pressure signals, temperature signals, flow signals, and vibration signals, which respectively represent the operation characteristics of the induced draft fan under different conditions. The pressure signal reflects the change in the intensity of the airflow, the temperature signal shows the thermal state of the equipment, the flow signal reveals the changes in the gas velocity and amount passing through the induced draft fan, and the vibration signal indicates the stability of the mechanical state.

[0018] Based on these multiple operating signals, feature selection is performed. Feature selection is the process of extracting the most representative and diagnostically valuable features from a large amount of data. A feature matrix is ​​constructed by analyzing the time domain features and frequency domain features of each signal. Next, the feature matrix is ​​processed using algorithms such as principal component analysis, the covariance matrix is ​​calculated, and eigenvalue decomposition is performed to extract the most important eigenvectors. The feature matrix is ​​projected onto these eigenvectors to determine multiple key operating features.

[0019] Furthermore, in the method provided in the embodiment of the application, feature selection is performed based on the multiple operation signals to determine multiple operation features, and further includes: Based on the multiple operating signals, time domain features and frequency domain features are extracted to construct a feature matrix; the feature matrix is ​​calculated using principal component analysis to obtain a covariance matrix; the covariance matrix is ​​subjected to eigenvalue decomposition to obtain M eigenvalues, where M is an integer greater than or equal to 1; the M eigenvalues ​​are arranged in descending order, and N eigenvalues ​​are extracted to obtain N eigenvectors, where N is an integer greater than 0 and less than M; the feature matrix is ​​projected onto the N eigenvectors to determine the multiple operating features.

[0020] In an embodiment of the present application, first, time domain features and frequency domain features are extracted from multiple operating signals such as vibration, temperature, pressure, current and voltage. For time domain features, statistical analysis methods are used, such as calculating the mean value, standard deviation, peak value and root mean square value of the signal. For frequency domain features, Fourier transform is used to convert the signal from time domain to frequency domain. Through Fourier transform, frequency domain features such as spectral density and main frequency components are extracted. Next, these extracted time domain features and frequency domain features are combined together to construct a feature matrix.

[0021] Then, principal component analysis is used to calculate the feature matrix. Specifically, the covariance matrix of the feature matrix is ​​first calculated to reflect the correlation between the features. Then, the covariance matrix is ​​decomposed by eigenvalue to obtain M eigenvalues. The eigenvalue represents the variance of the data in a specific direction.

[0022] Perform eigenvalue decomposition on the covariance matrix to obtain M eigenvalues. M is an integer greater than or equal to 1, representing the number of eigenvalues. Then, these eigenvalues ​​are sorted in descending order and the first N largest eigenvalues ​​are extracted. N is an integer greater than 0 and less than M, indicating that the most important N eigenvalues ​​are selected.

[0023] By extracting N eigenvalues, we obtain the corresponding eigenvectors, which represent the direction of the data in the feature space and reflect the main change pattern of the original data. We project the feature matrix onto these eigenvectors and map the original data into a low-dimensional space, thereby determining multiple key operating characteristics. Specifically, we select the eigenvectors corresponding to the first N eigenvalues ​​according to the eigenvalues ​​in descending order. We project the feature matrix onto the selected eigenvectors to complete the dimensionality reduction process and determine multiple key operating characteristics.

[0024] Step S300: constructing an operation monitoring channel of the boiler induced draft fan based on the multiple operation characteristics.

[0025] Furthermore, in the method provided in the embodiment of the application, an operation monitoring channel of the boiler induced draft fan is constructed based on the multiple operation characteristics, and further includes: Retrieve historical operation data record archives to extract multiple operation influencing parameters and associate them with the multiple operation characteristics to determine the operation influencing factors; assign weights to the multiple operation characteristics in combination with the operation influencing factors, and fuse them with the fault type information library according to the weight assignment results to generate a classification data set; train the classification data set based on a decision tree to generate training results, and construct a supervision data set and a test data set; adjust the training results according to the supervision data set, use the test data set to reciprocately verify the operation monitoring channel, and when the verification passes, output the operation monitoring channel.

[0026] In the embodiment of the present application, by retrieving the historical operation data record archive of the boiler induced draft fan, a plurality of operation influencing parameters related to the operation status of the induced draft fan are firstly extracted. These operation influencing parameters include ambient temperature, load change and maintenance records, etc., which can significantly affect the operation status of the induced draft fan. Next, these operation influencing parameters are correlated with the plurality of operation features extracted previously to determine the operation influencing factor of each feature.

[0027] After determining the operation influencing factors, weights are assigned to each operation feature in combination with the operation influencing factors. Weight assignment is to assign different weight values ​​to each feature based on its importance and its influence on the operation status of the induced draft fan. The weighted operation features are then fused with the fault type information library to generate a classification data set containing multiple operation statuses and fault types.

[0028] Next, the decision tree algorithm is used to train the classification data set. The classification data set is randomly divided into a training data set and a test data set. The classification data set is split in a ratio of 80:20, that is, 80% of the data is used to train the model and 20% of the data is used to test the model. The decision tree model is trained using the training data set. During the training process, the model learns how to predict the fault type based on the input features, gradually builds the structure of the decision tree, and generates preliminary training results, that is, the preliminary constructed decision tree model. Extract some data from the training data set to construct a supervised data set. Use the supervised data set to supervise the model and gradually adjust the model parameters so that it can more accurately describe and predict the actual operating status.

[0029] Use the test data set to verify the model and evaluate its classification accuracy and prediction ability. Use the test data set to repeatedly verify the model, and evaluate the stability and reliability of the model under different operating conditions through multiple verifications. Adjust the model according to the test results, correct potential problems, and ensure its effectiveness and reliability in practical applications. When the model passes multiple verifications and the verification results meet the predetermined accuracy and reliability requirements, the final operation monitoring channel is output.

[0030] Furthermore, the method provided in the application embodiment also includes: Fault records are performed based on multiple data sources to obtain multiple fault information sets, wherein the multiple fault information sets include multiple fault type information, multiple fault time information, and multiple fault location information; the multiple fault time information and the multiple fault location information are classified and trained according to the multiple fault type information to define multiple fault labels; a fault record table is constructed according to the multiple fault labels combined with a fault classification standard; the multiple fault information sets are classified and stored according to the fault record table to construct the fault type information library.

[0031] In an embodiment of the present application, fault record information of the induced draft fan is first obtained from multiple data sources. These data sources include sensor data, maintenance records, and operation logs. Each fault record contains fault type information, fault time information, and fault location information. By integrating this information, multiple fault information sets are constructed. Each fault information set contains one or more fault records, covering all relevant information, such as fault type, time, and location.

[0032] Next, the fault time information and fault location information are classified and trained according to the fault type information. Specifically, the fault types are classified into multiple categories. For each fault category, the corresponding time information and location information are analyzed to identify its characteristic patterns. Based on these characteristic patterns, multiple fault labels are defined, each label corresponding to a fault type and its characteristics.

[0033] According to the defined fault labels and preset fault classification standards, a detailed fault record table is constructed. The fault record table contains each fault label and its corresponding feature description; various fault types and their classification standards; and the typical time pattern and location characteristics of each fault type.

[0034] According to the constructed fault record table, multiple fault information sets are classified and stored. Specifically, each record in the fault information set is matched with the fault record table, and a corresponding fault label is assigned according to its characteristics. The fault information is classified and stored according to the fault label to form a structured database.

[0035] Finally, the classified and stored fault information sets are integrated into a comprehensive fault type information library. The specific steps include entering all fault labels and classification standards defined in the fault record table into the information library. Import the classified and stored fault information sets into the information library to ensure that all records are accurately classified and stored. Through the above steps, the construction of the fault type information library is completed.

[0036] Step S400: monitor the boiler induced draft fan using the operation monitoring channel, and synchronize the monitoring result to the operation early warning unit to generate an early warning signal.

[0037] Furthermore, in the method provided in the embodiment of the application, the operation monitoring channel is used to monitor the boiler induced draft fan, and the monitoring result is synchronized to the operation warning unit to generate a warning signal, and further includes: Parse multiple data sources to obtain a pressure data source, a temperature data source, a flow data source, and a vibration data source; synchronize the pressure data source, the temperature data source, the flow data source, and the vibration data source to the operation monitoring channel to monitor the boiler induced draft fan and generate multiple monitoring data, wherein the multiple monitoring data include operation pressure monitoring data, operation temperature monitoring data, operation flow monitoring data, and operation vibration monitoring data; traverse the fault type information library based on the operation pressure monitoring data, the operation temperature monitoring data, the operation flow monitoring data, and the operation vibration monitoring data to identify the fault and obtain a monitoring result, wherein the monitoring result includes an operation pressure fault identification, an operation temperature fault identification, an operation flow fault identification, and an operation vibration fault identification; sequentially extract the multiple monitoring data with the operation pressure fault identification, the operation temperature fault identification, the operation flow fault identification, and the operation vibration fault identification and compare them with a preset safety critical threshold; extract W monitoring data greater than or equal to the preset safety critical threshold, wherein the W monitoring data have W fault identifications, and W is an integer greater than or equal to 0; define multiple warning levels according to the W monitoring data combined with the W fault identifications, and generate a warning signal, wherein the warning signal has a warning level.

[0038] In the embodiment of the present application, a data analysis algorithm is used to obtain a pressure data source, a temperature data source, a flow data source, and a vibration data source for multiple data sources. The pressure data source is to monitor the gas pressure inside the induced draft fan through a pressure sensor. The temperature data source is to monitor the temperature of the induced draft fan components through a temperature sensor. The flow data source is to monitor the gas flow through the induced draft fan through a flow sensor. The vibration data source is to monitor the mechanical vibration state of the induced draft fan through a vibration sensor.

[0039] The parsed pressure, temperature, flow and vibration data sources are synchronized to the operation monitoring channel of the induced draft fan, and multiple monitoring data are generated through processing, including operation pressure monitoring data, operation temperature monitoring data, operation flow monitoring data, and operation vibration monitoring data.

[0040] Next, based on the monitoring data of operating pressure, temperature, flow and vibration, the fault type information library is traversed, and a pattern matching algorithm, such as a dynamic time warping algorithm, is used to compare multiple monitoring data with the standard patterns in the fault type information library to identify possible fault types and obtain monitoring results. The monitoring results include operating pressure fault identification, operating temperature fault identification, operating flow fault identification, and operating vibration fault identification.

[0041] Compare the monitoring data with fault identification with the preset critical safety threshold. The critical safety threshold is set according to the normal operating parameters of the induced draft fan. Exceeding this threshold indicates a potential risk. Use statistical analysis methods, such as Z-score analysis, to compare the real-time data with the preset threshold one by one, extract the over-limit data to obtain W monitoring data. These data have been identified as records with faults. Each monitoring data point corresponds to one or more fault identifications, such as pressure fault identification, temperature fault identification, flow fault identification, and vibration fault identification.

[0042] Next, define the warning level, for example, divided into low, medium, high, and critical. At the same time, assign a basic score to each fault indicator, for example, the pressure fault indicator is 20 points, the temperature fault indicator is 30 points, the flow fault indicator is 25 points, and the vibration fault indicator is 35 points. According to the W extracted monitoring data and their fault indicators, a comprehensive score is calculated by adding them. According to the comprehensive score, it is divided into different warning levels. A comprehensive score less than 30 is low, a comprehensive score greater than or equal to 30 and less than 50 is medium, a comprehensive score greater than or equal to 50 and less than 70 is high, and a comprehensive score greater than or equal to 70 is critical. Through the above process, a warning signal containing the warning level is generated.

[0043] Step S500: Tracing the abnormality of the boiler induced draft fan according to the early warning signal, determining multiple abnormal data sources, and transmitting the multiple abnormal data sources to the remote control terminal for alarm processing.

[0044] In an embodiment of the present application, when the monitoring system of the boiler induced draft fan generates an early warning signal, the abnormality tracing process is started. First, the fault information provided in the early warning signal is analyzed, the sensors and data sources involved are identified, and relevant historical data and real-time data are extracted from the monitoring system. Through comparative analysis and trend analysis, the abnormality of the data is verified and the accuracy of the fault is confirmed. The determined abnormal data sources are then integrated into an abnormal data packet. The abnormal data packet contains detailed abnormal data records, fault type, location and time, as well as data analysis results and suggestions. It is then transmitted to the remote control terminal via a network protocol.

[0045] After receiving and decrypting the abnormal data packet at the remote control terminal, the alarm is processed. By parsing the abnormal data packet, key information is extracted, and corresponding alarm measures are triggered according to the warning level and fault type. Finally, the maintenance personnel are notified in time through sound and light alarms, SMS notifications, email notifications, etc., and the alarm processing is completed.

[0046] Furthermore, in the method provided in the embodiment of the application, the abnormality of the boiler induced draft fan is traced according to the early warning signal to determine multiple abnormal data sources, and further includes: Determine whether the warning level of the warning signal is in a preset warning interval; if the warning level of the warning signal is in the preset warning interval, traverse the W monitoring data and combine the W fault identifiers to locate the fault, and determine the fault trigger time and the fault trigger position; based on the data credibility of the data sensor unit combined with the fault trigger time and the fault trigger position, check one by one, verify the checking results in real time, and determine the multiple abnormal data sources according to the verification results.

[0047] In an embodiment of the present application, a logical judgment algorithm is used to determine whether its warning level is in a preset alarm interval. For example, the preset alarm interval is defined as medium and above. If the warning level of the warning signal is in the preset alarm interval, the fault location is performed, W monitoring data are traversed, and the specific information of the fault is determined in combination with W fault identifiers. Specifically, a timestamp is used to record the fault trigger time corresponding to each fault identifier. The sensor position or the specific part of the induced draft fan corresponding to the fault identifier is determined by the sensor ID and position mapping table. For example, the vibration sensor detects an abnormality at a certain moment, and the sensor is located at the bearing part of the induced draft fan.

[0048] Next, check the data of each fault indicator one by one. First, use the data quality assessment algorithm to evaluate the credibility of each sensor data, such as evaluating the credibility of the sensor through data integrity check and sensor calibration history. Then, use the comparative analysis algorithm to compare the current data with the historical normal data to confirm whether there is an anomaly. For example, by comparing the mean of the current data with the mean of the historical data, determine whether the vibration sensor is abnormal. Finally, use real-time data stream processing technologies such as Apache Kafka to monitor data changes and verify whether the fault persists.

[0049] Finally, SQL aggregation functions are used to summarize all verification results, identify multiple abnormal data sources that trigger warning signals, and record this information in the database, including sensor type, location, and failure time.

[0050] Furthermore, in the method provided in the embodiment of the application, the plurality of abnormal data sources are transmitted to the remote control terminal for alarm processing, and further includes: The multiple abnormal data sources are combined with the W fault identifiers to encapsulate and obtain an abnormal data packet; the abnormal data packet is securely encrypted and transmitted to the remote control terminal for parsing, and it is determined whether the parsing result meets the preset alarm condition; if the parsing result meets the preset alarm condition, an alarm is triggered, and fault diagnosis is performed on the W monitoring data, and a maintenance strategy is formulated to repair the boiler induced draft fan.

[0051] In an embodiment of the present application, after multiple abnormal data sources are determined, these abnormal data sources are combined with W fault identifiers, data encapsulation is performed, and an abnormal data packet is generated. Data encapsulation refers to combining multiple data formats and contents together to form a complete data structure. In the present application, the abnormal data packet adopts the JSON format, which contains key information such as sensor ID, fault location, fault time, fault type and fault value. After the abnormal data packet is generated, the data packet is encrypted using the AES encryption algorithm to ensure the security of the data during transmission. The encrypted abnormal data packet is transmitted to the remote control terminal via the HTTPS protocol.

[0052] After receiving the data packet, the remote control terminal uses the same key to decrypt it and extracts fault information and abnormal data through the JSON parsing library. According to the parsing results, it is determined whether the preset alarm conditions are met, such as checking whether the pressure value exceeds the safety limit or whether the temperature rises abnormally. If the parsing result shows that the abnormal value exceeds the threshold, an alarm is triggered to notify the maintenance personnel. The alarm system includes sound and light alarms, SMS notifications, and email notifications to ensure that maintenance personnel receive alarm information in a timely manner.

[0053] Then, detailed fault diagnosis is performed on the W monitoring data. The fault tree analysis technology is used in the fault diagnosis process to find the root cause of the fault. Each monitoring data point is used as the bottom event of the fault tree, and the bottom events are summarized layer by layer through logic gates (such as AND gates and OR gates), and finally the top event (main fault) that triggers the warning signal is found. For example, assuming that the data of the vibration sensor and the temperature sensor are abnormal at the same time, the two bottom events are connected through the AND logic gate, indicating that the simultaneous occurrence of these two events may cause more serious faults. In this way, the fault tree is analyzed and constructed layer by layer, and the root cause of the warning signal is finally determined, such as bearing wear or fan imbalance.

[0054] After the diagnosis is completed, a detailed fault diagnosis report is generated, which includes the root cause, scope of impact and recommended treatment measures for each fault. The report also prioritizes the faults to determine which faults need to be handled immediately and which can be repaired within the planned maintenance cycle. Prioritization is based on the severity of the fault, the scope of impact and the criticality of the equipment operation. Based on the fault diagnosis report, a detailed inspection and maintenance strategy is formulated. The maintenance strategy includes replacing damaged parts, adjusting operating parameters, etc.

[0055] Through the above process, the alarm is triggered, fault diagnosis is performed, and a maintenance strategy is formulated to repair the boiler induced draft fan.

[0056] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application starts the boiler induced draft fan, performs real-time detection through a data sensing unit, and obtains a real-time operating parameter set of the boiler induced draft fan; collects the real-time operating parameter set through an operation control unit, obtains multiple operating signals, performs feature selection based on multiple operating signals, and determines multiple operating features; constructs an operation monitoring channel for the boiler induced draft fan based on multiple operating features; monitors the boiler induced draft fan using the operation monitoring channel, and synchronizes the monitoring results to the operation early warning unit to generate an early warning signal; traces the abnormality of the boiler induced draft fan according to the early warning signal, determines multiple abnormal data sources, and transmits the multiple abnormal data sources to the remote control terminal for alarm processing. The present invention solves the problem that the existing technology still has an imperfect early warning mechanism when dealing with a complex and changeable industrial environment. By constructing an operation monitoring channel, monitoring the induced draft fan in real time and generating an early warning signal, tracing the abnormality according to the early warning signal, determining and transmitting abnormal data to the remote control terminal for alarm processing, the technical effect of accurate early warning of the boiler induced draft fan is achieved.

[0057] Embodiment 2, based on the same inventive concept as the intelligent monitoring and early warning method for boiler induced draft fans in the previous embodiment, Figure 2As shown, the present application provides an intelligent monitoring and early warning system for a boiler induced draft fan, and the system and method embodiments in the present application are based on the same inventive concept. The system includes: An operation parameter acquisition module 11, wherein the operation parameter acquisition module 11 starts the boiler induced draft fan, performs real-time detection through a data sensing unit, and obtains a real-time operation parameter set of the boiler induced draft fan; a feature selection module 12, wherein the feature selection module 12 collects the real-time operation parameter set through an operation control unit, obtains multiple operation signals, performs feature selection based on the multiple operation signals, and determines multiple operation features; an operation monitoring channel construction module 13, wherein the operation monitoring channel construction module 13 constructs an operation monitoring channel for the boiler induced draft fan based on the multiple operation features; an early warning signal generation module 14, wherein the early warning signal generation module 14 uses the operation monitoring channel to monitor the boiler induced draft fan, and synchronizes the monitoring results to the operation early warning unit to generate an early warning signal; an alarm module 15, wherein the alarm module 15 traces the abnormality of the boiler induced draft fan according to the early warning signal, determines multiple abnormal data sources, and transmits the multiple abnormal data sources to the remote control terminal for alarm processing.

[0058] Furthermore, the system is also used to implement the following functions: Based on the multiple operating signals, time domain features and frequency domain features are extracted to construct a feature matrix; the feature matrix is ​​calculated using principal component analysis to obtain a covariance matrix; the covariance matrix is ​​subjected to eigenvalue decomposition to obtain M eigenvalues, where M is an integer greater than or equal to 1; the M eigenvalues ​​are arranged in descending order, and N eigenvalues ​​are extracted to obtain N eigenvectors, where N is an integer greater than 0 and less than M; the feature matrix is ​​projected onto the N eigenvectors to determine the multiple operating features.

[0059] Furthermore, the system is also used to implement the following functions: Retrieve historical operation data record archives to extract multiple operation influencing parameters and associate them with the multiple operation characteristics to determine the operation influencing factors; assign weights to the multiple operation characteristics in combination with the operation influencing factors, and fuse them with the fault type information library according to the weight assignment results to generate a classification data set; train the classification data set based on a decision tree to generate training results, and construct a supervision data set and a test data set; adjust the training results according to the supervision data set, use the test data set to reciprocately verify the operation monitoring channel, and when the verification passes, output the operation monitoring channel.

[0060] Furthermore, the system is also used to implement the following functions: Fault records are performed based on multiple data sources to obtain multiple fault information sets, wherein the multiple fault information sets include multiple fault type information, multiple fault time information, and multiple fault location information; the multiple fault time information and the multiple fault location information are classified and trained according to the multiple fault type information to define multiple fault labels; a fault record table is constructed according to the multiple fault labels combined with a fault classification standard; the multiple fault information sets are classified and stored according to the fault record table to construct the fault type information library.

[0061] Furthermore, the system is also used to implement the following functions: Parse multiple data sources to obtain a pressure data source, a temperature data source, a flow data source, and a vibration data source; synchronize the pressure data source, the temperature data source, the flow data source, and the vibration data source to the operation monitoring channel to monitor the boiler induced draft fan and generate multiple monitoring data, wherein the multiple monitoring data include operation pressure monitoring data, operation temperature monitoring data, operation flow monitoring data, and operation vibration monitoring data; traverse the fault type information library based on the operation pressure monitoring data, the operation temperature monitoring data, the operation flow monitoring data, and the operation vibration monitoring data to identify the fault and obtain a monitoring result, wherein the monitoring result includes an operation pressure fault identification, an operation temperature fault identification, an operation flow fault identification, and an operation vibration fault identification; sequentially extract the multiple monitoring data with the operation pressure fault identification, the operation temperature fault identification, the operation flow fault identification, and the operation vibration fault identification and compare them with a preset safety critical threshold; extract W monitoring data greater than or equal to the preset safety critical threshold, wherein the W monitoring data have W fault identifications, and W is an integer greater than or equal to 0; define multiple warning levels according to the W monitoring data combined with the W fault identifications, and generate a warning signal, wherein the warning signal has a warning level.

[0062] Furthermore, the system is also used to implement the following functions: Determine whether the warning level of the warning signal is in a preset warning interval; if the warning level of the warning signal is in the preset warning interval, traverse the W monitoring data and combine the W fault identifiers to locate the fault, and determine the fault trigger time and the fault trigger position; based on the data credibility of the data sensor unit combined with the fault trigger time and the fault trigger position, check one by one, verify the checking results in real time, and determine the multiple abnormal data sources according to the verification results.

[0063] Furthermore, the system is also used to implement the following functions: The multiple abnormal data sources are combined with the W fault identifiers to encapsulate and obtain an abnormal data packet; the abnormal data packet is securely encrypted and transmitted to the remote control terminal for parsing, and it is determined whether the parsing result meets the preset alarm condition; if the parsing result meets the preset alarm condition, an alarm is triggered, and fault diagnosis is performed on the W monitoring data, and a maintenance strategy is formulated to repair the boiler induced draft fan.

[0064] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0066] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An intelligent monitoring and early warning method for a boiler induced draft fan, characterized in that: The method comprises: Start the boiler induced draft fan, perform real-time detection through the data sensing unit, and obtain the real-time operating parameter set of the boiler induced draft fan; The real-time operation parameter set is collected by an operation control unit to obtain a plurality of operation signals, and feature selection is performed based on the plurality of operation signals to determine a plurality of operation features; constructing an operation monitoring channel of the boiler induced draft fan based on the multiple operation characteristics; The operation monitoring channel is used to monitor the boiler induced draft fan, and the monitoring result is synchronized to the operation early warning unit to generate an early warning signal; According to the early warning signal, the abnormality of the boiler induced draft fan is traced to the source, multiple abnormal data sources are determined, and the multiple abnormal data sources are transmitted to the remote control terminal for alarm processing.

2. The intelligent monitoring and early warning method for a boiler induced draft fan according to claim 1, characterized in that: Feature selection is performed based on the multiple operation signals to determine multiple operation features, the method comprising: Extracting time domain features and frequency domain features based on the multiple operating signals, and constructing a feature matrix; Calculating the characteristic matrix using principal component analysis to obtain a covariance matrix; Performing eigenvalue decomposition on the covariance matrix to obtain M eigenvalues, where M is an integer greater than or equal to 1; Arrange the M eigenvalues ​​in descending order, extract N eigenvalues ​​to obtain N eigenvectors, where N is an integer greater than 0 and less than M; The feature matrix is ​​projected onto the N feature vectors to determine the plurality of operating features.

3. The intelligent monitoring and early warning method for a boiler induced draft fan according to claim 1, characterized in that: An operation monitoring channel of a boiler induced draft fan is established based on the multiple operation characteristics, and the method includes: Retrieving historical operation data record archives to extract multiple operation influencing parameters and associate them with the multiple operation characteristics to determine the operation influencing factors; weighting the plurality of operation characteristics in combination with the operation influencing factors, and fusing the weighting results with the fault type information library to generate a classification data set; Training the classification data set based on the decision tree, generating training results, and constructing a supervision data set and a test data set; The training result is adjusted according to the supervision data set, and the operation monitoring channel is repeatedly verified using the test data set. When the verification passes, the operation monitoring channel is output.

4. The intelligent monitoring and early warning method for a boiler induced draft fan according to claim 3, characterized in that: The method comprises: Recording faults based on multiple data sources to obtain multiple fault information sets, wherein the multiple fault information sets include multiple fault type information, multiple fault time information, and multiple fault location information; Classify and train the multiple fault time information and the multiple fault location information according to the multiple fault type information to define multiple fault labels; Constructing a fault record table according to the multiple fault labels and the fault classification standard; The multiple fault information sets are classified and stored according to the fault record table to construct the fault type information library.

5. The intelligent monitoring and early warning method for a boiler induced draft fan according to claim 4, characterized in that: The operation monitoring channel is used to monitor the boiler induced draft fan, and the monitoring result is synchronized to the operation early warning unit to generate an early warning signal. The method includes: Analyze multiple data sources to obtain pressure data source, temperature data source, flow data source, and vibration data source; The pressure data source, the temperature data source, the flow data source, and the vibration data source are synchronized to the operation monitoring channel to monitor the boiler induced draft fan, and a plurality of monitoring data are generated, wherein the plurality of monitoring data include operation pressure monitoring data, operation temperature monitoring data, operation flow monitoring data, and operation vibration monitoring data; Based on the operating pressure monitoring data, the operating temperature monitoring data, the operating flow monitoring data, and the operating vibration monitoring data, the fault type information library is traversed to perform fault identification to obtain a monitoring result, wherein the monitoring result includes an operating pressure fault identification, an operating temperature fault identification, an operating flow fault identification, and an operating vibration fault identification; Sequentially extracting the plurality of monitoring data having the operating pressure fault mark, the operating temperature fault mark, the operating flow fault mark, and the operating vibration fault mark, and comparing them with a preset safety critical threshold value; Extracting W monitoring data greater than or equal to the preset safety critical threshold, wherein the W monitoring data have W fault identifiers, where W is an integer greater than or equal to 0; A plurality of warning levels are defined according to the W monitoring data in combination with the W fault identifiers, and a warning signal is generated, wherein the warning signal has a warning level.

6. The intelligent monitoring and early warning method for a boiler induced draft fan according to claim 5, characterized in that: According to the early warning signal, the abnormality of the boiler induced draft fan is traced to the source, and multiple abnormal data sources are determined. The method includes: Determining whether the warning level of the warning signal is within a preset warning range; If the warning level of the warning signal is within the preset warning interval, the W monitoring data are traversed in combination with the W fault identifiers to locate the fault and determine the fault triggering time and fault triggering position; Based on the data credibility of the data sensing unit combined with the fault triggering time and the fault triggering position, one by one is checked, the checking results are verified in real time, and the multiple abnormal data sources are determined according to the verification results.

7. The intelligent monitoring and early warning method for a boiler induced draft fan according to claim 5, characterized in that: The plurality of abnormal data sources are transmitted to a remote control terminal for alarm processing, the method comprising: Encapsulate the multiple abnormal data sources in combination with the W fault identifiers to obtain an abnormal data packet; The abnormal data packet is securely encrypted and transmitted to the remote control terminal for analysis, and it is determined whether the analysis result meets the preset alarm condition; If the analysis result reaches the preset alarm condition, an alarm is triggered, and fault diagnosis is performed on the W monitoring data, and a maintenance strategy is formulated to repair the boiler induced draft fan.

8. Intelligent monitoring and early warning system for boiler induced draft fan, characterized in that: The system comprises: An operation parameter acquisition module, wherein the operation parameter acquisition module starts the boiler induced draft fan and performs real-time detection through a data sensing unit to obtain a real-time operation parameter set of the boiler induced draft fan; A feature selection module, wherein the feature selection module collects the real-time operation parameter set through an operation control unit to obtain a plurality of operation signals, performs feature selection based on the plurality of operation signals, and determines a plurality of operation features; An operation monitoring channel construction module, wherein the operation monitoring channel construction module constructs an operation monitoring channel of the boiler induced draft fan based on the multiple operation characteristics; An early warning signal generation module, wherein the early warning signal generation module uses the operation monitoring channel to monitor the boiler induced draft fan, and synchronizes the monitoring result to the operation early warning unit to generate an early warning signal; An alarm module is used to trace the abnormality of the boiler induced draft fan according to the early warning signal, determine multiple abnormal data sources, and transmit the multiple abnormal data sources to the remote control terminal for alarm processing.

Citation Information

Patent Citations

  • Device and method for induced draft fan fault early warning based on power station operation data

    CN105114348A

  • Natural gas power generation early warning management method and system based on intelligent monitoring technology

    CN116110203A

  • Intelligent fire monitoring and alarming method and system for fire protection

    CN118097918A

  • Ship operation situation awareness system based on multi-source data fusion

    CN118626546A

  • Thermal power plant induced draft fan fault early warning device for multi-source signal data fusion analysis

    CN221482210U

Cited By

  • Method and device for monitoring external force of push-out door driven by motor

    CN120063392A

  • Automatic induced draft fan control system and method based on DCS

    CN120739728A

  • Intelligent monitoring and diagnosing method for operation state of boiler induced draft fan of thermal power plant

    CN121576294A

  • Intelligent monitoring and diagnosing method for running state of boiler induced draft fan in thermal power plant

    CN121576294B