A state monitoring method, device and medium based on an industrial dust removal system

Abstract

The application discloses a kind of based on industrial dust removal system's state monitoring method, equipment and medium, belong to industrial dust removal technical field, for solving the operation state monitoring of each module in existing welding industrial dust removal system is not enough in time, it is difficult to find and predict the fault problem existing in system, easily caused the technical problem of hysteresis of fault monitoring in dust removal system.Method includes: abnormal judgment about standard threshold and standard service life to real-time state data, determine single abnormal state data;The abnormal key module corresponding to single abnormal state data is associated with the joint state abnormal judgment of associated upstream and downstream module, to determine comprehensive abnormal state data;Cascade judgment based on fault location and fault type is carried out to the abnormal characteristics in comprehensive abnormal state data, to obtain actual fault type;Strategy solution template is matched with problem characteristics, to obtain fault solution strategy.

Description

Technical Field

[0001] This application relates to the field of industrial dust removal technology, and in particular to a method, equipment and medium for condition monitoring based on an industrial dust removal system. Background Technology

[0002] In welding factories, the management of welding fume pollution is particularly important. Welding fumes are not ordinary dust; they are complex mixtures formed by the condensation of molten and vaporized metal, containing a large number of harmful substances such as metal oxides, fluorides, ozone, carbon monoxide, and nitrogen oxides, which can easily affect the health of workers.

[0003] Given the aforementioned welding fumes, industrial dust removal equipment or systems become particularly important. Industrial dust removal systems can effectively control the fumes in the welding area, ensuring air quality within the welding zone. Furthermore, these systems enable the filtered air to be emitted without pollution, thus guaranteeing green industrial production.

[0004] Traditional welding industry dust removal equipment or systems suffer from untimely monitoring of key components and lack of intelligent fault and operational monitoring. Often, system malfunctions or deterioration in dust removal efficiency are only discovered manually or with specialized testing equipment, resulting in a lag in dust removal system fault monitoring. This makes it difficult to provide timely early warnings of dust removal faults and to monitor the operational status of each key module in the industrial dust removal system in real time. Summary of the Invention

[0005] This application provides a status monitoring method, device, and medium based on an industrial dust removal system to solve the following technical problem: the operation status monitoring of each module in the existing welding industry dust removal system is not timely enough, making it difficult to detect and predict faults in the system, which easily leads to the lag in fault monitoring in the dust removal system.

[0006] The embodiments of this application adopt the following technical solutions:

[0007] On one hand, this application provides a status monitoring method for an industrial dust removal system, comprising: real-time status monitoring of the entire process of the industrial dust removal system using a preset industrial multimodal sensor to obtain real-time status data of each key module; performing anomaly judgment on the real-time status data under relevant standard thresholds and standard service life to determine single abnormal status data; performing joint status anomaly judgment on the abnormal key module corresponding to the single abnormal status data with relevant upstream and downstream modules to determine comprehensive abnormal status data; using a dust removal fault identification model to perform cascade judgment on the abnormal features in the comprehensive abnormal status data based on fault location and fault type to obtain the actual fault type; performing problem feature matching processing on the strategy solution template according to the actual fault type to obtain a fault solution strategy; and integrating the actual fault type and the fault solution strategy to generate dust removal operation warning information.

[0008] This application's embodiments can quickly identify single abnormal state data, improving fault response speed. By associating the states of upstream and downstream modules, it provides more comprehensive fault diagnosis, reducing false positives and false negatives. Furthermore, through cascading judgments, it accurately determines the fault location and type, facilitating rapid identification of the problem's root cause. It can also match corresponding solution strategies based on the actual fault type, improving fault resolution efficiency. Moreover, through predictive analysis, it can proactively identify potential problems, reducing unexpected downtime and lowering maintenance costs. Timely maintenance and troubleshooting extend the lifespan of industrial dust collection systems. Simultaneously, it integrates actual fault types and solution strategies to generate intuitive operational warning messages. It also provides easy-to-understand data and visualization tools to help management make faster decisions.

[0009] In one feasible implementation, a pre-set industrial multimodal sensor is used to perform real-time status monitoring of the entire industrial dust removal system to obtain real-time status data for each key module. Specifically, this includes: pre-installing the industrial multimodal sensor in each key module of the industrial dust removal system; wherein the key modules include: an air inlet module, an air pressure module, a filter module, a fan module, and an air outlet module; starting the industrial dust removal system; generating a data curve from the sequential operating status data of each key module to obtain an operating status curve; wherein the operating status curve is the data value fluctuation curve under the actual operation of the key module; the sequential process is the operating order of each key module in the industrial dust removal system; calculating the average value between the peaks and troughs of the operating status curve of each key module to obtain stable operating parameters; wherein the stable operating parameters include: air pressure, emission concentration, voltage, current, and temperature; and determining the stable operating parameters of each key module as the real-time status data.

[0010] In one feasible implementation, the real-time status data is subjected to anomaly judgment based on relevant standard thresholds and standard service life to identify single abnormal status data. Specifically, this includes: comparing the intake air pressure value and intake valve opening in the intake module with the intake standard threshold to identify abnormal intake status data; comparing the service life of the pulse valve in the pressure module with the standard service life of the pressure module to identify abnormal pressure status data; comparing the filter cartridge differential pressure and emission concentration in the filter cartridge module with the standard threshold to identify abnormal filter cartridge usage status data; and comparing the filter cartridge service life in the filter cartridge module with... The abnormal status data of the filter cartridges are determined by comparing them with the standard service life of the filter cartridges; the abnormal status data of the blower are determined by comparing the inverter operating parameters in the fan module with the standard threshold values ​​of the inverter operating parameters; the abnormal status data of the outlet air pressure and the outlet valve opening in the outlet air module are determined by comparing them with the standard threshold values ​​of the outlet air pressure and the outlet valve opening; the abnormal status data of the inlet air, the abnormal status data of the air pressure, the abnormal status data of the filter cartridges, the abnormal status data of the filter cartridges, the abnormal status data of the filter cartridges, the abnormal status data of the fan, and the abnormal status data of the outlet air are combined to obtain the single abnormal status data.

[0011] In one feasible implementation, the abnormal key module corresponding to the single abnormal state data is subjected to joint state anomaly judgment of related upstream and downstream modules to determine comprehensive abnormal state data. Specifically, this includes: identifying the abnormal state data types in the single abnormal state data; if all abnormal state data types contain both air intake abnormal state data and air pressure abnormal state data, then the abnormal joint state data is judged as air pressure end comprehensive abnormal state data; if all abnormal state data types contain both filter cartridge usage abnormal state data and filter cartridge lifespan abnormal state data, then the abnormal joint state data is judged as filter cartridge end comprehensive abnormal state data; if all abnormal state data types contain both fan abnormal state data and air outlet abnormal state data, then the abnormal joint state data is judged as fan end comprehensive abnormal state data; wherein, the comprehensive abnormal state data includes: air pressure end comprehensive abnormal state data, filter cartridge end comprehensive abnormal state data, fan end comprehensive abnormal state data, and the single abnormal state data.

[0012] In one feasible implementation, before using a dust removal fault identification model to perform cascaded judgments based on fault location and fault type on the abnormal features in the comprehensive abnormal state data to obtain the actual fault type, the method further includes: sequentially labeling the key modules in the industrial dust removal system to obtain the location code of each key module; collecting historical fault type data corresponding to the location code; and using a preset LightGBM multi-classification model to perform high-dimensional time-series data consistency fitting training on the location code and the historical fault type data to obtain the dust removal fault identification model.

[0013] In one feasible implementation, a dust removal fault identification model is used to perform cascaded judgments on the abnormal features in the comprehensive abnormal state data based on fault location and fault type to obtain the actual fault type. Specifically, this includes: extracting the current abnormal features from the comprehensive abnormal state data; using the dust removal fault identification model to perform decision logic calculations on the current abnormal features at the fault location level to obtain current fault location data; using the dust removal fault identification model to perform decision logic calculations on the current abnormal features at the lower-level fault type level to obtain current fault type data; and inputting the current fault location data and the current fault location data into the dust removal fault identification model to generate the latest fault location data and the latest fault location data, respectively. The current fault location data is mapped to the latest fault location data to obtain a first mapping relationship; the current fault type data is mapped to the latest fault location data to obtain a second mapping relationship; if the first mapping relationship and the second mapping relationship are partially different, the data content in the partially different mapping relationships is re-applied to the dust removal fault identification model until the same mapping relationship is obtained; if the first mapping relationship and the second mapping relationship are the same, the current fault location data and the current fault type data are integrated to obtain the actual fault type; wherein, the actual fault type includes: low inlet air velocity, filter cartridge failure, filter cartridge blockage, fan failure, low air pressure cleaning pressure, and low outlet air velocity.

[0014] In one feasible implementation, based on the actual fault type, the strategy solution template is matched with problem features to obtain a fault solution strategy. Specifically, this includes: performing semantic extraction processing on the actual fault type to obtain fault keywords; combining the fault keywords with relevant problem statements and extracting the problem features; performing template matching processing on the problem features using a preset strategy solution template to determine the corresponding fault solution strategy template; and performing strategy information feedback processing on the fault solution strategy template to obtain the fault solution strategy.

[0015] In one feasible implementation, the actual fault type and the fault resolution strategy are integrated to generate dust removal operation warning information. Specifically, this includes: determining the fault severity level based on the actual fault type; color-coding the actual fault type and the corresponding fault resolution strategy according to the fault severity level; and integrating the color-coded actual fault type and the corresponding fault resolution strategy into a data table to generate the dust removal operation warning information under different color codes.

[0016] Secondly, embodiments of this application also provide a status monitoring device based on an industrial dust removal system, the device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to execute a status monitoring method based on an industrial dust removal system as described in any of the above embodiments.

[0017] Thirdly, embodiments of this application also provide a non-volatile computer storage medium, wherein the storage medium is a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium stores at least one program, each program including instructions, and the instructions, when executed by a terminal, cause the terminal to execute a status monitoring method based on an industrial dust removal system as described in any of the above embodiments.

[0018] This application provides a method, device, and medium for status monitoring of an industrial dust removal system. Compared with the prior art, the embodiments of this application have the following beneficial technical effects:

[0019] 1. Real-time monitoring and data analysis: Capable of real-time monitoring of the industrial dust removal system's operating status, ensuring the equipment is always in optimal working condition. By collecting and analyzing real-time status data, it provides data support to operators and maintenance personnel, helping them make more informed decisions.

[0020] 2. Anomaly Detection and Diagnosis: Capable of quickly identifying single abnormal data states, improving fault response speed. By correlating the status of upstream and downstream modules, it provides more comprehensive fault diagnosis, reducing false positives and false negatives.

[0021] 3. Fault Location and Resolution: Cascading diagnostics accurately pinpoint the location and type of fault, facilitating rapid identification of the problem's root cause. Matching appropriate resolution strategies to the actual fault type improves fault resolution efficiency.

[0022] 4. System Optimization and Maintenance: Predictive analytics helps identify potential problems early, reducing unexpected downtime and maintenance costs. Timely maintenance and troubleshooting extend the lifespan of industrial dust collection systems.

[0023] 5. Enhanced Safety: Ensures efficient operation of industrial dust removal systems, reduces dust emissions, and protects the environment. Reduces downtime and malfunctions, ensuring operator safety.

[0024] 6. Improved Operational Efficiency: By optimizing system operation, energy consumption is reduced and energy utilization efficiency is improved. Downtime due to equipment failure is reduced, thereby increasing overall production efficiency.

[0025] 7. Information Integration and Visualization: Integrates actual fault types and resolution strategies to generate intuitive operational warning messages. Provides easy-to-understand data and visualization tools to help management make faster decisions. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0027] Figure 1 A flowchart of a status monitoring method based on an industrial dust removal system is provided for embodiments of this application;

[0028] Figure 2 This is a schematic diagram of the structure of a condition monitoring device based on an industrial dust removal system, provided as an embodiment of this application. Detailed Implementation

[0029] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0030] It should be noted that the dust removal process in an industrial dust removal system is expressed in the following order: air inlet module, air pressure module, filter module, fan module, and air outlet module. That is, the air inlet module performs dust removal by utilizing the air pressure changes in the air pressure module, then the filter module filters the air, and finally the fan module blows the filtered air through the air outlet module to complete the purification of the air before it is discharged.

[0031] This application provides a method for monitoring the condition of an industrial dust removal system, such as... Figure 1 As shown, the condition monitoring method based on an industrial dust removal system specifically includes steps S101-S106:

[0032] S101. Real-time status monitoring of the entire process of the industrial dust removal system is carried out through preset industrial multimodal sensors to obtain real-time status data of each key module.

[0033] Specifically, industrial multimodal sensors are first pre-installed in each key module of the industrial dust removal system. These key modules include: air inlet module, air pressure module, filter module, fan module, and air outlet module.

[0034] Next, the industrial dust removal system is started. Data curves are generated from the sequential operational status data of each key module, resulting in an operational status curve. This operational status curve represents the data value fluctuation curve under the actual operation of the key module. The sequential process refers to the operating order of each key module in the industrial dust removal system.

[0035] Furthermore, the mean values ​​between the peaks and troughs of the operating status curves of each key module are calculated to obtain stable operating parameters. These stable operating parameters include: wind pressure, emission concentration, voltage, current, and temperature. These stable operating parameters for each key module are then defined as real-time status data.

[0036] As a feasible implementation method, industrial multimodal sensors suitable for industrial environments are selected. These sensors should be able to simultaneously detect multiple parameters, such as temperature, pressure, current, and voltage. Then, the key modules of the industrial dust removal system are identified, including the air inlet module, air pressure module, filter module, fan module, and air outlet module. Next, corresponding multimodal sensors are pre-installed in each of these key modules to ensure that the sensors can collect real-time data from each module. The industrial dust removal system is then started, ensuring that all modules operate normally in a predetermined sequence. Operating status data for each key module, including air pressure, emission concentration, voltage, current, and temperature, are collected through sensors. Real-time processing of the operating status data for each key module is also required to generate data curves reflecting the numerical fluctuations of each module under sequential flow. The operating status curves of each module are analyzed, and the average value between the peaks and troughs is calculated to obtain stable operating parameters, such as air pressure, emission concentration, voltage, current, and temperature. Finally, the calculated stable operating parameters are stored as real-time status data in the central processing system.

[0037] S102. Perform anomaly judgment on real-time status data under relevant standard thresholds and standard service life to identify single abnormal status data.

[0038] Specifically, it is also necessary to compare the intake air pressure value in the intake module with the intake valve opening to determine the intake standard threshold value and identify abnormal intake status data.

[0039] In one embodiment, the intake air pressure of the intake module is judged according to the standard threshold: if the air pressure is abnormal, an abnormality is marked, and the intake air pressure abnormality status data is obtained. Then, the status of the intake valve is further checked. If there is an abnormality, the intake valve abnormality status data is obtained.

[0040] Furthermore, the service life of the pulse valve in the pneumatic module is compared with the standard service life of the pneumatic pressure to determine the data of abnormal pneumatic pressure conditions.

[0041] In one embodiment, the pulse valve in the pneumatic module is subjected to a threshold judgment of the usage time to obtain abnormal state data. If there is no abnormal state data of the lifespan, the gas source pressure sensor is further judged to be abnormal. Finally, the abnormal state data is combined and determined to be the pneumatic pressure abnormal state data.

[0042] Furthermore, the filter cartridge pressure differential and emission concentration in the filter cartridge module are compared with the standard threshold values ​​of the filter cartridges to identify abnormal filter cartridge usage data. The service life of the filter cartridges in the filter cartridge module is also compared with the standard service life of the filter cartridges to identify abnormal filter cartridge lifespan data.

[0043] In one embodiment, the filter cartridge is first judged against the standard threshold of filter cartridge differential pressure and emission concentration. Abnormal states are respectively the abnormal state of needing to replace the filter cartridge and the abnormal state of filter cartridge damage. Then, it is further judged whether the service life of the filter cartridge has reached the standard service life. Finally, the abnormal judgment is completed, and the abnormal state data of filter cartridge use and abnormal state data of filter cartridge life are obtained.

[0044] Furthermore, the operating parameters of the frequency converter in the fan module are compared with the standard threshold values ​​of the frequency converter to determine the abnormal state data of the fan.

[0045] In one embodiment, it is also necessary to determine the abnormal operating status of the wind turbine, identify the fault parameters or operating parameters of the frequency converter in the wind turbine module, and then compare them with the standard thresholds to finally determine the abnormal state data of the wind turbine.

[0046] Furthermore, the air pressure value in the air outlet module is compared with the air outlet valve opening to determine the standard threshold value and identify abnormal air outlet status data.

[0047] In one embodiment, the intake air pressure of the air outlet module is judged according to the standard threshold: if the air pressure is abnormal, an abnormality is marked, and the abnormal status data of the air outlet air pressure is obtained. Then, the status of the air outlet valve is further checked. If there is an abnormality, the abnormal status data of the air outlet valve is obtained.

[0048] Furthermore, the abnormal air intake status data, abnormal air pressure status data, abnormal filter cartridge usage status data, abnormal filter cartridge lifespan status data, abnormal fan status data, and abnormal air outlet status data are combined to obtain single abnormal status data.

[0049] S103. Perform joint state anomaly judgment on the critical modules corresponding to the single abnormal state data and related upstream and downstream modules to determine the comprehensive abnormal state data.

[0050] Specifically, first identify the abnormal state data type in a single abnormal state data.

[0051] If all abnormal status data types contain both inlet air abnormality data and air pressure abnormality data, then the combined abnormal status data is identified as comprehensive air pressure-side abnormal status data. If all abnormal status data types contain both filter cartridge usage abnormality data and filter cartridge lifespan abnormality data, then the combined abnormal status data is identified as comprehensive filter cartridge-side abnormal status data. If all abnormal status data types contain both fan abnormality data and outlet air abnormality data, then the combined abnormal status data is identified as comprehensive fan-side abnormal status data.

[0052] The comprehensive abnormal status data includes: comprehensive abnormal status data of the air pressure end, comprehensive abnormal status data of the filter cartridge end, comprehensive abnormal status data of the fan end, and single abnormal status data.

[0053] As a feasible implementation method, in actual anomaly judgment, anomalies in related modules often lead to their impact, causing even normal upstream and downstream modules to trigger anomaly alarms. For example, in the comprehensive anomaly data at the air pressure end, if both intake air anomaly data and air pressure anomaly data are present, it is generally an anomaly in the air pressure module, and the problem is primarily marked as an air pressure end issue, with subsequent auxiliary repairs to the intake air module. Similarly, in the comprehensive anomaly data at the filter cartridge end, if both filter cartridge usage anomaly data and filter cartridge lifespan anomaly data are present, it is generally a core anomaly at the filter cartridge end, i.e., the filter cartridge module has experienced a critical anomaly. Subsequent repairs will focus on the filter cartridge itself, without further detailed analysis.

[0054] S104. Using the dust removal fault identification model, perform cascade judgments on the abnormal features in the comprehensive abnormal state data based on the fault location and fault type to obtain the actual fault type.

[0055] Specifically, it is also necessary to sequentially label the key modules in the industrial dust removal system in advance to obtain the location code of each key module. Historical fault type data corresponding to the location codes should then be collected.

[0056] Furthermore, by using the pre-defined LightGBM multi-classification model, the location coding and historical fault type data are subjected to high-dimensional time-series data consistency fitting training to obtain the dust removal fault identification model.

[0057] In one embodiment, the historical fault type data corresponding to the collected location codes are standardized in data format, and then abnormal features are extracted, including: time-domain features: mean, variance, peak factor; frequency-domain features: the amplitude of the dominant frequency after FFT transformation; and spatiotemporal correlation features: for example, the time delay correlation between inlet pressure difference and outlet dust concentration, and the phase difference between zone current fluctuation and cleaning cycle, etc. Then, based on the label definition method of location code-fault area-fault manifestation, a fault location identification model is constructed. Then, combined with the LightGBM multi-classification model, and after limiting the depth to prevent overfitting calculations, the model training and validation operations are completed, and finally, a dust removal fault identification model is generated.

[0058] Furthermore, the current abnormal features are extracted from the comprehensive abnormal state data.

[0059] Furthermore, using the dust removal fault identification model, decision logic calculations are performed on the current abnormal features at the fault location level to obtain the current fault location data. Also, using the dust removal fault identification model, decision logic calculations are performed on the current abnormal features at the lower-level fault type level to obtain the current fault type data.

[0060] In one embodiment, the following decision logic code can be used to implement the model calculation of the cascaded decision engine, for example:

[0061] def cascade_diagnosis(input_data):

[0062] #Level 1: Fault Location Determination

[0063] position=position_model.predict(input_data)

[0064] #Level 2: Invoke the corresponding fault type model

[0065] if position == "P02": # Filter bucket area

[0066] if bag_break_model.predict_proba(input_data)>0.85:

[0067] return "Filter cartridge damage fault"

[0068] elif bag_clog_model.predict(input_data)==True:

[0069] The message "Filter cartridge differential pressure fault" is displayed.

[0070] else:

[0071] return "Filter cartridge malfunction"

[0072] elif position=="P03":#Inlet air area

[0073] return pulse_valve_diagnosis(input_data) # Air intake module

[0074] Then, based on the anti-conflict mechanism, the decision logic calculations at the lower fault type level and fault location level are performed on the current abnormal features, so that the above dust removal fault identification model finally outputs the current fault location data and the current fault type data.

[0075] Furthermore, the current fault location data and the latest fault location data are input into the dust removal fault identification model, respectively, to generate the latest fault location data and the latest fault location data. Then, the current fault location data and the latest fault location data are mapped to obtain a first mapping relationship. Finally, the current fault type data and the latest fault location data are mapped to obtain a second mapping relationship.

[0076] If the first mapping relationship and the second mapping relationship have some differences, the data in the differences will be re-applied to the dust removal fault identification model until the same mapping relationship is obtained. If the first mapping relationship and the second mapping relationship are the same, the current fault location data and the current fault type data will be integrated to obtain the actual fault type. The actual fault types include: low inlet air velocity, filter cartridge failure, filter cartridge blockage, fan failure, low air pressure cleaning pressure, and low outlet air velocity.

[0077] As a feasible implementation method, mapping relationship comparison and integration are utilized. Specifically, the first and second mapping relationships are compared to check for any partially different mapping relationships. If partially different mapping relationships exist, the fault identification model is reprocessed on this part of the data until the same mapping relationship is obtained. If the first and second mapping relationships are identical, the current fault location data and the current fault type data are integrated to obtain the actual fault type. Finally, the integrated data is output as the actual fault type, such as low inlet air velocity, filter cartridge failure, filter cartridge blockage, fan failure, low air pressure cleaning pressure, and low outlet air velocity.

[0078] In other words, by comparing and re-identifying the mapping relationships, the accuracy of fault types can be ensured, reducing misdiagnosis. Furthermore, real-time data integration and mapping processing can quickly determine the actual fault type, improving fault diagnosis efficiency. Accurate fault type output helps in developing more effective fault response strategies, reducing maintenance time and costs. And through continuous monitoring and rapid response, the reliability of industrial dust removal systems can be enhanced, reducing downtime.

[0079] S105. Based on the actual fault type, perform problem feature matching processing on the strategy solution template to obtain the fault solution strategy.

[0080] Specifically, it is also necessary to perform semantic extraction processing on the actual fault types to obtain fault keywords, combine the fault keywords with relevant problem statements, and extract problem features.

[0081] Furthermore, by using a pre-defined strategy solution template, the problem characteristics are matched to determine the corresponding fault solution template. The fault solution template is then processed with strategy information feedback to obtain the fault resolution strategy.

[0082] S106. Integrate the actual fault types and fault resolution strategies into data to generate dust removal operation warning information.

[0083] Specifically, the severity level of the fault is determined based on the actual fault type. According to the severity level, the actual fault type and corresponding fault resolution strategy are color-coded. The color-coded fault types and corresponding fault resolution strategies are then integrated into a tabular data structure to generate dust collector operation warning messages under different color codes.

[0084] In addition, embodiments of this application also provide a condition monitoring device based on an industrial dust removal system, such as... Figure 2 As shown, the condition monitoring device 200 based on the industrial dust removal system specifically includes:

[0085] At least one processor 201. And a memory 202 communicatively connected to the at least one processor 201. The memory 202 stores instructions executable by the at least one processor 201, enabling the at least one processor 201 to execute:

[0086] By using pre-set industrial multimodal sensors, the real-time status monitoring of the industrial dust removal system is carried out throughout the entire process, and the real-time status data of each key module is obtained.

[0087] Anomalies are identified in real-time status data under relevant standard thresholds and standard service life.

[0088] The abnormal key modules corresponding to a single abnormal state data are subjected to joint state abnormality judgment of related upstream and downstream modules to determine the comprehensive abnormal state data.

[0089] By using the dust removal fault identification model, the abnormal features in the comprehensive abnormal state data are cascaded based on the fault location and fault type to obtain the actual fault type.

[0090] Based on the actual fault type, the problem characteristics of the strategy solution template are matched to obtain the fault solution strategy.

[0091] By integrating the actual fault types and fault resolution strategies, dust removal operation warning information is generated.

[0092] This application's embodiments can quickly identify single abnormal state data, improving fault response speed. By associating the states of upstream and downstream modules, it provides more comprehensive fault diagnosis, reducing false positives and false negatives. Furthermore, through cascading judgments, it accurately determines the fault location and type, facilitating rapid identification of the problem's root cause. It can also match corresponding solution strategies based on the actual fault type, improving fault resolution efficiency. Moreover, through predictive analysis, it can proactively identify potential problems, reducing unexpected downtime and lowering maintenance costs. Timely maintenance and troubleshooting extend the lifespan of industrial dust collection systems. Simultaneously, it integrates actual fault types and solution strategies to generate intuitive operational warning messages. It also provides easy-to-understand data and visualization tools to help management make faster decisions.

[0093] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device and medium embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the description of the method embodiments.

[0094] The devices and media provided in this application are one-to-one with the methods. Therefore, the devices and media also have similar beneficial technical effects as their corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be repeated here.

[0095] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0097] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0099] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0100] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0101] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0102] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0103] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of this specification.

Claims

1. A condition monitoring method based on an industrial dust removal system, characterized in that, The method includes: By using pre-set industrial multimodal sensors, the real-time status monitoring of the industrial dust removal system is carried out throughout the entire process, and the real-time status data of each key module is obtained. The real-time status data is subjected to anomaly judgment under relevant standard thresholds and standard service life to identify single abnormal status data. The abnormal key module corresponding to the single abnormal state data is subjected to joint state abnormality judgment of related upstream and downstream modules to determine the comprehensive abnormal state data. By using a dust removal fault identification model, the abnormal features in the comprehensive abnormal state data are subjected to cascaded judgment based on fault location and fault type to obtain the actual fault type, specifically including: Extract the current abnormal features from the comprehensive abnormal state data; Using the dust removal fault identification model, decision logic calculations are performed on the current abnormal features at the fault location level to obtain the current fault location data. Using the dust removal fault identification model, the decision logic calculations for the current abnormal features at the relevant lower-level fault type hierarchy are performed to obtain the current fault type data. The current fault location data and the current fault location data are respectively input into the dust removal fault identification model to generate the latest fault location data and the latest fault location data respectively; The current fault location data is mapped to the latest fault location data to obtain a first mapping relationship; and the current fault type data is mapped to the latest fault location data to obtain a second mapping relationship. If the first mapping relationship and the second mapping relationship are partially different, the data content in the partially different mapping relationships will be re-applied to the dust removal fault identification model until the same mapping relationship is obtained; If the first mapping relationship and the second mapping relationship are the same, then the current fault location data and the current fault type data are integrated to obtain the actual fault type; wherein, the actual fault type includes: low inlet air velocity, filter cartridge failure, filter cartridge blockage, fan failure, low air pressure cleaning pressure, and low outlet air velocity; Based on the actual fault type, the problem feature matching process is performed on the strategy solution template to obtain the fault solution strategy; The actual fault types and fault resolution strategies are integrated to generate dust removal operation warning information.

2. The condition monitoring method based on an industrial dust removal system according to claim 1, characterized in that, By using pre-set industrial multimodal sensors, the industrial dust removal system is monitored in real time throughout the entire process, obtaining real-time status data for each key module, specifically including: The industrial multimodal sensor is pre-installed in each key module of the industrial dust removal system; wherein the key modules include: air inlet module, air pressure module, filter module, fan module and air outlet module; Start the industrial dust removal system; For each of the key modules, data curves are generated by processing the sequential process data of the operating status to obtain an operating status curve; wherein, the operating status curve is the data value fluctuation curve of the key module under actual operation; the sequential process is the operating order of each key module in the industrial dust removal system; The mean between the peaks and troughs of the operating status curves of each key module is calculated to obtain stable operating parameters; wherein, the stable operating parameters include: wind pressure, emission concentration, voltage, current and temperature; The stable operating parameters of each key module are determined as the real-time status data.

3. The condition monitoring method based on an industrial dust removal system according to claim 1, characterized in that, The real-time status data is subjected to anomaly detection based on relevant standard thresholds and standard lifespan to identify single abnormal status data, specifically including: The intake air pressure value in the intake module is compared with the intake valve opening to determine the intake standard threshold value and identify abnormal intake status data. The service life of the pulse valve in the pneumatic module is compared with the standard service life of the pneumatic module to determine the data of abnormal pneumatic conditions. The filter cartridge pressure difference and emission concentration in the filter cartridge module are compared with the standard threshold of the filter cartridge to determine the abnormal state data of the filter cartridge; and the service life of the filter cartridge in the filter cartridge module is compared with the standard service life of the filter cartridge to determine the abnormal state data of the filter cartridge life. The abnormal status data of the fan are determined by comparing the operating parameters of the frequency converter in the fan module with the standard threshold of the frequency converter. The air pressure value in the air outlet module is compared with the air outlet valve opening to determine the standard threshold value and identify abnormal air outlet status data. The abnormal air intake status data, the abnormal air pressure status data, the abnormal filter cartridge usage status data, the abnormal filter cartridge lifespan status data, the abnormal fan status data, and the abnormal air outlet status data are combined to obtain the single abnormal status data.

4. The condition monitoring method based on an industrial dust removal system according to claim 1, characterized in that, The abnormal key module corresponding to the single abnormal state data is subjected to joint state abnormality judgment of related upstream and downstream modules to determine the comprehensive abnormal state data, specifically including: Identify the data type of the abnormal state in the single abnormal state data; If all of the above abnormal state data types contain both air intake abnormal state data and air pressure abnormal state data, then the combined abnormal state data will be judged as air pressure end comprehensive abnormal state data. If all of the above abnormal status data types contain abnormal status data for filter cartridge use and abnormal status data for filter cartridge life, then the combined abnormal status data will be judged as comprehensive abnormal status data for the filter cartridge end. If all of the above abnormal state data types contain both fan abnormal state data and air outlet abnormal state data, then the combined abnormal state data will be judged as the comprehensive abnormal state data of the fan end. The comprehensive abnormal status data includes: comprehensive abnormal status data of the air pressure end, comprehensive abnormal status data of the filter barrel end, comprehensive abnormal status data of the fan end, and single abnormal status data.

5. The condition monitoring method based on an industrial dust removal system according to claim 1, characterized in that, Before obtaining the actual fault type by performing a cascaded judgment based on fault location and fault type on the abnormal features in the comprehensive abnormal state data using a dust removal fault identification model, the method further includes: The key modules in the industrial dust removal system are sequentially numbered to obtain the position code of each key module. Collect historical fault type data corresponding to the location code; The dust removal fault identification model is obtained by using a preset LightGBM multi-classification model to perform high-dimensional time-series data consistency fitting training on the location code and the historical fault type data.

6. The condition monitoring method based on an industrial dust removal system according to claim 1, characterized in that, Based on the actual fault type, the strategy solution template is matched with problem features to obtain a fault solution strategy, which specifically includes: Semantic extraction processing is performed on the actual fault types to obtain fault keywords; The fault keywords are combined with relevant problem statements, and the problem features are extracted. By using a preset strategy solution template, the problem characteristics are subjected to template matching processing to determine the corresponding fault solution template. The fault resolution strategy template is processed by feedback of strategy information to obtain the fault resolution strategy.

7. The condition monitoring method based on an industrial dust removal system according to claim 1, characterized in that, The actual fault types and fault resolution strategies are integrated to generate dust removal operation warning information, specifically including: Based on the actual fault type, the severity level of the fault is determined; Based on the severity level of the fault, the actual fault type and the corresponding fault resolution strategy are marked with color information. The actual fault types marked with colors and the corresponding fault resolution strategies are integrated into a data table to generate the dust removal operation warning information under different color markings.

8. A condition monitoring device based on an industrial dust removal system, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, enabling the at least one processor to perform a status monitoring method based on an industrial dust removal system according to any one of claims 1-7.

9. A non-volatile computer storage medium, characterized in that, The storage medium is a non-volatile computer-readable storage medium that stores at least one program, each program including instructions that, when executed by a terminal, cause the terminal to perform a status monitoring method based on an industrial dust removal system according to any one of claims 1-7.

Images