Multi-strategy-based rod control device working condition labeling method, device and equipment

Abstract

The application discloses a multi-strategy-based rod control rod position equipment working condition labeling method and device and equipment, relates to the technical field of industrial time series data processing, and aims at the operation data of a rod control rod position equipment. The working conditions are automatically labeled according to preset rules through rule knowledge discrimination. The working conditions that need to be labeled by professional personnel are efficiently identified and labeled in a man-machine interactive mode. The labeling process is more efficient, fast and accurate. The labeled data can form a data set for constructing a fault diagnosis model, and the model precision is improved. The method comprises the following steps: acquiring working condition operation data of a rod control rod position equipment to be labeled; adopting a sliding window mode to label the working condition operation data to be labeled according to a preset working condition recognition rule; sending the labeled working condition operation data to a terminal held by a user, so that the user corrects and labels the labeled working condition operation data; and acquiring working condition operation data labeling results fed back by the user based on the terminal and storing the working condition operation data labeling results in a database.

Description

Technical Field

[0001] The present application relates to the fields of industrial time series data processing, sequence labeling and computer technology, and in particular to a method, device and equipment for labeling the working conditions of rod control and rod position equipment based on multiple strategies. Background Art

[0002] With the development of nuclear reactors, electromechanical equipment such as control rod drive mechanisms (CRDMs) are key components of reactors and primary circuit systems. The accuracy and timeliness of their operation are directly related to reactor power control and operational safety. With the rise of machine learning and artificial intelligence technologies in nuclear reactors, the mining of massive amounts of drive mechanism operating data and the development of intelligent fault diagnosis models for rod position control equipment can improve the accuracy of fault warnings. This can minimize unplanned downtime and maintenance time while ensuring safe and reliable equipment operation, ultimately ensuring safe and reliable management of critical reactor equipment.

[0003] In the related art, the machine learning and deep learning models used are mainly supervised learning methods, such as random forests, CNN (Convolutional Neural Networks), LSTM (Long Short-Term Memory), BERT (Bidirectional Encoder Representations from Transformers, a deep bidirectional language representation model based on the Transformer architecture), attention mechanisms, etc. Taking the drive mechanism as an example as the research object of the rod control and rod position equipment, the drive mechanism operation data collected from nuclear power plants or prototype tests are generally multi-signal sequence data with time stamps, such as current and vibration. In order to build an intelligent fault diagnosis model, the collected operation data needs to be manually labeled for normal working conditions and fault conditions. However, the applicant recognizes that there are many types of drive mechanism operating conditions and their identification requires professional knowledge and experience. Manual identification is difficult, and the amount of data to be labeled is large. Manual data labeling is time-consuming, labor-intensive, and inefficient, resulting in low accuracy and completeness of sample labels, which in turn leads to low accuracy of the intelligent fault diagnosis model. Summary of the Invention

[0004] In view of this, the present application provides a method, device and equipment for labeling the working conditions of rod-controlled rod position equipment based on multiple strategies. The main purpose is to solve the following problems: the manual identification of the operating conditions of the driving mechanism is difficult, the amount of data to be labeled is large, it is time-consuming and labor-intensive, and the efficiency is low, resulting in the accuracy and completeness of the sample labels being low, and thus causing the low accuracy of the intelligent fault diagnosis model.

[0005] According to the first aspect of the present application, a method for labeling working conditions of a rod control and rod position device based on multiple strategies is provided, the method comprising:

[0006] Obtain the operating data of the rod control and rod position equipment to be marked;

[0007] Obtaining a sliding window method and a preset working condition identification rule, and using the sliding window method to perform working condition labeling on the working condition operation data to be labeled according to the preset working condition identification rule to obtain the labeled working condition operation data;

[0008] Sending the annotated working condition data to a terminal held by a user, so that the user can correct and annotate the annotated working condition data;

[0009] Obtain the working condition operation data annotation result fed back by the user based on the terminal held by the user, and store the working condition operation data annotation result in a database.

[0010] Optionally, the obtaining of the operating data of the rod control position device to be marked includes:

[0011] Reading the operation information of the rod control and rod position device from the database;

[0012] The operation information is visualized using a visualization tool to obtain the operating data of the rod control rod position device to be marked.

[0013] Optionally, the step of performing working condition labeling on the working condition operation data to be labeled according to the preset working condition identification rule in the sliding window manner to obtain the labeled working condition operation data includes:

[0014] Determining an initial time point of the operating condition data to be labeled, wherein the operating condition data to be labeled is time series data;

[0015] Starting from the initial time point, the preset operating condition identification rule is used to detect the operating condition data of each sliding window until the time series detection is completed to obtain the labeled operating condition data. The sliding window slides above the operating condition data to be labeled. The preset operating condition identification rule includes multiple operating condition types and operating condition characteristics corresponding to each operating condition type.

[0016] Optionally, detecting the operating condition data of each sliding window using the preset operating condition identification rule starting from the initial time point includes:

[0017] Detecting the operating condition data of the first sliding window;

[0018] When the working condition operation data of the first sliding window belongs to a non-stationary waveform working condition, the preset working condition recognition rule is used to detect the working condition operation data of the first sliding window;

[0019] If the working condition operation data of the first sliding window meets the preset working condition recognition rule, a first working condition feature corresponding to the working condition operation data of the first sliding window is determined, and a working condition label corresponding to the first working condition feature is read in the preset working condition recognition rule;

[0020] The working condition operation data of the first sliding window is labeled by using the working condition label corresponding to the first working condition feature;

[0021] A preset labeling period is obtained, the first sliding window is moved in a direction away from the initial time point by the preset labeling period to obtain a second sliding window, and the working condition operation data of the second sliding window is detected.

[0022] Optionally, after the working condition operation data of the first sliding window is detected by using the preset working condition recognition rule, the method further includes:

[0023] If the working condition operation data of the first sliding window does not meet the preset working condition recognition rule, a preset window sliding time is obtained, the first sliding window is moved in a direction away from the initial time point by the preset window sliding time to obtain a third sliding window;

[0024] The working condition operation data of the third sliding window is detected.

[0025] Optionally, after the working condition operation data of the first sliding window is detected, the method further includes:

[0026] When the working condition operation data of the first sliding window belongs to a stationary waveform working condition and the working condition operation data of the first sliding window meets the preset working condition recognition rule, a second working condition feature corresponding to the working condition operation data of the first sliding window is determined, a working condition label corresponding to the second working condition feature is read in the preset working condition recognition rule, and the working condition operation data of the first sliding window is labeled by using the working condition label corresponding to the second working condition feature;

[0027] A preset offset is obtained, the first sliding window is moved in a direction away from the initial time point by the preset offset to obtain a fourth sliding window, and a third working condition feature corresponding to the working condition operation data of the fourth sliding window is determined;

[0028] The second working condition feature is compared with the third working condition feature;

[0029] If the second working condition feature is the same as the third working condition feature, the working condition running data of the fourth sliding window is labeled by using the working condition label corresponding to the second working condition feature.

[0030] The fourth sliding window is moved by the preset offset away from the initial time point to obtain a fifth sliding window.

[0031] The second working condition feature is compared with a fourth working condition feature corresponding to the working condition running data of the fifth sliding window.

[0032] Optionally, after the second working condition feature is compared with the third working condition feature, the method further includes:

[0033] If the second working condition feature is different from the third working condition feature, a preset window sliding time is obtained, the fourth sliding window is moved by the preset window sliding time away from the initial time point to obtain a sixth sliding window.

[0034] The working condition running data of the sixth sliding window is detected.

[0035] Optionally, the obtaining of the working condition running data labeling result fed back by the user based on the terminal includes:

[0036] The corrected working condition running data fed back by the user based on the terminal is obtained, the corrected working condition running data is labeled with a target working condition type, the target working condition type does not belong to the plurality of working condition types included in the preset working condition recognition rule, and the target working condition type is associated with the plurality of working condition types.

[0037] The starting time point, the ending time point and the target working condition label selected by the user based on the terminal are obtained, and a target time sequence between the starting time point and the ending time point is determined.

[0038] Starting from the starting time point, the corrected working condition running data in a sliding window is labeled by using the target working condition label, the sliding window is moved by a preset labeling period away from the starting time point to obtain a next sliding window, the corrected working condition running data in the next sliding window is labeled by using the target working condition label, and the operation is repeated until the target time sequence labeling is completed, so as to obtain the working condition running data labeling result.

[0039] According to the second aspect of the present application, a working condition labeling device for a rod control and rod position device based on multiple strategies is provided, and the device includes:

[0040] An acquisition module is configured to acquire working condition running data of a rod control and rod position device to be labeled.

[0041] a labeling module, configured to obtain a sliding window method and a preset working condition identification rule, and perform working condition labeling on the working condition operation data to be labeled using the sliding window method according to the preset working condition identification rule to obtain the labeled working condition operation data;

[0042] A review module, configured to send the annotated operating condition data to a terminal held by a user, so that the user can correct and annotate the annotated operating condition data;

[0043] The storage module is used to obtain the working condition operation data annotation results fed back by the user based on the terminal held by the user, and store the working condition operation data annotation results in a database.

[0044] Optionally, the acquisition module is used to read the operating information of the rod control and rod position device from the database; and use a visualization tool to visualize the operating information to obtain the operating data of the working condition to be marked of the rod control and rod position device.

[0045] Optionally, the labeling module is used to determine the initial time point of the operating condition data to be labeled, and the operating condition data to be labeled is time series data; starting from the initial time point, the operating condition data of each sliding window is detected using the preset operating condition identification rule until the time series detection is completed to obtain the labeled operating condition data, and the sliding window slides above the operating condition data to be labeled, and the preset operating condition identification rule includes multiple operating condition types and operating condition characteristics corresponding to each operating condition type.

[0046] Optionally, the labeling module is used to detect the operating condition data of the first sliding window; when the operating condition data of the first sliding window belongs to a non-stationary waveform operating condition, the preset operating condition identification rule is used to detect the operating condition data of the first sliding window; if the operating condition data of the first sliding window meets the preset operating condition identification rule, the first operating condition feature corresponding to the operating condition data of the first sliding window is determined, and the operating condition label corresponding to the first operating condition feature is read in the preset operating condition identification rule; the operating condition data of the first sliding window is labeled with the operating condition label corresponding to the first operating condition feature; a preset labeling period is obtained, the first sliding window is moved away from the initial time point by the preset labeling period, a second sliding window is obtained, and the operating condition data of the second sliding window is detected.

[0047] Optionally, the labeling module is used to obtain the preset window sliding time if the operating condition data of the first sliding window does not comply with the preset operating condition identification rule, move the first sliding window away from the initial time point by the preset window sliding time, and obtain a third sliding window; and detect the operating condition data of the third sliding window.

[0048] Optionally, the labeling module is configured to, when the operating condition data of the first sliding window belongs to a steady waveform operating condition and the operating condition data of the first sliding window meets the preset operating condition identification rule, determine a second operating condition feature corresponding to the operating condition data of the first sliding window, read a operating condition label corresponding to the second operating condition feature in the preset operating condition identification rule, and label the operating condition data of the first sliding window using the operating condition label corresponding to the second operating condition feature; obtain a preset offset, move the first sliding window away from the initial time point by the preset offset to obtain a fourth sliding window, determine a third operating condition feature corresponding to the operating condition data of the fourth sliding window; compare the second operating condition feature with the third operating condition feature; if the second operating condition feature is the same as the third operating condition feature, label the operating condition data of the fourth sliding window using the operating condition label corresponding to the second operating condition feature; move the fourth sliding window away from the initial time point by the preset offset to obtain a fifth sliding window; and compare the second operating condition feature with the fourth operating condition feature corresponding to the operating condition data of the fifth sliding window.

[0049] Optionally, the labeling module is used to obtain a preset window sliding time if the second operating condition characteristic is different from the third operating condition characteristic, move the fourth sliding window away from the initial time point by the preset window sliding time, and obtain a sixth sliding window; and detect the operating condition operation data of the sixth sliding window.

[0050] Optionally, the storage module is used to obtain the corrected operating condition data fed back by the user based on the terminal held, the corrected operating condition data is marked with a target operating condition type, the target operating condition type does not belong to the multiple operating condition types included in the preset operating condition identification rule, and the target operating condition type is associated with multiple operating condition types; obtain the starting time point, ending time point and target operating condition label selected by the user based on the terminal held, and determine the target time series between the starting time point and the ending time point; starting from the starting time point, use the target operating condition label to mark the corrected operating condition data in the sliding window, move the sliding window in the direction away from the starting time point by a preset marking period, obtain the next sliding window, use the target operating condition label to mark the corrected operating condition data in the next sliding window, repeat the operation until the target time series marking is completed, and obtain the operating condition data marking result.

[0051] According to a third aspect of the present application, a device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the methods described in the first aspect when executing the computer program.

[0052] By means of the above technical solution, the present application provides a method, device and equipment for labeling the working condition of rod control and rod position equipment based on multiple strategies. The present application obtains the working condition operation data to be labeled of the rod control and rod position equipment, obtains a sliding window method and preset working condition identification rules, uses a sliding window method to label the working condition of the working condition operation data to be labeled according to the preset working condition identification rules, obtains the labeled working condition operation data, sends the labeled working condition operation data to the terminal held by the user, so that the user can correct and label the labeled working condition operation data, obtains the working condition operation data labeling results based on the user's feedback on the terminal held, and stores the working condition operation data labeling results in a database. For the operating data collected from nuclear power plants or prototype tests, the working conditions identified by rule knowledge are periodically and automatically labeled according to preset rules. Some working conditions that require professional personnel to label based on experience are efficiently identified and labeled in a human-computer interactive manner through visualization technology. The labeled data is stored in a relational database for on-demand use when building an intelligent model. The operating condition labeling of the timing data of the rod control and rod position equipment based on human-machine collaboration is more efficient, fast and accurate than manual processing. The labeled data can be stored in a standardized manner and constitute a data set for constructing the fault diagnosis intelligent model, thereby improving the accuracy of the fault diagnosis model.

[0053] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present application. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0055] Figure 1 A schematic flow chart of a method for marking working conditions of a rod control and rod position device based on multiple strategies provided in an embodiment of the present application is shown;

[0056] Figure 2A A schematic flow chart of another method for marking working conditions of a rod control and rod position device based on multiple strategies provided in an embodiment of the present application is shown;

[0057] Figure 2B A schematic diagram of operating data of a working condition to be marked provided in an embodiment of the present application is shown;

[0058] Figure 2C A schematic diagram of the result of automatic labeling based on preset working condition identification rules provided by an embodiment of the present application is shown;

[0059] Figure 2D A schematic diagram of overlapping sampling results of a single holding working condition provided by an embodiment of the present application, which is automatically labeled based on preset working condition identification rules, is shown;

[0060] Figure 2E A schematic diagram showing the result of correcting the lifting to a sliding bar condition during an expert review and annotation correction process provided by an embodiment of the present application is shown;

[0061] Figure 2F A schematic diagram of expert annotation results provided by an embodiment of the present application is shown;

[0062] Figure 2G A flowchart of a method for labeling working conditions of a rod control and rod position device based on multiple strategies provided in an embodiment of the present application is shown;

[0063] Figure 3 A schematic structural diagram of a multi-strategy rod control and rod position equipment working condition marking method provided in an embodiment of the present application is shown;

[0064] Figure 4 A schematic diagram of the device structure of a device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0065] The following describes exemplary embodiments of the present application in more detail with reference to the accompanying drawings. Although exemplary embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present application and to fully convey the scope of the present application to those skilled in the art.

[0066] The embodiment of the present application provides a method for marking the working condition of a rod control and rod position device based on multiple strategies, such as Figure 1 As shown, the method includes:

[0067] 101. Obtain the operating data of the rod control rod position equipment under the working condition to be marked.

[0068] Drive mechanisms and other electromechanical equipment are key main equipment for reactors and primary circuit systems. The accuracy and timeliness of their actions are directly related to the power control and operational safety of the reactor. However, traditional control decisions and fault diagnosis rely too much on diagnostic experts and professional technicians, resulting in problems such as a relatively scarce number of diagnostic experts and low diagnostic timeliness. Currently, machine learning and deep learning intelligent models are mostly used for intelligent diagnosis. Therefore, the accuracy of the true labels of samples and the integrity of the sample set are very important for improving the intelligent model. Due to the wide variety of operating conditions of the drive mechanism (for example, normal conditions include lifting, lowering, single holding, and rod dropping, and fault conditions include sliding rods, stuck rods, and inability to lift rods), and the high difficulty of manual identification, which requires professional knowledge and experience, there is currently a lack of working condition labeling methods and tools for rod control and rod position equipment. This results in time-consuming and labor-intensive data labeling, difficulty in constructing sample sets, small sample sizes, and errors in sample labels, resulting in low accuracy of intelligent fault diagnosis models.

[0069] To solve this problem, this application proposes a multi-strategy-based working condition labeling method for rod control and rod position equipment. The execution subject of this application can be a working condition labeling system, which relies on the computing power of the server to provide services to users. The server can be an independent server or a server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), as well as servers for basic cloud computing such as big data and artificial intelligence platforms, so that the working condition labeling system can accurately and efficiently complete the labeling of various working conditions.

[0070] In an embodiment of the present application, the operating condition labeling system obtains the operating condition data to be labeled of the rod control rod position equipment, wherein the operating condition data to be labeled is usually in the format of csv (Comma-Separated Values, plain text file), excel (spreadsheet), etc., or is time series data directly read from the database, and then displayed through a visualization tool page such as Echarts (a data visualization chart library based on JavaScript).

[0071] 102. Obtain a sliding window method and a preset working condition identification rule, use the sliding window method to perform working condition labeling on the working condition operation data to be labeled according to the preset working condition identification rule, and obtain the labeled working condition operation data.

[0072] In an embodiment of the present application, the operating condition labeling system utilizes a sliding window method and preset operating condition identification rules, employing the sliding window method to label the operating condition data to be labeled according to the preset operating condition identification rules, thereby obtaining the labeled operating condition data. Normal operating conditions of the rod control and rod position equipment operating data can be identified and interpreted using specialized rule knowledge. Therefore, the present application provides preset operating condition identification rules, converting the preset operating condition identification rules into a computer operating condition interpretation program, thereby enabling periodic automated operating condition labeling of the operating condition labeling system.

[0073] 103. Send the marked working condition operation data to the terminal held by the user, so that the user can correct and mark the marked working condition operation data.

[0074] In an embodiment of the present application, for some working conditions that require professional personnel to mark based on experience, such as fault conditions such as stuck rods, sliding rods, and inability to lift the driving mechanism, the working condition marking system sends the marked working condition operation data to the terminal held by the user, so that the user can correct and mark the marked working condition operation data, which is more efficient and accurate than manual processing.

[0075] 104. Obtain the working condition operation data annotation results fed back by the user based on the terminal held by the user, and store the working condition operation data annotation results in the database.

[0076] In this embodiment of the present application, the operating condition annotation system obtains the operating condition data annotation results provided by the user based on the terminal they hold and stores them in a database. By storing the data and corresponding labels in a relational database as structured data, it can provide high-quality datasets to support the construction of various intelligent fault diagnosis models and facilitate data reuse.

[0077] The method provided in the embodiments of the present application obtains operating data of rod control and position equipment to be labeled, acquires a sliding window method and preset operating condition identification rules, uses the sliding window method to label the operating data according to the preset operating condition identification rules, obtains the labeled operating data, sends the labeled operating data to a user's terminal, and allows the user to correct and label the labeled operating data, obtains the labeling results of the operating data based on the user's terminal feedback, and stores the labeling results in a database. For operating data collected from nuclear power plants or prototype tests, operating conditions identified by rule knowledge are periodically and automatically labeled according to preset rules. Some operating conditions that require professional experience to label are efficiently identified and labeled using visualization technology in a human-computer interactive manner. The labeled data is stored in a relational database for on-demand use in intelligent model construction. This human-computer collaborative approach to labeling operating conditions for time-series data of rod control and position equipment is more efficient, rapid, and accurate than manual processing. Furthermore, the labeled data can be stored in a standardized manner and constitute a dataset for constructing an intelligent fault diagnosis model, thereby improving the accuracy of the fault diagnosis model.

[0078] Furthermore, as a refinement and expansion of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, the embodiment of the present application provides another method for marking the working condition of the rod control rod position equipment based on multiple strategies, such as Figure 2A As shown, the method includes:

[0079] 201. Obtain the operating data of the rod control rod position device under the working condition to be marked.

[0080] In the embodiment of the present application, the working condition marking system reads the operating information of the rod control rod position device from the database, wherein the operating information is data in formats such as csv, excel, or time series data directly read from the database. Then, the working condition marking system uses visualization tools such as Echarts to visualize the operating information and obtain the working condition operating data to be marked for the rod control rod position device, such as Figure 2B The time series data of multiple signals are shown, where LC represents the lifting coil current signal, MC represents the moving coil current signal, SC represents the holding coil current signal, and ZD1 represents the vibration signal.

[0081] 202. Determine the initial time point of the operating data of the working condition to be marked.

[0082] In an embodiment of the present application, the operating condition labeling system needs to determine the initial time point of the operating condition data to be labeled, so that the sliding window starts sliding from the initial time point (i.e., 0ms) to improve data processing efficiency.

[0083] The operating condition labeling system then uses preset operating condition identification rules to detect the operating condition data in each sliding window starting from the initial time point until the time series detection is complete, obtaining the labeled operating condition data. The sliding window slides over the operating condition data to be labeled. The preset operating condition identification rules include multiple operating condition types and corresponding operating condition characteristics for each operating condition type. The specific process is as follows: Steps 203 to 210.

[0084] 203. Detect the operating condition data of the first sliding window; when the operating condition data of the first sliding window belongs to a non-stationary waveform condition, execute the following step 204; when the operating condition data of the first sliding window belongs to a stationary waveform condition and the operating condition data of the first sliding window meets the preset operating condition identification rules, execute the following step 207.

[0085] In an embodiment of the present application, the operating condition annotation system detects the operating condition data of the first sliding window, where the first sliding window refers to the sliding window at the initial time point; when the operating condition data of the first sliding window belongs to a non-stationary waveform operating condition, the following step 204 is executed; when the operating condition data of the first sliding window belongs to a stationary waveform operating condition and the operating condition data of the first sliding window meets the preset operating condition identification rules, the following step 207 is executed. Among them, non-stationary waveform operating conditions include the driving mechanism lifting operating condition, the lowering operating condition, the rod dropping operating condition, etc., and the stationary waveform operating condition includes the driving mechanism single-hold operating condition, the double-hold operating condition, etc., indicating that the signal quantity remains unchanged, or only changes in the initial stage but after a long waiting time, the signals tend to remain stable.

[0086] Specifically, the preset working condition identification rules are as follows:

[0087] Lifting condition: When the moving coil current signal and the holding coil current signal rise at the same time, it is regarded as the starting point of the lifting state cycle, divided into 833ms periods. If the period is less than 833ms, it is discarded;

[0088] Insertion working condition: the holding coil current signal rises first, and then after 20ms, the lifting coil current signal rises and changes to the insertion state. The time when the holding coil current signal just rises is regarded as the starting point of the insertion state cycle, which is divided into 833ms periods. If the period is less than 833ms, it will be discarded;

[0089] Single hold mode: The holding coil current signal is approximately 4.7A (theoretical value is 4.7A, actual measurement may have signal disturbance fluctuations). When the lifting and moving coil current signals are both 0, the single hold mode is in operation. The duration of the single hold mode is manually controlled and divided into 833ms periods. If the period is less than 833ms, the state is abandoned.

[0090] Double holding working condition: the mobile coil current signal and the holding coil current signal are about 4.7 A, which is the double holding state. The duration of the double holding state is controlled by the user, and is divided by 833 ms as a period. If the period is less than 833 ms, it is discarded.

[0091] Drop rod working condition: the holding coil current signal decreases from 4.7 A until the vibration signal tends to be flat, which is the drop rod state. The holding coil current signal just decreases from 4.7 A is taken as the starting point of the drop rod state period, and is divided by 833 ms as a period.

[0092] It should be noted that since the input requirement of the machine learning and deep learning model construction is fixed dimension, the sequence should be labeled according to the fixed period length during data labeling. Taking the driving mechanism running condition as an example, the standard running speed is 0.833 seconds / step, which is specifically that when the sequence data frequency is 1 khz, every 833 consecutive time points are a period, so the labeling period is preset to 833. Therefore, the running data to be labeled should meet the preset labeling period frequency and contain the signals involved in the preset working condition recognition rule. In the embodiment of the present application, for the rod control rod position equipment driving mechanism, the data frequency needs to be 1 KHZ, and the mobile coil current signal, the holding coil current signal and the vibration signal are included.

[0093] 204、When the working condition running data of the first sliding window belongs to a non-stationary waveform working condition, the preset working condition recognition rule is used to detect the working condition running data of the first sliding window; if the working condition running data of the first sliding window meets the preset working condition recognition rule, step 205 is executed; if the working condition running data of the first sliding window does not meet the preset working condition recognition rule, step 206 is executed.

[0094] In the embodiment of the present application, when the working condition running data of the first sliding window belongs to a non-stationary waveform working condition, i.e. driving mechanism lifting working condition, down-inserting working condition, drop rod working condition, etc., the preset working condition recognition rule is used to detect the working condition running data of the first sliding window; if the working condition running data of the first sliding window meets the preset working condition recognition rule, the working condition is labeled, and step 205 is executed; if the working condition running data of the first sliding window does not meet the preset working condition recognition rule, the sliding window is moved, and step 206 is executed.

[0095] 205、If the working condition running data of the first sliding window meets the preset working condition recognition rule, the working condition running data of the first sliding window is labeled by using the preset working condition recognition rule, and the first sliding window is moved by a preset labeling period away from the initial time point to obtain a second sliding window; when the time sequence detection is completed, step 211 is executed; when the time sequence detection is not completed, step 203 is executed by using the second sliding window.

[0096] In an embodiment of the present application, if the operating condition data of the first sliding window meets the preset operating condition identification rules, the operating condition labeling system determines the first operating condition feature corresponding to the operating condition data of the first sliding window and reads the operating condition label corresponding to the first operating condition feature from the preset operating condition identification rules. For example, if the operating condition feature is that the moving coil current signal and the holding coil current signal rise simultaneously, the operating condition labeling system determines the corresponding operating condition label as a lifting condition from the preset operating condition identification rules based on this operating condition feature. Next, the operating condition labeling system labels the operating condition data of the first sliding window using the operating condition label corresponding to the first operating condition feature. The operating condition labeling system then shifts the first sliding window away from the initial time point by a preset labeling period to obtain a second sliding window, where the second sliding window refers to a sliding window at a time point one preset labeling period away from the initial time point. It should be noted that if the sliding window slides to the end of the time series for less than the preset period, it is directly discarded, and the subsequent process is similar. If the operating condition labeling system has already used the sliding window to detect the entire time series of the operating condition data to be labeled, the labeled operating condition data is obtained, and step 211 is executed below to send the labeled operating condition data to professionals for re-inspection and correction. If the operating condition labeling system has not detected the entire time series of the operating condition data to be labeled, step 203 is executed using a second sliding window, i.e., the operating condition data in the second sliding window is re-detected.

[0097] 206. If the operating condition data of the first sliding window does not meet the preset operating condition identification rules, the first sliding window is moved away from the initial time point by the preset window sliding time to obtain a third sliding window; when the time series detection is completed, the following step 211 is executed; when the time series detection is not completed, the third sliding window is used to execute step 203.

[0098] In an embodiment of the present application, if the operating condition data of the first sliding window does not meet the preset operating condition identification rules, the operating condition labeling system obtains the preset window sliding time, which can be 1ms, and then moves the first sliding window in the direction away from the initial time point by the preset window sliding time to obtain a third sliding window, wherein the third sliding window refers to a sliding window at a time point 1ms away from the initial time point. If the operating condition labeling system has used the sliding window to detect the entire time series of the operating condition data to be labeled at this time, the labeled operating condition data is obtained, and the following step 211 is executed to send the labeled operating condition data to professionals for re-inspection and correction. If the operating condition labeling system has not detected the entire time series of the operating condition data to be labeled at this time, the third sliding window is used to execute step 203, that is, the operating condition data of the third sliding window is re-detected until the operating condition that meets the preset operating condition identification rules is identified.

[0099] Based on steps 204 to 206, for the working condition for which the preset working condition identification rules have been set, the working condition labeling system can realize automatic working condition labeling based on the identification rules, and obtain the following Figure 2C The results of automatic labeling based on the preset working condition identification rules are shown. There is no need for the user to filter or specify which specific rule to execute. The system automatically performs judgments in sequence according to the computer working condition interpretation program of the preset working condition identification rules, thereby improving the accuracy of working condition labeling.

[0100] 207. When the operating condition data of the first sliding window belongs to a steady waveform operating condition and meets a preset operating condition identification rule, the operating condition data of the first sliding window is marked using the preset operating condition identification rule, and the first sliding window is moved away from the initial time point by a preset offset to obtain a fourth sliding window.

[0101] For steady waveform working conditions, such as single-hold and double-hold working conditions of the drive mechanism, the working condition labeling system provides an overlapping sampling algorithm based on the labeling based on the preset working condition identification rules to achieve the expansion of this type of working condition data. In an embodiment of the present application, when the working condition operation data of the first sliding window belongs to a steady waveform working condition and the working condition operation data of the first sliding window meets the preset working condition identification rules, the working condition labeling system determines the second working condition feature corresponding to the working condition operation data of the first sliding window, then reads the working condition label corresponding to the second working condition feature in the preset working condition identification rules, and uses the working condition label corresponding to the second working condition feature to label the working condition operation data of the first sliding window. Then, the working condition labeling system obtains a preset offset and moves the first sliding window away from the initial time point by the preset offset to obtain a fourth sliding window.

[0102] 208. Determine the third operating condition characteristic corresponding to the operating condition data of the fourth sliding window, and compare the second operating condition characteristic with the third operating condition characteristic; if the second operating condition characteristic is the same as the third operating condition characteristic, execute the following step 209; if the second operating condition characteristic is different from the third operating condition characteristic, execute the following step 210.

[0103] In an embodiment of the present application, the operating condition labeling system determines the third operating condition feature corresponding to the operating condition operation data of the fourth sliding window, and compares the second operating condition feature with the third operating condition feature; if the second operating condition feature is the same as the third operating condition feature, execute the following step 209; if the second operating condition feature is different from the third operating condition feature, execute the following step 210.

[0104] 209. If the second operating condition feature is the same as the third operating condition feature, the operating condition label corresponding to the second operating condition feature is used to mark the operating condition data of the fourth sliding window, and the fourth sliding window is moved away from the initial time point by a preset offset to obtain a fifth sliding window; when the time series detection is completed, execute the following step 211; when the time series detection is not completed, use the fifth sliding window to execute step 208.

[0105] In an embodiment of the present application, if the second operating condition characteristic is the same as the third operating condition characteristic, the operating condition labeling system uses the operating condition label corresponding to the second operating condition characteristic to label the operating condition data in the fourth sliding window, and moves the fourth sliding window away from the initial time point by a preset offset to obtain a fifth sliding window. If the operating condition labeling system has already used the sliding window to detect the entire time series of the operating condition data to be labeled, the labeled operating condition data is obtained, and the following step 211 is executed to send the labeled operating condition data to a professional for re-inspection and correction. If the operating condition labeling system has not detected the entire time series of the operating condition data to be labeled, the fifth sliding window is used to execute step 208, that is, the second operating condition characteristic is re-compared with the fourth operating condition characteristic corresponding to the operating condition data in the fifth sliding window.

[0106] 210. If the second operating condition characteristic is different from the third operating condition characteristic, obtain a preset window sliding time, move the fourth sliding window away from the initial time point by the preset window sliding time, and obtain a sixth sliding window; when the time series detection is completed, execute the following step 211; when the time series detection is not completed, use the sixth sliding window to execute step 203.

[0107] In an embodiment of the present application, if the second operating condition characteristic is different from the third operating condition characteristic, the operating condition labeling system obtains a preset window sliding time, which may be 1ms, and then moves the fourth sliding window away from the initial time point by the preset window sliding time to obtain a sixth sliding window. If at this time the operating condition labeling system has already used the sliding window to detect the entire time series of the operating condition operation data to be labeled, the labeled operating condition operation data is obtained, and the following step 211 is executed to send the labeled operating condition operation data to a professional for re-inspection and correction. If at this time the operating condition labeling system has not detected the entire time series of the operating condition operation data to be labeled, the sixth sliding window is used to execute step 203, that is, the operating condition operation data of the sixth sliding window is re-detected until a working condition that meets the preset operating condition identification rules is identified.

[0108] Based on steps 207 to 210, the following is obtained: Figure 2D The single-hold operating condition overlapping sampling results shown are automatically labeled based on preset operating condition identification rules to automatically, efficiently and accurately implement operating data condition labeling.

[0109] To summarize, for the operating data collected from typical equipment or prototype tests of nuclear reactors in nuclear power plants, this application uses preset rules to automatically mark the operating conditions (such as normal operating conditions such as lifting, lowering, holding, double holding, and rod dropping of the driving mechanism) on a periodic basis, with a marking accuracy of 100%.

[0110] 211. Send the marked working condition operation data to the terminal held by the user, so that the user can correct and mark the marked working condition operation data.

[0111] For some working conditions that cannot be automatically labeled according to preset rules and are related to the working conditions set in the preset working condition identification rules, the working condition labeling can be automatically performed based on the preset working condition identification rules, and then the experts can check the working conditions that are incorrectly labeled by the system based on their experience and knowledge, and the experts can manually correct them to the correct working condition labels. In an embodiment of the present application, the working condition labeling system sends the labeled working condition operation data to the terminal held by the user, so that the user can correct and label the labeled working condition operation data. For example, working conditions such as the driving mechanism sliding rod, unable to lift, lifting card rod, and lowering card rod usually appear in lifting or lowering working conditions. During the expert review, the working conditions that are incorrectly labeled as lifting or lowering by the system will be modified to the corresponding correct working conditions.

[0112] For some working conditions that cannot be automatically labeled based on the preset working condition identification rules and are not related to the working conditions set in the preset working condition identification rules, such as the working conditions of equipment that executes non-periodic instructions such as regulators and steam generators, manual labeling is performed based on visual interactive pages such as Echarts.

[0113] 212. Obtain the working condition operation data annotation results fed back by the user based on the terminal held by the user, and store the working condition operation data annotation results in the database.

[0114] In the embodiment of the present application, the working condition labeling system obtains the corrected working condition operation data based on the feedback of the user's terminal, wherein the result of correcting the improvement to the sliding bar working condition is as follows: Figure 2E As shown, the corrected operating condition data is marked with a target operating condition type. The target operating condition type does not belong to the multiple operating condition types included in the preset operating condition identification rules, and the target operating condition type is associated with multiple operating condition types, including driving mechanism sliding rod, unable to lift, lifting card rod, lowering card rod and other operating conditions.

[0115] Next, the operating condition labeling system obtains the start time point, end time point and target operating condition label selected by the user based on the terminal held, and determines the target time series between the start time point and the end time point. Then, starting from the start time point, the operating condition labeling system uses the target operating condition label to label the corrected operating condition data in the sliding window, moves the sliding window away from the start time point for a preset labeling cycle, obtains the next sliding window, and uses the target operating condition label to label the corrected operating condition data in the next sliding window. Repeat the operation until the target time series labeling is completed. The obtained operating condition data labeling result is as follows: Figure 2F As shown, through human-computer visual interaction, professionals can manually identify and quickly label working conditions (such as drive mechanism failures such as stuck rods, sliding rods, and inability to lift), which is more efficient and accurate than manual processing. Finally, the working condition labeling system stores the labeling results of the working condition operation data in a database. By storing the data and corresponding labels in a relational database as structured data, it can provide high-quality data sets to support the construction of various intelligent fault diagnosis models and facilitate data reuse.

[0116] It should be noted that for some working conditions that cannot be automatically labeled based on the preset working condition identification rules and are not related to the working conditions set in the preset working condition identification rules, such as the working conditions of equipment such as regulators and steam generators that execute non-periodic instructions, the working condition labeling system uses the preset labeling period as a sliding window to cut the start time of the selected sequence in sequence, and labels the cut bands with working condition information. If the selected sequence is less than the preset labeling period, a prompt is generated that the labeling cannot be performed; if the remaining sequence after the selected sequence is cut in sequence is less than the preset labeling period, it will be discarded and not labeled. Therefore, if there is no period setting requirement for the selected sequence, the preset labeling period can be set to the same as the time length of the selected sequence.

[0117] In summary, the flowchart of a multi-strategy-based rod control and rod position equipment working condition labeling method proposed in the embodiment of the present application is as follows:

[0118] like Figure 2G As shown in the figure, existing working condition interpretation rules include working condition interpretation rules for multiple working conditions. They can automatically annotate the working condition of the data to be annotated based on rule knowledge. Some working conditions that require professional annotation based on experience (such as fault conditions) can be reviewed and annotated by experts through human-computer interaction through visualization technology. Finally, the annotated sequence data segments and corresponding labels are stored in a structured database. Multi-strategy annotation methods such as automated annotation based on rule knowledge, re-inspection annotation of annotated data, and human-computer visual interactive annotation can automatically, efficiently, and accurately annotate the working conditions of operating data. The annotated data can be stored in a standardized manner and organized into a dataset for constructing an intelligent fault diagnosis model, thereby improving the accuracy of the fault diagnosis model.

[0119] The method provided in the embodiments of the present application obtains operating data of rod control and position equipment to be labeled, acquires a sliding window method and preset operating condition identification rules, uses the sliding window method to label the operating data according to the preset operating condition identification rules, obtains the labeled operating data, sends the labeled operating data to a user's terminal, and allows the user to correct and label the labeled operating data, obtains the labeling results of the operating data based on the user's terminal feedback, and stores the labeling results in a database. For operating data collected from nuclear power plants or prototype tests, operating conditions identified by rule knowledge are periodically and automatically labeled according to preset rules. Some operating conditions that require professional experience to label are efficiently identified and labeled using visualization technology in a human-computer interactive manner. The labeled data is stored in a relational database for on-demand use in intelligent model construction. This human-computer collaborative approach to labeling operating conditions for time-series data of rod control and position equipment is more efficient, rapid, and accurate than manual processing. Furthermore, the labeled data can be stored in a standardized manner and constitute a dataset for constructing an intelligent fault diagnosis model, thereby improving the accuracy of the fault diagnosis model.

[0120] Further, as Figure 1 The specific implementation of the method, the embodiment of the present application provides a device for marking the working condition of a rod control rod position device based on multiple strategies, such as Figure 3 As shown, the device includes: an acquisition module 301, a marking module 302, a review module 303 and a storage module 304.

[0121] An acquisition module 301 is used to acquire the operating data of the rod control and rod position device to be marked;

[0122] The labeling module 302 is configured to obtain a sliding window method and a preset working condition identification rule, and use the sliding window method to label the working condition operation data to be labeled according to the preset working condition identification rule to obtain the labeled working condition operation data;

[0123] A review module 303 is configured to send the annotated operating condition data to a terminal held by a user, so that the user can correct and annotate the annotated operating condition data;

[0124] The storage module 304 is configured to obtain the operating condition data annotation results fed back by the user based on the terminal held by the user, and store the operating condition data annotation results in a database.

[0125] In a specific application scenario, the acquisition module 301 is used to read the operating information of the rod control and rod position device in the database; use a visualization tool to visualize the operating information to obtain the operating data of the rod control and rod position device to be marked.

[0126] In a specific application scenario, the labeling module 302 is used to determine the initial time point of the operating condition data to be labeled, and the operating condition data to be labeled is time series data; starting from the initial time point, the operating condition data of each sliding window is detected using the preset operating condition identification rule until the time series detection is completed to obtain the labeled operating condition data, and the sliding window slides above the operating condition data to be labeled, and the preset operating condition identification rule includes multiple operating condition types and operating condition characteristics corresponding to each operating condition type.

[0127] In a specific application scenario, the labeling module 302 is used to detect the operating condition data of the first sliding window; when the operating condition data of the first sliding window belongs to a non-stationary waveform operating condition, the preset operating condition identification rule is used to detect the operating condition data of the first sliding window; if the operating condition data of the first sliding window meets the preset operating condition identification rule, the first operating condition feature corresponding to the operating condition data of the first sliding window is determined, and the operating condition label corresponding to the first operating condition feature is read in the preset operating condition identification rule; the operating condition data of the first sliding window is labeled with the operating condition label corresponding to the first operating condition feature; a preset labeling period is obtained, the first sliding window is moved away from the initial time point by the preset labeling period, a second sliding window is obtained, and the operating condition data of the second sliding window is detected.

[0128] In a specific application scenario, the labeling module 302 is used to obtain the preset window sliding time if the operating condition data of the first sliding window does not meet the preset operating condition identification rules, move the first sliding window away from the initial time point by the preset window sliding time, and obtain a third sliding window; and detect the operating condition data of the third sliding window.

[0129] In a specific application scenario, the labeling module 302 is configured to, when the operating condition data of the first sliding window belongs to a steady waveform operating condition and the operating condition data of the first sliding window meets the preset operating condition identification rule, determine a second operating condition feature corresponding to the operating condition data of the first sliding window, read the operating condition label corresponding to the second operating condition feature in the preset operating condition identification rule, and label the operating condition data of the first sliding window using the operating condition label corresponding to the second operating condition feature; obtain a preset offset, move the first sliding window away from the initial time point by the preset offset to obtain a fourth sliding window, determine a third operating condition feature corresponding to the operating condition data of the fourth sliding window; compare the second operating condition feature with the third operating condition feature; if the second operating condition feature is the same as the third operating condition feature, label the operating condition data of the fourth sliding window using the operating condition label corresponding to the second operating condition feature; move the fourth sliding window away from the initial time point by the preset offset to obtain a fifth sliding window; and compare the second operating condition feature with the fourth operating condition feature corresponding to the operating condition data of the fifth sliding window.

[0130] In a specific application scenario, the labeling module 302 is used to obtain the preset window sliding time if the second operating condition characteristic is different from the third operating condition characteristic, move the fourth sliding window away from the initial time point by the preset window sliding time, and obtain a sixth sliding window; and detect the operating condition operation data of the sixth sliding window.

[0131] In a specific application scenario, the storage module 304 is used to obtain the corrected operating condition data fed back by the user based on the terminal held, the corrected operating condition data being marked with a target operating condition type, the target operating condition type not belonging to the multiple operating condition types included in the preset operating condition identification rule, and the target operating condition type being associated with multiple operating condition types; obtaining the starting time point, ending time point and target operating condition label selected by the user based on the terminal held, and determining the target time sequence between the starting time point and the ending time point; starting from the starting time point, using the target operating condition label to mark the corrected operating condition data in the sliding window, moving the sliding window in a direction away from the starting time point by a preset marking period to obtain the next sliding window, using the target operating condition label to mark the corrected operating condition data in the next sliding window, repeating the operation until the target time series marking is completed, and obtaining the operating condition data marking result.

[0132] The apparatus provided in the embodiments of the present application acquires operating data of rod control and position equipment to be labeled, acquires a sliding window method and preset operating condition identification rules, and uses the sliding window method to label the operating data according to the preset operating condition identification rules, obtaining the labeled operating data. The labeled operating data is then sent to a user's terminal, allowing the user to correct and label the labeled operating data. The labeled operating data labeling results based on user feedback from the terminal are obtained and stored in a database. For operating data collected from nuclear power plants or prototype tests, operating conditions identified through rule knowledge are periodically and automatically labeled according to preset rules. Some operating conditions that require professional experience to label are efficiently identified and labeled using visualization technology in a human-computer interactive manner. The labeled data is stored in a relational database for on-demand use in intelligent model construction. This human-computer collaborative approach to labeling operating conditions for time-series data of rod control and position equipment is more efficient, faster, and more accurate than manual processing. Furthermore, the labeled data can be stored in a standardized manner and constitute a dataset for constructing an intelligent fault diagnosis model, thereby improving the accuracy of the fault diagnosis model.

[0133] It should be noted that for other corresponding descriptions of the functional units involved in the multi-strategy rod control and rod position equipment working condition marking device provided in the embodiment of the present application, please refer to Figure 1 and Figures 2A to 2G The corresponding description in will not be repeated here.

[0134] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0135] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0136] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

[0137] In an exemplary embodiment, see Figure 4A device is also provided, comprising a bus, a processor, a memory, and a communication interface. The device may also include an input / output interface and a display device, wherein the various functional units can communicate with each other via the bus. The memory stores a computer program, and the processor is configured to execute the program stored in the memory and implement the multi-strategy-based rod control and rod position device operating condition labeling method described in the above embodiment.

[0138] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the multi-strategy-based rod control and rod position equipment working condition labeling method.

[0139] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented through hardware or by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0140] Those skilled in the art will understand that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required to implement the present application.

[0141] Those skilled in the art will appreciate that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the implementation scenario description, or can be modified accordingly and located in one or more devices different from the implementation scenario. The modules in the above implementation scenario can be combined into one module or further split into multiple submodules.

[0142] The above application serial numbers are for description only and do not represent the advantages or disadvantages of the implementation scenarios.

[0143] The above disclosure only describes several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the scope of protection of the present application.

Claims

1. A multi-strategy based method for marking working conditions of rod control and rod position equipment, characterized in that: include: Obtain the operating data of the rod control and rod position equipment to be marked; Obtaining a sliding window method and a preset working condition identification rule, using the sliding window method to perform working condition labeling on the working condition operation data to be labeled according to the preset working condition identification rule, to obtain the labeled working condition operation data, wherein the preset working condition identification rule includes multiple working condition types and working condition characteristics corresponding to each working condition type, and the preset working condition identification rule is used to detect whether the working condition operation data of the sliding window meets the working condition characteristics corresponding to the multiple working condition types, wherein the multiple working condition types include lifting working condition, inserting working condition, single holding working condition, double holding working condition, and rod dropping working condition; Sending the annotated working condition data to a terminal held by a user, so that the user can correct and annotate the annotated working condition data; Obtain the working condition operation data annotation result fed back by the user based on the terminal held by the user, and store the working condition operation data annotation result in a database.

2. The method according to claim 1, characterized in that The obtaining of the operating data of the rod control position device to be marked includes: Reading the operation information of the rod control and rod position device from the database; The operation information is visualized using a visualization tool to obtain the operating data of the rod control rod position device to be marked.

3. The method according to claim 1, characterized in that The step of performing working condition labeling on the working condition operation data to be labeled according to the preset working condition identification rule by using the sliding window method to obtain the labeled working condition operation data includes: Determining an initial time point of the operating condition data to be labeled, wherein the operating condition data to be labeled is time series data; Starting from the initial time point, the preset operating condition identification rule is used to detect the operating condition data of each sliding window until the time series detection is completed to obtain the labeled operating condition data. The sliding window slides above the operating condition data to be labeled. The preset operating condition identification rule includes multiple operating condition types and operating condition characteristics corresponding to each operating condition type.

4. The method according to claim 3, characterized in that The detecting the operating condition data of each sliding window using the preset operating condition identification rule starting from the initial time point includes: Detecting the operating condition data of the first sliding window; When the operating condition data of the first sliding window belongs to a non-stationary waveform operating condition, detecting the operating condition data of the first sliding window using the preset operating condition identification rule; If the operating condition data of the first sliding window meets the preset operating condition identification rule, determining a first operating condition feature corresponding to the operating condition data of the first sliding window, and reading an operating condition tag corresponding to the first operating condition feature in the preset operating condition identification rule; Annotating the operating condition data of the first sliding window using the operating condition label corresponding to the first operating condition feature; A preset marking period is obtained, the first sliding window is moved in a direction away from the initial time point by the preset marking period to obtain a second sliding window, and the working condition operation data of the second sliding window is detected.

5. The method according to claim 4, characterized in that After detecting the operating condition data of the first sliding window using the preset operating condition identification rule, the method further includes: If the operating condition data of the first sliding window does not conform to the preset operating condition identification rule, obtaining a preset window sliding time, and moving the first sliding window away from the initial time point by the preset window sliding time to obtain a third sliding window; The operating condition data of the third sliding window is detected.

6. The method according to claim 4, characterized in that After detecting the operating condition data of the first sliding window, the method further includes: When the operating condition data of the first sliding window belongs to a steady waveform operating condition and the operating condition data of the first sliding window meets the preset operating condition identification rule, determining a second operating condition feature corresponding to the operating condition data of the first sliding window, reading an operating condition label corresponding to the second operating condition feature in the preset operating condition identification rule, and labeling the operating condition data of the first sliding window using the operating condition label corresponding to the second operating condition feature; Obtaining a preset offset, moving the first sliding window in a direction away from the initial time point by the preset offset to obtain a fourth sliding window, and determining a third operating condition feature corresponding to the operating condition data of the fourth sliding window; comparing the second operating condition characteristic with the third operating condition characteristic; If the second operating condition feature is the same as the third operating condition feature, the operating condition label corresponding to the second operating condition feature is used to mark the operating condition data of the fourth sliding window; Moving the fourth sliding window in a direction away from the initial time point by the preset offset to obtain a fifth sliding window; The second operating condition feature is compared with a fourth operating condition feature corresponding to the operating condition data of the fifth sliding window.

7. The method according to claim 6, characterized in that After comparing the second operating condition characteristic with the third operating condition characteristic, the method further includes: If the second operating condition characteristic is different from the third operating condition characteristic, obtaining a preset window sliding time, and moving the fourth sliding window away from the initial time point by the preset window sliding time to obtain a sixth sliding window; The operating condition data of the sixth sliding window is detected.

8. The method according to claim 1, characterized in that The obtaining of the working condition operation data annotation result based on the user's feedback of the terminal includes: Obtaining corrected operating condition data fed back by the user based on the terminal held by the user, wherein the corrected operating condition data is marked with a target operating condition type, the target operating condition type does not belong to the multiple operating condition types included in the preset operating condition identification rule, and the target operating condition type is associated with multiple operating condition types; Obtaining a start time point, an end time point, and a target operating condition tag selected by the user based on the terminal held, and determining a target time sequence between the start time point and the end time point; Starting from the starting time point, the target operating condition label is used to mark the corrected operating condition data in the sliding window, and the sliding window is moved away from the starting time point by a preset marking period to obtain the next sliding window, and the target operating condition label is used to mark the corrected operating condition data in the next sliding window. The operation is repeated until the target time series marking is completed to obtain the operating condition data marking result.

9. A device for marking working conditions of rod control and rod position equipment based on multiple strategies, applied to the method of claim 1, characterized in that: include: An acquisition module is used to obtain the operating data of the rod control rod position device to be marked; a labeling module, configured to obtain a sliding window method and a preset operating condition identification rule, perform operating condition labeling on the operating condition operation data to be labeled using the sliding window method according to the preset operating condition identification rule, and obtain labeled operating condition operation data, wherein the preset operating condition identification rule includes multiple operating condition types and operating condition characteristics corresponding to each operating condition type, and the preset operating condition identification rule is configured to detect whether the operating condition operation data of the sliding window meets the operating condition characteristics corresponding to the multiple operating condition types, wherein the multiple operating condition types include lifting operating condition, inserting operating condition, single holding operating condition, double holding operating condition, and rod dropping operating condition; A review module, configured to send the annotated operating condition data to a terminal held by a user, so that the user can correct and annotate the annotated operating condition data; The storage module is used to obtain the working condition operation data annotation results fed back by the user based on the terminal held by the user, and store the working condition operation data annotation results in a database.

10. A device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images