A converter tilting mechanism state monitoring method and system

Abstract

This invention belongs to the technical field of converter tilting mechanisms. It provides a method and system for monitoring the condition of converter tilting mechanisms. The method involves real-time acquisition of condition monitoring parameter data, fluctuation analysis of the data, and extraction of parameters with large fluctuations as those to be monitored. Based on these parameters, a two-dimensional analysis of time and trend is performed, considering the operating conditions of the converter tilting mechanism, to determine the degree of synchronicity between fluctuations and changes in operating conditions. This invention identifies abnormal fluctuations not driven by operating conditions by screening parameters of interest, performing a two-dimensional synchronicity analysis of operating conditions over time and trends, and comparing historical normal data. This significantly reduces the false judgment rate and minimizes unnecessary shutdowns caused by misjudging fluctuations caused by normal load changes as abnormal. Furthermore, for key parameters with significant fluctuations, it reduces interference from irrelevant data, improving the overall efficiency and accuracy of condition monitoring.

Description

Technical Field

[0001] This invention belongs to the technical field of converter tilting mechanism, specifically a method and system for monitoring the status of converter tilting mechanism. Background Technology

[0002] In the metallurgical industry, the converter tilting mechanism is a key component of the converter tilting system. Its performance directly affects the normal operation of the converter. Through the coordinated transmission of the motor, coupling, reducer, tilting trunnion, and support ring, the converter is driven to tilt from 0° to 360° to complete core processes such as charging, smelting, and tapping. Its stable operation is the key to ensuring the steelmaking rhythm and avoiding equipment accidents.

[0003] However, when the converter is tilted, the load changes with the tilting angle (e.g., the load increases and then decreases when the converter tilting angle changes from 0° to 90°), causing normal fluctuations in vibration and current. Currently, when monitoring the condition of the converter tilting mechanism, a fixed threshold method is often used, which sets a single parameter threshold (e.g., vibration acceleration ≥ 0.5g alarm, current ≥ 200A shutdown). This lacks a refined mechanism to distinguish between normal fluctuations and abnormal trends. Vague fluctuations can lead to misjudgments. If normal fluctuations are judged as abnormal, it can lead to unnecessary shutdowns and disrupt the continuity of production.

[0004] Therefore, the present invention provides a method and system for monitoring the state of a converter tilting mechanism. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: a method for monitoring the state of a converter tilting mechanism, comprising the following steps:

[0007] Real-time acquisition of condition monitoring parameter data of converter tilting mechanism, and fluctuation analysis of condition monitoring parameter data, extracting condition monitoring parameter data with large fluctuations as fluctuation parameters to be monitored;

[0008] Based on the fluctuation parameters to be monitored, a two-dimensional analysis of time and trend is conducted in conjunction with the working conditions of the converter tilting mechanism to determine the degree of synchronicity between fluctuations and changes in working conditions.

[0009] If the degree of synchronization is low, analyze the fluctuation parameters to be monitored based on historical normal data to determine whether they meet the standards of historical normal fluctuations and distinguish whether the equipment status is normal or abnormal.

[0010] If the equipment status fluctuates abnormally, the abnormality level will be classified and a graded warning will be issued.

[0011] A converter tilting mechanism condition monitoring system, the system comprising:

[0012] Parameter fluctuation analysis module: Real-time acquisition of status monitoring parameter data of converter tilting mechanism, and analysis of fluctuation of status monitoring parameter data, extracting status monitoring parameter data with large fluctuations as fluctuation parameters to be monitored;

[0013] Fluctuation Condition Synchronization Analysis Module: Based on the fluctuation parameters of interest, and combined with the operating conditions of the converter tilting mechanism, a two-dimensional analysis of time and trend is conducted to determine the degree of synchronicity between fluctuations and changes in operating conditions.

[0014] Historical data comparison module: If the synchronization level is low, it analyzes whether the fluctuation parameters of concern meet the standards of historical normal fluctuations based on historical normal data, and distinguishes whether the equipment status is normal or abnormal fluctuation.

[0015] Anomaly Level Classification Module: If the device status fluctuates abnormally, the anomaly level is classified and a graded warning is issued.

[0016] The beneficial effects of this invention are as follows:

[0017] This invention identifies abnormal fluctuations not driven by operating conditions by screening fluctuation parameters of interest, performing synchronous analysis of operating time and trends, and comparing with historical normal data. This significantly reduces the false judgment rate and minimizes unnecessary downtime caused by misjudging fluctuations caused by normal load changes as abnormal. At the same time, it reduces interference from irrelevant data for key parameters with significant fluctuations, thereby improving the efficiency and accuracy of overall status monitoring. Attached Figure Description

[0018] The invention will now be further described with reference to the accompanying drawings.

[0019] Figure 1 This is a flowchart of the steps of a converter tilting mechanism status monitoring method according to the present invention;

[0020] Figure 2 This is an architecture diagram of a converter tilting mechanism status monitoring system according to the present invention. Detailed Implementation

[0021] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0022] Example 1

[0023] Please see Figure 1 As shown in the figure, the method for monitoring the status of a converter tilting mechanism according to an embodiment of the present invention includes the following steps:

[0024] Step S10: Collect the status monitoring parameter data of the converter tilting mechanism in real time, perform fluctuation analysis on the status monitoring parameter data, and extract the status monitoring parameter data with large fluctuations as fluctuation parameters to be monitored;

[0025] In this step, the status monitoring parameter data of the converter tilting mechanism are collected in real time;

[0026] The condition monitoring parameters include, but are not limited to: mechanical vibration parameters, temperature parameters, electrical operating parameters, and hydraulic / lubrication parameters;

[0027] For example, the mechanical vibration parameters are: bearing housing vibration acceleration (X / Y / Z three directions) and vibration displacement;

[0028] The temperature parameters are: motor stator temperature, reducer lubricating oil temperature, and trunnion bearing temperature.

[0029] The electrical operating parameters are: motor operating current, voltage, power, and power factor;

[0030] The hydraulic / lubrication parameters are: hydraulic system working pressure, lubricating oil flow rate, and oil contamination level.

[0031] The collected status monitoring parameter data is preprocessed, including at least the following: noise, outliers, missing values, and data standardization.

[0032] The process of extracting highly fluctuating state monitoring parameter data using the sliding analysis window method is as follows:

[0033] A preset sliding analysis window is provided, the size of which is determined by those skilled in the art based on the acquisition frequency and the operating cycle of the converter tilting mechanism.

[0034] For example, when the converter tilts at 0.5 r / min, a 30-second analysis window is set (covering 1 / 4 of the tilting cycle), and the window is slid every 10 seconds to output the fluctuation index (fluctuation amplitude) of the status monitoring parameters in each window in real time.

[0035] Based on any single status monitoring parameter;

[0036] The fluctuation amplitude of the state monitoring parameter within each sliding analysis window is calculated to reflect the range of change of the state monitoring parameter.

[0037] The calculation process for the fluctuation amplitude is as follows: extract the maximum and minimum values ​​of the state monitoring parameters within the sliding analysis window, calculate the difference between the maximum and minimum values ​​of the state monitoring parameters, and obtain the fluctuation amplitude.

[0038] The fluctuation range is compared with a preset fluctuation range threshold, which is set by a person skilled in the art based on statistical historical normal operation data and in combination with industry standards.

[0039] If the fluctuation amplitude is less than the preset fluctuation amplitude threshold, it indicates that the fluctuation range of the corresponding sliding analysis window is small, and it is judged as a small fluctuation window.

[0040] If the fluctuation amplitude is greater than or equal to the preset fluctuation amplitude threshold, it indicates that the fluctuation range of the corresponding sliding analysis window is large, and it is judged as a large fluctuation window.

[0041] If three consecutive sliding analysis windows are marked as large fluctuation windows or the number of large fluctuation windows exceeds the window number threshold, the corresponding status monitoring parameter will be marked as a fluctuation parameter to be concerned.

[0042] If there are no three consecutive sliding analysis windows marked as large fluctuation windows or the number of large fluctuation windows is greater than the window number threshold, then the corresponding state monitoring parameter is marked as a parameter that does not need to be concerned.

[0043] The threshold for the number of windows is set by those skilled in the art based on the operating characteristics of the converter tilting mechanism and the statistical analysis of fluctuation windows in historical normal operating data.

[0044] The purpose of extracting the fluctuation parameters of interest is to: focus on the condition monitoring parameters with significant fluctuation amplitudes, provide targeted analysis objects for subsequent condition analysis, reduce the interference of parameters with small fluctuations and weak impact on equipment condition in the monitoring data analysis process, and thus improve the efficiency and accuracy of overall condition monitoring.

[0045] Step S20: Based on the fluctuation parameters to be monitored, and combined with the working conditions of the converter tilting mechanism, conduct a two-dimensional analysis of time and trend to determine the degree of synchronicity between the fluctuation and the changes in working conditions.

[0046] In this step, the operating parameter data of the converter tilting mechanism are obtained based on the fluctuation parameters of interest;

[0047] The operating condition parameter data includes: tilting action status parameters, load status parameters, operation command parameters, and time node parameters.

[0048] Specifically, the tilting motion state parameters include at least: tilting direction, speed, and angle, which are used to reflect whether the mechanism is in dynamic operation and motion characteristics;

[0049] The load condition parameters include at least: the weight of molten steel in the furnace and the tilting load torque, which are used to reflect the magnitude and changes of the load borne by the mechanism;

[0050] The operating command parameters include at least: start, stop, emergency stop signals, and gear shifting signals, used to capture sudden changes in operating conditions caused by human or automatic control;

[0051] The time node parameters should include at least: the switching time of each operating condition, which serves as a precise time reference for marking changes in operating conditions;

[0052] In this step, the temporal synchronicity between the fluctuation parameter of interest and the changes in operating conditions is determined based on the time dimension. The specific process is as follows:

[0053] For any large fluctuation window corresponding to any fluctuation parameter to be monitored;

[0054] Extract the moment when the fluctuation amplitude exceeds the threshold as the start moment of the fluctuation;

[0055] Extract the moment of sudden change in operating conditions from the operating condition parameters as the moment of change in operating conditions;

[0056] For any fluctuation start time, calculate the difference between the fluctuation start time and the most recent operating condition change time, and take the absolute value as the time difference;

[0057] Extract fluctuation windows with time differences less than the time difference limit as strong time correlation windows;

[0058] The time difference limit is set by those skilled in the art based on historical normal operation data and combined with work experience;

[0059] The total number of strongly correlated time windows is counted, and the proportion of the total number of strongly correlated time windows to the total number of large fluctuation windows is calculated as a value representing the time synchronization performance.

[0060] In this step, the trend synchronization between the fluctuation parameter of interest and the changes in operating conditions is determined based on the trend dimension. The specific process is as follows:

[0061] For any fluctuation parameter to be monitored;

[0062] Based on the time series of the fluctuation parameters and operating parameters to be monitored, a preset analysis interval is defined for each analysis interval;

[0063] Calculate the difference between the value of the fluctuation parameter to be monitored at the end of the analysis interval and the value of the fluctuation parameter to be monitored at the beginning of the analysis interval, and then calculate the ratio of the difference to the analysis interval, which is taken as the rate of change of the fluctuation parameter interval.

[0064] Calculate the difference between the operating condition parameter value at the end of the analysis interval and the operating condition parameter value at the beginning of the analysis interval, and then calculate the ratio of the difference to the analysis interval as the rate of change of the operating condition parameter interval.

[0065] Calculate the ratio of the rate of change of the fluctuation parameter interval to the rate of change of the operating condition parameter interval within the same analysis interval, and take the absolute value of the ratio as the interval rate of change ratio;

[0066] The standard interval rate of change ratio is obtained by averaging all interval rate of change ratios.

[0067] Calculate the difference between each interval rate of change ratio and the standard interval rate of change ratio, and take the absolute value of the difference to obtain the interval rate of change ratio deviation value.

[0068] The analysis intervals where the rate of change of the interval is greater than the deviation threshold are extracted and used as asynchronous analysis intervals.

[0069] The deviation threshold is set by those skilled in the art based on historical normal operation data and combined with work experience;

[0070] It is important to note that if the difference between the operating condition parameter value at the end of the analysis interval and the operating condition parameter value at the beginning of the analysis interval is 0, that is, the rate of change of the operating condition parameter interval is 0, then it belongs to "abnormal fluctuation without operating condition drive". Its rate of change should be significantly greater than the normal range to reflect the asynchrony of the trend, and it is used as an asynchronous analysis interval.

[0071] The total number of asynchronous analysis intervals is counted, and the proportion of the total number of asynchronous analysis intervals to the total number of analysis intervals is calculated as the trend synchronization performance value.

[0072] In this step, the process of determining the degree of synchronicity between fluctuations and changes in operating conditions is as follows:

[0073] The synchronization judgment value is obtained by multiplying the time synchronization performance value and the trend synchronization performance value.

[0074] It should be explained that the synchronicity judgment value is a quantitative indicator of the comprehensive degree of coordination between the fluctuation parameters of concern to the converter tilting mechanism (such as vibration, current, etc.) and changes in operating conditions (such as tilting angle, load torque, operating commands, etc.) in terms of time correlation and trend consistency. The time synchronicity performance value reflects the degree of overlap between the occurrence time of parameter fluctuations and the time of abrupt changes in operating conditions. Parameter fluctuations driven by normal operating conditions (such as the increase in motor current when the tilting angle increases from 0° to 30°) will be highly synchronized with changes in operating conditions in time (the difference between the start time of the fluctuation and the time of abrupt changes in operating conditions is minimal). Abnormal fluctuations (such as sudden vibrations without changes in operating conditions) will show obvious time misalignment; the trend synchronization performance value reflects the matching degree between the change law of parameter fluctuation and the change law of operating conditions. During normal fluctuations, the rate of change of parameters (such as the rate of increase of current) will keep in sync with the rate of change of operating conditions (such as the rate of increase of tilt angle and the rate of increase of load torque) (for example, when the tilt speed increases, the rate of increase of current increases synchronously); if there is wear of transmission components, sensor failure, etc., there will be a trend misalignment where the operating conditions change slowly but the parameter fluctuations are fast or the operating conditions change fast but the parameter fluctuations are lagging.

[0075] If the synchronicity judgment value is less than the synchronicity judgment threshold, it indicates that the synchronicity between parameter fluctuations and changes in operating conditions is low, and the synchronicity is judged to be low.

[0076] If the synchronicity judgment value is greater than or equal to the synchronicity judgment threshold, it indicates that the parameter fluctuation and the change of operating conditions are highly synchronized, and thus it is judged as having a high degree of synchronicity.

[0077] Step S30: If the synchronization is low, analyze whether the fluctuation parameters to be monitored meet the standards of historical normal fluctuations based on historical normal data, and distinguish whether the equipment status is normal fluctuation or abnormal fluctuation.

[0078] In this step, the degree of synchronicity indicates that the fluctuation parameter to be monitored has poor coordination with the time and trend of the change in operating conditions. It may be a fluctuation not driven by operating conditions. It needs to be compared with historical normal data to distinguish between normal fluctuations and abnormal fluctuations.

[0079] Specifically, firstly, historical normal operation data consistent with the current working conditions of the converter tilting mechanism are extracted from the historical database to construct a historical normal operation dataset;

[0080] Operating condition matching dimensions: correspond to the operating condition parameter data in step S20, including:

[0081] Tilt motion status matching: The current tilt direction (e.g., forward / reverse), speed (e.g., 0.5 r / min), and angle range (e.g., 30°-60°) must be consistent with historical data;

[0082] Load condition matching: The current range of molten steel weight in the furnace (e.g., 80-100 tons) and the range of tilting load torque must match historical data;

[0083] Operation command matching: The current operation mode (such as automatic / manual tilt) and the status of special commands such as no emergency stop / gear shifting must be consistent with historical data;

[0084] Perform the same preprocessing as in step S10 on the historical normal dataset and the current fluctuation parameter data to be monitored;

[0085] Based on the preprocessed historical normal dataset, a historical normal standard for the fluctuation parameter of interest is constructed.

[0086] Specifically, historical normal standards include the standard range of fluctuation amplitude and the standard range of fluctuation trend;

[0087] Using the sliding analysis window method in step S10 (using the same sliding analysis window size), calculate the fluctuation amplitude of the fluctuation parameter of interest in each sliding analysis window in the historical normal dataset (i.e., the maximum value of the parameter in the window minus the minimum value);

[0088] Take the 95th percentile of all fluctuation ranges as the upper limit of the historical normal fluctuation range, and take the 10th percentile as the lower limit of the historical normal fluctuation range to obtain the range of historical normal fluctuation range;

[0089] For example: if the 95th percentile of the fluctuation amplitude of bearing vibration acceleration in historical normal dataset is 0.4g, then whether the current fluctuation amplitude is ≤0.4g can be used to preliminarily determine whether the fluctuation is normal.

[0090] Calculate the ratio of the interval change rate of the fluctuation parameter of concern to be monitored to the corresponding operating condition parameter in the historical normal dataset;

[0091] The mean of the ratios of change across all intervals, ± 2 standard deviations, is taken as the range of historical normal trend deviation.

[0092] For example: if the mean of the rate of change of current and tilt angle in historical normal data is 1.2 and the standard deviation is 0.3, then the normal range is between 0.6 (1.2-2×0.3) and 1.8 (1.2+2×0.3).

[0093] The real-time fluctuation parameters to be monitored are compared with historical normal standards using all sliding analysis windows:

[0094] Comparison 1: Is the fluctuation range of the current sliding analysis window within the historical normal fluctuation range?

[0095] Comparison 2: Whether the rate of change of the fluctuation parameters and operating parameters in the current sliding analysis window is within the deviation range of the historical normal trend;

[0096] If the sliding analysis window simultaneously meets the criteria for both fluctuation amplitude and rate of change ratio, it is marked as a historical matching window.

[0097] If the sliding analysis window does not meet the criteria for both fluctuation amplitude and rate of change, it is marked as a historical mismatch window.

[0098] Count the number of consecutive historical mismatch windows and compare it with the consecutive mismatch threshold;

[0099] The consecutive mismatch threshold is set by those skilled in the art based on their work experience and the number of consecutive mismatch windows before historical failures.

[0100] If the number of consecutive historical mismatch windows is less than or equal to the consecutive mismatch threshold, then the current fluctuation is considered normal.

[0101] If the number of consecutive historical mismatch windows exceeds the consecutive mismatch threshold, it indicates that the current fluctuation is an abnormal fluctuation.

[0102] Step S40: If the equipment status fluctuates abnormally, the abnormality level is classified and a graded warning is issued, which can reduce the situation of misjudging normal fluctuations as abnormal fluctuations and improve the reliability of status monitoring.

[0103] In this step, if the device status is abnormally fluctuating, the number of consecutive historical mismatch windows is obtained, and the duration of the mismatch is converted into the product of the number of consecutive historical mismatch windows and the window size according to the size of each window.

[0104] Get the current fluctuation range, calculate the ratio between the current fluctuation range and the upper limit of the historical normal fluctuation range to get the fluctuation range exceeding the standard multiple, extract the maximum value of the fluctuation range exceeding the standard multiple corresponding to the mismatch window, and the fluctuation intensity value;

[0105] If the mismatch duration is less than the mismatch duration threshold and the fluctuation intensity value is less than the fluctuation intensity threshold, it is marked as a low-level anomaly.

[0106] If the duration of the mismatch is greater than or equal to the mismatch duration threshold or the fluctuation intensity value is greater than or equal to the fluctuation intensity threshold, it is marked as a high-level anomaly.

[0107] Based on low-level anomalies, the early warning strategy includes, but is not limited to: displaying a blue warning icon on the system monitoring interface, pushing notifications only to the mobile APP of the equipment inspection team leader, not activating audible and visual alarms, not interrupting the converter tilting operation, and only automatically recording the changing trend of abnormal parameters. Inspection personnel must verify the data on-site within 24 hours and re-measure parameters such as vibration and temperature using portable instruments. If the verification confirms that there is no actual anomaly, the system will automatically include the data in the historical normal dataset for subsequent threshold optimization. If minor issues are found (such as slightly low lubricating oil level), they can be addressed uniformly during the next planned shutdown.

[0108] Based on high-level anomalies, early warning strategies include, but are not limited to: the system immediately activates a yellow audible and visual alarm (simultaneously sounding in the workshop and central control room), and simultaneously pushes an emergency notification to the production scheduler and equipment maintenance supervisor; if a veto is triggered due to critical safety parameters, the system will also automatically pop up a shutdown prompt window (operators must manually confirm whether to shut down), and technicians must analyze the cause of the anomaly, locate the problem, and prioritize shutdown for maintenance within 8 hours.

[0109] This embodiment has at least the following functions:

[0110] This invention effectively distinguishes between normal fluctuations caused by load changes and abnormal trends caused by equipment failures during converter tilting through a progressive process of fluctuation parameter screening, operating condition synchronization analysis, and historical data comparison. This reduces the risk of misjudging normal fluctuations as abnormalities and causing unnecessary shutdowns, or missing real abnormalities and causing equipment accidents, thereby improving the reliability of condition monitoring.

[0111] By using sliding window analysis to extract the fluctuation parameters of interest, parameters with significant fluctuation amplitudes and key impacts on equipment status are extracted, while irrelevant parameters with minor fluctuations are excluded, making subsequent analysis more targeted and improving monitoring efficiency.

[0112] Based on accurate anomaly identification and graded early warning strategies, low-level anomalies do not require interruption of converter tilting operations and only require planned handling; high-level anomalies require emergency response, but due to accurate identification, excessive downtime can be reduced, thereby stabilizing the steelmaking rhythm and ensuring production continuity.

[0113] The tiered early warning mechanism corresponds to differentiated response strategies: low-level anomalies are only notified to inspection personnel for verification, while high-level anomalies are linked to production scheduling and maintenance supervisors for priority handling, avoiding resource waste, allowing maintenance resources to focus on truly urgent faults, and improving fault handling efficiency.

[0114] Example 2

[0115] Based on the same inventive concept as the converter tilting mechanism state monitoring method in the foregoing embodiments, such as Figure 2 As shown, this application provides a converter tilting mechanism condition monitoring system, wherein the system specifically includes:

[0116] Parameter fluctuation analysis module: Real-time acquisition of status monitoring parameter data of converter tilting mechanism, and analysis of fluctuation of status monitoring parameter data, extracting status monitoring parameter data with large fluctuations as fluctuation parameters to be monitored;

[0117] Fluctuation Condition Synchronization Analysis Module: Based on the fluctuation parameters of interest, and combined with the operating conditions of the converter tilting mechanism, a two-dimensional analysis of time and trend is conducted to determine the degree of synchronicity between fluctuations and changes in operating conditions.

[0118] Historical data comparison module: If the synchronization level is low, it analyzes whether the fluctuation parameters of concern meet the standards of historical normal fluctuations based on historical normal data, and distinguishes whether the equipment status is normal or abnormal fluctuation.

[0119] Anomaly Level Classification Module: If the device status fluctuates abnormally, the anomaly level is classified and graded warnings are issued. This can reduce the situation where normal fluctuations are misjudged as abnormal fluctuations and improve the reliability of status monitoring.

[0120] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method of monitoring the condition of a tilting mechanism of a converter, characterized in that: Includes the following steps: Real-time acquisition of condition monitoring parameter data of converter tilting mechanism, and fluctuation analysis of condition monitoring parameter data, extracting condition monitoring parameter data with large fluctuations as fluctuation parameters to be monitored; Based on the fluctuation parameters to be monitored, a two-dimensional analysis of time and trend is conducted in conjunction with the working conditions of the converter tilting mechanism to determine the degree of synchronicity between fluctuations and changes in working conditions. The process for determining the degree of synchronicity between fluctuations and changes in operating conditions is as follows: The operating conditions of the fluctuation parameters and the converter tilting mechanism were analyzed to obtain the time synchronization performance value and the trend synchronization performance value. The synchronization judgment value is obtained by multiplying the time synchronization performance value and the trend synchronization performance value. If the synchronization judgment value is less than the synchronization judgment threshold, it is judged as low synchronization level; The process for obtaining the time synchronization performance value is as follows: For each fluctuation window corresponding to each fluctuation parameter of interest, extract the starting time when the fluctuation amplitude exceeds the threshold as the fluctuation start time, and extract the time of sudden change in operating conditions from the operating condition parameters as the operating condition change time. For any fluctuation start time, calculate the difference between the fluctuation start time and the most recent operating condition change time, and take the absolute value as the time difference; Extract fluctuation windows with time differences less than the time difference limit as strong time correlation windows; The total number of strongly correlated time windows is counted, and the proportion of the total number of strongly correlated time windows to the total number of large fluctuation windows is calculated as a value representing the time synchronization performance. The process for obtaining the trend synchronicity performance value is as follows: For any fluctuation parameter of interest, based on the time series of the fluctuation parameter of interest and the operating condition parameter, for each analysis interval, the interval change rate of the fluctuation parameter and the interval change rate of the operating condition parameter are calculated by analysis. Calculate the ratio of the rate of change of the fluctuation parameter interval to the rate of change of the operating condition parameter interval within the same analysis interval, and take the absolute value of the ratio as the interval rate of change ratio; The standard interval rate of change ratio is obtained by averaging all interval rate of change ratios. Calculate the difference between each interval rate of change ratio and the standard interval rate of change ratio, and take the absolute value of the difference to obtain the interval rate of change ratio deviation value. The analysis intervals where the rate of change of the interval is greater than the deviation threshold are extracted and used as asynchronous analysis intervals. The total number of asynchronous analysis intervals is counted, and the proportion of the total number of asynchronous analysis intervals to the total number of analysis intervals is calculated as the trend synchronization performance value. If the degree of synchronization is low, analyze the fluctuation parameters to be monitored based on historical normal data to determine whether they meet the standards of historical normal fluctuations and distinguish whether the equipment status is normal or abnormal. If the equipment status fluctuates abnormally, the abnormality level will be classified and a graded warning will be issued.

2. A method of monitoring the condition of a tilting mechanism of a vessel as claimed in claim 1, characterized in that: The process for obtaining the fluctuation parameters to be monitored is as follows: Based on any one state monitoring parameter, calculate the fluctuation amplitude of the state monitoring parameter within each sliding analysis window; The calculation process for the fluctuation amplitude is as follows: extract the maximum and minimum values ​​of the state monitoring parameters within the sliding analysis window, calculate the difference between the maximum and minimum values ​​of the state monitoring parameters, and obtain the fluctuation amplitude. If the fluctuation range exceeds the limit, it is judged as a large fluctuation window; If three consecutive sliding analysis windows are marked as large fluctuation windows or the number of large fluctuation windows exceeds the window number threshold, the corresponding state monitoring parameter will be marked as a fluctuation parameter to be monitored.

3. The method for monitoring the state of a converter tilting mechanism according to claim 1, characterized in that: The process for obtaining the rate of change of the fluctuation parameter interval and the rate of change of the operating condition parameter interval is as follows: Calculate the difference between the value of the fluctuation parameter to be monitored at the end of the analysis interval and the value of the fluctuation parameter to be monitored at the beginning of the analysis interval, and then calculate the ratio of the difference to the analysis interval, which is taken as the rate of change of the fluctuation parameter interval. Calculate the difference between the operating parameter value at the end of the analysis interval and the operating parameter value at the beginning of the analysis interval, and then calculate the ratio of the difference to the analysis interval, which is taken as the rate of change of the operating parameter interval.

4. The method for monitoring the state of a converter tilting mechanism according to claim 1, characterized in that: The process of distinguishing whether the device status is fluctuating normally or abnormally is as follows: The real-time fluctuation parameters to be monitored are compared with historical normal standards using all sliding analysis windows: Compare whether the fluctuation range of the current sliding analysis window is within the historical normal fluctuation range; Compare the rate of change of the fluctuation parameters and operating parameters in the current sliding analysis window to see if they are within the range of deviation from the historical normal trend. If the sliding analysis window simultaneously meets the criteria for both fluctuation amplitude and rate of change ratio, it is marked as a historical matching window; otherwise, it is marked as a historical mismatch window. If the number of consecutive historical mismatch windows is less than or equal to the consecutive mismatch threshold, the current fluctuation is considered normal; otherwise, it is considered abnormal.

5. The method for monitoring the state of a converter tilting mechanism according to claim 4, characterized in that: The process for obtaining the historical normal standards is as follows: Extract historical normal operation data that are consistent with the current working conditions of the converter tilting mechanism from the historical database, construct a historical normal dataset, and construct historical normal standards, including standard ranges for fluctuation amplitude and standard ranges for fluctuation trends. Calculate the fluctuation amplitude of the fluctuation parameter of interest within each sliding analysis window in the historical normal dataset; Take the 95th percentile of all fluctuation ranges as the upper limit of the historical normal fluctuation range, and take the 10th percentile as the lower limit of the historical normal fluctuation range to obtain the range of historical normal fluctuation range; Calculate the ratio of the interval change rate of the fluctuation parameter of concern to be monitored to the corresponding operating condition parameter in the historical normal dataset; The mean of the ratios of change across all intervals, ± 2 standard deviations, is taken as the range of historical normal trend deviation.

6. The method for monitoring the state of a converter tilting mechanism according to claim 1, characterized in that: The process of classifying anomaly levels is as follows: If the device status is abnormally fluctuating, obtain the number of consecutive historical mismatch windows and convert them into the mismatch duration according to the size of each window, which is the product of the number of consecutive historical mismatch windows and the window size. Get the current fluctuation range, calculate the ratio between the current fluctuation range and the upper limit of the historical normal fluctuation range to get the fluctuation range exceeding the standard multiple, extract the maximum value of the fluctuation range exceeding the standard multiple corresponding to the mismatch window, and the fluctuation intensity value; If the duration of the mismatch is less than the mismatch duration threshold and the fluctuation intensity value is less than the fluctuation intensity threshold, it is marked as a low-level anomaly; otherwise, it is marked as a high-level anomaly.

7. A state monitoring system for a converter tilting mechanism, characterized in that, The system is used to perform the method according to any one of claims 1-6, and the system comprises: Parameter fluctuation analysis module: Real-time acquisition of status monitoring parameter data of converter tilting mechanism, and analysis of fluctuation of status monitoring parameter data, extracting status monitoring parameter data with large fluctuations as fluctuation parameters to be monitored; Fluctuation Condition Synchronization Analysis Module: Based on the fluctuation parameters of interest, and combined with the operating conditions of the converter tilting mechanism, a two-dimensional analysis of time and trend is conducted to determine the degree of synchronicity between fluctuations and changes in operating conditions. Historical data comparison module: If the synchronization level is low, it analyzes whether the fluctuation parameters of concern meet the standards of historical normal fluctuations based on historical normal data, and distinguishes whether the equipment status is normal or abnormal fluctuation. Anomaly Level Classification Module: If the device status fluctuates abnormally, the anomaly level is classified and a graded warning is issued.

Images