Method and system for predicting service life of fire valve

Abstract

The invention discloses a method and system for predicting the service life of a fire valve, and relates to the field of prediction maintenance, and the method comprises the steps: obtaining a multi-dimensional monitoring data matrix which comprises the sealing performance data, operation torque data and corrosion degree data of the fire valve continuously monitored in a preset time period; determining a parameter incidence matrix according to the multi-dimensional monitoring data matrix; determining a plurality of principal component vectors and corresponding characteristic values according to the parameter incidence matrix through a principal component analysis algorithm, and determining a current health state score of the fire valve according to the characteristic values corresponding to the plurality of principal component vectors; determining a degradation law function of the fire-fighting valve according to the historical operation record of the fire-fighting valve; and predicting the service life of the fire valve according to the current health state score and the degradation law function. According to the scheme, the accuracy of life prediction of the fire valve is improved, and the problem that the accuracy of life prediction of the fire valve is low is solved.

Description

Technical Field

[0001] This application relates to the field of predictive maintenance, and more specifically, to a method and system for predicting the service life of a fire valve. Background Art

[0002] Industrial fire valves, as core components in fluid control systems, play a crucial role in modern industrial systems. Whether in the petrochemical, power, or water treatment industries, the performance of valves directly affects the operating efficiency and safety of the entire system. With the development of industrial equipment towards large-scale and continuous operation, the requirements for the long-term stable operation of valves have become increasingly stringent. Therefore, how to accurately predict the service life of valves has become an urgent problem to be solved.

[0003] Currently, traditional valve maintenance methods mainly rely on regular inspections and experience-based judgments. Although this method is simple and easy to implement, it has obvious deficiencies. On the one hand, planned maintenance often fails to accurately reflect the actual state of the equipment, easily leading to over-maintenance or untimely maintenance; on the other hand, prediction methods based on single failure modes are difficult to comprehensively capture the true degradation process of valves. In addition, simple statistical analysis methods usually ignore the influence of individual equipment differences and changes in operating environments, which further limits the accuracy of prediction results.

[0004] In practical applications, there are complex interaction relationships among the performance parameters of valves (such as sealing performance, operating torque, and corrosion degree). For example, a decrease in sealing performance may cause internal leakage problems, while wear and corrosion will change the operating torque, and the aggravation of corrosion degree will have a negative impact on the sealing effect and mechanical properties. Since the changes in these parameters have the characteristics of co-degradation, monitoring a single parameter alone can no longer meet the accurate assessment requirements for the overall health status of valves, which also brings new challenges to service life prediction.

[0005] In view of the related art, there is currently no effective solution to the problem of low accuracy in predicting the service life of fire valves.

[0006] Therefore, it is necessary to improve the related art to overcome the above-mentioned defects in the related art. Summary of the Invention

[0007] Embodiments of this application provide a method and system for predicting the service life of a fire valve to at least solve the problem of low accuracy in predicting the service life of a fire valve.

[0008] According to one aspect of the embodiments of the present application, a method for predicting the service life of a fire protection valve is provided, including: obtaining a multi-dimensional monitoring data matrix, wherein the multi-dimensional monitoring data matrix has sealing performance data, operating torque data, and corrosion degree data of the fire protection valve continuously monitored within a preset time period; determining a parameter correlation matrix according to the multi-dimensional monitoring data matrix, wherein the parameter correlation matrix is used to represent the correlation between any two of a plurality of parameters, and the plurality of parameters include: sealing performance, operating torque, and corrosion degree; determining a plurality of principal component vectors and corresponding eigenvalues according to the parameter correlation matrix through a principal component analysis algorithm, and determining the current health state score of the fire protection valve according to the eigenvalues corresponding to the plurality of principal component vectors; and determining a degradation law function of the fire protection valve according to the historical operation records of the fire protection valve, wherein the degradation law function is used to reflect the change relationship of the health state score of the fire protection valve over time; predicting the service life of the fire protection valve according to the current health state score and the degradation law function.

[0009] In an exemplary embodiment, after obtaining the multi-dimensional monitoring data matrix, the method further includes: processing each time series data in the multi-dimensional monitoring data matrix in the following manner to process the multi-dimensional monitoring data matrix, wherein the multi-dimensional monitoring data matrix includes: sealing performance time series data, operating torque time series data, and corrosion degree time series; when processing each time series data, each time series data is target time series data, and the target time series data includes data at different time points: using a median filtering algorithm to smooth the target time series data to obtain the smoothed target time series data; determining abnormal data in the smoothed target time series data based on a standard deviation-based outlier detection algorithm, and using a linear interpolation algorithm to correct the abnormal data to obtain the corrected target time series data; using a spline interpolation algorithm to perform curve smoothing on the corrected target time series data.

[0010] In an exemplary embodiment, after using a spline interpolation algorithm to perform curve smoothing on the corrected target time series data, the method further includes: determining a target window size and a target moving step; processing the curve-smoothed target time series data according to the target window size and the target moving step through a window translation technique to obtain standardized time series data, wherein the standardized time series data includes the weighted average value of the target time series data within each target window size.

[0011] In an exemplary embodiment, determining a parameter correlation matrix according to the multi-dimensional monitoring data matrix includes: calculating the Pearson correlation coefficient between the sealing performance and the operating torque according to the sealing performance time series data and the operating torque time series data in the multi-dimensional monitoring data matrix to obtain a first Pearson correlation coefficient; calculating the Pearson correlation coefficient between the operating torque and the corrosion degree according to the operating torque time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain a second Pearson correlation coefficient; calculating the Pearson correlation coefficient between the sealing performance and the corrosion degree according to the sealing performance time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain a third Pearson correlation coefficient; and determining the parameter correlation matrix according to the first Pearson correlation coefficient, the second Pearson correlation coefficient, and the third Pearson correlation coefficient.

[0012] In an exemplary embodiment, determining the current health status score of the fire protection valve according to the eigenvalues corresponding to the multiple principal component vectors includes: determining the weight values corresponding to the multiple principal component vectors according to the eigenvalues corresponding to the multiple principal component vectors, wherein the weight value corresponding to the target principal component vector among the multiple principal component vectors is equal to the eigenvalue corresponding to the target principal component vector divided by the sum of the eigenvalues corresponding to the multiple principal component vectors; and performing weighted summation on the weight values and the eigenvalues corresponding to the multiple principal component vectors to obtain the current health status score of the fire protection valve.

[0013] In an exemplary embodiment, determining the degradation law function of the fire protection valve according to the historical operation records of the fire protection valve includes: processing the historical operation records by using a time series analysis method to obtain health index time series data, wherein the historical operation records include operation records, failure records, and maintenance records, and the health index time series data includes the health status scores at different time points; in the case where the health index time series data is linearly varying data, calculating the slope of the change of the health status score with time according to the health index time series data by using a linear regression method to obtain a degradation rate; determining the degradation law function according to the initial health status score of the fire protection valve and the degradation rate; or in the case where the health index time series data is non-linearly varying data, using a Weibull distribution model to determine the degradation law function according to the health index time series data.

[0014] In an exemplary embodiment, predicting the service life of the fire protection valve according to the current health status score and the degradation law function includes: determining a specified health status score when the fire protection valve reaches the service life; determining a first time according to the current health status score and the degradation law function, and determining a second time according to the specified health status score and the degradation law function; determining the remaining service life of the fire protection valve according to the first time and the second time.

[0015] According to another aspect of the embodiments of the present application, there is also provided a service life prediction system for a fire protection valve, including: an acquisition module for acquiring a multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix has sealing performance data, operating torque data, and corrosion degree data of the fire protection valve continuously monitored within a preset time period; a first determination module for determining a parameter correlation matrix according to the multi-dimensional monitoring data matrix, where the parameter correlation matrix is used to represent the correlation between any two of a plurality of parameters, and the plurality of parameters include: sealing performance, operating torque, and corrosion degree; a second determination module for determining a plurality of principal component vectors and corresponding eigenvalues according to the parameter correlation matrix by a principal component analysis algorithm, and determining the current health status score of the fire protection valve according to the eigenvalues corresponding to the plurality of principal component vectors; a third determination module for determining the degradation law function of the fire protection valve according to the historical operation record of the fire protection valve, where the degradation law function is used to reflect the change relationship of the health status score of the fire protection valve with time; a prediction module for predicting the service life of the fire protection valve according to the current health status score and the degradation law function.

[0016] According to yet another aspect of the embodiments of the present application, there is also provided an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that when the processor executes the computer program, the steps of the above-mentioned service life prediction method for a fire protection valve are implemented.

[0017] According to yet another aspect of the embodiments of the present application, there is also provided a computer-readable storage medium, in which a computer program is stored, where the computer program is configured to execute the above-mentioned service life prediction method for a fire protection valve when running.

[0018] According to yet another aspect of the embodiments of the present application, there is also provided a computer program product, including a computer program, where the computer program, when executed by a processor, is the above-mentioned service life prediction method for a fire protection valve.

[0019] In this application, key parameter data such as sealing performance, operating torque, and corrosion degree are comprehensively obtained using a multi-dimensional monitoring data matrix. Compared with traditional single-parameter methods, it can more completely reflect the operating state of the valve and its complex degradation process. Secondly, by constructing a parameter correlation matrix and calculating the correlation between any two parameters, and using the principal component analysis algorithm to extract the main component vectors and their eigenvalues, the current health status score is calculated, which not only reduces data redundancy but also highlights the key factors that have the greatest impact on valve degradation. In addition, a degradation law function is established by combining historical operation records, taking into account the change relationship of the health status score over time, and based on the current health status score and the degradation law function, the life prediction is carried out, greatly improving the accuracy of the life prediction of the fire protection valve and providing a scientific basis for equipment maintenance decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0021] To more clearly illustrate the embodiments of this application or the technologies in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0022] Figure 1 It is a schematic flowchart of a method for predicting the service life of a fire protection valve according to an embodiment of this application; Figure 2 It is a structural block diagram of a system for predicting the service life of a fire protection valve according to an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0023] In order to enable those skilled in the art to better understand this application, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of this application.

[0024] It should be noted that the terms "first", "second", etc. in the description, claims and the above drawings of this application are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0025] In this embodiment, a method for predicting the service life of a fire valve is provided. Figure 1 It is a schematic flowchart of a method for predicting the service life of a fire valve according to an embodiment of the present application, as Figure 1 shown, the process includes the following steps S102 - step S110: Step S102: Obtain a multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix has seal performance data, operating torque data, and corrosion degree data of the fire valve continuously monitored within a preset time period. Optionally, the seal performance data, operating torque data, and corrosion degree data can be obtained from the operation of the fire valve through multi-sensor synchronous acquisition technology. The time synchronization protocol is used to add timestamps to the collected sensor data to obtain the original multi-dimensional data set. The original multi-dimensional data set is segmented according to a preset sampling frequency, the interpolation algorithm is used to complement the missing data points, and the data consistency verification algorithm is used to judge the time alignment of the sensor data, so as to obtain the multi-dimensional monitoring data matrix after consistency verification.

[0026] Exemplarily, for the multi-sensor synchronous acquisition technology, the seal performance, operating torque, and corrosion degree data can be collected in real time through pressure sensors, torque sensors, and corrosion sensors installed on the fire valve. The pressure sensor monitors the pressure change of the sealing surface of the fire valve, the torque sensor records the torque fluctuation during the operation process, and the corrosion sensor detects the corrosion rate of the surface of the fire valve by an electrochemical method. Assuming that the sampling frequency is 10 times per second, each sensor collects data at the same moment to ensure data synchronization. It should be noted that this multi-dimensional data acquisition can comprehensively reflect the valve operation state and provide a reliable basis for subsequent analysis.

[0027] Optionally, a time synchronization protocol (including but not limited to the NTP protocol) can be used to add timestamps to sensor data. Suppose a certain collection is made at 02:54:00.123 on June 4, 2025. The pressure sensor records the sealing pressure as 2.5 MPa, the torque sensor records the operating torque as 150 Nm, and the corrosion sensor records the corrosion rate as 0.02 mm / year. The timestamp is accurate to milliseconds to ensure the alignment of multi-dimensional data in the time dimension, facilitating the subsequent analysis of the valve operation trend and abnormal points.

[0028] Optionally, when processing the original multi-dimensional data set in segments, the data can be segmented with a time window of one minute. Suppose a certain segment of data is missing the pressure value, then a linear interpolation algorithm can be used to complete it.

[0029] Exemplarily, in a certain minute, the pressure in the previous second is 2.5 MPa, and in the next second is 2.7 MPa. The missing point can be interpolated to 2.6 MPa. This method can ensure data continuity and improve the analysis accuracy.

[0030] It should be noted that the data consistency verification algorithm can judge the time alignment by checking whether the timestamp deviation of each sensor is less than 10 milliseconds. If the deviation is large, it means there is a problem with sensor synchronization, and the device needs to be calibrated. This verification can ensure the reliability of the multi-dimensional monitoring data matrix and provide accurate input for anomaly detection.

[0031] In an exemplary embodiment, if the value of any sensor data in the multi-dimensional monitoring data matrix exceeds the preset threshold, the sliding window algorithm is used to extract the abnormal data segment, and the distribution characteristics of the abnormal data segment are determined through statistical analysis to obtain the anomaly detection result. The multi-dimensional monitoring data matrix and the anomaly detection result are stored in the database through the real-time monitoring database interface, and the database is partitioned and managed using distributed storage technology to obtain the real-time updated valve operation status data.

[0032] Exemplarily, in anomaly detection, suppose the pressure threshold is 3.0 MPa. If the pressure value reaches 3.2 MPa during a certain period, the sliding window algorithm (window size set to 30 seconds) can be used to extract the abnormal segment. Statistical analysis shows that the average pressure of the abnormal segment is 3.15 MPa and the standard deviation is 0.05, indicating that the anomaly is concentrated and has little fluctuation, which may be caused by the wear of the sealing surface. This analysis helps to accurately locate the cause of the fault.

[0033] Optionally, through the real-time monitoring database interface, the multi-dimensional monitoring data matrix and the anomaly detection result can be stored in a distributed database. Using distributed storage technology, the data is stored in partitions according to time. For example, the data for each hour is stored in a partition, which is convenient for quick retrieval and analysis. This method improves the data access efficiency and supports real-time monitoring of the valve status.

[0034] It should be noted that the real-time updated operation status data of the fire protection valve can be displayed through the dashboard. For example, the trend charts of pressure, torque, and corrosion data can intuitively reflect the performance changes of the fire protection valve. If the corrosion rate continues to rise, it indicates that maintenance needs to be carried out in advance. This real-time monitoring and data management method significantly improves the operation reliability of the valve and reduces the failure risk.

[0035] In an exemplary embodiment, after the above step S102, the method further includes: processing each time series data in the multi-dimensional monitoring data matrix through the following steps S11 - S13 to process the multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix includes: seal performance time series data, operating torque time series data, and corrosion degree time series; when processing each time series data, each time series data is target time series data, and the target time series data includes data at different time points: Step S11: Smooth the target time series data by using a median filtering algorithm to obtain the smoothed target time series data; Exemplarily, assume that the fire protection valve monitoring system collects data 10 times per second, and a certain time series contains pressure data from 03:00:00 to 03:01:00 on June 4, 2025, with recorded values of 2.4 MPa, 2.45 MPa, 2.5 MPa, etc. The original data may contain abnormal fluctuations due to sensor noise or environmental interference, such as a sudden change to 3.5 MPa at a certain point. The median filtering algorithm can effectively remove such noise.

[0036] Specifically, select a time window (such as 5 sampling points), sort the data within the window, and replace the central point with the median. For example, for a window of data 2.4 MPa, 2.45 MPa, 3.5 MPa, 2.5 MPa, 2.48 MPa, the median after sorting is 2.48 MPa, and this value is used to replace 3.5 MPa to obtain the smoothed time series data. This method can retain the data trend, reduce noise interference, and improve the accuracy of subsequent analysis.

[0037] Step S12: Determine the abnormal data in the smoothed target time series data based on the standard deviation-based outlier detection algorithm, and correct the abnormal data by using a linear interpolation algorithm to obtain the corrected target time series data; Optionally, for the smoothed target time series data, an outlier detection algorithm based on standard deviation can further identify outliers. Suppose the mean pressure of a certain time series is 2.5 MPa and the standard deviation is 0.1. If the preset threshold is the mean ± 2 standard deviations (i.e., from 2.3 MPa to 2.7 MPa), then the data point 3.0 MPa will be marked as an outlier. After marking, time series data with outlier marks is obtained. This method quantifies outliers through statistical characteristics, ensuring the objectivity and reliability of the detection results, and helps to discover potential faults of fire valves.

[0038] Exemplarily, for the value correction of abnormal data points, a linear interpolation algorithm can be used to fill in the outliers. Suppose the time of an abnormal point is 03:00:05 and the pressure value is 3.0 MPa. The normal points before and after are 2.5 MPa at 03:00:04 and 2.7 MPa at 03:00:06 respectively. The corrected value at 03:00:05 is calculated by linear interpolation to be 2.6 MPa. This method derives a reasonable value through the data before and after, maintaining the continuity of the time series and providing a smooth input for subsequent analysis.

[0039] Step S13: Perform curve smoothing on the corrected target time series data using a spline interpolation algorithm.

[0040] It should be noted that for the corrected time series data, the spline interpolation algorithm can further optimize the curve smoothness. Among them, spline interpolation fits the data points through piecewise polynomials to generate a smooth curve.

[0041] Exemplarily, for the pressure data sequence 2.5 MPa, 2.6 MPa, 2.7 MPa, the algorithm generates a continuous curve, and updates the data in the time series data to the data on the continuous curve to eliminate minor fluctuations. This method can better reflect the operation trend of the valve and is convenient for long-term state analysis.

[0042] It should be noted that the above methods progress step by step from noise filtering to curve smoothing, ensuring data quality. Each step takes the operation data of the fire valve as the core, supports each other, and forms a complete data cleaning process, which helps to improve the reliability of the monitoring system.

[0043] In an exemplary embodiment, after the above step S13, the target time series data after curve smoothing can also be standardized, specifically, including the following steps S21 - S22: Step S21: Determine the target window size and the target moving step size; Step S22: By means of window translation technology, process the target time series data after curve smoothing according to the target window size and the target moving step length to obtain standardized time series data, where the standardized time series data includes the weighted average value of the target time series data within each target window size.

[0044] Optionally, the data can be further processed by a sliding window method to capture local features. The sliding window method analyzes the time series segment by segment by setting a fixed window size and a moving step length.

[0045] For example, assume that a valve monitoring system collects 10 pressure data points per second, and the time series covers from 03:00:00 to 03:01:00 on June 4, 2025, containing 600 data points. The window size is set to 5 sampling points and the step length is 1. The window starts from the first data point and covers 5 pressure values from 03:00:00 to 03:00:00.4, such as 2.50 MPa, 2.52 MPa, 2.51 MPa, 2.49 MPa, 2.53 MPa, and slides backward in turn. This method can effectively extract local trends and facilitate subsequent analysis.

[0046] Optionally, for each data segment within the window, a weighted average formula is used to calculate the smoothed value. The weights can be assigned according to the time distance or stability of the data points. For example, data points closer to the center of the window are given higher weights, and the weights of the edge points are lower. Assume that the weight of the center point 2.51 MPa in the above window is 0.4, the weights of the two side points 2.50 MPa and 2.52 MPa are 0.25, and the weights of the edge points 2.49 MPa and 2.53 MPa are 0.15. After calculating the weighted average value, the smoothed value is approximately 2.51 MPa, which is used to replace the data at the center point of the window. This method can effectively smooth local fluctuations and retain the main trend.

[0047] Exemplarily, for the smoothed segment data, if there are still random fluctuations, they can be filtered by comparing the weighted average values of adjacent windows with a preset threshold. Assume that the weighted average values of adjacent windows are 2.51 MPa and 2.55 MPa respectively, and the preset threshold is 0.05 MPa. The difference of 0.04 MPa is within the threshold, which is determined to be a normal fluctuation. If the difference of a certain window reaches 0.1 MPa, it is marked as a random fluctuation and excluded. This method reduces abnormal interference through comparison between windows and ensures data consistency.

[0048] It should be noted that the data segments after fluctuation filtering need to be integrated into a standardized time data series. By using the mean calculation method, the weighted averages of all windows are aggregated to generate the final time data series. Exemplarily, after aggregating the weighted averages of 600 windows, a mean value of 2.52 MPa is obtained as the reference value of the standardized series, and it is adjusted in combination with the data of each window to form a smooth and consistent series. This method eliminates local deviations through multi-window mean integration and provides reliable input for the state analysis of fire valves.

[0049] Step S104: Determine a parameter correlation matrix according to the multi-dimensional monitoring data matrix, where the parameter correlation matrix is used to represent the correlation between any two parameters among a plurality of parameters, and the plurality of parameters include: sealing performance, operating torque, and corrosion degree; In an exemplary embodiment, the above step S104 can be implemented through the following steps S31 - S34: Step S31: Calculate the Pearson correlation coefficient between the sealing performance and the operating torque according to the sealing performance time series data and the operating torque time series data in the multi-dimensional monitoring data matrix to obtain a first Pearson correlation coefficient; Step S32: Calculate the Pearson correlation coefficient between the operating torque and the corrosion degree according to the operating torque time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain a second Pearson correlation coefficient; Step S33: Calculate the Pearson correlation coefficient between the sealing performance and the corrosion degree according to the sealing performance time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain a third Pearson correlation coefficient; It should be noted that there is no execution sequence among the above steps S31 - S33.

[0050] Step S34: Determine the parameter correlation matrix according to the first Pearson correlation coefficient, the second Pearson correlation coefficient, and the third Pearson correlation coefficient.

[0051] It should be noted that the formula for calculating the first Pearson correlation coefficient is as follows:

[0052] where r is the first Pearson correlation coefficient, x i and y i are respectively the time series data points of the sealing performance and the operating torque at the i-th moment, μ x and μ y are the corresponding means, and σ x and σ y are the corresponding standard deviations.

[0053] It should be noted that the first Pearson correlation coefficient can reflect the strength of the correlation between the sealing performance and the operating torque.

[0054] It should be noted that the calculation methods of the second Pearson correlation coefficient and the third Pearson correlation coefficient are the same as those of the first Pearson correlation coefficient, and thus will not be elaborated herein in this application.

[0055] It should be noted that the parameter correlation matrix is as follows:

[0056] where r 12 represents the correlation between the sealing performance and the operating torque (the first Pearson correlation coefficient), r 23 represents the correlation between the operating torque and the corrosion degree (the second Pearson correlation coefficient); r 13 represents the correlation between the sealing performance and the corrosion degree (the third Pearson correlation coefficient); the values on the diagonal are all 1, indicating the correlation of itself with itself.

[0057] It should be noted that since the correlation coefficient matrix is symmetric, r 12 is equal to r 21 , r 23 is equal to r 32 , and r 13 is equal to r 31 .

[0058] Exemplarily, the parameter correlation matrix shows that the correlation between the sealing performance and the operating torque is 0.85, the correlation between the operating torque and the corrosion degree is 0.65, and the correlation between the sealing performance and the corrosion degree is 0.45. This matrix structure clearly presents the relationships among multiple parameters, facilitating the comprehensive analysis of the valve operating state. When constructing the matrix, it can be sorted according to the strength of the correlation, and priority can be given to paying attention to the influence of high-correlation parameter pairs on the valve performance.

[0059] Exemplarily, seal performance time series data and operating torque time series data are extracted from the multi-dimensional monitoring data matrix for analyzing the correlation between the two. The Pearson correlation coefficient is a method for measuring the linear correlation between two variables. By comparing the change trends of the seal performance data and the operating torque data, it is determined whether the two change synchronously. Suppose a valve system collected 600 data points from 03:00:00 to 03:01:00 on June 4, 2025. Among them, the seal performance data is represented by the leakage rate, with the unit of mL / min, and the operating torque data is represented by N·m. Select a certain sequence. The seal performance data is 0.02, 0.03, 0.04 mL / min, and the operating torque data is 50, 52, 55 N·m. Calculated through the Pearson correlation coefficient formula, the correlation coefficient is approximately 0.85, indicating that there may be a strong positive correlation between the two, that is, an increase in torque may lead to an increase in the leakage rate. This method provides a basis for subsequent analysis by quantifying the relationship between variables.

[0060] In a possible implementation, for the second data subset of the operating torque and the degree of corrosion, the Pearson correlation coefficient is also used to analyze the correlation. The degree of corrosion can be represented by the proportion of the surface corrosion area.

[0061] For example, select 100 data points in the same time period. The operating torque data is 50, 51, 53 N·m, and the degree of corrosion data is 2.1%, 2.3%, 2.5%. The calculated correlation coefficient is approximately 0.65, indicating that there is a certain positive correlation between the torque and the degree of corrosion, and it may be due to the intensification of corrosion that leads to an increase in torque requirements. This analysis helps to identify the impact of corrosion on valve operation and provides a reference for maintenance decisions.

[0062] In an exemplary embodiment, if the absolute value of the first Pearson correlation coefficient is greater than the preset threshold of 0.7, then the significance of the correlation can be further evaluated through a t-test. The t-test determines whether the correlation is caused by random factors through statistical methods. Suppose the data subset contains 50 data points and the correlation coefficient is 0.85. After calculating the t-value, the p-value is 0.01, which is less than the preset threshold of 0.05, indicating that the association between the seal performance and the operating torque is statistically significant. This method quantifies the reliability of the result through the p-value to ensure the credibility of the analysis conclusion.

[0063] In an exemplary embodiment, according to the parameter correlation matrix, the optimization effect of interpolation correction and sliding window smoothing on the calculation of the Pearson correlation coefficient can be verified. If the standard error corresponding to the Pearson correlation coefficient is less than the preset error threshold, the effectiveness of the data preprocessing is confirmed, and an optimized parameter correlation matrix is obtained.

[0064] It should be noted that the calculation formula of the standard error is as follows:

[0065] where SE r is the standard error, r is the Pearson correlation coefficient, and n is the corresponding number of samples.

[0066] Step S106: By using the principal component analysis algorithm, determine multiple principal component vectors and corresponding eigenvalues according to the parameter correlation matrix, and determine the current health status score of the fire valve according to the eigenvalues corresponding to the multiple principal component vectors; It should be noted that principal component analysis (PCA) is a key data reduction and feature extraction tool. Through the principal component analysis algorithm, a complex multi-dimensional parameter correlation matrix can be transformed into a set of new orthogonal variables (i.e., principal component vectors), which can retain the information in the original data to the greatest extent while reducing redundancy. The core of principal component analysis is to extract the main component vectors reflecting the overall degradation state of the valve and calculate the contribution rate weights of each principal component vector. This step helps to extract the core features from high-dimensional data and reduce the complexity of subsequent calculations.

[0067] It should be noted that the eigenvalue is a key indicator in principal component analysis to measure the importance of each principal component vector. The larger the eigenvalue, the stronger the explanatory ability of the principal component vector to the overall data change.

[0068] In an exemplary embodiment, the above-mentioned determining the current health status score of the fire valve according to the eigenvalues corresponding to the multiple principal component vectors can be implemented through the following steps S41 - S42: Step S41: Determine the weight values corresponding to the multiple principal component vectors according to the eigenvalues corresponding to the multiple principal component vectors, where the weight value corresponding to the target principal component vector among the multiple principal component vectors is equal to the eigenvalue corresponding to the target principal component vector divided by the sum of the eigenvalues corresponding to the multiple principal component vectors; Step S42: Perform a weighted sum of the weight values and eigenvalues corresponding to the multiple principal component vectors to obtain the current health status score of the fire valve.

[0069] Exemplarily, assume that the data collected by a certain valve system contains 600 time points. The sealing performance is represented by the leakage rate, with the unit of mL / min, the operating torque is represented by N·m, and the degree of corrosion is represented by the proportion of the surface corrosion area. Through principal component analysis, the parameter correlation matrix of these three parameters can be transformed into eigenvalues and eigenvectors. The eigenvalue reflects the explanatory ability of each principal component to the data variation, and the eigenvector defines the linear combination method of the original variables. Suppose the calculated three eigenvalues are 4.5, 1.2, and 0.3 respectively. The first principal component explains most of the variation, indicating that it has the strongest representativeness for the degradation characteristics of the fire valve.

[0070] It should be noted that the calculation of the contribution rate (equivalent to the above weight value) is an important step in principal component analysis. The contribution rate reflects the proportion of the total variation explained by each principal component. For example, the sum of the eigenvalues 4.5, 1.2, and 0.3 is 6.0. The contribution rate of the first principal component is 4.5 / 6.0 = 75%, the second is 1.2 / 6.0 = 20%, and the third is 0.3 / 6.0 = 5%. This indicates that the first principal component captures 75% of the information and is the core for analyzing the degradation characteristics of the fire valve. The principal components with higher contribution rates account for a larger proportion in the calculation of the current health status score, ensuring that the result is closer to the actual degradation trend.

[0071] It should be noted that the current health status score is calculated by the weighted linear combination method, combining the principal component vectors and the contribution rate weights. Exemplarily, assume that the eigenvalue of the first principal component vector reflects a strong correlation between the sealing performance and the operating torque, with a weight of 0.75; the eigenvalue of the second principal component vector mainly reflects the degree of corrosion, with a weight of 0.20. When calculating, multiply the eigenvalue of each principal component vector by the corresponding weight and sum them to obtain the current health status score.

[0072] Optionally, assume that the current health status score of the fire valve is 0.65, while the preset health threshold is 0.70, which indicates that the fire valve may enter the degradation state. The evaluation report will record this score and the degradation state, prompting the maintenance requirements.

[0073] It should be noted that the generation of the evaluation report needs to integrate multi-dimensional information. For example, the report may include a time series graph showing the change trend of the sealing performance over time, such as the leakage rate increasing from 0.02 mL / min to 0.05 mL / min, combined with the change in the operating torque from 50 N·m to 60 N·m, indicating a possible risk of seal failure. The report may also mark the change in the degree of corrosion, such as the proportion of the surface corrosion area increasing from 5% to 8%, providing a basis for maintenance decisions. This comprehensive analysis of multiple parameters ensures the comprehensiveness of the degradation state judgment.

[0074] In an exemplary embodiment, the dimensionality reduction result of principal component analysis can also be used to optimize the monitoring strategy. For example, by analyzing the principal component vectors, it is found that the operating torque contributes more to the degradation characteristics. The torque monitoring frequency can be adjusted preferentially, and unnecessary corrosion degree detections can be reduced, thereby optimizing the resource allocation.

[0075] Optionally, the health status score can be calculated at regular intervals and then compared with historical data. For example, if the current health status score drops from 0.75 to 0.65, it indicates an exacerbation of the degradation trend. The evaluation report can include suggestions such as checking the seals or replacing the lubricant. This continuous monitoring and report generation method provides dynamic support for the maintenance of the fire valve.

[0076] Step S108: Determine the degradation law function of the fire protection valve according to the historical operation records of the fire protection valve, where the degradation law function is used to reflect the change relationship of the health status score of the fire protection valve over time; It should be noted that there is no execution sequence between the above steps S102 - S106 and the above step S108.

[0077] In an exemplary embodiment, the above step S108 can be implemented through the following steps S51 - S53: Step S51: Process the historical operation records by using a time series analysis method to obtain health index time series data, where the historical operation records include: operation records, fault records, and maintenance records, and the health index time series data includes the health status scores at different time points; It should be noted that obtaining a data subset containing fault records and maintenance records from the historical operation records can provide reliable basic data for time series analysis. Historical operation records usually contain the operation parameters of the valve at different time points, such as operating pressure, temperature, leakage rate, etc., as well as the time, type of faults, and maintenance records, such as seal replacement or lubricant addition. The screening of the data subset needs to ensure data integrity and representativeness. Exemplarily, select the operation data of an industrial valve system in the past year, including records at 1200 time points, covering the operating pressure from 2.0 MPa to 3.5 MPa, the temperature from 20°C to 80°C, and fault records such as two minor leaks and one major seal failure.

[0078] It should be noted that the time series analysis method processes the data subset and calculates the time series values of the health status score to reveal the valve degradation trend. The health status score is usually a comprehensive quantitative index based on multiple parameters, for example, combining the changes in leakage rate, operating pressure, and temperature.

[0079] Optionally, the calculation method of the health status score in the health index time series data and the calculation method of the current health status score according to the parameter correlation matrix above can be the same or different. In the case of different calculation methods, the final calculation will be standardized so that the health status scores obtained by different calculation methods can be compared.

[0080] Step S52: In the case where the health index time series data is linearly varying data, use a linear regression method to calculate the slope of the change of the health status score over time according to the health index time series data to obtain the degradation rate; determine the degradation law function according to the initial health status score of the fire protection valve and the degradation rate; It should be noted that in the case where the health index time series data is linearly varying data, linear regression analysis can be used to calculate the change slope of the health index over time, that is, the degradation rate.

[0081] Exemplarily, if the health index time series data is analyzed and it is found that the health status score decreases by 0.02 per month, the degradation rate is 0.02 / month. This reflects the degradation speed of the fire valve under the current operating conditions. The level of the degradation rate is closely related to the operating conditions. It should be noted that under high-pressure conditions, the degradation rate may increase to 0.03 / month, while under low-temperature and low-pressure conditions, it may decrease to 0.01 / month.

[0082] Optionally, the degradation law function can be:

[0083] where HI(t) is the health status score at time t, HI0 is the initial health status score, DR is the degradation rate, and t is the time.

[0084] In an exemplary embodiment, multiple regression analysis can also be used to establish the relationship between the degradation rate and the operating condition parameters. The operating pressure and temperature are used as key operating condition parameters and directly affect the valve degradation. For example, in a certain scenario, data analysis shows that for every 0.5 MPa increase in the operating pressure, the degradation rate increases by 0.005 / month; for every 10 °C increase in the temperature, the degradation rate increases by 0.003 / month. This relationship is quantified by a multiple regression model, clearly revealing the impact of the operating conditions on the degradation. In practical applications, the operating personnel can adjust the operating pressure or optimize the cooling measures according to this model to delay the degradation process.

[0085] Step S53: In the case where the health index time series data is non-linearly varying data, use the Weibull distribution model to determine the degradation law function according to the health index time series data.

[0086] Optionally, the Weibull distribution model can characterize the distribution characteristics of the degradation rate through its shape parameter and scale parameter. For example, when analyzing the data of a certain valve, the shape parameter of the fitted Weibull distribution is 2.5 and the scale parameter is 1500 hours, which indicates that the degradation trend accelerates after running for about 1000 hours. The advantage of this model is that it can effectively capture the non-linear degradation law and provide support for the prediction of the remaining service life.

[0087] Optionally, based on the fitted Weibull distribution model, the probability density function of the remaining useful life can be calculated, which describes the probability distribution of the valve failing in the future. For example, the probability density function of a certain valve shows that the probability of it failing within the next 500 hours is 20%, and the probability of it failing within 1000 hours is 60%. By calculating the expected value of this function using numerical integration methods, the expected value of the remaining useful life can be obtained. For example, the calculation result shows that the expected value is 800 hours, indicating that the valve can still operate for an average of 800 hours, and this result intuitively reflects the remaining life of the valve.

[0088] Step S110: Predict the service life of the fire protection valve based on the current health status score and the degradation law function.

[0089] In an exemplary embodiment, the above step S110 includes: determining the specified health status score when the fire protection valve reaches the service life; determining the first time according to the current health status score and the degradation law function, and determining the second time according to the specified health status score and the degradation law function; determining the remaining service life of the fire protection valve according to the first time and the second time.

[0090] It should be noted that the specified health status score when the fire protection valve reaches the service life needs to be set based on industry standards or actual experience, indicating the state where the valve performance deteriorates to the point where it cannot meet the requirements for safe operation. For example, assuming that when the health status score of a certain fire protection valve drops below 0.3, the valve will not work properly, then 0.3 can be used as the specified health status score.

[0091] The current health status score reflects the actual health condition of the fire protection valve at the current time point, while the degradation law function describes the trend of the health status score changing over time. By substituting the current health status score into the degradation law function, the time point corresponding to the current moment, that is, the first time, can be calculated. For example, if the current health status score is 0.7 and the degradation law function is HI(t)=0.9 - 0.02t; then the first time t1 = 10 (unit: month) can be obtained by solving the equation 0.7 = 0.9 - 0.02y.

[0092] Furthermore, determine the second time according to the specified health status score and the degradation law function. By substituting the specified health status score into the degradation law function, the time point when the health status score of the fire protection valve reaches the specified value, that is, the second time, can be calculated. For example, if the specified health status score is 0.3 and the above degradation law function HI(t)=0.9 - 0.02t is still used, then the second time t2 = 30 (unit: month) can be obtained by solving the equation 0.3 = 0.9 - 0.02t.

[0093] Finally, the remaining service life = t2 - t1 = 30 - 10 = 20 months. That is, from the current time point, the fire valve is expected to operate normally for another 20 months.

[0094] It should be noted that this prediction method based on the current health status score and the degradation law function can not only provide accurate life assessment, but also help the operation and maintenance personnel formulate maintenance plans in advance to avoid downtime losses caused by sudden failures.

[0095] In an exemplary embodiment, after the above step S110, the method further includes: integrating the newly obtained real-time monitoring data into the life prediction result through the Bayesian update algorithm, dynamically correcting the correlation coefficients of the degradation law function, and then obtaining a continuously optimized predicted value of the remaining life of the fire valve.

[0096] It should be noted that the real-time monitoring data includes the latest sealing performance, operating torque, and corrosion degree data obtained from the multi-sensor synchronous acquisition technology. After adding timestamps to these data through the time synchronization protocol, they are integrated into the existing multi-dimensional monitoring data matrix.

[0097] In the above steps S102 - S110, the multi-dimensional monitoring data matrix is used to comprehensively obtain key parameter data such as sealing performance, operating torque, and corrosion degree. Compared with the traditional single-parameter method, it can more completely reflect the operating state of the valve and its complex degradation process. Secondly, by constructing a parameter correlation matrix and calculating the correlation between any two parameters, and using the principal component analysis algorithm to extract the main component vectors and their eigenvalues, the current health status score is calculated, which not only reduces data redundancy but also highlights the key factors that have the greatest impact on valve degradation. In addition, by establishing a degradation law function in combination with historical operation records, taking into account the change relationship of the health status score over time, and performing life prediction based on the current health status score and the degradation law function, the accuracy of the fire valve life prediction is greatly improved, providing a scientific basis for equipment maintenance decisions.

[0098] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases, the former is a better implementation method. Based on such an understanding, the technical essence of the present application or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), including several instructions for causing a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of the present application.

[0099] In this embodiment, a service life prediction system for a fire protection valve is further provided. The service life prediction system for the fire protection valve is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be repeated. As used hereinafter, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0100] Figure 2 FIG. 4 is a structural block diagram of a service life prediction system for a fire protection valve according to an embodiment of the present application. The system includes: An acquisition module 202, configured to acquire a multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix includes sealing performance data, operating torque data, and corrosion degree data of the fire protection valve continuously monitored within a preset time period; A first determination module 204, configured to determine a parameter correlation matrix according to the multi-dimensional monitoring data matrix, where the parameter correlation matrix is used to represent the correlation between any two parameters among a plurality of parameters, and the plurality of parameters include: sealing performance, operating torque, and corrosion degree; A second determination module 206, configured to determine a plurality of principal component vectors and corresponding eigenvalues according to the parameter correlation matrix through a principal component analysis algorithm, and determine the current health state score of the fire protection valve according to the eigenvalues corresponding to the plurality of principal component vectors; A third determination module 208, configured to determine a degradation law function of the fire protection valve according to the historical operation record of the fire protection valve, where the degradation law function is used to reflect the change relationship of the health state score of the fire protection valve over time; A prediction module 210, configured to predict the service life of the fire protection valve according to the current health state score and the degradation law function.

[0101] The above system comprehensively acquires key parameter data such as sealing performance, operating torque, and corrosion degree by using a multi-dimensional monitoring data matrix. Compared with the traditional single-parameter method, it can more completely reflect the operating state of the valve and its complex degradation process. Secondly, by constructing a parameter correlation matrix and calculating the correlation between any two parameters, and using the principal component analysis algorithm to extract the main component vectors and their eigenvalues, and calculating the current health state score, it not only reduces data redundancy, but also highlights the key factors that have the greatest impact on valve degradation. In addition, by combining the historical operation record to establish a degradation law function, taking into account the change relationship of the health state score over time, and performing life prediction based on the current health state score and the degradation law function, the accuracy of the fire protection valve life prediction is greatly improved, providing a scientific basis for equipment maintenance decision-making.

[0102] In an exemplary embodiment, the system further includes: a processing module, which, after obtaining the multi-dimensional monitoring data matrix, the method further includes: processing each time series data in the multi-dimensional monitoring data matrix in the following manner to process the multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix includes: seal performance time series data, operating torque time series data, and corrosion degree time series; when processing each time series data, each time series data is target time series data, and the target time series data includes data at different time points: using a median filtering algorithm to smooth the target time series data to obtain the smoothed target time series data; determining abnormal data in the smoothed target time series data based on a standard deviation-based outlier detection algorithm, and using a linear interpolation algorithm to correct the abnormal data to obtain the corrected target time series data; using a spline interpolation algorithm to perform curve smoothing on the corrected target time series data.

[0103] In an exemplary embodiment, the processing module is further configured to, after performing curve smoothing on the corrected target time series data using a spline interpolation algorithm, determine a target window size and a target moving step; through window translation technology, process the curve-smoothed target time series data according to the target window size and the target moving step to obtain standardized time series data, where the standardized time series data includes the weighted average of the target time series data within each target window size.

[0104] In an exemplary embodiment, the first determination module is further configured to calculate the Pearson correlation coefficient between the seal performance and the operating torque based on the seal performance time series data and the operating torque time series data in the multi-dimensional monitoring data matrix to obtain a first Pearson correlation coefficient; calculate the Pearson correlation coefficient between the operating torque and the corrosion degree based on the operating torque time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain a second Pearson correlation coefficient; calculate the Pearson correlation coefficient between the seal performance and the corrosion degree based on the seal performance time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain a third Pearson correlation coefficient; determine the parameter correlation matrix based on the first Pearson correlation coefficient, the second Pearson correlation coefficient, and the third Pearson correlation coefficient.

[0105] In an exemplary embodiment, the second determination module is further configured to determine weight values corresponding to the multiple principal component vectors according to eigenvalues corresponding to the multiple principal component vectors, where the weight value corresponding to a target principal component vector among the multiple principal component vectors is equal to the eigenvalue corresponding to the target principal component vector divided by the sum of the eigenvalues corresponding to the multiple principal component vectors; perform weighted summation on the weight values and eigenvalues corresponding to the multiple principal component vectors to obtain the current health status score of the fire valve.

[0106] In an exemplary embodiment, the third determination module is further configured to process the historical operation records by using a time series analysis method to obtain health index time series data, where the historical operation records include: operation records, fault records, and maintenance records, and the health index time series data includes health status scores at different time points; in the case where the health index time series data is linearly varying data, calculate the slope of the change of the health status score with time according to the health index time series data by using a linear regression method to obtain a degradation rate; determine the degradation law function according to the initial health status score and the degradation rate of the fire valve; or in the case where the health index time series data is non-linearly varying data, use a Weibull distribution model to determine the degradation law function according to the health index time series data.

[0107] In an exemplary embodiment, the prediction module is further configured to determine a specified health status score when the fire valve reaches its service life; determine a first time according to the current health status score and the degradation law function, and determine a second time according to the specified health status score and the degradation law function; determine the remaining service life of the fire valve according to the first time and the second time.

[0108] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, where the computer program is configured to execute the steps in any one of the above method embodiments when running.

[0109] Optionally, in this embodiment, the above storage medium may be configured to store a computer program for executing the following steps.

[0110] In an exemplary embodiment, the above computer-readable storage medium may include, but is not limited to: various media such as a USB flash drive, a read-only memory (ROM for short), a random access memory (RAM for short), a mobile hard disk, a magnetic disk, or an optical disc that can store a computer program.

[0111] For the specific examples in this embodiment, reference may be made to the examples described in the above embodiments and the exemplary embodiments, and they will not be elaborated herein.

[0112] An embodiment of the present application also provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0113] Another embodiment of the present application also provides a computer program product, including a non-volatile computer-readable storage medium, where the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are implemented.

[0114] An embodiment of the present application also provides an electronic device, which includes a memory and a processor. A computer program is stored in the memory, and the processor is configured to execute the steps in any one of the above method embodiments through the computer program.

[0115] For the specific examples in this embodiment, reference may be made to the examples described in the above embodiments and the exemplary embodiments, and they will not be elaborated herein.

[0116] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. They can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a different order than here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module to implement. In this way, the present application is not limited to any specific combination of hardware and software.

[0117] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A method for predicting the service life of a fire valve, characterized in that, Including: Obtain a multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix has data on the sealing performance, operating torque, and corrosion degree of a fire valve continuously monitored within a preset time period; Determine a parameter correlation matrix according to the multi-dimensional monitoring data matrix, where the parameter correlation matrix is used to represent the correlation between any two parameters among multiple parameters, and the multiple parameters include: sealing performance, operating torque, and corrosion degree; Through the principal component analysis algorithm, determine multiple principal component vectors and corresponding eigenvalues according to the parameter correlation matrix, and determine the current health status score of the fire valve according to the eigenvalues corresponding to the multiple principal component vectors; Determine the degradation law function of the fire valve according to the historical operation records of the fire valve, where the degradation law function is used to reflect the change relationship of the health status score of the fire valve over time; Predict the service life of the fire valve according to the current health status score and the degradation law function.

2. The method according to claim 1, characterized in that After obtaining the multi-dimensional monitoring data matrix, the method further includes: Process each time series data in the multi-dimensional monitoring data matrix in the following manner to process the multi-dimensional monitoring data matrix, where the multi-dimensional monitoring data matrix includes: sealing performance time series data, operating torque time series data, and corrosion degree time series data; when processing each time series data, each time series data is target time series data, and the target time series data includes data at different time points: Use the median filtering algorithm to smooth the target time series data to obtain the smoothed target time series data; Determine the abnormal data in the smoothed target time series data based on the outlier detection algorithm based on the standard deviation, and use the linear interpolation algorithm to correct the abnormal data to obtain the corrected target time series data; Use the spline interpolation algorithm to perform curve smoothing on the corrected target time series data.

3. The method according to claim 2, wherein After using the spline interpolation algorithm to perform curve smoothing on the corrected target time series data, the method further includes: Determine the target window size and the target moving step; Through the window translation technique, process the curve-smoothed target time series data according to the target window size and the target moving step to obtain the standardized time series data, where the standardized time series data includes the weighted average value of the target time series data within each target window size.

4. The method according to claim 1, characterized in that, Determining the parameter correlation matrix according to the multi-dimensional monitoring data matrix includes: Calculate the Pearson correlation coefficient between the sealing performance and the operating torque according to the sealing performance time series data and the operating torque time series data in the multi-dimensional monitoring data matrix to obtain the first Pearson correlation coefficient; Calculate the Pearson correlation coefficient between the operating torque and the corrosion degree according to the operating torque time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix to obtain the second Pearson correlation coefficient; Calculate the Pearson correlation coefficient between the sealing performance and the corrosion degree according to the sealing performance time series data and the corrosion degree time series data in the multi-dimensional monitoring data matrix, and obtain the third Pearson correlation coefficient; Determine the parameter correlation matrix according to the first Pearson correlation coefficient, the second Pearson correlation coefficient and the third Pearson correlation coefficient.

5. The method according to claim 1, wherein Determine the current health status score of the fire protection valve according to the eigenvalues corresponding to the multiple principal component vectors, including: Determine the weight values corresponding to the multiple principal component vectors according to the eigenvalues corresponding to the multiple principal component vectors, wherein the weight value corresponding to the target principal component vector in the multiple principal component vectors is equal to the eigenvalue corresponding to the target principal component vector divided by the sum of the eigenvalues corresponding to the multiple principal component vectors; Perform weighted summation on the weight values and eigenvalues corresponding to the multiple principal component vectors to obtain the current health status score of the fire protection valve.

6. The method according to claim 1, wherein Determine the degradation law function of the fire protection valve according to the historical operation records of the fire protection valve, including: Process the historical operation records by using a time series analysis method to obtain health index time series data, wherein the historical operation records include operation records, fault records and maintenance records, and the health index time series data includes health status scores at different time points; In the case where the health index time series data is linearly varying data, use a linear regression method to calculate the slope of the change of the health status score with time according to the health index time series data to obtain the degradation rate; determine the degradation law function according to the initial health status score and the degradation rate of the fire protection valve; or In the case where the health index time series data is non-linearly varying data, use a Weibull distribution model to determine the degradation law function according to the health index time series data.

7. The method according to claim 1, wherein Predict the service life of the fire protection valve according to the current health status score and the degradation law function, including: Determine the specified health status score when the fire protection valve reaches the service life; Determine the first time according to the current health status score and the degradation law function, and determine the second time according to the specified health status score and the degradation law function; Determine the remaining service life of the fire protection valve according to the first time and the second time.

8. A service life prediction system for a fire protection valve, characterized in that, Including: An acquisition module for acquiring a multi-dimensional monitoring data matrix, wherein the multi-dimensional monitoring data matrix has sealing performance data, operating torque data and corrosion degree data of a fire protection valve continuously monitored within a preset time period; A first determination module for determining a parameter correlation matrix according to the multi-dimensional monitoring data matrix, wherein the parameter correlation matrix is used to represent the correlation between any two parameters among a plurality of parameters, and the plurality of parameters include: sealing performance, operating torque and corrosion degree; A second determination module for determining a plurality of principal component vectors and corresponding eigenvalues according to the parameter correlation matrix by using a principal component analysis algorithm, and determining the current health status score of the fire protection valve according to the eigenvalues corresponding to the plurality of principal component vectors; A third determination module, configured to determine a degradation law function of the fire protection valve according to the historical operation record of the fire protection valve, wherein the degradation law function is used to reflect the variation relationship of the health state score of the fire protection valve with time; A prediction module, configured to predict the service life of the fire protection valve according to the current health state score and the degradation law function.

9. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, wherein when the computer program is executed by a processor, the steps of the method described in any one of claims 1 to 7 are implemented.

10. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method described in any one of claims 1 to 7 are implemented.

Images