Computer monitoring system applied to water conservancy project

Abstract

The invention discloses a computer monitoring system applied to a water conservancy project, and particularly relates to the technical field of computer monitoring. By monitoring the running state of water level sensor equipment, collecting sensor output data and combining a time sequence analysis method, the overall performance stability of a sensor and the consistency of current water level data and historical data are evaluated; through a fault diagnosis algorithm and an early warning rule, sensor faults can be identified and processed in time, adverse effects of the faults on a water conservancy project are reduced, a standby sensor switching module is designed, automatic switching to a standby sensor and starting can be realized, continuity and stability of a monitoring system are ensured, and reliability of the monitoring system is improved. It can be guaranteed that the monitoring system can continuously and stably operate under various environments and working conditions, and therefore safe operation and management of the water conservancy project are guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of computer monitoring, and in particular to a computer monitoring system applied to water conservancy projects. Background Art

[0002] The computer monitoring technology of water conservancy projects has become quite mature, mainly including sensor technology, real-time data acquisition and transmission technology, database and data management technology, data analysis and decision support technology, remote monitoring and control technology, artificial intelligence and big data technology, etc. The comprehensive application of these technologies enables water conservancy projects to achieve real-time monitoring of various parameters, efficient management and analysis of data, remote monitoring and control, and intelligent decision support, providing strong technical support for the safe operation and effective management of water conservancy projects. However, due to environmental factors or equipment aging, some water level sensors may fail after long-term operation, resulting in the monitoring system being unable to accurately read water level data, thereby affecting the accurate monitoring and early warning of water level changes. Summary of the invention

[0003] The purpose of the present invention is to provide a computer monitoring system applied to water conservancy projects to solve the shortcomings of the background technology.

[0004] In order to achieve the above object, the present invention provides the following technical solution: A computer monitoring system applied to water conservancy projects, characterized in that it includes a sensor monitoring module, a water level data quality assessment module, a fault diagnosis and early warning module and a standby sensor switching module; Sensor monitoring module: monitors the equipment operation status of the water level sensor, collects the output data of the sensor, analyzes the output data trend of the sensor using the time series analysis method, and evaluates the overall performance stability of the sensor; Water level data quality assessment module: compare and analyze the currently collected water level data with historical data, determine the degree of difference between the data, and evaluate the consistency between the current water level data and historical data; Fault diagnosis and early warning module: conducts comprehensive analysis on the overall performance stability of the sensor and the consistency between the current water level data and the historical data. By setting the fault diagnosis algorithm and early warning rules, the sensor faults found in the monitoring system are diagnosed and early warning notifications are issued in a timely manner. Backup sensor switching module: Set up backup sensors and design switching logic. When a sensor fails, it automatically switches to the backup sensor and is enabled.

[0005] In a preferred embodiment, in the sensor monitoring module, the equipment operation status of the water level sensor is monitored, and the output data of the sensor is collected, and the output data trend of the sensor is analyzed using a time series analysis method to evaluate the overall performance stability of the sensor, including the following: Stationarity calculation: ADF test is used to test the stationarity of time series data. The null hypothesis of ADF test is that the sequence has a unit root, that is, non-stationary. If the p-value is less than the set significance level, the null hypothesis is rejected and the sequence is considered to be stationary. Periodicity calculation: Use the autocorrelation function to analyze whether the data has periodicity, calculate the autocorrelation coefficient, and observe whether there is a significant periodic peak; Trend calculation: Use linear regression to test whether there is a trend in the data; perform linear fitting on time series data, observe the significance of the regression coefficient and the slope of the fitting curve, and determine whether there is a trend change; Seasonality calculation: Use decomposition methods to detect whether the data is affected by seasonality, perform seasonal decomposition on the data, and calculate seasonal indicators; Normalize the results of stability, periodicity, trend and seasonality, and comprehensively calculate the overall performance stability coefficient of the sensor through the calculation results of the normalized stability, periodicity, trend and seasonality; The overall performance stability coefficient of the sensor is compared with the standard threshold. If the overall performance stability coefficient of the sensor is greater than or equal to the standard threshold, a normal device signal is issued; if the overall performance stability coefficient of the sensor is less than the standard threshold, an abnormal device signal is issued.

[0006] In a preferred embodiment, in the water level data quality assessment module, the consistency between the current water level data and the historical data is assessed, including the following contents; Collect the currently collected water level data and historical water level data, organize them into a time series format, and align the currently collected water level data with the historical data; The aligned data is used for difference calculation, which directly calculates the difference between the current data and the historical data, that is, , where For the current data, For historical data; Calculate the ratio of the difference between current data and historical data to historical data, that is, ; Draw a time series diagram of the difference data changes over time, calculate the statistical indicators of the difference data; draw a histogram of the difference data to show the distribution of the data; Observe the distribution and peak shape of the data through histograms and kernel density plots to determine whether there is skewness and peak phenomenon; Based on the results of difference analysis and anomaly detection, the consistency between current water level data and historical data is comprehensively evaluated; If the consistency is high and there are no obvious anomalies, the current data is considered to be consistent with the historical data; if the consistency is low and there are anomalies, the current data is considered to be inconsistent with the historical data.

[0007] In a preferred embodiment, in the fault diagnosis and early warning module, a comprehensive analysis is performed on the overall performance stability of the sensor and the consistency between the current water level data and the historical data, and specific calculation steps for supervised learning model training using a support vector machine include: Select support vector machine as the supervised learning model and use labeled data to train the SVM model; The goal of SVM is to find an optimal hyperplane that maximizes the interval between abnormal data and normal data; The optimization objective of SVM is expressed as: , the constraints are: , where i=1,2,...,n, w is the normal vector of the hyperplane, b is the intercept, C is the penalty parameter, is the slack variable, It's a label. is the eigenvector.

[0008] In a preferred embodiment, the trained model is evaluated using cross validation, and the trained SVM model is used to predict the current data to obtain the abnormal probability; according to the prediction result and the set threshold, an early warning notification is triggered, including: Divide historical data into two subsets with higher stability and lower stability; Use each subset to evaluate the model and calculate the accuracy, recall, and F1-score of the model on different subsets; Use the trained model to predict the current data and obtain the abnormal probability p; Set a standard threshold θ to convert the abnormal probability of continuous values ​​into the classification results of marked abnormalities and marked normals; If the probability p is greater than the threshold θ, the data is marked as abnormal and a warning signal is issued; if the probability p is less than or equal to the threshold θ, the data is marked as normal and no warning signal is issued.

[0009] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. The present invention monitors the equipment operation status of the water level sensor, collects the output data of the sensor, analyzes the trend of the output data of the sensor using the time series analysis method, evaluates the overall performance stability of the sensor, compares and analyzes the currently collected water level data with the historical data, determines the degree of difference between the data, and evaluates the consistency between the current water level data and the historical data. It aims to improve the data accuracy of the monitoring system, achieve early fault warning, optimize maintenance strategies, thereby improving the reliability and stability of the system, reducing maintenance costs, and ensuring the safe operation and management of water conservancy projects.

[0010] 2. The present invention conducts a comprehensive analysis of the overall performance stability of the sensor and the consistency between the current water level data and the historical data, and diagnoses the sensor faults found in the monitoring system by setting the fault diagnosis algorithm and early warning rules, and promptly issues early warning notifications. By setting the fault diagnosis algorithm and early warning rules, the sensor faults can be identified and processed in time to reduce the impact of the faults on the water conservancy project; a backup sensor is set, and the switching logic is designed. When the sensor fails, it automatically switches to the backup sensor and enables it. It can effectively improve the reliability and robustness of the monitoring system, so that it can operate continuously and stably in various environments and working conditions, and ensure the safe operation and management of the water conservancy project. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0012] Figure 1 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] For examples, see Figure 1 As shown, the computer monitoring system applied to water conservancy projects described in this embodiment includes: a sensor monitoring module, a water level data quality assessment module, a fault diagnosis and early warning module and a standby sensor switching module; Sensor monitoring module: monitors the equipment operation status of the water level sensor, collects the output data of the sensor, analyzes the output data trend of the sensor using the time series analysis method, and evaluates the overall performance stability of the sensor; Water level data quality assessment module: compare and analyze the currently collected water level data with historical data, determine the degree of difference between the data, and evaluate the consistency between the current water level data and historical data; Fault diagnosis and early warning module: conducts comprehensive analysis on the overall performance stability of the sensor and the consistency between the current water level data and the historical data. By setting the fault diagnosis algorithm and early warning rules, the sensor faults found in the monitoring system are diagnosed and early warning notifications are issued in a timely manner. Backup sensor switching module: Set up backup sensors and design switching logic. When a sensor fails, it automatically switches to the backup sensor and is enabled.

[0015] Among them, in the sensor monitoring module, the equipment operation status of the water level sensor is monitored, and the output data of the sensor is collected. The output data trend of the sensor is analyzed using the time series analysis method to evaluate the overall performance stability of the sensor. Specifically: Set the acquisition frequency, obtain data from the water level sensor regularly or periodically, record the collected water level data, and establish a time series data set; Stationarity calculation: Use the unit root test (ADF test) to test the stationarity of time series data. The null hypothesis of the ADF test is that the sequence has a unit root, that is, non-stationary; if the p-value is less than the set significance level (usually 0.05), the null hypothesis is rejected and the sequence is considered to be stationary. The specific calculation expression can be: ADF= , where represents the autocorrelation coefficient of the time series, represents the residual variance of the series, represents the sample size; Periodicity calculation: Use the autocorrelation function to analyze whether the data has periodicity; calculate the autocorrelation coefficient and observe whether there is a significant periodic peak. The specific calculation expression is: rk= , where represents the observed value of the time series, represents the mean of the observations, represents the lag period, rk is the autocorrelation coefficient; Trend calculation: Use linear regression to test whether there is a trend in the data; perform linear fitting on the time series data, observe the significance of the regression coefficient and the slope of the fitting curve, and determine whether there is a trend change. The specific calculation formula can be: ; In the formula, represents the observed value of the time series, t represents time, c and c1t represent regression coefficients, and et represents the error term; Seasonal calculation: Use decomposition methods (such as additive model or multiplicative model) to detect whether the data is affected by seasonality, decompose the data seasonally, and calculate seasonal indicators, such as the change amplitude of the seasonal component or the significance of the seasonal indicator. The specific calculation formula can be: yt=Tt+St+Et, where Tt represents the trend component, St represents the seasonal component, Et represents the error term, and yt is the seasonal decomposition coefficient; The overall performance stability of the sensor is calculated by weighted summing up the calculation results of stationarity, periodicity, trend, and seasonality: Normalize the results of stability, periodicity, trend and seasonality, and comprehensively calculate the overall performance stability coefficient of the sensor through the calculation results of the normalized stability, periodicity, trend and seasonality; For example, the present invention can use the following formula to calculate the overall performance stability coefficient of the sensor, and the calculation expression is: = + yt, where is the stability coefficient, , , , is the unit root, autocorrelation coefficient, the observed value of the time series, and the proportionality coefficient of the seasonal decomposition coefficient, and 0; The overall performance stability coefficient of the sensor is compared with the standard threshold. If the overall performance stability coefficient of the sensor is greater than or equal to the standard threshold, it means that the overall performance stability of the sensor is higher, and a normal device signal is issued at this time; if the overall performance stability coefficient of the sensor is less than the standard threshold, it means that the overall performance stability of the sensor is lower, and an abnormal device signal is issued at this time.

[0016] The greater the overall performance stability of the sensor, the better the sensor's working condition and data output stability over a long period of time, indicating that the sensor is of high design and manufacturing quality and the operating environment is well controlled and managed. Therefore, the greater the overall performance stability, the lower the risk of sensor failure, specifically: The overall performance stability of a sensor reflects the quality of its design and manufacturing. Sensors with high stability tend to have more sophisticated designs and strict manufacturing processes, so the probability of failure in long-term operation is relatively low.

[0017] The improvement of the overall performance stability of the sensor means that its data output is more stable. Stable data output means that the sensor's measurement results are more reliable and accurate, reducing the probability of data anomalies and errors, thereby reducing the risk of failure.

[0018] The improvement in the overall performance stability of the sensor may reflect that the operating environment is better controlled and managed. A good operating environment can reduce the impact of external factors on the sensor and reduce the possibility of sensor failure.

[0019] The higher the overall performance stability, the more stable the sensor will be in long-term operation. This means that the sensor will remain reliable in the face of various working conditions and environmental changes, reducing the risk of failure caused by long-term use.

[0020] Sensors with high stability usually require less maintenance and care because their working state is relatively stable and the possibility of failure is less. This reduces maintenance costs and workload, and further reduces the risk of sensor failure.

[0021] In the water level data quality assessment module, the currently collected water level data is compared and analyzed with the historical data to determine the degree of difference between the data and to assess the consistency between the current water level data and the historical data, including: Collect the current water level data and historical water level data and organize them into time series; Align the currently collected water level data with the historical data to ensure the consistency of time points. Data can be aligned by timestamp or time index; The aligned data is used for difference calculation, which can calculate the absolute difference or relative difference between the current data and the historical data. Common calculation methods include: Absolute difference: directly calculate the difference between the current data and the historical data, that is, , where For the current data, For historical data; Relative difference: Calculate the ratio of the difference between current data and historical data to historical data, that is, ; Analyze the calculated difference data and observe the distribution trend of the difference, including: Draw the change of difference data over time into a time series diagram to observe its fluctuation and trend change.

[0022] The time series diagram can be used to intuitively discover the distribution trend of difference data, including whether there are periodic fluctuations, long-term trends, etc.

[0023] Calculate statistical indicators of difference data, such as mean, standard deviation, maximum value, minimum value, etc.

[0024] Statistical indicators can be used to understand the distribution characteristics of difference data, as well as the degree of fluctuation and dispersion of data; Draw a histogram or kernel density plot of the difference data to show the distribution of the data.

[0025] The distribution shape and peak state of the data can be observed through histograms or kernel density plots, as well as whether there is skewness or peak phenomenon; Perform distribution fitting on the difference data and try to fit common probability distribution models, such as normal distribution, exponential distribution, lognormal distribution, etc. By fitting the distribution model, we can understand the distribution form of the data and whether it conforms to a specific probability distribution; Perform time series analysis on difference data, including trend analysis, periodicity analysis, and seasonality analysis.

[0026] Through time series analysis, we can understand the long-term trend, cyclical fluctuations and seasonal changes of data, and then analyze the distribution trend of differential data; If the difference is small and the fluctuation is stable, it means that the current data is highly consistent with the historical data; If the difference is large or there is significant fluctuation, it may indicate that there is a large inconsistency between the current data and the historical data; Perform anomaly detection on the difference data to identify whether there are any anomalies. Statistical methods or machine learning algorithms can be used for anomaly detection; Based on the results of difference analysis and anomaly detection, the consistency between current water level data and historical data is comprehensively evaluated; If the consistency is high and there are no obvious anomalies, the current data can be considered consistent with the historical data. If the consistency is low or there are anomalies, further analysis and processing are required.

[0027] The higher the consistency between the current water level data and the historical data, the more consistent the sensor output is over different time periods, indicating that the sensor has high measurement stability and accuracy, specifically: A sensor with high consistency usually means that it has good stability, that is, the sensor can maintain good performance in long-term operation and is not prone to abnormalities or failures.

[0028] Highly consistent data indicates that the sensor's measurement results are consistent with historical data, and the sensor's working state is relatively stable, so the risk of sensor failure is relatively low.

[0029] The output of sensors with high consistency has high credibility and can be used as a stable and reliable monitoring indicator, which helps to improve the reliability and data quality of the monitoring system and reduce data errors.

[0030] The monitoring system can use the consistency between current water level data and historical data as one of the early warning indicators. When the consistency decreases, it may indicate that the possibility of sensor failure has increased. Early warnings can be issued in time to reduce the impact of failures on water conservancy projects.

[0031] Highly consistent sensor data can be used as a reference for maintenance strategies, such as planning maintenance cycles in advance or adjusting maintenance frequency based on changes in data consistency to ensure long-term stable operation of the sensor.

[0032] In the fault diagnosis and early warning module, a comprehensive analysis is conducted on the overall performance stability of the sensor and the consistency between the current water level data and the historical data. By setting the fault diagnosis algorithm and early warning rules, the sensor faults found in the monitoring system are diagnosed and early warning notifications are issued in a timely manner. Specifically: Collect the overall performance stability data of the sensor and the consistency data between the current water level data and the historical data, and perform pre-processing operations such as data cleaning, missing value processing, feature selection, etc. for model training; The data is marked based on the overall performance stability data of the sensor in the historical data and the consistency data between the current water level data and the historical data. Data with low stability or inconsistent with historical data can be marked as abnormal and marked as 1, otherwise it is marked as normal and marked as 0; Use labeled data to train supervised learning models, such as support vector machines (SVMs), random forests, etc. The specific calculation steps for using support vector machines (SVMs) to train supervised learning models are as follows: Preprocess the data, including feature scaling, feature selection, etc. Support vector machine (SVM) is selected as the supervised learning model. SVM is suitable for binary classification problems and performs well when processing high-dimensional data; Use labeled data to train the SVM model; The goal of SVM is to find an optimal hyperplane that maximizes the interval between abnormal data and normal data; The optimization objective of SVM can be expressed as: , the constraints are: , where i=1,2,...,n, w is the normal vector of the hyperplane, b is the intercept, C is the penalty parameter, is the slack variable, It's a label. is the eigenvector; You can use methods such as cross-validation to select a suitable penalty parameter C, as well as a suitable kernel function and kernel function parameters (if a kernel function is used); Cross-validation can help evaluate the performance of the model under different parameters and select the optimal parameter combination; Use validation sets or cross-validation to evaluate the trained model and calculate performance indicators such as accuracy, recall, and F1-score.

[0033] Use the trained SVM model to predict the current data and obtain the anomaly probability or anomaly score. According to the prediction results and the set threshold, trigger the warning notification, which mainly includes: Divide historical data into two subsets with higher stability and lower stability; Use each subset to evaluate the model and calculate the performance indicators such as accuracy, recall, F1-score, etc. of the model on different subsets; Among them, the calculation accuracy is calculated by the following specific formula: = , where is the recall accuracy, er is the number of samples correctly predicted as abnormal, that is, the number of abnormal samples correctly predicted as abnormal by the model, rt is the number of samples correctly predicted as normal, that is, the number of normal samples correctly predicted as normal by the model, kt is the number of samples incorrectly predicted as abnormal, that is, the number of normal samples incorrectly predicted as abnormal by the model, lp is the number of samples incorrectly predicted as normal, that is, the number of abnormal samples incorrectly predicted as normal by the model; Calculate the recall rate. The specific calculation formula is: = , where is the recall rate; calculate the F1-score, the specific calculation expression is: , the value range of F1-score is between 0 and 1, and the closer it is to 1, the better the performance of the model; Use the trained model to predict the current data and obtain the abnormal probability p; Set a standard threshold θ to convert the abnormal probability of continuous values ​​into the classification results of marked abnormalities and marked normals; If the probability p is greater than the threshold θ, the data is marked as abnormal, and an early warning signal is issued to notify relevant personnel to handle and repair it to avoid the fault from causing serious impact on the water conservancy project; if the probability p is less than or equal to the threshold θ, the data is marked as normal, no early warning signal is issued, and no immediate processing is required.

[0034] It should be noted that the specific value of the threshold value θ is set by those skilled in the art according to specific sensor performance requirements and will not be elaborated here.

[0035] Among them, in the backup sensor switching module, a backup sensor is set and the switching logic is designed. When a sensor fails, it automatically switches to the backup sensor and enables it, which mainly includes: Set up a backup sensor in the system and make sure it has the same functions and measurement capabilities as the primary sensor. The backup sensor should be consistent with the primary sensor so that the switchover is seamless.

[0036] Design switching logic to monitor the status of the primary sensor and automatically switch to the backup sensor when needed. The switching logic should include fault detection, switching trigger conditions, and switching actions.

[0037] Regularly monitor the status of major sensors, including data quality, communication status, etc. Design fault detection algorithms to identify whether major sensors have faults, such as data anomalies, communication interruptions, etc.

[0038] Determine the conditions that trigger the switch, such as the main sensor data continuously exceeding the normal range, the communication interruption time exceeding the set threshold, etc. Based on the monitored fault conditions, determine whether the switch trigger conditions are met.

[0039] When the switching trigger conditions are met, the switching action is executed to switch the monitoring system to the backup sensor and enable it. The switching action should ensure a smooth transition to avoid data loss and monitoring interruption.

[0040] Continuously monitor the status of the backup sensor to ensure that it is working properly. If the backup sensor also fails, send an alarm notification in time for manual intervention and repair.

[0041] In this embodiment, by monitoring the equipment operation status of the water level sensor and collecting the output data of the sensor, the output data trend of the sensor is analyzed by using the time series analysis method, the overall performance stability of the sensor is evaluated, the currently collected water level data is compared and analyzed with the historical data, the degree of difference between the data is judged, the consistency between the current water level data and the historical data is evaluated, and the overall performance stability of the sensor and the consistency between the current water level data and the historical data are comprehensively analyzed. By setting the fault diagnosis algorithm and early warning rules, the sensor fault found in the monitoring system is diagnosed, and the early warning notification is issued in time. By setting the fault diagnosis algorithm and early warning rules, the sensor fault is timely identified and processed to reduce the impact of the fault on the water conservancy project; a backup sensor is set, and the switching logic is designed. When the sensor fails, it automatically switches to the backup sensor and enables it. It can effectively improve the reliability and robustness of the monitoring system, so that it can operate continuously and stably under various environments and working conditions, and ensure the safe operation and management of the water conservancy project.

[0042] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0043] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0044] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0045] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0046] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0047] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0048] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0049] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0050] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0051] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A computer monitoring system applied to water conservancy projects, characterized in that: It includes sensor monitoring module, water level data quality assessment module, fault diagnosis and early warning module and standby sensor switching module; Sensor monitoring module: monitors the equipment operation status of the water level sensor, collects the output data of the sensor, analyzes the output data trend of the sensor using the time series analysis method, and evaluates the overall performance stability of the sensor; The sensor monitoring module specifically includes: Stationarity calculation: ADF test is used to test the stationarity of time series data. The null hypothesis of ADF test is that the sequence has a unit root, that is, non-stationary. If the p-value is less than the set significance level, the null hypothesis is rejected and the sequence is considered to be stationary. Periodicity calculation: Use the autocorrelation function to analyze whether the data has periodicity, calculate the autocorrelation coefficient, and observe whether there is a significant periodic peak; Trend calculation: Use linear regression to test whether there is a trend in the data; perform linear fitting on time series data, observe the significance of the regression coefficient and the slope of the fitting curve, and determine whether there is a trend change; Seasonality calculation: Use decomposition methods to detect whether the data is affected by seasonality, perform seasonal decomposition on the data, and calculate seasonal indicators; Normalize the results of stability, periodicity, trend and seasonality, and comprehensively calculate the overall performance stability coefficient of the sensor through the calculation results of the normalized stability, periodicity, trend and seasonality; The overall performance stability coefficient of the sensor is compared with the standard threshold. If the overall performance stability coefficient of the sensor is greater than or equal to the standard threshold, a normal signal of the device is issued; if the overall performance stability coefficient of the sensor is less than the standard threshold, an abnormal signal of the device is issued; Water level data quality assessment module: compare and analyze the currently collected water level data with historical data, determine the degree of difference between the data, and evaluate the consistency between the current water level data and historical data; Fault diagnosis and early warning module: conducts a comprehensive analysis of the overall performance stability of the sensor and the consistency between the current water level data and the historical data. By setting the fault diagnosis algorithm and early warning rules, the sensor faults found in the monitoring system are diagnosed and early warning notifications are issued in a timely manner; Backup sensor switching module: Set up backup sensors and design switching logic. When a sensor fails, it automatically switches to the backup sensor and is enabled.

2. A computer monitoring system for water conservancy projects according to claim 1, characterized in that: In the water level data quality assessment module, the consistency between the current water level data and the historical data is assessed, including the following contents; Collect the currently collected water level data and historical water level data, organize them into a time series format, and align the currently collected water level data with the historical data; The aligned data is used for difference calculation, which directly calculates the difference between the current data and the historical data, that is, , where For the current data, For historical data; Calculate the ratio of the difference between current data and historical data to historical data, that is, ; Draw a time series diagram of the difference data changes over time, calculate the statistical indicators of the difference data; draw a histogram of the difference data to show the distribution of the data; Observe the distribution and peak shape of the data through histograms and kernel density plots to determine whether there is skewness and peak phenomenon; Based on the results of difference analysis and anomaly detection, the consistency between current water level data and historical data is comprehensively evaluated; If the consistency is high and there are no obvious anomalies, the current data is considered to be consistent with the historical data; if the consistency is low and there are anomalies, the current data is considered to be inconsistent with the historical data.

3. A computer monitoring system for water conservancy projects according to claim 2, characterized in that: In the fault diagnosis and early warning module, the overall performance stability of the sensor and the consistency between the current water level data and the historical data are comprehensively analyzed. The specific calculation steps of using the support vector machine for supervised learning model training include: selecting the support vector machine as the supervised learning model and using the labeled data to train the SVM model; The goal of SVM is to find an optimal hyperplane that maximizes the interval between abnormal data and normal data; The optimization objective of SVM is expressed as: , the constraints are: ,in, , w is the normal vector of the hyperplane, b is the intercept, C is the penalty parameter, is the slack variable, It's a label. is the eigenvector.

4. A computer monitoring system for water conservancy projects according to claim 3, characterized in that: Use cross-validation to evaluate the trained model, and use the trained SVM model to predict the current data to obtain the anomaly probability; Based on the prediction results and the set thresholds, early warning notifications are triggered, including: Divide historical data into two subsets: high stability and low stability; Use each subset to evaluate the model and calculate the accuracy, recall, and F1-score of the model on different subsets; Use the trained model to predict the current data and obtain the abnormal probability p; Set a standard threshold θ to convert the abnormal probability of continuous values ​​into the classification results of marked abnormalities and marked normals; If the probability p is greater than the threshold θ, the data is marked as abnormal and a warning signal is issued; if the probability p is less than or equal to the threshold θ, the data is marked as normal and no warning signal is issued.

Images