Hydrological condition monitoring fault early warning method and system based on artificial intelligence

Abstract

The present application relates to the technical field of fault alarm, in particular to a hydrological condition monitoring fault early warning method and system based on artificial intelligence, comprising: establishing a monitoring value linkage model between a calibration unit and each hydrological monitoring unit; obtaining a calibration monitoring value measured by the calibration unit, and calculating a predicted detection value of each hydrological monitoring unit based on the calibration monitoring value and the monitoring value linkage model; calculating a deviation value of the predicted detection value and a real-time monitoring parameter actually detected by each hydrological monitoring unit to obtain historical monitoring data of a target hydrological monitoring unit and a nearest same-type hydrological monitoring unit; obtaining data fluctuation characteristics based on the change degree of the historical monitoring data; comparing the data fluctuation characteristics of the current monitoring unit and the nearest same-type hydrological monitoring unit to determine whether the target monitoring unit has a fault, so that the fault condition of the monitoring unit can be more accurately determined.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault alarm, in particular to a hydrological condition monitoring fault early warning method and system based on artificial intelligence. BACKGROUND

[0002] Hydrological condition monitoring involves comprehensive observation and analysis of physical, chemical and biological properties of water bodies such as rivers, lakes, reservoirs and groundwater, and is an important means to understand water resource status, ensure flood control safety, promote agricultural irrigation, assist water conservancy planning and operation management, and protect the environment. This process includes determining water flow velocity and volume to calculate flow, assessing water quality (such as pollutant concentration, dissolved oxygen, pH value and temperature) to ensure compliance with environmental quality standards, monitoring water level changes using water level gauges to warn of flood risk, recording precipitation and its distribution to understand water resource replenishment dynamics in the watershed, measuring water surface evaporation rate to study regional water resource circulation, and focusing on water level and water quality conditions of groundwater resources.

[0003] The existing hydrological condition monitoring needs to set up multiple monitoring units in the monitoring area. Due to the complexity of the outdoor environment, there are influences of temperature, humidity, atmospheric pressure, animal interference, etc., so that the failure probability of the monitoring unit is high. The existing method of judging whether the monitoring unit is faulty by monitoring the extreme value of the data of a single monitoring unit is relatively single, and cannot accurately judge the case where the error of the monitoring unit becomes large but does not reach the extreme value, thus having monitoring risks. SUMMARY

[0004] The present application aims to provide a hydrological condition monitoring fault early warning method and system based on artificial intelligence, which can judge the error condition of the monitoring unit by combining reference data and historical data, so as to more accurately judge the fault condition of the monitoring unit.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides a hydrological condition monitoring fault early warning method based on artificial intelligence, comprising establishing a monitoring value linkage model between a calibration unit and each hydrological monitoring unit under the condition of a preset environmental parameter, the environmental parameter including temperature, humidity, altitude data, the calibration unit being arranged between the multiple hydrological monitoring units;

[0006] obtaining calibration monitoring values measured by the calibration unit every preset time interval in the monitoring environment, and calculating predicted detection values of each hydrological monitoring unit based on the calibration monitoring values and the monitoring value linkage model;

[0007] The deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit is calculated, whether the deviation value is within a predetermined target range is judged, when the deviation value is not within the target range, the historical monitoring data of the target hydrological monitoring unit and the hydrological monitoring unit of the same type closest to the target hydrological monitoring unit are obtained;

[0008] The first data fluctuation feature of the target monitoring unit and the second data fluctuation feature of the hydrological monitoring unit closest to the target monitoring unit are established based on the change degree of the historical monitoring data;

[0009] The first data fluctuation feature of the target monitoring unit and the second data fluctuation feature of the hydrological monitoring unit closest to the target monitoring unit are compared, and if the difference value exceeds a preset threshold, it is determined that the target monitoring unit has a fault;

[0010] The feature information of the hydrological monitoring unit with the fault is obtained and relevant personnel are notified;

[0011] The fault type of the hydrological monitoring unit is classified based on the feature information;

[0012] The replacement time period of the hydrological monitoring unit is matched according to the fault type.

[0013] The type of the hydrological monitoring unit includes a rain gauge, a flow meter and a water quality analyzer.

[0014] The specific steps of establishing the monitoring value linkage model between the calibration unit and each hydrological monitoring unit under the condition of the preset environmental parameter include:

[0015] The placement position of the calibration unit is calculated based on the position information of all hydrological monitoring units;

[0016] After the calibration unit is placed at the placement position, calibration data is collected every preset time period;

[0017] The monitoring data of all hydrological monitoring units with the same timestamp as the calibration data is obtained;

[0018] The relationship between the calibration data and the monitoring data is obtained by polynomial regression analysis, and the monitoring value linkage model is obtained.

[0019] The specific steps of obtaining the relationship between the calibration data and the monitoring data by polynomial regression analysis include:

[0020] Temperature, humidity and atmospheric pressure data are obtained and used as input features to predict the difference value rule between the calibration data and the monitoring data;

[0021] The data is divided into a training set and a test set;

[0022] A polynomial regression model is fitted on the training set using a linear regression algorithm;

[0023] The polynomial regression model is tested and adjusted by using the test set to obtain a monitoring value linkage model.

[0024] The specific step of collecting calibration data every preset time interval after the calibration unit is placed in the placement position comprises:

[0025] A calibration period is set;

[0026] The calibration data is collected after the calibration unit runs in the actual environment for a preset time interval.

[0027] The specific step of calculating the deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit, judging whether the deviation value is within the predetermined target range, and obtaining the historical monitoring data of the target hydrological monitoring unit and the hydrological monitoring unit of the same type closest to the target hydrological monitoring unit when the deviation value is not within the target range comprises:

[0028] The predicted detection value and the real-time monitoring parameter at the corresponding time point are obtained.

[0029] The abnormal values and missing values in the real-time monitoring parameter are removed, and the predicted detection value at the corresponding time point is removed.

[0030] The deviation value of the predicted detection value and the real-time monitoring parameter at the corresponding time point is calculated.

[0031] When the deviation value is not within the target range, the historical monitoring data of the target hydrological monitoring unit and the hydrological monitoring unit of the same type closest to the target hydrological monitoring unit are obtained.

[0032] The specific step of obtaining the feature information of the faulty hydrological monitoring unit and notifying the relevant personnel comprises:

[0033] The basic information of the faulty hydrological monitoring unit is extracted from the database, and the basic information comprises the device model, the installation position, the running time, and the maintenance history.

[0034] The notification information is generated based on the basic information.

[0035] The notification information is sent to the relevant processing personnel through multiple communication channels.

[0036] The specific step of classifying the fault type of the hydrological monitoring unit based on the feature information comprises:

[0037] The historical working state data from the hydrological monitoring unit is collected, including data under normal operating conditions and data under known fault conditions.

[0038] Extract fault classification features based on historical working state data to obtain a training data set, the fault classification features including temperature, humidity, flow rate, and pressure;

[0039] Train the support vector machine model using the training data set to obtain a discrimination model;

[0040] Obtain the fault type based on the discrimination model and the fault classification feature data of the currently detected hydrological monitoring unit.

[0041] In a second aspect, the present application also provides a hydrological condition monitoring fault early warning system, comprising a calibration unit, a plurality of hydrological monitoring units, a linkage model generation module, a detection value prediction module, a first deviation calculation module, a data feature calculation module, a second deviation calculation module, and a fault alarm module.

[0042] The linkage model generation module is configured to establish a monitoring value linkage model between the calibration unit and each hydrological monitoring unit under the condition of preset environmental parameters, the environmental parameters including temperature, humidity, and altitude data, and the calibration unit is arranged between the plurality of hydrological monitoring units.

[0043] The detection value prediction module is configured to obtain calibration monitoring values measured by the calibration unit every preset time period in a monitoring environment, and calculate predicted detection values of each hydrological monitoring unit based on the calibration monitoring values and the monitoring value linkage model.

[0044] The first deviation calculation module is configured to calculate deviation values of the predicted detection values and real-time monitoring parameters actually detected by each hydrological monitoring unit, judge whether the deviation values are within a predetermined target range, and when the deviation values are not within the target range, obtain historical monitoring data of the target hydrological monitoring unit and a nearest same-type hydrological monitoring unit.

[0045] The data feature calculation module is configured to establish a first data fluctuation feature of the target monitoring unit and a second data fluctuation feature of the nearest same-type hydrological monitoring unit based on the variation degree of the historical monitoring data.

[0046] The second deviation calculation module is configured to compare the first data fluctuation feature of the current monitoring unit with the second data fluctuation feature of the nearest same-type hydrological monitoring unit, and if the difference value exceeds a preset threshold value, determine that the target monitoring unit has a fault.

[0047] The fault alarm module is configured to obtain feature information of the hydrological monitoring unit with a fault and notify relevant personnel.

[0048] The artificial intelligence-based hydrological condition monitoring fault early warning method and system of the present application establishes a linkage model connecting the calibration unit and multiple distributed hydrological monitoring units under the set environmental conditions (such as specific temperature, humidity and altitude). This model allows the values of each monitoring point to be related to each other, thereby more accurately reflecting the hydrological changes in the region. The calibration unit is strategically placed between different hydrological monitoring units to ensure the consistency and accuracy of the entire monitoring network.

[0049] The calibration unit enters the actual monitoring environment for measurement at a preset time interval and records these calibration monitoring values. Subsequently, using the pre-established monitoring value linkage model, combined with the latest calibration monitoring values, the expected detection values of each hydrological monitoring unit are calculated. This process allows us to predict the theoretical readings of each monitoring point, providing a basis for subsequent deviation analysis.

[0050] The predicted detection values calculated above are compared with the actual monitoring data reported by the hydrological monitoring units in real time, and the deviation value between the two is calculated. If the deviation of a certain monitoring unit exceeds the pre-defined target range, it indicates that there may be an abnormal situation. At this time, the system will further analyze by obtaining the historical monitoring data of the problem monitoring unit and its adjacent same type monitoring units to establish their respective data fluctuation characteristics.

[0051] By comparing the first data fluctuation characteristics of the target monitoring unit with the second data fluctuation characteristics of its nearest neighbor same type unit, the difference between the two can be more carefully evaluated. If the difference exceeds the set threshold, it can be reasonably suspected that the target monitoring unit has failed. This step effectively improves the accuracy of fault judgment and avoids false positives due to data anomalies at a single time point.

[0052] Once it is confirmed that a monitoring unit has failed, the system automatically collects relevant feature information (such as location, model, fault type, etc.) of the unit and immediately notifies the relevant maintenance personnel so that they can respond quickly and take necessary repair measures. Thus, the error situation of the monitoring unit can be judged in combination with reference data and historical data to more accurately judge the fault condition of the monitoring unit, which not only helps to improve the stability of the hydrological monitoring system, but also greatly reduces the negative impact of faults and ensures the smooth development of water resource management work. BRIEF DESCRIPTION OF DRAWINGS

[0053] In order to make the technical solutions in the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced. Obviously, the accompanying drawings in the following description only only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0054] Figure 1 is a flow chart of a hydrological condition monitoring fault early warning method based on artificial intelligence.

[0055] Figure 2 is a flow chart of the present application for establishing a monitoring value linkage model between the calibration unit and each hydrological monitoring unit under the condition of preset environmental parameters.

[0056] Figure 3 is a flow chart of the present application for obtaining the relationship between calibration data and monitoring data by using polynomial regression analysis to obtain the monitoring value linkage model.

[0057] Figure 4 is a flow chart of the present application for collecting calibration data every preset time interval after placing the calibration unit in the placement position.

[0058] Figure 5 is a flow chart of the present application for calculating the deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit, judging whether the deviation value is within the predetermined target range, and when the deviation value is not within the target range, obtaining the historical monitoring data of the target hydrological monitoring unit and the closest hydrological monitoring unit of the same type.

[0059] Figure 6 is a flow chart of the present application for obtaining the characteristic information of the fault hydrological monitoring unit and notifying the relevant personnel.

[0060] Figure 7 is a flow chart of the present application for classifying the fault type of the hydrological monitoring unit based on the characteristic information.

[0061] Figure 8 is a flow chart of the present application for matching the replacement time period of the hydrological monitoring unit according to the fault type.

[0062] Figure 9 is a structural diagram of a hydrological condition monitoring fault early warning system of the present application.

[0063] Calibration unit 101, hydrological monitoring unit 102, linkage model generation module 103, detection value prediction module 104, first deviation calculation module 105, data characteristic calculation module 106, second deviation calculation module 107, fault alarm module 108. DETAILED DESCRIPTION

[0064] Embodiments of the present application are described below in detail with reference to the accompanying drawings, wherein like or similar elements are denoted by the same or similar reference signs throughout the drawings. The embodiments described below are exemplary and are intended to explain the present application, and are not to be understood as limiting the present application.

[0065] First embodiment

[0066] Referring to Figures 1-8 The present application provides a hydrological condition monitoring fault early warning method based on artificial intelligence, comprising:

[0067] S101, under the condition of a preset environmental parameter, a monitoring value linkage model between a calibration unit and each hydrological monitoring unit is established, the environmental parameter includes temperature, humidity, and altitude data, and the calibration unit is arranged between the plurality of hydrological monitoring units;

[0068] Under the condition of a preset environmental parameter (such as temperature, humidity, and altitude data), a monitoring value linkage model between a calibration unit and a plurality of hydrological monitoring units is established. These hydrological monitoring units include but are not limited to rain gauges, flowmeters, and water quality analyzers, which are distributed in different geographical locations for collecting various hydrological data. The calibration unit will be strategically placed between these monitoring units to ensure that it can represent the environmental conditions of most monitoring points.

[0069] The specific steps include:

[0070] S201, the placement position of the calibration unit is calculated based on the position information of all hydrological monitoring units;

[0071] The precise geographic coordinates (latitude and longitude) of all hydrological monitoring units are collected and converted into coordinates (x, y) in a plane rectangular coordinate system. If high-precision calculation is required, the influence of the curvature of the earth should be considered.

[0072] An initial guess position is selected as the starting position of the calibration unit. This can be the position of any hydrological monitoring unit or the average value of the coordinates of all hydrological monitoring units.

[0073] Weiszfeld's algorithm is used to find the optimal position. Specifically, set the initial estimate P_0P0 as the position of the calibration unit.

[0074] For each iteration k, update the position P k For the new estimate P k+1 , the formula is as follows:

[0075]

[0076] Here P i represents the position of the i-th hydrological monitoring unit,‖P k -P i ‖ represents the Euclidean distance from the current estimate P k to the monitoring unit P i .

[0077] Check for convergence. If‖P k+1 -P k ‖<∈, where ∈ is a small positive number preset, stop iteration; otherwise continue iteration.

[0078] Finally, verify whether the final position obtained is reasonable and ensure that it indeed minimizes the sum of distances to all hydrological monitoring units. Cross-validation can be performed by comparing the results obtained under different initialization conditions.

[0079] S202 collects calibration data every preset time interval after placing the calibration unit at the placement position;

[0080] The specific steps include:

[0081] S401 sets a calibration period;

[0082] According to the characteristics of the equipment and the monitoring needs, set a reasonable calibration period (e.g., daily, weekly, or monthly). This period should be able to capture changes in a timely manner without increasing the workload too frequently.

[0083] S402 collects calibration data after the calibration unit runs in the actual environment for a preset time period.

[0084] After the calibration unit enters the actual monitoring environment, give it enough time to adapt to the new environmental conditions and reduce errors caused by sudden changes in temperature, humidity, and other factors.

[0085] Ensure that the timestamps of the calibration unit and all other hydrological monitoring units are synchronized to facilitate subsequent data comparison and analysis.

[0086] Start the data collection process of the calibration unit and all other hydrological monitoring units within the same time period to ensure that all devices are measuring under the same conditions to obtain comparable data sets.

[0087] S203 obtains monitoring data of all hydrological monitoring units with the same timestamp as the calibration data;

[0088] In this stage, real-time monitoring parameters with the same timestamp as the calibration data will be obtained from all hydrological monitoring units. This step is crucial because it ensures that we are comparing data at the same time, thereby improving the accuracy of the analysis results.

[0089] S204 obtains the relationship between the calibration data and the monitoring data using polynomial regression analysis to obtain a monitoring value linkage model.

[0090] The specific steps include:

[0091] S301 obtains temperature, humidity, and atmospheric pressure data as input features for predicting the difference pattern between the calibration data and the monitoring data.

[0092] The environmental condition information (such as temperature, humidity, and atmospheric pressure) around each monitoring unit is collected as input features for predicting the difference pattern between the calibration data and the monitoring data.

[0093] S302 divides the data into a training set and a test set.

[0094] The data set is divided into a training set and a test set in a ratio of 70%-80%, the former is used for training the model, and the latter is used for evaluating the model performance. Ensure that the training set can fully represent the entire data distribution, and the test set is large enough to accurately reflect the model generalization ability.

[0095] S303 uses a linear regression algorithm to fit a polynomial regression model on the training set.

[0096] Select the order of the quadratic polynomial and fit the polynomial regression model on the training set by least squares method or other methods. The goal of this step is to find a set of optimal parameters that minimize the error between the predicted output and the actual observed values.

[0097] S304 uses the test set to test and adjust the polynomial regression model to obtain a monitoring value linkage model.

[0098] The fitted polynomial regression model is verified using the test set to check its prediction accuracy and stability. If the model performs poorly, it needs to return to the previous steps to adjust the feature selection, algorithm selection, or hyperparameter settings until the most suitable configuration is found. To improve the model generalization ability, the K-fold cross-validation method can be used to repeatedly train and test the model on different subsets, and finally take the average score as the performance indicator of the model.

[0099] S102 obtains calibration monitoring values measured by the calibration unit every preset time period in the monitoring environment, and calculates the predicted detection values of each hydrological monitoring unit based on the calibration monitoring values and the monitoring value linkage model.

[0100] According to the device characteristics and monitoring requirements, a reasonable calibration period (such as daily, weekly, or monthly) is set. This period should be able to capture potential changes in a timely manner without increasing unnecessary workload too frequently.

[0101] According to the predetermined schedule, move the calibration unit to the actual monitoring environment. This step involves physically transporting or remotely adjusting its location. Allow the calibration unit to run for a period of time (e.g., several minutes to half an hour) in the new environment to adapt to the new environmental conditions and reduce errors caused by sudden changes in temperature, humidity, etc. Ensure that the timestamps of the calibration unit and all other hydrological monitoring units are synchronized to facilitate subsequent data comparison and analysis.

[0102] Start the data collection process for the calibration unit and all other hydrological monitoring units within the same time period, ensuring that all devices measure under the same conditions to obtain comparable data sets. Record all data obtained during this calibration process, including timestamps, geographic locations, measurement results, etc., to provide a basis for subsequent analysis. Call the previously established monitoring value linkage model, which has been trained and can accurately predict the behavior of other hydrological monitoring units. Input the data obtained in the current calibration into the linkage model to directly calculate the predicted values.

[0103] S103 Calculate the deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit, and determine whether the deviation value is within the predetermined target range. When the deviation value is not within the target range, obtain the historical monitoring data of the target hydrological monitoring unit and the closest hydrological monitoring unit of the same type;

[0104] The specific steps include:

[0105] S501 Obtain the predicted detection value and the real-time monitoring parameter at the corresponding time point;

[0106] Ensure that the calibration unit and all other hydrological monitoring units collect data within the same time period to obtain data sets with the same timestamps. Extract the predicted detection value and the corresponding real-time monitoring parameter of each hydrological monitoring unit at a specific time point from the database. This includes but is not limited to temperature, humidity, precipitation, flow rate, etc. Key indicators.

[0107] S502 Remove outliers and missing values in the real-time monitoring parameter, and remove the predicted detection value at the corresponding time point;

[0108] Use statistical methods (such as Z-score, IQR) to identify and mark outliers. For obviously incorrect data points, they should be deleted. For missing data points, interpolation methods (such as linear interpolation, spline interpolation) or average values based on adjacent data points can be used to fill them. At the same time, remove the predicted detection values of those time points that cannot be reliably filled.

[0109] S503 Calculate the deviation value of the predicted detection value and the real-time monitoring parameter at the corresponding time point;

[0110] The percentage error is selected to calculate the deviation value for each hydrological monitoring unit, and the deviation value between the predicted detection value and the actual detection value is calculated in batches. Save these deviation values and their corresponding time stamps for subsequent analysis. The relationship between the predicted value and the actual value can be visually displayed using charts such as line graphs, scatter plots, heat maps, etc., to help more clearly understand the similarities and differences between the two.

[0111] S504 When the deviation value is not within the target range, obtain the historical monitoring data of the target hydrological monitoring unit and the nearest hydrological monitoring unit of the same type.

[0112] According to historical data, domain knowledge or expert advice, set a reasonable "normal" range (e.g. ±5%) for each monitoring parameter. This range can distinguish between normal fluctuations and abnormal situations. For each hydrological monitoring unit, compare its deviation value with the pre-set target range one by one. When the deviation value of a certain monitoring unit is found to be outside the pre-set range, immediately mark the unit as "potential failure" and record the specific deviation degree and time point. Through geographic information system (GIS) tools, calculate the spatial distance between the target hydrological monitoring unit and other monitoring units of the same type, and find the nearest one. Extract all monitoring records of the two units in the past period (e.g. the past week or month) from the database. Ensure that the data covers measurement values under different seasons, weather conditions and working environments.

[0113] S104 Establishing a first data fluctuation feature of the target monitoring unit based on the variation degree of the historical monitoring data and a second data fluctuation feature of the nearest hydrological monitoring unit of the same type;

[0114] Extract all monitoring records of the target hydrological monitoring unit and the nearest hydrological monitoring unit of the same type in the past period (e.g. the past week or month) from the database. Ensure that the data covers measurement values under different seasons, weather conditions and working environments.

[0115] Calculate the data variation (such as difference) in adjacent time periods to capture the trend of data change over time. Calculate the moving average line (such as 7 days, 30 days) of different window sizes to smooth short-term fluctuations and highlight long-term trends. Use linear regression or other trend analysis methods to assess the trend of data change over time and identify patterns of rising, falling or stability.

[0116] According to the above analysis results, define a series of key features that can reflect the data fluctuation characteristics of the target monitoring unit. This includes but is not limited to average change rate, fluctuation amplitude, cycle length, trend slope, etc.

[0117] Quantify each feature into a specific numerical value to form a feature vector. This feature vector will be used for subsequent comparative analysis.

[0118] Repeat the above process: perform the same process on the nearest hydrological monitoring unit of the same type, extract its second data fluctuation feature.

[0119] S105 compares the first data fluctuation feature of the current monitoring unit with the second data fluctuation feature of the nearest hydrological monitoring unit of the same type. If the difference value exceeds the preset threshold, it is determined that the target monitoring unit has failed;

[0120] In this step, we will compare the first data fluctuation feature of the current monitoring unit (target monitoring unit) with the second data fluctuation feature of the nearest hydrological monitoring unit of the same type in detail. By quantifying the difference between the two and comparing it with the preset threshold, we can effectively identify whether the target monitoring unit has failed. The following are the detailed implementation steps:

[0121] According to the actual situation, select the cosine similarity index to quantify the difference between the two feature vectors. For each feature dimension, batch calculate the difference value between the target monitoring unit and the adjacent same type unit. Save these difference values and their corresponding time stamps for subsequent analysis.

[0122] According to historical data, determine a reasonable preset threshold. This threshold should be able to distinguish between normal fluctuations and abnormal situations, but not be too sensitive to cause frequent false alarms.

[0123] Compare each time point's difference value with the preset threshold one by one. If the difference value of a certain time point exceeds the preset threshold, immediately mark that time point as the moment of "potential failure" of the target monitoring unit, and record the specific deviation degree and time point.

[0124] For the time points marked as "potential failure", further comprehensive evaluation of the data performance in a period of time before and after it, to exclude the influence of accidental and short-term fluctuations. If multiple consecutive time points exceed the threshold, it is more inclined to believe that there is indeed a failure.

[0125] S106 obtains the feature information of the failed hydrological monitoring unit and notifies the relevant personnel.

[0126] The specific steps include:

[0127] S601 extracts the basic information of the failed hydrological monitoring unit from the database, including device model, installation location, running time, maintenance history;

[0128] Device Model Records the specific model and version number of the hydrological monitoring unit, which is crucial for assessing device performance and selecting appropriate maintenance methods. Installation Location Obtains the exact geographical location (longitude, latitude) of the unit and a physical environment description (such as river name, monitoring station number) where it is located. This helps field engineers quickly locate the device.

[0129] Running Time Records the length of time the unit has been in operation since installation, including total running time and time since the last restart. Long periods of continuous operation can cause the device to age or wear out. Organize the unit's historical maintenance records, including the last calibration date, replacement parts list, fault repair log, etc. These information can help analyze the cause of the current fault and provide reference for future preventive maintenance.

[0130] S602 generates notification information based on basic information;

[0131] Design a standardized notification template, including title, body, attachments, etc. Ensure that the notification content is concise and clear while covering all key information. According to the information extracted from the database, automatically fill in the variable fields in the notification template, such as device model, installation location, running time, maintenance history, etc. If necessary, additional explanations or suggestions can be added to the notification, such as preliminary diagnosis results, tool and spare parts list to carry, etc.

[0132] S603 sends notification information to relevant processing personnel through multiple communication channels.

[0133] For emergency situations, send a brief notification message to designated contacts through the SMS platform to ensure that attention is drawn at the first time.

[0134] Use the company's internal email system or external services to send formal notification emails, accompanied by detailed fault reports and support files. This method is suitable for conveying more complex information and documents.

[0135] Use enterprise-level instant messaging software (such as WeChat, DingTalk, Slack, etc.) to create a dedicated fault response group to share the latest progress and discuss solutions in real time.

[0136] For particularly serious situations, directly call the responsible person's phone for oral notification and confirm that the other party has received and understood the relevant information.

[0137] If the organization has a dedicated mobile application or management platform, you can send alerts to relevant personnel through the application's push notification function, reminding them to check the latest fault information.

[0138] S107 classifies the fault type of the hydrological monitoring unit based on feature information.

[0139] The specific steps include:

[0140] S701 Collect historical working state data from the hydrological monitoring unit, including data under normal operating conditions and data under known fault conditions;

[0141] A comprehensive set of historical working state data is established, which includes data records of the hydrological monitoring unit under normal operating conditions and under different known fault conditions. These data come from multiple dimensions, ensuring the diversity and representativeness of the samples. Specifically, we will collect the following types of data:

[0142] S702 Extract fault classification features based on historical working state data to obtain a training data set, the fault classification features including temperature, humidity, flow rate, and pressure;

[0143] Next, we extract feature values that can effectively represent fault types from the collected historical data. The "fault classification features" mentioned here refer to key indicators that best reflect the characteristics of faults, such as temperature, humidity, flow rate, and pressure. Determine which features are the most important through statistical analysis, signal processing techniques, or the experience of experts in the field, and use them as input variables for subsequent model training.

[0144] We combine the extracted features into a feature vector, each vector representing a snapshot of the hydrological monitoring unit state at a specific time. Then we label these feature vectors according to whether they correspond to fault states (such as 0 for normal and 1 for a certain fault), forming our training data set.

[0145] S703 Train the support vector machine model using the training data set to obtain a discrimination model;

[0146] With the prepared training data set, we can begin to train the support vector machine model. SVM is a supervised learning method that is good at handling classification problems in high-dimensional space. It finds an optimal hyperplane that maximizes the separation between two classes of data to distinguish different fault categories. During training, model parameters are adjusted to optimize classification results until a satisfactory accuracy is achieved.

[0147] S704 Obtain the fault type based on the discrimination model and the fault classification feature data of the currently detected hydrological monitoring unit.

[0148] After completing model training, we obtain a well-validated discrimination model. When a new hydrological monitoring unit appears with suspected faults, we can collect its current feature data in real time and input it into this discrimination model for prediction. The model will output the most likely fault type, helping maintenance personnel quickly locate the problem and take appropriate maintenance measures.

[0149] S108 matches the replacement time period of the hydrological monitoring unit according to the fault type.

[0150] The specific steps include:

[0151] S801 Obtain the life prediction data provided by the equipment manufacturer to evaluate the expected life of each hydrological monitoring unit;

[0152] First, obtain the official life prediction data of each hydrological monitoring unit from the equipment manufacturer. These data are usually based on laboratory tests and long-term field use experience, including but not limited to mean time between failures (MTBF), maximum service life, and performance degradation rate under different environmental conditions.

[0153] S802 Set the replacement schedule based on the expected life and fault type;

[0154] All known faults are divided into predictable faults (such as natural wear and tear, aging) and sudden faults (such as natural disasters, accidental damage). For each fault mode, analyze its frequency of occurrence, impact range, and possible consequences. For predictable faults, determine a reasonable preventive replacement cycle based on the expected life evaluation results. This cycle should take into account the safety margin to ensure that replacement is completed before the expected failure occurs, thereby avoiding service interruption. Generate a schedule that can be updated in real time based on the latest evaluation results. For example, if it is found that the actual life of a certain type of equipment is longer or shorter than expected, the corresponding schedule should be adjusted in a timely manner.

[0155] Where the maintenance time interval of the sensor can be calculated using the following formula:

[0156] R sensor = k * T

[0157] R sensor is the sensor maintenance or replacement time interval (hours); k is the proportional coefficient, which can be determined according to the sensor type and environmental conditions, for example, for water quality sensors, k = 0.05.

[0158] Assuming a water quality sensor has been used for 3000 hours and the proportional coefficient k = 0.05: then

[0159] R sensor = 0.05.3000 = 150 hours, so cleaning or replacement is recommended within 150 hours.

[0160] If it is a circuit failure, the following formula can be used to calculate:

[0161]

[0162] Wherein, MTBF is the mean time between failures (hours); S is the safety time window; τ is the time constant. T WF F is the rate of entering metastable state. D P is the probability of failure detection.

[0163] Assuming the parameters of the circuit are S = 10, T = 5, T WF = 0.01, F D = 0.1, the calculated result is that it is recommended to perform a circuit inspection every 7389 hours.

[0164] If it is a communication failure, the following formula can be used to calculate the maintenance time interval:

[0165]

[0166] Wherein, R comm is the maintenance time interval of the communication equipment (hours); λ is the failure rate (unit: times / hour). Assuming the failure rate of the communication module is λ = 0.001 times / hour, it is recommended to perform a communication module inspection every 1000 hours.

[0167] S803 obtains the current fault type and usage time, and matches the corresponding replacement time period.

[0168] When a fault alarm occurs, a preliminary diagnosis is quickly performed to determine the specific type of the fault. If it is a sudden failure, an emergency response process is immediately started; if it is a predictable failure, it is further checked whether it is close to the preset replacement time point. Then, according to the current fault type and the cumulative usage time of the equipment, the pre-set replacement time table is searched to find the most suitable replacement time period. For the equipment that is about to expire, the replacement task is planned and arranged in advance; for the equipment that has exceeded the safety margin, the replacement should be arranged as soon as possible.

[0169] Second embodiment

[0170] Please refer to Figure 9The application also provides a hydrological condition monitoring fault early warning system, comprising a calibration unit 101, a plurality of hydrological monitoring units 102, a linkage model generation module 103, a detection value prediction module 104, a first deviation calculation module 105, a data feature calculation module 106, a second deviation calculation module 107 and a fault alarm module 108; the linkage model generation module 103 is used for establishing a monitoring value linkage model between the calibration unit 101 and each hydrological monitoring unit 102 under the condition of preset environmental parameters, the environmental parameters include temperature, humidity and altitude data, and the calibration unit 101 is arranged between the plurality of hydrological monitoring units 102; the detection value prediction module 104 is used for obtaining calibration monitoring values measured by the calibration unit 101 every preset time period in a monitoring environment, and calculating predicted detection values of each hydrological monitoring unit 102 based on the calibration monitoring values and the monitoring value linkage model; the first deviation calculation module 105 is used for calculating deviation values of the predicted detection values and real-time monitoring parameters actually detected by each hydrological monitoring unit 102, judging whether the deviation values are within a predetermined target range, when the deviation values are not within the target range, obtaining historical monitoring data of a target hydrological monitoring unit 102 and a hydrological monitoring unit 102 of the same type closest to the target hydrological monitoring unit 102; the data feature calculation module 106 is used for establishing a first data fluctuation feature of the target monitoring unit and a second data fluctuation feature of the hydrological monitoring unit 102 of the same type closest to the target monitoring unit based on a change degree of the historical monitoring data; the second deviation calculation module 107 is used for comparing the first data fluctuation feature of the current monitoring unit with the second data fluctuation feature of the hydrological monitoring unit 102 of the same type closest to the target monitoring unit, and if a difference value exceeds a preset threshold value, it is determined that the target monitoring unit has a fault; and the fault alarm module 108 is used for obtaining feature information of the hydrological monitoring unit 102 with the fault and notifying relevant personnel.

[0171] In the embodiment, the calibration unit 101 is arranged between the plurality of hydrological monitoring units 102, and mainly provides a reference standard to ensure the data accuracy of all monitoring devices. It will enter the monitoring environment for measurement regularly, and adjust other hydrological monitoring units 102 according to the obtained results.

[0172] The hydrological monitoring units 102 are distributed in different geographical positions to collect important hydrological parameters such as water flow velocity, water level height and water quality composition in real time.

[0173] Considering that factors such as temperature, humidity and altitude in the natural environment will affect the hydrological data, the linkage model generation module 103 constructs a mathematical model capable of reflecting the relationship between these factors and the readings of each monitoring point. In this way, even under changing external conditions, the consistency of the predicted value and the actual observation value can be maintained.

[0174] The detection value prediction module 104 can predict the value of each hydrological monitoring point at a future time based on the previously established linkage model and the fresh correction information from the calibration unit 101. This step is crucial for early detection of potential problems.

[0175] The first deviation calculation module 105 compares the predicted value with the actual readings obtained from each hydrological monitoring unit 102 to calculate the difference between the two. If the error at a certain point exceeds the set safety limit, it means there is a risk of failure, which needs further investigation.

[0176] The data feature calculation module 106 obtains the behavior patterns of the target monitoring point and similar types of sites near it over a period of time to find unique trends or rules indicating faults after a preliminary judgment of abnormality.

[0177] On this basis, the second deviation calculation module 107 compares the behavior characteristics of the above two sites. If it finds that the difference between them deviates significantly from the normal range, it means that the target monitoring point has a hardware failure or software error, and immediate measures need to be taken.

[0178] Finally, once it is confirmed that a hydrological monitoring unit 102 has a problem, the fault alarm module 108 will automatically trigger the alarm mechanism to quickly inform the relevant personnel about the specific location of the fault and other related information, so that they can intervene as soon as possible to prevent the situation from worsening.

[0179] This system not only improves the efficiency and accuracy of hydrological monitoring, but also reduces the need for manual intervention through intelligent means, making water resource management and disaster prevention more scientific and reasonable.

[0180] The above disclosure is only a preferred embodiment of the present application, and of course cannot limit the scope of the invention. Those skilled in the art can understand that the implementation of all or part of the above-mentioned embodiments and the equivalent changes made according to the claims of the present application still belong to the scope covered by the invention.

Claims

1. A method for monitoring and early warning of hydrological condition faults based on artificial intelligence, Characterized in that It comprises: under the condition of preset environmental parameters, the monitoring value linkage model between the calibration unit and each hydrological monitoring unit is established, the environmental parameters include temperature, humidity, altitude data, and the calibration unit is arranged between the plurality of hydrological monitoring units; The calibration monitoring value measured by the calibration unit every preset time interval entering the monitoring environment is obtained, and the predicted detection value of each hydrological monitoring unit is calculated based on the calibration monitoring value and the monitoring value linkage model; The deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit is calculated, whether the deviation value is within the predetermined target range is judged, when the deviation value is not within the target range, the historical monitoring data of the target hydrological monitoring unit and the nearest same type hydrological monitoring unit is obtained; The first data fluctuation feature of the target monitoring unit and the second data fluctuation feature of the nearest same type hydrological monitoring unit are established based on the change degree of the historical monitoring data; The first data fluctuation feature of the current monitoring unit and the second data fluctuation feature of the nearest same type hydrological monitoring unit are compared, if the difference value exceeds the preset threshold value, it is judged that the target monitoring unit fails; The feature information of the failed hydrological monitoring unit is obtained and the relevant personnel is informed; The fault type of the hydrological monitoring unit is classified based on the feature information; The replacement time period of the hydrological monitoring unit is matched according to the fault type; The specific steps of establishing the monitoring value linkage model between the calibration unit and each hydrological monitoring unit under the condition of preset environmental parameters include: The placement position of the calibration unit is calculated based on the position information of all hydrological monitoring units; After placing the calibration unit at the placement position, calibration data is collected every preset time interval; The monitoring data of all hydrological monitoring units with the same timestamp as the calibration data is obtained; The relationship between the calibration data and the monitoring data is obtained by polynomial regression analysis to obtain the monitoring value linkage model; The specific steps of obtaining the relationship between the calibration data and the monitoring data by polynomial regression analysis to obtain the monitoring value linkage model include: Temperature, humidity, atmospheric pressure data are obtained and used as input features to predict the difference value rule of calibration data and monitoring data; The data is divided into training set and test set; A polynomial regression model is fitted on the training set using linear regression algorithm; The test set is used to test and adjust the polynomial regression model to obtain the monitoring value linkage model.

2. The hydrological condition monitoring fault early warning method based on artificial intelligence according to claim 1, characterized in that The types of the hydrological monitoring units include rain gauge, flow meter and water quality analyzer.

3. The hydrological condition monitoring fault early warning method based on artificial intelligence according to claim 2, characterized in that The specific steps of collecting calibration data every preset time interval after placing the calibration unit at the placement position include: A calibration period is set; The calibration unit is allowed to run in the actual environment for a preset time interval before collecting calibration data.

4. The hydrological condition monitoring fault early warning method based on artificial intelligence according to claim 3, characterized in that, The deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit is calculated, it is judged whether the deviation value is in the target range, when the deviation value is not in the target range, the specific steps of obtaining the historical monitoring data of the target hydrological monitoring unit and the closest same type hydrological monitoring unit include: Obtain the predicted detection value and the real-time monitoring parameter at the corresponding time point; Remove the abnormal value and the missing value in the real-time monitoring parameter, and remove the predicted detection value at the corresponding time point; Calculate the deviation value of the predicted detection value and the real-time monitoring parameter at the corresponding time point; When the deviation value is not in the target range, obtain the historical monitoring data of the target hydrological monitoring unit and the closest same type hydrological monitoring unit.

5. The hydrological condition monitoring fault early warning method based on artificial intelligence according to claim 4, wherein the specific steps of obtaining the feature information of the fault hydrological monitoring unit and notifying the relevant personnel include: Extract the basic information of the fault hydrological monitoring unit from the database, and the basic information includes the device model, the installation position, the running time, and the maintenance history; Generate notification information based on the basic information; Send the notification information to the relevant processing personnel through multiple communication channels.

6. The hydrological condition monitoring fault early warning method based on artificial intelligence according to claim 5, wherein the specific steps of classifying the fault type of the hydrological monitoring unit based on the feature information include: Collect historical working state data from the hydrological monitoring unit, including data under normal operating conditions and data under known fault conditions; Extract fault classification features based on the historical working state data to obtain a training data set, and the fault classification features include temperature, humidity, flow rate, and pressure; Train a support vector machine model using the training data set to obtain a discrimination model; Obtain the fault type based on the discrimination model and the fault classification feature data of the currently detected hydrological monitoring unit. It includes a calibration unit, a plurality of hydrological monitoring units, a linkage model generation module, a detection value prediction module, a first deviation calculation module, a data feature calculation module, a second deviation calculation module, and a fault alarm module. The linkage model generation module is used to establish a monitoring value linkage model between the calibration unit and each hydrological monitoring unit under the condition of preset environmental parameters, the environmental parameters include temperature, humidity, and altitude data, and the calibration unit is arranged between the plurality of hydrological monitoring units.

7. A hydrological condition monitoring fault early warning system applied to the hydrological condition monitoring fault early warning method based on artificial intelligence in claims 1-6, characterized in that, The detection value prediction module is used to obtain calibration monitoring values measured by the calibration unit every preset time period in a monitoring environment, and calculate predicted detection values of each hydrological monitoring unit based on the calibration monitoring values and the monitoring value linkage model; The first deviation calculation module is used to calculate the deviation value of the predicted detection value and the real-time monitoring parameter actually detected by each hydrological monitoring unit, judge whether the deviation value is in the target range, when the deviation value is not in the target range, obtain the historical monitoring data of the target hydrological monitoring unit and the closest same type hydrological monitoring unit. ​ ​ ​ The data feature calculation module is configured to establish a first data fluctuation feature of the target monitoring unit and a second data fluctuation feature of a hydrological monitoring unit of the same type closest to the target monitoring unit based on a variation degree of historical monitoring data; The second deviation calculation module is configured to compare the first data fluctuation feature of the current monitoring unit with the second data fluctuation feature of the hydrological monitoring unit of the same type closest to the target monitoring unit, and determine that the target monitoring unit is faulty if a difference value exceeds a preset threshold value; The fault alarm module is configured to acquire feature information of the faulty hydrological monitoring unit and notify relevant personnel.

Images