Water conservancy data acquisition device

Abstract

The invention discloses a water conservancy data acquisition device, and relates to the technical field of Internet of Things, and the device comprises the following steps: dividing a target water area into a plurality of monitoring water areas; performing characteristic factor analysis on each monitored water area to obtain a characteristic factor analysis result; adjusting the installation number and positions of the sensors in the monitored water area according to the characteristic factor analysis result, and obtaining hydrological data monitored in real time; performing edge calculation analysis on the hydrological data monitored in real time to obtain an edge analysis result; performing data synchronization on the edge analysis result and uploading the edge analysis result to a cloud platform to obtain cloud data; performing intelligent analysis on the cloud data to obtain a water area dynamic report; performing risk early warning according to the water area dynamic report; the acquisition device collects and analyzes hydrological data of different water areas by combining real-time data acquisition of a sensor and an edge calculation and analysis method. According to the method, loss and errors of hydrological data are reduced, and consistency, high efficiency, accuracy and timeliness of the hydrological data are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of the Internet of Things, and particularly to a water conservancy data acquisition device. Background Art

[0002] With the increasingly strict requirements for water resources management and environmental protection, water quality monitoring and hydrological data acquisition have become particularly important. Currently, water conservancy data acquisition devices mainly rely on a variety of sensors and monitoring equipment to monitor the water quality, hydrology, climate, etc. of water areas in real time. These data provide decision-making support for water area management and ecological protection after being collected, transmitted, and processed. However, although devices such as sensors and current meters in the prior art can collect a large amount of water quality and hydrological data, due to the accuracy differences of different devices and the non-uniform data formats, the data integration process is still cumbersome and prone to data loss or deviation. Therefore, in the data preprocessing process, the prior art is difficult to ensure the consistency and efficiency of data, affecting the accuracy of subsequent analysis. In the prior art, the analysis of water machine data relies too much on traditional analysis methods, and the ability to detect complex water area change patterns is weak. Especially when dealing with high-dimensional data and large-scale data sets, there are often cases of missed reports or false reports, and potential abnormal events cannot be discovered in a timely and effective manner. Summary of the Invention

[0003] Aiming at the deficiencies of the prior art, the present invention provides a water conservancy data acquisition device to solve the problems raised in the above background art.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] In a first aspect, an embodiment of the present invention provides a water conservancy data acquisition device. When in use, the water conservancy data acquisition device includes the following steps:

[0006] S1. Divide the target water area into multiple monitoring water areas;

[0007] Delimit different monitoring water areas through remote sensing technology, geographic information system GIS, and historical hydrological data;

[0008] S2. Conduct characteristic factor analysis on each monitoring water area to obtain a characteristic factor analysis result;

[0009] According to the monitoring water areas, install sensors in each monitoring water area, and conduct characteristic factor analysis on the hydrological environment in each monitoring water area based on the acquisition results of the sensors to obtain a characteristic factor analysis result;

[0010] S3. Adjust the installation quantity and position of sensors in the monitoring water area according to the characteristic factor analysis result, and obtain the real-time monitored hydrological data;

[0011] According to the analysis results of characteristic factors of each monitored water area, combined with the actual monitoring requirements and hydrological data quality requirements, adjust the current sensor installation positions to obtain real-time monitored hydrological data;

[0012] S4. Conduct edge computing analysis on the real-time monitored hydrological data to obtain edge analysis results;

[0013] Conduct preliminary analysis on the real-time monitored hydrological data through edge computing technology to obtain edge analysis results;

[0014] S5. Synchronize the edge analysis results and upload them to the cloud platform to obtain cloud data;

[0015] Transmit the edge analysis results to the cloud platform through a communication network, store and manage the edge analysis results on the cloud platform to obtain cloud data;

[0016] S6. Conduct intelligent analysis on the cloud data to obtain a water area dynamic report;

[0017] Conduct intelligent analysis on the cloud data through artificial intelligence algorithms to obtain a water area dynamic report;

[0018] S7. Conduct risk early warning according to the water area dynamic report;

[0019] According to the water area dynamic report, combined with the hydrological changes and water quality analysis results of the water area, predict potential risks and abnormal phenomena. After processing these risks and abnormal phenomena through anomaly detection algorithms, conduct risk early warning for each monitored water area.

[0020] Further optimize this technical solution. The characteristic factors in step S2 include water quality factors, water temperature factors, water flow velocity and flow factors, precipitation factors, pollution source factors, and hydrological slope factors.

[0021] Further optimize this technical solution. The steps of edge computing analysis in step S4 include:

[0022] Use the edge calculation formula model for calculation:

[0023] This model is based on the idea of multi-dimensional data fusion and threshold judgment, and conducts edge computing analysis by comprehensively considering water quality factors, water temperature factors, water flow velocity and flow factors, precipitation factors, pollution source factors, and hydrological slope factors; model formula:

[0024] ;

[0025] Among them,

[0026] is a weighting function;

[0027] is the score of water quality factors, which are converted into scores through a standardization formula;

[0028] is the score of water flow velocity and flow rate factors, which are converted into scores through a standardization formula;

[0029] is the score of precipitation factors, which are converted into scores through a standardization formula;

[0030] is the score of water temperature factors, which are converted into scores through a standardization formula;

[0031] is the score of pollution source factors, which are converted into scores through a standardization formula;

[0032] is the hydrological slope factor, which is converted into a score through a standardization formula;

[0033] is the abnormal score of the water body state.

[0034] To further optimize this technical solution, in step S5, synchronizing the edge analysis results and uploading them to the cloud platform includes:

[0035] Combining big data analysis with the cloud platform, through comparative analysis of historical data and real-time data, the cloud platform generates a dynamic water area health status report, providing water quality change trends and flow velocity fluctuation information for different time periods to obtain cloud data.

[0036] To further optimize this technical solution, the steps of intelligent analysis in step S6 include data input, machine learning model training and prediction, and dynamic report generation.

[0037] To further optimize this technical solution, the machine learning algorithm for machine learning model training and prediction is a multi-task random forest model, and the training and prediction steps of this model include data integration and feature construction, time series feature enhancement algorithm, spatial weighting and spatial feature enhancement algorithm, random forest model and anomaly detection algorithm;

[0038] Data integration and feature construction include:

[0039] Obtaining the score of water quality factors from the edge analysis results , the score of water temperature factors , the score of water flow velocity and flow rate factors , the score of precipitation factors , the score of pollution source factors , the score of the hydrological slope factor ;

[0040] Calculate these hydrological data through the data integration formula, and the integration formula is:

[0041] ;

[0042] Among them,

[0043] is the weight of each feature,

[0044] is the standardized score of each feature,

[0045] is the comprehensive feature value, which is the input value of the random forest algorithm;

[0046] The time series feature enhancement algorithm includes:

[0047] Time series enhancement formula:

[0048] ;

[0049] Among them,

[0050] is the time weight coefficient;

[0051] is the scoring value of each feature at time

[0052] is the scoring value of each feature at time

[0053] is the enhanced time series feature value, which is also the input value of the random forest algorithm;

[0054] The spatial weighting and spatial feature enhancement algorithm includes:

[0055] Spatial weighting formula:

[0056] ;

[0057] Among them,

[0058] is the spatial weight decay factor;

[0059] is the number of monitoring points;

[0060] is the monitoring point and the monitoring point distance;

[0061] For the monitoring points Each characteristic factor measured;

[0062] Is the spatially weighted characteristic score;

[0063] The random forest model and the anomaly detection algorithm include:

[0064] The random forest model training algorithm, and the training formula is:

[0065] ;

[0066] Among them,

[0067] Is the prediction output at the current moment Of,

[0068] Is the prediction of the th decision tree,

[0069] Is the total number of trees,

[0070] Is the characteristic score including time series enhancement and spatial weighting at the current moment Of;

[0071] Anomaly detection judgment formula:

[0072] ;

[0073] Among them,

[0074] Is the threshold of anomaly detection,

[0075] Is the anomaly flag. When the absolute value of the difference between the predicted value And the actual observed value Is greater than , The anomaly flag value is 1, otherwise it is 0;

[0076] When Is 1, it means an anomaly is detected, and when it is 0, there is no anomaly.

[0077] To further optimize this technical solution, the anomaly detection algorithm includes a model of the anomaly detection algorithm, which is divided into two modules: a spatio-temporal correlation analysis module and a local anomaly analysis module.

[0078] To further optimize this technical solution, the spatio-temporal correlation analysis module includes: temporal feature analysis, feature construction, dynamic threshold setting, spatial correlation analysis, and spatial weighting mechanism.

[0079] To further optimize this technical solution, the content of the dynamic report includes: current water quality status, water flow dynamics analysis, ecological impact assessment, and pollution source early warning.

[0080] To further optimize this technical solution, the risk early warning in step S7 includes: abnormal alarm, real-time early warning, and event classification analysis.

[0081] In a second aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program instructions are executed by the processor, the usage steps of a water conservancy data acquisition device as described in the first aspect of the present invention are implemented.

[0082] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program instructions are executed by the processor, the usage steps of a water conservancy data acquisition device as described in the first aspect of the present invention are implemented.

[0083] Compared with the prior art, the present invention provides a water conservancy data acquisition device, which has the following beneficial effects:

[0084] By setting an edge analysis algorithm and an anomaly detection algorithm and combining the anomaly detection mechanism of the random forest model and IsolationForest, this water conservancy data acquisition device can accurately process multi-source heterogeneous data and effectively identify abnormal changes in water areas, automatically optimize the data integration process, reduce data loss and errors, and enhance the consistency, efficiency, and accuracy of data.

[0085] By introducing a temporal feature enhancement algorithm, a spatial weighting and spatial feature enhancement algorithm, the monitoring ability of complex water area changes is strengthened. In particular, it can issue early warnings in a timely manner under sudden pollution or extreme climate conditions, providing more accurate decision-making support for water area management and ecological protection, which enhances the readiness of data and the timeliness of handling abnormal events. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0087] Figure 1Schematic flow chart of a water conservancy data acquisition device proposed by the present invention;

[0088] Figure 2 Flow chart of the intelligent analysis of cloud data by the artificial intelligence algorithm of a water conservancy data acquisition device proposed by the present invention;

[0089] Figure 3 Schematic diagram of the anomaly detection algorithm module of a water conservancy data acquisition device proposed by the present invention;

[0090] Figure 4 Schematic diagram of the risk warning system of a water conservancy data acquisition device proposed by the present invention; Specific implementation manners

[0091] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the specific implementation manners of the present invention will be described in detail below with reference to the accompanying drawings of the specification.

[0092] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0093] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" appearing in different places in this specification does not all refer to the same embodiment, nor is it an embodiment that is separately or selectively mutually exclusive with other embodiments.

[0094] Embodiment 1:

[0095] Refer to Figures 1 to 4 , which is the first embodiment of the present invention. This embodiment provides a water conservancy data acquisition device. When the water conservancy data acquisition device is in use, it includes the following steps:

[0096] S1. Divide the target water area into multiple monitoring water areas;

[0097] Define different monitoring water areas through remote sensing technology, geographic information system GIS, and historical hydrological data.

[0098] In this embodiment, before collecting water conservancy data, it is first necessary to reasonably divide the target water area to be measured to ensure the comprehensiveness and accuracy of the monitoring data. These monitored water areas can include multiple parts of different water areas such as reservoirs, rivers, and lakes. When dividing different regions, factors such as possible hydrological changes, such as seasonal water level changes, precipitation, evaporation, etc., should also be considered to better formulate a data collection plan. Through this division, a structured regional basis can be provided for subsequent efficient monitoring, data collection, and analysis, reducing errors in the data collection process and improving the accuracy and pertinence of monitoring.

[0099] S2. Conduct characteristic factor analysis on each monitored water area to obtain the characteristic factor analysis results;

[0100] According to the monitored water areas, install sensors in each monitored water area, and conduct characteristic factor analysis on the hydrological environment in each monitored water area based on the acquisition results of the sensors to obtain the characteristic factor analysis results;

[0101] Characteristic factors include water quality factors, water temperature factors, water flow velocity and flow factors, precipitation factors, pollution source factors, and hydrological slope factors;

[0102] Water quality factors: By deploying a water quality sensor network with water quality monitoring sensors, chemical and physical characteristic factors such as dissolved oxygen (DO), pH value, conductivity (EC), turbidity, suspended solids (SS), ammonia nitrogen (NH3-N), etc. in the water area are monitored in real time.

[0103] In this embodiment,

[0104] Water temperature factors: Monitor the water area temperature through temperature sensors.

[0105] Water flow velocity and flow factors: Use current meters (such as ultrasonic current meters, radar current meters) to monitor the water flow velocity in the water area in real time.

[0106] Precipitation factors: Monitor the precipitation and evaporation in the water area by installing meteorological sensors and meteorological satellite remote sensing.

[0107] Pollution source factors: Judge the pollution source by collecting sediments and monitor the change amount of the pollution source through water quality monitoring sensors.

[0108] Hydrological slope factors: Obtained through the Geographic Information System (GIS).

[0109] S3. Adjust the installation quantity and location of sensors in the monitored water area according to the characteristic factor analysis results, and obtain the hydrological data monitored in real time;

[0110] According to the analysis results of characteristic factors of each monitored water area, combined with the actual monitoring requirements and the quality requirements of hydrological data, adjust the current sensor installation positions to obtain real-time monitored hydrological data;

[0111] In this embodiment, for areas with intensive pollution sources, the deployment density of water quality sensors can be increased; while in areas with relatively small changes in flow velocity, the number of current meters can be reduced. In addition, the technical specifications and sampling frequencies of the sensors should also be adjusted according to data requirements to ensure the accuracy and real-time nature of the data.

[0112] S4. Conduct edge computing analysis on the real-time monitored hydrological data to obtain edge analysis results;

[0113] Conduct preliminary analysis on the real-time monitored hydrological data through edge computing technology to obtain edge analysis results;

[0114] The steps of edge computing analysis include:

[0115] Use the edge calculation formula model for calculation.

[0116] This model is based on the idea of multi-dimensional data fusion and threshold judgment, and conducts edge computing analysis by comprehensively considering water quality factors, water temperature factors, water flow velocity and flow factors, precipitation factors, pollution source factors, and hydrological slope factors; model formula:

[0117] ;

[0118] Among them,

[0119] is the weighting function,

[0120] is the score of water quality factors. It is obtained by converting water quality factors into scores through a standardization formula, and its standardization formula is:

[0121] ;

[0122] Among them, is the value of water quality factors, is the standard value of water quality;

[0123] is the score of water flow velocity and flow factors. It is obtained by converting water flow velocity and flow factors into scores through a standardization formula, and its standardization formula is:

[0124] ;

[0125] Among them, is the value of water flow velocity and flow factors, is the standard value of water flow velocity and flow;

[0126] is the precipitation factor score, which is obtained by converting the precipitation factor into a score through a normalization formula. The normalization formula is:

[0127] ;

[0128] where is the precipitation factor value, is the standard value of precipitation;

[0129] is the water temperature factor score, which is obtained by converting the water temperature factor into a score through a normalization formula. The normalization formula is:

[0130] ;

[0131] where is the water temperature factor value, is the standard value of water temperature;

[0132] is the pollution source factor score, which is obtained by converting the pollution source factor into a score through a normalization formula. The normalization formula is:

[0133] ;

[0134] where is the pollution source factor value, is the pollution source concentration threshold;

[0135] is the hydrological slope factor, which is obtained by converting the hydrological slope factor into a score through a normalization formula. The normalization formula is:

[0136] ;

[0137] where is the hydrological slope factor value, is the hydrological slope standard value;

[0138] is the anomaly score of the water body state.

[0139] In this embodiment, the use of the formula includes four parts: data input, score calculation, threshold judgment, and real-time response.

[0140] Data input: Obtain data on water quality factors, water temperature factors, water flow velocity and flow factors, precipitation factors, pollution source factors, and hydrological slope factors through various sensors, and input these data into the model formula.

[0141] Score calculation: Based on the calculation results of each factor, i.e., in the formula through the weighted function , calculate the comprehensive anomaly score.

[0142] Threshold judgment: Set the threshold of the anomaly score according to historical data or preset standards. If the anomaly score exceeds a certain preset threshold (e.g., the anomaly score is greater than 0.8), it is considered that the water body state is abnormal and an alarm is triggered immediately.

[0143] Real-time response: The edge computing device evaluates the water quality and hydrological state of each monitoring area in real time. Once an anomaly is detected (e.g., pollution source leakage, abnormal flow rate, etc.), the system will trigger an alarm and provide treatment suggestions.

[0144] S5. Synchronize the edge analysis results and upload them to the cloud platform to obtain cloud data;

[0145] Transmit the edge analysis results to the cloud platform through the communication network, store and manage the edge analysis results on the cloud platform to obtain cloud data;

[0146] Combine big data analysis with the cloud platform. Through the comparative analysis of historical data and real-time data, the cloud platform generates a dynamic water area health status report, providing water quality change trends and flow rate fluctuation information for different time periods to obtain cloud data.

[0147] In this embodiment, the obtained edge computing analysis results are transmitted to the cloud platform through an efficient communication network. On the cloud platform, data is uniformly stored, managed, and comprehensively analyzed. Through the powerful computing power of the cloud platform, multi-regional and multi-dimensional data can be combined for in-depth data mining and trend prediction.

[0148] S6. Conduct intelligent analysis on the cloud data to obtain a water area dynamic report;

[0149] Conduct intelligent analysis on the cloud data through artificial intelligence algorithms to obtain a water area dynamic report;

[0150] Intelligent analysis includes data input, machine learning model training and prediction, and dynamic report generation;

[0151] Data input: Input the water quality factor score, water temperature factor score, water flow velocity and flow factor score, precipitation factor score, pollution source factor score, and hydrological slope factor score for subsequent analysis;

[0152] Machine learning model training and prediction. The training and prediction steps of this model include data integration and feature construction, time series feature enhancement algorithm, spatial weighting and spatial feature enhancement algorithm, random forest model and anomaly detection algorithm. Data integration and feature construction include:

[0153] Obtain the water quality factor score from the edge analysis results , the water temperature factor score , the water flow velocity and flow rate factor score , the precipitation factor score , the pollution source factor score , the hydrological slope factor score ;

[0154] Calculate these hydrological data through the data integration formula, and the integration formula is:

[0155] ;

[0156] Among them,

[0157] is the weight of each feature,

[0158] is the standardized score of each feature,

[0159] is the comprehensive feature value, which is the input value of the random forest algorithm;

[0160] The time series feature enhancement algorithm includes:

[0161] Time series enhancement formula:

[0162] ;

[0163] Among them,

[0164] is the time weight coefficient;

[0165] is the scoring value of each feature at time

[0166] is the scoring value of each feature at time

[0167] is the enhanced time series feature value, which is also the input value of the random forest algorithm;

[0168] The spatial weighting and spatial feature enhancement algorithm includes:

[0169] Spatial weighting formula:

[0170] ;

[0171] Among them,

[0172] is the spatial weight decay factor;

[0173] is the number of monitoring points;

[0174] is the monitoring point and the monitoring point distance;

[0175] is the monitoring point measured characteristic factors;

[0176] is the spatially weighted characteristic score;

[0177] The random forest model and anomaly detection algorithm include:

[0178] The random forest model training algorithm, and the training formula is:

[0179] ;

[0180] where

[0181] is the predicted output at the current moment ;

[0182] is the prediction of the k-th decision tree,

[0183] is the total number of trees,

[0184] is the characteristic score including temporal enhancement and spatial weighting at the current moment ;

[0185] Anomaly detection judgment formula:

[0186] ;

[0187] where

[0188] is the threshold for anomaly detection,

[0189] is the anomaly flag. When the absolute value of the difference between the predicted value and the actual observed value is greater than , the anomaly flag value is 1, otherwise it is 0;

[0190] When is 1, it means an anomaly is detected, and when it is 0, there is no anomaly.

[0191] In this embodiment, based on the data sources of different sensors (water quality, water flow velocity, precipitation), the collected data is standardized and weighted and synthesized, and the comprehensive eigenvalue is calculated through the data integration formula.

[0192] In the random forest model, by simultaneously using the time series feature enhancement algorithm for multiple decision trees, the change trend between the current time point and the past time point can be calculated, and finally an enhanced time series feature is calculated. The spatial weighted and spatial feature enhancement algorithms are used to introduce spatial correlation, calculate the distances and spatial weights between monitoring points for multiple decision trees simultaneously, and calculate the spatially weighted feature scores through the use of the spatial weighted formula. The spatially weighted feature scores of multiple decision trees, the enhanced time series features and the comprehensive eigenvalue are substituted into the training formula to calculate the predicted value of the water area state at the current moment. Finally, the calculated predicted value is compared with the threshold (this threshold is set by the relevant water conservancy department according to the actual situation) to determine whether the monitored water area is abnormal.

[0193] The anomaly detection algorithm, including the model of the anomaly detection algorithm, is divided into two modules: the spatio-temporal correlation analysis module and the local anomaly analysis module;

[0194] The spatio-temporal correlation analysis module: Utilize the correlation of the time series and spatial features between hydrological data to detect long-term trend changes and sudden fluctuations;

[0195] The correlation of the time series and spatial features between hydrological data, the content of which is: time series feature analysis, feature construction, dynamic threshold setting, spatial correlation analysis, spatial weighting mechanism;

[0196] In this embodiment,

[0197] Time series feature analysis: Refer to the hydrological data of multiple time points and analyze the differences between the current hydrological data and the historical hydrological data;

[0198] Feature construction: For the data at time t, by performing weighted calculations on the data at times t-1, t-2,..., t-n, the historical trend score is obtained;

[0199] Dynamic threshold setting: Dynamically adjust the threshold according to the standard deviation and volatility of the historical hydrological data;

[0200] Spatial correlation analysis: Consider the spatial distribution of multiple monitoring points and analyze the correlation of water quality or flow velocity between adjacent monitoring points.

[0201] Spatial weighting mechanism: The water quality change of each monitoring point is weighted according to its distance from neighboring points.

[0202] Local Anomaly Analysis Module: Combining the IsolationForest algorithm, it deeply detects local anomaly patterns. IsolationForest (IForest) is a tree-based anomaly detection method, especially suitable for high-dimensional data and large datasets. In this embodiment, we use IsolationForest to detect local anomalies in water area monitoring data, such as sudden pollution source emissions, drastic fluctuations in flow velocity, etc.

[0203] This module includes local anomaly degree calculation, anomaly scoring, integration mechanism, and local anomaly pattern.

[0204] Local Anomaly Degree Calculation: Train a model through IsolationForest to capture potential abnormal samples in the data. Since IsolationForest can effectively handle the non-linear relationships of data, for complex water area data (including water quality, flow velocity, precipitation, etc.), it can better identify extreme abnormal samples.

[0205] Anomaly Scoring: For the data of each monitoring point, the IForest model will calculate its anomaly score. The higher the score, the more abnormal the data at this point is compared with other data.

[0206] Integration Mechanism: Combining spatio-temporal features and the output results of IsolationForest to further confirm whether a data point belongs to an anomaly.

[0207] Local Anomaly Pattern: Detect potential local anomaly patterns in the data. For example, in a monitoring point, when the water flow score shows a sudden change and the flow velocity in the surrounding area does not change significantly, the system may judge it as an abnormal pattern. At this time, in-depth analysis is carried out through IsolationForest to identify potential pollution sources or illegal emissions.

[0208] Dynamic Report Generation: Generate a real-time water area dynamic report based on the analysis results of the multi-task random forest algorithm and display the report content to the user. The report content includes: current water quality status, water flow dynamic analysis, ecological impact assessment, pollution source warning.

[0209] Current Water Quality Status: Water quality status data obtained based on water quality factor scores;

[0210] Water Flow Dynamic Analysis: Includes the current water flow velocity, direction, and predicted flow trend;

[0211] Ecological Impact Assessment: Evaluate the changing trend of the water area ecological environment based on an ecological model;

[0212] Pollution Source Warning: Based on the anomaly detection results, give the specific location and change situation of the pollution source;

[0213] Its data display forms include tabular display, dynamic curve graph, heat map and spatial distribution map, and future trend prediction map.

[0214] Tabular display: List in detail the data such as water quality, water flow, precipitation, etc. of each monitoring point, which is convenient for viewing historical trends and current status.

[0215] Dynamic curve graph: Display the trend curves of water quality, water flow velocity, temperature, etc. changing over time, which can intuitively reflect the changes in the water area.

[0216] Heat map and spatial distribution map: Through the GIS system, display the status of each monitoring point in the water area on the map, highlighting the problem areas.

[0217] Future trend prediction map: Based on the prediction model, display the predicted trends of water quality, water flow, etc. in a future period of time.

[0218] Based on this embodiment, Table 1 is drawn according to the scores of each characteristic factor collected.

[0219] Monitoring point Water quality score Water flow velocity and velocity score Precipitation score Temperature score Pollution source score Hydrological slope score A 0.72 0.55 0.55 0.80 0.50 0.68 B 0.88 0.60 0.60 0.75 0.45 0.62 C 0.55 0.40 0.40 0.65 0.90 0.70

[0220] Table 1

[0221] Based on Table 1, the model can generate the following several intelligent analysis results:

[0222] Water quality analysis: The water quality score of monitoring point C is significantly lower than that of other monitoring points, and there may be pollution sources or other abnormal situations.

[0223] Abnormal flow velocity: The flow velocity of monitoring point A is relatively low, and there may be water flow stagnation, affecting the self-purification ability of the water area.

[0224] Correlation analysis of precipitation and temperature: The changes in precipitation and temperature scores show seasonal change trends at different monitoring points, which may affect the changes in water flow and water quality.

[0225] Pollution source detection: The pollution source score of monitoring point C is 0.90, indicating the existence of a pollution source, and the system will issue a warning.

[0226] Display of water area dynamic report:

[0227] Tabular display: As shown in the above table, the system automatically calculates the water quality score, flow velocity, etc. of each monitoring point and marks the abnormal situations.

[0228] Dynamic curve graph: The X-axis is time and the Y-axis is score.

[0229] Draw the change curves of water quality scores, flow velocity scores, etc. of each monitoring point over time to intuitively understand the trends.

[0230] For example, assume that the water quality score has been gradually decreasing in the past 24 hours. Users can see this changing trend through the dynamic curve.

[0231] Heat map and spatial distribution map:

[0232] Based on real-time data, each monitoring point of the water area is displayed on the map, and the status such as water quality and flow rate is marked with colors.

[0233] High-risk areas (such as pollution source points) are highlighted in red, and low-risk areas are shown in green.

[0234] Future trend prediction map:

[0235] Based on the prediction of the machine learning model, it shows the predicted trends of water flow, water quality, etc. in the next 7 days to help decision-makers plan in advance.

[0236] S7. Conduct risk early warning according to the water area dynamic report;

[0237] According to the water area dynamic report, combining the hydrological changes and water quality analysis results of the water area, predict potential risks and abnormal phenomena. After processing these risks and abnormal phenomena through the anomaly detection algorithm, conduct risk early warning for each monitored water area.

[0238] Risk early warning includes: anomaly alarm, real-time early warning, and event classification analysis;

[0239] In this embodiment,

[0240] Anomaly alarm: Send an alarm message to the management personnel according to the detected abnormal data; the alarm message includes the specific location (monitoring point) of the anomaly, the type of the anomaly (such as water quality pollution, flow rate fluctuation, etc.), and the possible cause of the anomaly occurrence.

[0241] Real-time early warning: Provide real-time anomaly alarms by weighted fusion of the results of the spatio-temporal correlation analysis module and the local anomaly analysis module, reducing the situation of missed alarms or false alarms;

[0242] Event classification analysis: Classify the abnormal time after the anomaly occurs. For example:

[0243] Pollution source emission: Identify possible pollution sources through the mutation of water quality data combined with spatial analysis.

[0244] Abnormal water flow fluctuation: Identify possible water flow interference factors or natural disasters through the abnormal fluctuation of flow rate data.

[0245] Impact of climate change: When climate characteristics such as precipitation change violently, the model will analyze whether there are water area anomalies caused by climate change.

[0246] Embodiment 2:

[0247] This embodiment also provides a computer device, which is applicable to a water conservancy data acquisition device. It includes a memory and a processor. The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the usage steps of a water conservancy data acquisition device as proposed in the above embodiment.

[0248] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the usage steps of a water conservancy data acquisition device as proposed in the above embodiment.

[0249] The computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a carrier network, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.

[0250] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. And the aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs, etc., which can store program codes.

[0251] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered as a definable sequence list of executable instructions for implementing logical functions, which can be embodied specifically in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in conjunction with these instruction execution systems, apparatuses, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0252] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, a computer-readable medium can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0253] It should be understood that various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0254] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A water conservancy data acquisition device, characterized in that When the water conservancy data acquisition device is in use, it includes the following steps: S1. Divide the target water area into multiple monitoring water areas; Use remote sensing technology, Geographic Information System (GIS), and historical hydrological data to delimit different monitoring water areas; S2. Conduct characteristic factor analysis on each monitoring water area to obtain the characteristic factor analysis result; According to the monitoring water areas, install sensors in each monitoring water area, and conduct characteristic factor analysis on the hydrological environment in each monitoring water area based on the acquisition results of the sensors to obtain the characteristic factor analysis result; S3. Adjust the installation quantity and location of sensors in the monitoring water area according to the characteristic factor analysis result, and obtain the real-time monitored hydrological data; According to the characteristic factor analysis result of each monitoring water area, combined with the actual monitoring requirements and hydrological data quality requirements, adjust the current sensor installation location, and obtain the real-time monitored hydrological data; S4. Conduct edge computing analysis on the real-time monitored hydrological data to obtain the edge analysis result; Use edge computing technology to conduct preliminary analysis on the real-time monitored hydrological data to obtain the edge analysis result; S5. Synchronize the edge analysis result and upload it to the cloud platform to obtain the cloud data; Transmit the edge analysis result to the cloud platform through the communication network, store and manage the edge analysis result on the cloud platform to obtain the cloud data; S6. Conduct intelligent analysis on the cloud data to obtain the water area dynamic report; Use artificial intelligence algorithms to conduct intelligent analysis on the cloud data to obtain the water area dynamic report; S7. Conduct risk warning according to the water area dynamic report; According to the water area dynamic report, combined with the hydrological changes and water quality analysis results of the water area, predict potential risks and abnormal phenomena. After processing the risks and abnormal phenomena through the anomaly detection algorithm, conduct risk warning on each monitoring water area.

2. The water conservancy data acquisition device according to claim 1, characterized in that The characteristic factors in step S2 include: Water quality factor; Water temperature factor; Water flow velocity and flow rate factor; Precipitation factor; Pollution source factor; Hydrological slope factor.

3. A water conservancy data acquisition device according to claim 1, characterized in that, The steps of edge computing analysis in step S4 include: Use the edge calculation formula model for calculation: This model is based on the idea of multi-dimensional data fusion and threshold judgment, and conducts edge computing analysis by comprehensively considering the water quality factor, water temperature factor, water flow velocity and flow rate factor, precipitation factor, pollution source factor, and hydrological slope factor; model formula: ； Among them, is a weighting function; is the water quality factor score, which is obtained by converting the water quality factor into a score through a standardization formula; is the water flow velocity and flow rate factor score, which is obtained by converting the water flow velocity and flow rate factor into a score through a normalization formula; is the precipitation factor score, which is obtained by converting the precipitation factor into a score using a normalization formula; is the water temperature factor score, which is obtained by converting the water temperature factor into a score using a normalization formula; is the pollution source factor score, which is obtained by converting the pollution source factor into a score through a standardization formula; is the hydrological slope factor, which is converted into a score by a standardization formula; is the anomaly score of the water body state.

4. The water conservancy data acquisition device according to claim 1, characterized in that Synchronizing the edge analysis result and uploading it to the cloud platform in step S5 includes: Combining big data analysis with the cloud platform. Through the comparative analysis of historical data and real-time data, the cloud platform generates a dynamic water area health status report, providing the water quality change trend and flow velocity fluctuation information in different time periods to obtain the cloud data.

5. The water conservancy data acquisition device according to claim 1, characterized in that, The steps of intelligent analysis in step S6 include: Data input; Machine learning model training and prediction; Dynamic report generation.

6. A water conservancy data acquisition device according to claim 5, characterized in that, The machine learning algorithm for machine learning model training and prediction is the multi-task random forest model. The training and prediction steps of this model include data integration and feature construction, time series feature enhancement algorithm, spatial weighting and spatial feature enhancement algorithm, random forest model, and anomaly detection algorithm; Data integration and feature construction include: Obtain the water quality factor score from the edge analysis results , the water temperature factor score , the water flow velocity and flow rate factor score , the precipitation factor score , the pollution source factor score , the hydrological slope factor score ; Calculate the hydrological data through the data integration formula, and the integration formula is: ； Among them, is the weight of each feature; is the standardized score for each feature; is the comprehensive eigenvalue, which is the input value of the random forest algorithm; The time series feature enhancement algorithm includes: Time series enhancement formula: ； Among them, is the time weight coefficient; For the scoring value of each feature at a moment; For the scoring value of each feature at a moment; It is an enhanced timing feature value, which is also the input value of the random forest algorithm; The spatial weighting and spatial feature enhancement algorithm includes: Spatial weighting formula: ； Among them, is the spatial weight decay factor; is the number of monitoring points; is the monitoring point and the monitoring point distance; Monitoring point Measured characteristic factors is the spatially weighted feature score; The random forest model and anomaly detection algorithm include: Random forest model training algorithm, and the training formula is: ； Among them, is the predicted output at the current moment of For the prediction of the th decision tree, is the total number of trees, for the current moment feature score that includes temporal enhancement and spatial weighting; Anomaly detection judgment formula: ； Among them, is the threshold for anomaly detection, is an anomaly flag. When the predicted value and the actual observed value the absolute value of the difference is greater than , the anomaly flag value is 1; otherwise it is 0. When being 1 indicates that an abnormality is detected, and being 0 indicates no abnormality.

7. The water conservancy data acquisition device according to claim 6, characterized in that, The anomaly detection algorithm includes the model of the anomaly detection algorithm, which is divided into two modules: Spatio-temporal correlation analysis module; Local anomaly analysis module.

8. A water conservancy data acquisition device according to claim 7, characterized in that, The spatio-temporal correlation analysis module includes: Time series feature analysis; Feature construction; Dynamic threshold setting; Spatial correlation analysis; Spatial weighting mechanism.

9. The water conservancy data acquisition device according to claim 5, wherein The content of the dynamic report includes: Current water quality status; Water flow dynamic analysis; Ecological impact assessment; Pollution source early warning.

10. A water conservancy data acquisition device according to claim 1, characterized in that, The risk early warning in step S7 includes: Anomaly alarm; Real-time early warning; Event classification analysis.

Images