A floating monitoring platform and hydrological monitoring method for light wave water environments

Abstract

This invention relates to the field of hydrological monitoring technology, specifically a floating monitoring platform for light wave environments. The platform comprises: a water pipe with a suction head installed at its first end, the suction end of which has a filter screen; an input end of a water pump connected to the second end of the water pipe for pumping water; a water pipe release device with a slow-rotating storage device, the main body of which is coiled and stored on the slow-rotating storage device; a flow velocity sensor with its monitoring end installed on the water pipe for detecting the water flow velocity within the pipe; and the sensor being installed on the water pipe at a distance from the suction head that is a preset height from the underwater silt. This floating monitoring platform for light wave environments solves the problem of accurately controlling the underwater height of the sensor from the silt. Furthermore, this invention proposes a hydrological monitoring method using the aforementioned floating monitoring platform for light wave environments.

Description

Technical Field

[0001] This invention relates to the field of hydrological monitoring technology, and in particular to a floating monitoring platform and hydrological monitoring method for light wave water environments. Background Technology

[0002] Existing hydrological monitoring systems are generally suspended on the water surface. An important issue in hydrological monitoring is the monitoring of underwater silt deposition direction. General monitoring equipment has certain limitations. Sensors for monitoring silt need to be placed at a specific height, usually at a certain height distance from the silt. Ordinary underwater equipment cannot accurately detect the depth of silt, especially floating platforms used in rivers and lakes. The distance of underwater silt from the water surface varies in different locations. If the height of the detection area is insufficient, it can have a significant impact on the overall data acquisition of underwater monitoring, especially on the accuracy and completeness of the data.

[0003] In hydrological monitoring, the distribution characteristics of sedimentation directly affect the sensor deployment scheme, and the rationality of the deployment scheme determines the accuracy and completeness of the monitoring data. Sedimentation areas typically exhibit a non-uniform spatial distribution, with some areas experiencing rapid sedimentation rates while others remain relatively stable. Since variations in sediment thickness are closely related to factors such as water flow velocity, sediment content, and topographic features, sensor deployment must fully consider these factors to ensure that monitoring points accurately reflect the true state of sedimentation. However, a technical dilemma in practical deployment is that while overly dense sensor deployment can improve data accuracy, it significantly increases equipment costs and maintenance difficulty; conversely, sparse deployment may lead to localized monitoring data, failing to comprehensively capture the dynamic changes in sedimentation. Furthermore, the matching between the sensor measurement range and the sedimentation depth is also a critical issue; a measurement range that is too small may result in missing data, while a range that is too large may reduce data precision. These issues need to be comprehensively considered in the deployment scheme. Summary of the Invention

[0004] To address the aforementioned problems, this invention proposes a floating monitoring platform for light-wave water environments, which solves the difficulty in accurately controlling the underwater distance between the sensor and the silt. Furthermore, this invention proposes a hydrological monitoring method utilizing the aforementioned floating monitoring platform for light-wave water environments.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] In the first technical solution, a floating monitoring platform for light wave water environments includes:

[0007] A water pipe, wherein a water suction head is installed at the first end of the water pipe, and the water suction end of the water suction head has a filter screen;

[0008] A water pump, the input end of which is connected to the second end of a water pipe, for pumping water.

[0009] A water pipe release device, wherein the water pipe release device has a slow-moving rotating storage device, and the main body of the water pipe is coiled and stored on the slow-moving rotating storage device;

[0010] A flow velocity sensor, wherein the monitoring end of the flow velocity sensor is installed on the water pipe, and the flow velocity sensor is used to detect the flow velocity of water in the water pipe;

[0011] The sensor is installed on the water pipe at a distance from the suction head that is a preset height from the underwater silt.

[0012] In the first technical solution, preferably, the water pipe is connected to the water pump input end via a sealed bearing.

[0013] In the first technical solution, as a preferred embodiment, the water pipe release device includes:

[0014] Rotary power unit;

[0015] An output shaft is connected to the output end of the rotational power device, and the output shaft is vertically arranged.

[0016] The upper end of the output shaft is connected to the rotation center of the tray;

[0017] A limiting plate is installed above the support plate, and a hole is provided on the limiting plate for the water pipe to pass through;

[0018] The water pump is installed above the limiting plate.

[0019] In the first technical solution, as a preferred embodiment, the outer side of the water pipe release device is also provided with a set of water pipe correction rollers for guiding the water pipe to move vertically downward.

[0020] In the first technical solution, preferably, the water suction head includes a water suction plate and several branch pipes. The water suction plate is provided with not less than 2 water suction holes. Each water suction hole is provided with a filter screen. The first ends of the several branch pipes are respectively connected to the water suction holes, and the second ends of the branch pipes are collected and connected to the water pipe.

[0021] In the first technical solution, as a preferred embodiment, the method of using the floating monitoring platform for light wave water environments is as follows:

[0022] Install the suction head onto the water pipe, connect the second end of the water pipe to the input end of the water pump, install the sensor on the water pipe at a predetermined distance from the suction head, and place the suction head underwater to complete the preparation work.

[0023] Rotate the water pipe release device to slowly release the water pipe. At the same time, turn on the water pump and activate the flow rate sensor. When the flow rate sensor detects that the water flow rate in the water pipe is less than the threshold, control the water pipe release device to stop releasing the water pipe. At this time, the suction head contacts the silt, the silt seals the suction hole, the water flow rate in the water pipe decreases, and the sensor reaches the predetermined distance from the silt at the bottom of the water.

[0024] In the second technical solution, a hydrological monitoring method utilizes the floating monitoring platform for light wave water environments described in the first technical solution, the method comprising:

[0025] Acquire historical sedimentation data of the target water area, analyze the spatial characteristics of sedimentation distribution and the temporal trend of sedimentation rate, and identify high-incidence areas of sedimentation and key points of rate change.

[0026] Based on the characteristics of siltation distribution and the trend of rate change, the initial locations and number of sensors are preset to ensure coverage of high-incidence areas and key points;

[0027] By using sensor measurement range and accuracy parameters, the monitoring coverage and error range of each preset location are calculated, and the deployment location and number are adjusted to optimize the coverage effect of the monitoring network.

[0028] By simulating the trend of sedimentation rate changes, we can predict possible changes in future sedimentation distribution and determine whether it is necessary to increase or adjust the location and number of sensors.

[0029] Clustering analysis, a machine learning algorithm, is used to cluster historical sedimentation data, identify the main patterns and changes in sedimentation distribution, and assist in optimizing sensor deployment schemes.

[0030] By using regression analysis in machine learning algorithms, a model of the relationship between sedimentation rate and time is established to predict the future trend of sedimentation rate changes, providing a basis for dynamic adjustment of sensor deployment.

[0031] In the second technical solution, as a preferred embodiment, the acquisition of historical siltation data of the target water area, analysis of the spatial characteristics of siltation distribution and the temporal trend of siltation rate, and determination of high-incidence siltation areas and key points of rate change include:

[0032] Acquire historical siltation data for the target water area and define the dataset boundaries based on the water area's extent;

[0033] Spatial analysis methods are used to calculate the spatial characteristics of siltation distribution and generate a spatial distribution map;

[0034] The time-varying trend of sedimentation rate is calculated using time analysis methods, and a time trend graph is generated.

[0035] By combining spatial distribution maps, determine the boundaries and locations of areas prone to siltation;

[0036] Based on the time trend chart, identify the key time points of change in the siltation rate;

[0037] By integrating high-incidence areas and key time points, a time-series distribution map of high-incidence areas of siltation is generated;

[0038] Based on time series distribution maps, a predictive model for areas prone to siltation is established;

[0039] The method of pre-setting the initial location and number of sensors based on the distribution characteristics and rate change trends of siltation to ensure coverage of high-incidence areas and key points includes:

[0040] Acquire historical sedimentation data for the target area and extract the spatial characteristics of sedimentation distribution;

[0041] Analyze time series data to determine the changing trend pattern of sedimentation rate;

[0042] Based on spatial characteristics and trends, draw a spatial distribution map of areas prone to siltation;

[0043] Based on the distribution characteristics of high-incidence areas, determine the preset sensor deployment locations;

[0044] If the siltation rate changes drastically, increase the number of preset sensors deployed.

[0045] Spatial density calculation methods are used to optimize sensor deployment schemes and ensure coverage of key locations;

[0046] Data collection is carried out through a pre-set plan to establish a siltation monitoring and analysis model.

[0047] In the second technical solution, preferably, the step of using sensor measurement range and accuracy parameters to calculate the monitoring coverage and error range of each preset location, adjusting the deployment location and number, and optimizing the coverage effect of the monitoring network includes:

[0048] Obtain the measurement range and accuracy parameters of the sensor, and calculate the monitoring coverage area for each preset location using a spatial coverage algorithm;

[0049] Based on the accuracy parameters, the error range for each preset position is calculated using an error propagation model;

[0050] Determine whether the monitoring coverage area meets the preset threshold; if not, adjust the deployment location and quantity.

[0051] Using network optimization algorithms, the monitoring coverage and error range are recalculated for the adjusted deployment locations and quantities;

[0052] Based on the recalculated results, a comprehensive evaluation model is used to determine the coverage effect of the monitoring network;

[0053] If the coverage effect is not optimal, return to the step of adjusting the deployment location and quantity;

[0054] Determine the final deployment plan and output the optimized monitoring network coverage effect;

[0055] In the second technical solution, as a preferred embodiment, the use of cluster analysis methods in machine learning algorithms to cluster historical sedimentation data, identify the main patterns and variation laws of sedimentation distribution, and assist in optimizing the sensor deployment scheme includes:

[0056] The raw dataset is obtained from historical backlog data, and preprocessed to remove noise and standardize the data size.

[0057] The K-means clustering algorithm was used to perform cluster analysis on the processed data, and Euclidean distance was selected as the distance metric.

[0058] The number of cluster centers is determined based on data characteristics, and a cluster distribution map and change rate statistics are generated.

[0059] If the clustering results show significant regional differences in data distribution, then adjust the sensor deployment locations;

[0060] By optimizing the deployment locations, the monitoring coverage of the sensors for siltation data can be improved;

[0061] Generate a sensor distribution map based on the final deployment plan to achieve efficient monitoring;

[0062] The aforementioned method utilizes regression analysis in machine learning algorithms to establish a model relating siltation rate to time, predicting future trends in siltation rate changes and providing a dynamic adjustment basis for sensor deployment. This includes:

[0063] Obtain the rate values ​​on the historical timeline and use linear regression to establish the relationship between the rate values ​​and the timeline.

[0064] Calculate the predicted value based on the relationship, and determine the amount of change in the rate value;

[0065] Obtain the location information of the sensor deployment points and calculate the adjustment amount by combining the predicted values;

[0066] Determine the relationship between the change and the preset threshold. If the change exceeds the threshold, update the location information of the deployment point.

[0067] Based on the updated deployment point information, the adjustment amount is recalculated to determine a new sensor deployment scheme.

[0068] The decision tree algorithm is used to classify the adjustment quantities and determine the dynamic indicators;

[0069] The model quantities are optimized based on dynamic indicators, and the final sensor deployment scheme is output.

[0070] The beneficial effects of this solution are explained in detail below through examples. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the overall floating monitoring platform for light wave water environment according to the present invention.

[0072] Figure 2 for Figure 1 A schematic diagram of the functional device components.

[0073] Figure 3 for Figure 2 Side view of the water pipe straightening roller.

[0074] Figure 4 This is a schematic diagram of the water suction head.

[0075] Figure 5 This is a schematic diagram of the water suction tray.

[0076] Figure 6 This is a flowchart of hydrological monitoring methods.

[0077] The reference numerals in the figures include:

[0078] 10-Floating body, 20-Functional equipment components, 30-Equipment box, 40-Photovoltaic system, 50-Camera, 60-Communication system;

[0079] 21-Rotating power unit, 22-Output shaft, 23-Panel, 24-Limiting plate, 25-Water pipe, 26-Water pump, 27-Sealed bearing, 28-Flow rate sensor, 29-Supporting platform, 210-Water pipe correction roller, 211-Suction head, 2111-Suction tray, 2112-Branch pipe, 212-Sensor. Detailed Implementation

[0080] To make the objectives, technical solutions, and advantages of this technical solution clearer, the following detailed description, in conjunction with specific embodiments, further illustrates this technical solution. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of this technical solution.

[0081] Example 1

[0082] To address the problem of difficulty in controlling the arrangement distance between the sensor 212 and the silt surface in existing technologies, such as... Figures 1-5 As shown in the figure, this embodiment proposes a floating monitoring platform for light wave water environments.

[0083] like Figure 1As shown, this floating monitoring platform for light wave water environments includes a float 10, a solid structure that serves as the main buoyancy provider for the platform. The upper part of the float 10 is the working platform; specifically, the top surface of the float 10 forms the working platform, which is surrounded by guardrails. A photovoltaic system 40, comprising two photovoltaic panels, is installed on the platform. The photovoltaic system 40 has a built-in battery to store electrical energy and power other electronic and electrical equipment on the platform. A functional device component 20 is also installed on the platform to control the height of the sensor 212, ensuring that the sensor 212 is at a specified distance from the underwater silt.

[0084] A support pole is erected on the working platform, and an equipment box 30 is mounted on the pole. The equipment box 30 stores necessary tools for easy maintenance of the floating monitoring platform. A camera 50 and a communication system 60 are also mounted on the support pole. The camera 50 is used for monitoring, and the communication equipment is used to communicate with external devices, transmitting data and control signals in real time. All of the aforementioned functional equipment components 20, camera 50, and communication system 60 are electrically connected to the photovoltaic system 40.

[0085] like Figure 2 As shown, in this embodiment, the functional device component 20 includes a rotational power device 21, which includes a motor and a reducer. The output end of the motor is connected to the input end of the reducer, and the output end of the reducer is connected to an output shaft 22. In this embodiment, the output shaft 22 is vertically arranged. The output shaft 22 is connected to the rotation center of the lower end face of the circular support plate 23. A limiting plate 24 is installed above the support plate 23, and the limiting plate 24 is connected to the support plate 23 by a small support rod. The limiting plate 24 is a circular plate with a hollowed-out through hole at its center. A water pump 26 is installed above the limiting plate 24.

[0086] The above structure forms the basic structure of the functional equipment component 20. The water pipe 25 is coiled and positioned between the support plate 23 and the limiting plate 24. The first end of the water pipe 25 is connected to the suction head 211, and the second end of the water pipe 25 passes through the through hole on the limiting plate 24 and is connected to the input end of the water pump 26. Preferably, when the rotating power device 21 drives the water pipe 25 to rotate around the axis of the output shaft 22, a sealed bearing 27 is used to connect the water pipe 25 and the input end of the water pump 26 to prevent the water pipe 25 from twisting and deforming.

[0087] In addition, in this embodiment, the function of the limiting plate 24 is to prevent the water pipe 25 from being compressed and flattened, which would make it difficult to draw water. Due to the combination of the support plate 23 and the limiting plate 24, even if the water pipe 25 is subjected to radial pressure and its cross-sectional shape shrinks from a circle to a square, it can still maintain the effect of water flow. Furthermore, when the water pump 26 draws water, the water inside it can also have a certain effect of resisting the flattening of the water pipe 25.

[0088] In one feasible embodiment, when the support plate 23 and the limiting plate 24 rotate to release the water pipe 25, the water pipe 25 is guided by two water pipe straightening rollers 210 disposed on the support platform 29, so that the water pipe 25 can only move downward in the vertical direction. The structure of the water pipe straightening rollers 210 is as follows: Figure 3 As shown, its wheel surface has a concave curved structure, and the curvature of the curved surface matches the outer diameter of the water pipe 25.

[0089] like Figure 4 and Figure 5 As shown, in this embodiment, the suction head 211 includes a suction plate 2111 and several branch pipes 2112. The suction plate 2111 has three suction holes, and each suction hole has a filter screen inside. The first end of each branch pipe 2112 is connected to a suction hole, and the second end of each branch pipe 2112 converges and connects to the water pipe 25. The purpose of having multiple suction holes on the suction plate 2111 is to prevent debris in the water from clogging a single suction hole, which could trigger the flow rate sensor and cause incorrect measurement of the silt depth.

[0090] Based on the above technical solutions, the method of using the floating monitoring platform for light wave water environments proposed in this embodiment is as follows:

[0091] Install the suction head 211 onto the water pipe 25, connect the second end of the water pipe 25 to the input end of the water pump 26, install the sensor 212 on the water pipe 25 at a predetermined distance from the suction head 211, and place the suction head 211 underwater to complete the preparation work.

[0092] The water release device 25 is rotated, and the water pump 26 and flow rate sensor 212 are turned on simultaneously. The water pipe 25 is slowly released. When the flow rate sensor 212 detects that the water flow velocity in the water pipe 25 is less than a threshold, the sealing mesh with the suction hole is blocked by silt. At this time, the water release device 25 is controlled to stop releasing the water pipe 25. The suction head 211 then contacts the silt, the silt seals the suction hole, the water flow velocity in the water pipe 25 decreases, and the sensor 212 reaches a predetermined distance from the silt at the bottom of the water.

[0093] Example 2

[0094] This embodiment proposes a hydrological monitoring method using the floating monitoring platform for light wave environments described in Embodiment 1. Since the floating monitoring platform for light wave environments can accurately control the distance between the sensor 212 and the silt, combining this method with the floating monitoring platform for light wave environments described in Embodiment 1 can achieve the expected detection results more accurately. Figure 6 As shown, specific hydrological monitoring methods may include:

[0095] Step S101: Obtain historical siltation data of the target water area, analyze the spatial characteristics of siltation distribution and the temporal trend of siltation rate, and identify high-incidence areas of siltation and key points of rate change.

[0096] Historical siltation data for the target water area is acquired, and the dataset boundaries are defined based on the water area's extent. Spatial analysis methods are used to calculate the spatial characteristics of siltation distribution, generating a spatial distribution map. Temporal analysis methods are used to calculate the temporal trend of siltation rate, generating a time trend map. Combined with the spatial distribution map, the boundaries and locations of high-incidence siltation areas are determined. Based on the time trend map, key time points of siltation rate change are identified. High-incidence areas and key time points are integrated to generate a time series distribution map of high-incidence siltation areas. Based on the time series distribution map, a predictive model for high-incidence siltation areas is established.

[0097] Specifically, obtaining historical siltation data for the target waterway can be achieved by consulting hydrological yearbooks, historical channel maps, historical depth measurement reports, and other materials to digitally extract key information. For example, dredging volume data for each section can be extracted from the annual maintenance dredging project reports of a certain deep-water channel from 1980 to 2020. Combined with the section spacing (assuming the distance between section A and section B is 5000 meters) and the channel width (assuming it is 500 meters), the annual average siltation thickness of each section can be calculated (for example, if the dredging volume of section A in 1985 was 5 million cubic meters, then the annual average siltation thickness would be 5 million cubic meters / (5000 meters * 500 meters) = 2 meters). Then, using the Kriging interpolation algorithm, based on existing siltation thickness data for each cross-section, a siltation thickness distribution map of the entire target water area (assumed to be 100 square kilometers) is generated with a spatial resolution of 100 meters × 100 meters. Inverse distance weighting (IDW) is used for supplementary interpolation to improve the accuracy of the spatial distribution. Analyzing the spatial characteristics of the siltation distribution allows for the calculation of the spatial variation coefficient of siltation thickness (e.g., a coefficient of variation of 6 for the entire area indicates significant spatial differences in siltation). Areas with siltation thickness greater than 3 meters are delineated as high-incidence areas of siltation (e.g., a region south of the main channel is found to have a perennial siltation thickness exceeding 3 meters). Simultaneously, historical channel maps are digitized, and GIS technology is used to calculate the channel centerline offset for different years. The correlation between the offset and siltation thickness is analyzed (e.g., a positive correlation is found between the southward offset of the channel centerline and the siltation thickness on the south side, with a correlation coefficient of 8), further corroborating the high-incidence areas of siltation. To analyze the temporal trend of sedimentation rate, time series analysis can be performed on the annual average sedimentation thickness data of each cross section (e.g., cross section A). Moving averages (e.g., a 5-year moving average) can be used to smooth the data, eliminating short-term fluctuations and observing long-term trends. Furthermore, the Mann-Kendall trend test can be used to determine if there is a significant trend in sedimentation rate (e.g., the sedimentation rate of cross section A shows a significant downward trend after 1995, with a significance level of p < 0.5). Combined with wavelet analysis, the periodic characteristics of sedimentation rate changes can be identified (e.g., a periodic fluctuation of approximately 10 years is found in the sedimentation rate of cross section A), with 1998 and 2010 being key years for changes in sedimentation rate.

[0098] Step S102: Based on the siltation distribution characteristics and rate change trend, preset the initial location and number of sensors 212 to ensure coverage of high-incidence areas and key points.

[0099] Historical siltation data for the target area was acquired, and the spatial characteristics of siltation distribution were extracted. Time-series data was analyzed to determine the changing trend of siltation rate. Based on the spatial characteristics and trends, a spatial distribution map of high-incidence siltation areas was drawn. The locations of pre-set sensor 212 deployment points were determined according to the distribution characteristics of high-incidence areas. If the siltation rate changes drastically, the number of pre-set sensor 212 deployment points was increased. A spatial density calculation method was used to optimize the sensor 212 deployment scheme to ensure coverage of key points. Data collection was implemented through the pre-set scheme to establish a siltation monitoring and analysis model.

[0100] Specifically, firstly, historical sedimentation data, such as the annual sedimentation volume and extent of a reservoir over the past 10 years, are used to create sedimentation distribution contour maps and rate change trend maps using geographic information system software such as ArcGIS. Assuming the analysis results show that the average annual sedimentation rate in upstream region A is 5 meters / year, the average annual sedimentation rate in downstream region B is 1 meter / year, and due to topographical influence, a local high-value area with a sedimentation rate of 8 meters / year appears in the central region C. Based on this, stratified sampling combined with Kriging interpolation is used to initially determine the sensor 212 deployment scheme: In region A, where the sedimentation rate variation is relatively uniform, one sensor 212 is deployed every 2 square kilometers; in region B, where the sedimentation rate is lower and the variation is gradual, one sensor 212 is deployed every 5 square kilometers; and in region C, as a high-incidence sedimentation area, one sensor 212 is deployed every 5 square kilometers, with additional sensors 212 deployed at locations with the largest sedimentation rate gradient (determined by calculating the rate change rate, for example, locations with a change rate exceeding 2 meters / year / square kilometer). Simultaneously, combining numerical simulation of water flow, software such as HEC-RAS was used to simulate the flow velocity and direction under different inflow conditions to predict sediment transport paths. At the predicted main sediment transport channels, such as point D (coordinates X = 123456, Y = 789012) at a river bend, additional sensors 212 were installed to capture sedimentation at key points. Thus, through data and model analysis, a preliminary deployment of 30 sensors 212 was determined: 5 in area A, 4 in area B, 18 in area C, and 3 at point D and the sediment transport channel. Time series analysis of sedimentation rates was performed, for example, using the ARIMA model to predict changes in sedimentation rates over the next 5 years, and wavelet analysis was combined to identify periodic fluctuations in sedimentation rates. The sensor deployment density was dynamically adjusted based on the prediction results and periodic fluctuation characteristics. The software was used for automated execution and analysis.

[0101] Step S103: Using the measurement range and accuracy parameters of sensor 212, calculate the monitoring coverage and error range of each preset location, adjust the deployment location and number, and optimize the coverage effect of the monitoring network.

[0102] Obtain the measurement range and accuracy parameters of sensor 212. For each preset location, calculate the monitoring coverage range using a spatial coverage algorithm. Based on the accuracy parameters, calculate the error range for each preset location using an error propagation model. Determine if the monitoring coverage range meets a preset threshold; if not, adjust the deployment location and quantity. Using a network optimization algorithm, recalculate the monitoring coverage range and error range for the adjusted deployment location and quantity. Based on the recalculation results, use a comprehensive evaluation model to assess the monitoring network coverage effect. If the coverage effect is not optimal, return to the step of adjusting the deployment location and quantity. Determine the final deployment scheme and output the optimized monitoring network coverage effect.

[0103] Specifically, obtaining the measurement range and accuracy parameters of sensor 212 is a crucial first step in optimizing the deployment scheme. Taking a certain type of acoustic Doppler current profiler (ADCP) as an example, its measurement range can reach 100 meters, with an accuracy of ±1%. For each preset location, a spatial coverage algorithm is used to calculate the monitoring coverage area. For example, using the Voronoi diagram algorithm, the reservoir area is divided into multiple polygons, each representing the coverage area of ​​one sensor 212. Based on the accuracy parameters, an error propagation model is used to calculate the error range for each preset location. If the water depth measured at a certain location is 50 meters, considering the ±1% accuracy of ADCP, the error range for that location can be calculated to be ±0.5 meters. This step helps to assess the reliability of the data and provides a basis for subsequent decisions. It is then determined whether the monitoring coverage area meets the preset threshold. If not, the deployment location and number are adjusted. Assuming the preset threshold is 90% coverage, calculations show that the initial deployment scheme only achieves 85% coverage, requiring adjustment. Consideration can be given to adding sensors 212 in areas with low coverage, or moving the existing sensor 212 locations to expand the coverage area. A network optimization algorithm is employed to recalculate the monitoring coverage and error range based on the adjusted deployment locations and quantities. For example, a genetic algorithm is used to optimize the sensor 212 layout, iterating multiple times to find the optimal deployment scheme. Each iteration evaluates the coverage and error range, progressively optimizing until the preset target is achieved. Based on the recalculated results, a comprehensive evaluation model is used to judge the monitoring network coverage effect. A multi-objective evaluation function incorporating factors such as coverage, error range, and cost can be constructed. For example, a weight of 0.5 for coverage, 0.3 for error range, and 0.2 for cost can be set to calculate a comprehensive score. If the score does not reach a preset threshold, the process returns to adjusting the deployment locations and quantities. This iterative optimization process not only improves the monitoring network coverage effect but also balances cost and benefit. For example, optimization may reveal that adding a small number of high-precision sensors 212 in certain areas is more effective than adding a large number of low-precision sensors 212. This method maximizes resource utilization efficiency while ensuring monitoring quality. After determining the final deployment scheme, the optimized monitoring network coverage effect is output. This may include increasing coverage to 95%, reducing the average error range to ±0.3 meters, and saving 20% ​​in cost compared to the initial approach. This optimization not only improves monitoring accuracy but also enhances data reliability and representativeness, providing a more solid foundation for reservoir sedimentation monitoring and management decisions. Through these steps, an efficient and accurate reservoir sedimentation monitoring network can be established. This approach considers not only spatial coverage but also data quality and cost-effectiveness, providing reliable technical support for water resource management. The optimized monitoring network can better capture dynamic changes in sedimentation, providing timely and accurate data support for reservoir operation and sediment management, thereby improving the long-term operational efficiency and safety of the reservoir.

[0104] Step S104: By simulating the trend of sedimentation rate change, predict the possible changes in future sedimentation distribution, and determine whether it is necessary to increase or adjust the placement and number of sensors 212.

[0105] The system acquires changes in historical siltation data and analyzes its trend characteristics. Time series analysis is used to process these trend characteristics, yielding simulated values. These simulated values ​​are then matched with a pre-defined siltation model to calculate predicted values. A siltation distribution map is generated based on the predicted values, and characteristic values ​​are identified within the map. The locations and number of sensor 212 deployment points are determined based on these characteristic values. If a characteristic value exceeds a preset threshold, the required adjustment amount is calculated. By combining the adjustment amount with the existing location relationships of the deployment points, the required deployment of sensor 212 is assessed.

[0106] Specifically, assuming historical siltation data of a reservoir shows that over the past 10 years, the average siltation rate in area A was 15 meters per year, in area B it was 1 meter per year, and in area C it was 0.5 meters per year. In this embodiment, the ARIMA (Autoregressive Integral Moving Average) model from time series analysis is used to simulate the siltation rate in each area. First, the siltation rate data for area A (assumed to be 14, 16, 15, 17, 14, 15, 16, 18, 15, 14) is analyzed. The model parameters are determined by calculating the autocorrelation function (ACF) and partial autocorrelation function (PACF), assuming an ARIMA(1,0,1) model, i.e., a first-order autoregressive and first-order moving average model. The model parameters are estimated using the least squares method or the maximum likelihood estimation method, resulting in the model equation: Siltation rate (t) = 6 * Siltation rate (t-1) + 2 * Error (t-1) + Error (t) + 15. Similarly, models are established for the data in areas B and C. Next, the established ARIMA model is used to predict the siltation rate for the next 5 years. The predicted values ​​for region A over the next 5 years are assumed to be 16 meters, 17 meters, 18 meters, 17 meters, and 16 meters. Corresponding predictions are also made for regions B and C. The predicted siltation rate data is input into a two-dimensional mathematical model of water and sediment transport based on the finite element method or finite difference method. This model considers factors such as water flow, sediment transport, and riverbed deformation, and simulates the transport and deposition process of sediment within the reservoir through numerical calculations. It is assumed that the model calculation results show that the siltation thickness in region A will significantly increase due to the increased siltation rate in the future. Then, the cumulative siltation amount in each region over the next 5 years is calculated, and a siltation thickness distribution map is generated. Comparing this with the current sensor 212 distribution map (assuming there are 2 sensors 212 in region A, 3 in region B, and 2 in region C), it is found that the number of sensors 212 in region A is relatively small compared to its predicted siltation amount, and the distribution may not cover severely silted areas. Therefore, the system automatically generates a report, adds one sensor 212 to area A, and adjusts the position of one of the sensors 212 from its original coordinates (x1, y1) to (x2, y2) to more accurately monitor changes in sedimentation. Simultaneously, the system synchronizes the sensor 212 adjustment information to the database and updates the sensor 212 distribution map on the visualization interface.

[0107] Step S105: Cluster analysis method in machine learning algorithm is used to cluster historical siltation data, identify the main patterns and changes in siltation distribution, and assist in optimizing the deployment scheme of sensor 212.

[0108] The raw dataset is obtained from historical sedimentation data. Preprocessing is performed to remove noise and standardize the data size. K-means clustering is used to cluster the processed data, with Euclidean distance selected as the distance metric. The number of cluster centroids is determined based on data characteristics, generating sedimentation distribution maps and change rate statistics. If the clustering results show significant regional differences in data distribution, the deployment locations of sensors 212 are adjusted. By optimizing the deployment locations, the monitoring coverage of sedimentation data by sensors 212 is improved. A sensor 212 distribution map is generated based on the final deployment plan, achieving efficient monitoring.

[0109] Specifically, this embodiment assumes a set of historical siltation data, recording the monthly siltation thickness (in centimeters) at 10 different locations in a reservoir over the past 5 years. The data is stored in a CSV file named "sediment.csv," containing four columns: "Location ID," "Year," "Month," and "Siltation Thickness." To identify siltation distribution patterns and optimize sensor 212 deployment, this embodiment employs the K-means clustering algorithm. First, the data is read using Python's Pandas library and preprocessed. The "Year" and "Month" are merged into a timestamp, forming a new feature "Year-Month." Then, a pivot table is constructed with "Location ID" as the row, "Year-Month" as the column, and "Siltation Thickness" as the value, filling missing values ​​with 0. Next, cluster analysis is performed using the KMeans class from the Scikit-learn library. Considering that there may be 3-5 main types of reservoir siltation patterns, this embodiment attempts to set the number of clusters n_clusters to 3, 4, and 5, and calculates the silhouette coefficient under different cluster numbers. For example, when n_clusters = 4, the calculated silhouette coefficient is 65, which is higher than the results when n_clusters = 3 (52) and n_clusters = 5 (48), indicating that the effect is best when the number of clusters is 4. Through cluster analysis, in this embodiment, 10 locations are divided into 4 groups: Group 1 includes locations 1, 5, and 8; Group 2 includes locations 2 and 6; Group 3 includes locations 3, 7, and 9; and Group 4 includes locations 4 and 10. Observing the curve of the sedimentation thickness of each location in each group over time, it is found that the average sedimentation thickness of Group 1 is the largest and shows obvious seasonal fluctuations; the sedimentation thickness of Group 2 is lower and changes more gently; and the sedimentation thickness and change trend of Groups 3 and 4 are in between. Based on the clustering results, in order to more effectively monitor the sedimentation of the reservoir, in this embodiment, a representative location is selected in each group to deploy sensor 212: location 5 is selected for Group 1 (the change in sedimentation thickness is the most representative), location 2 is selected for Group 2, location 9 is selected for Group 3, and location 4 is selected for Group 4. In this way, the main siltation patterns can be covered by four sensors 212, enabling effective monitoring of the siltation situation of the entire reservoir. Compared with the original deployment of sensors 212 at 10 locations, this saves resources and improves monitoring efficiency. Simultaneously, using Pandas' groupby function, the average and standard deviation of siltation thickness for each group are calculated. These statistical results are then correlated with the sensor 212 deployment locations and written to a new file "sensor_location.csv", containing four columns: "Group ID", "Representative Location ID", "Average Siltation Thickness", and "Standard Deviation of Siltation Thickness", facilitating subsequent analysis and decision-making.

[0110] Step S106: By using regression analysis in machine learning algorithms, a model of the relationship between siltation rate and time is established to predict the future trend of siltation rate changes, providing a basis for dynamic adjustment of sensor 212 deployment.

[0111] The system acquires rate values ​​from the historical timeline and establishes a relationship between rate values ​​and the timeline using linear regression. Based on this relationship, predicted values ​​are calculated to determine the rate change. Location information of sensor 212 deployment points is obtained, and adjustment amounts are calculated using the predicted values. The relationship between the change and a preset threshold is compared; if the change exceeds the threshold, the deployment point location information is updated. The adjustment amounts are recalculated based on the updated deployment point information to determine a new sensor 212 deployment scheme. A decision tree algorithm is used to classify the adjustment amounts and determine dynamic indicators. The model is optimized based on these dynamic indicators, and the final sensor 212 deployment scheme is output.

[0112] Specifically, assuming that sedimentation rate data for the past 10 years has been obtained through some means, measured annually, with the data (unit: cm / year) as follows: 0, 2, 5, 9, 2, 4, 7, 0, 2, 4. To establish a model of the relationship between sedimentation rate and time and predict future trends, a linear regression analysis method can be used. First, the years are numbered, for example, 2014 is 1, 2015 is 2, and so on, with 2023 being 10. Then, the year is used as the independent variable (x), and the sedimentation rate as the dependent variable (y), and a linear regression model is constructed using Python's Scikit-learn library. The specific implementation is as follows: Import `LinearRegression` from `sklearn.linear_model`, create a model instance `model=LinearRegression()`, and use `model.fit(X,y)` to train the model, where X is a two-dimensional array of years `[[1],[2],[3],[4],[5],[6],[7],[8],[9],

[10] ]`, and y is a one-dimensional array of siltation rates `[0,2,5,9,2,4,7,0,2,4]`. After training, the slope is obtained through `model.coef_`, and the intercept is obtained through `model.intercept_`. Assuming the obtained slope is 25 and the intercept is 75, the linear regression equation is y=25x+75. In order to predict the siltation rate for the next 3 years, the years 11, 12, and 13 are substituted into the equation respectively, and the predicted values ​​are 5 cm / year, 75 cm / year, and 0 cm / year respectively. Based on the predictions and historical data analysis, the sedimentation rate shows a trend of increasing year by year, with the growth rate remaining relatively stable. Therefore, these predicted values ​​can be used as a reference for sensor 212 deployment. For example, if sensor 212 needs maintenance or replacement when the sedimentation rate reaches 0 cm / year, it can be estimated that the corresponding operation will be performed around the 13th year. Simultaneously, to ensure the accuracy of the model, new sedimentation rate data needs to be acquired periodically (e.g., every six months or one year), the model retrained, the model parameters updated, and the sensor 212 deployment strategy dynamically adjusted based on the new prediction results, such as adjusting the sensor 212's location or replacement cycle.

[0113] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

[0114] It should be noted that, in the description of this invention, unless otherwise stated, "a plurality of" means two or more; the terms "upper," "lower," "left," "right," "inner," "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. At the same time, in the description of this invention, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0115] The above content is only a preferred embodiment of the present invention. For those skilled in the art, many changes can be made in the specific implementation and application scope based on the ideas of the present invention. As long as these changes do not depart from the concept of the present invention, they all fall within the protection scope of this patent.

Claims

1. A floating monitoring platform for light wave water environments, characterized in that: include: A water pipe, wherein a water suction head is installed at the first end of the water pipe, and the water suction end of the water suction head has a filter screen; A water pump, the input end of which is connected to the second end of a water pipe, for pumping water. A water pipe release device, wherein the water pipe release device has a slow-moving rotating storage device, and the main body of the water pipe is coiled and stored on the slow-moving rotating storage device; A flow velocity sensor, wherein the monitoring end of the flow velocity sensor is installed on the water pipe, and the flow velocity sensor is used to detect the flow velocity of water in the water pipe; A sensor for monitoring silt, wherein the sensor for monitoring silt is installed on a water pipe and the distance from the water suction head is a preset height of the sensor for monitoring silt from the underwater silt; The method of using the floating monitoring platform for light wave water environment is as follows: Install the suction head onto the water pipe, connect the second end of the water pipe to the input end of the water pump, install the sludge monitoring sensor on the water pipe at a predetermined distance from the suction head, and place the suction head underwater to complete the preparation work. Rotate the water pipe release device to slowly release the water pipe. At the same time, turn on the water pump and activate the flow rate sensor. When the flow rate sensor detects that the water flow rate in the water pipe is less than the threshold, control the water pipe release device to stop releasing the water pipe. At this time, the suction head contacts the silt, the silt seals the suction hole, the water flow rate in the water pipe decreases, and the sensor monitoring the silt reaches the predetermined distance from the silt at the bottom of the water.

2. The floating monitoring platform for light wave water environments according to claim 1, characterized in that: The water pipe is connected to the water pump input end via a sealed bearing.

3. The floating monitoring platform for light wave water environments according to claim 2, characterized in that: The water pipe release device includes: Rotary power unit; An output shaft is connected to the output end of the rotational power device, and the output shaft is vertically arranged. The upper end of the output shaft is connected to the rotation center of the tray; A limiting plate is installed above the support plate, and a hole is provided on the limiting plate for the water pipe to pass through; The water pump is installed above the limiting plate.

4. The floating monitoring platform for light wave water environments according to claim 1, characterized in that: The outer side of the water pipe release device is also equipped with a set of water pipe correction rollers for guiding the water pipe to move vertically downward.

5. The floating monitoring platform for light wave water environments according to claim 1, characterized in that: The water suction head includes a water suction plate and several branch pipes. The water suction plate is provided with no less than 2 water suction holes. Each water suction hole is provided with a filter screen. The first ends of the several branch pipes are respectively connected to the water suction holes, and the second ends of the branch pipes are collected and connected to the water pipe.

6. A hydrological monitoring method, using a floating monitoring platform for light wave water environments as described in any one of claims 1-5, characterized in that: The method includes: Acquire historical sedimentation data of the target water area, analyze the spatial characteristics of sedimentation distribution and the temporal trend of sedimentation rate, and identify high-incidence areas of sedimentation and key points of rate change. Based on the characteristics of siltation distribution and the trend of rate change, the initial locations and number of sensors for monitoring siltation are preset to ensure coverage of high-incidence areas and key points. Based on the measurement range and accuracy parameters of the sensors used to monitor silt, the monitoring coverage and error range of each preset location are calculated, and the deployment location and number are adjusted to optimize the coverage effect of the monitoring network. By simulating the trend of sedimentation rate changes, we can predict possible changes in future sedimentation distribution and determine whether it is necessary to increase or adjust the location and number of sensors for monitoring sediment. Clustering analysis, a machine learning algorithm, is used to cluster historical sedimentation data, identify the main patterns and changes in sedimentation distribution, and assist in optimizing the sensor deployment scheme for monitoring sediment. By using regression analysis in machine learning algorithms, a model of the relationship between sedimentation rate and time is established to predict the future trend of sedimentation rate changes, providing a basis for dynamic adjustment of sensor deployment for monitoring sediment.

7. The hydrological monitoring method according to claim 6, characterized in that: The process involves acquiring historical siltation data of the target water area, analyzing the spatial characteristics of siltation distribution and the temporal trend of siltation rate, and identifying high-incidence siltation areas and key points of rate change, including: Acquire historical siltation data for the target water area and define the dataset boundaries based on the water area's extent; Spatial analysis methods are used to calculate the spatial characteristics of siltation distribution and generate a spatial distribution map; The time-varying trend of sedimentation rate is calculated using time analysis methods, and a time trend graph is generated. By combining spatial distribution maps, determine the boundaries and locations of areas prone to siltation; Based on the time trend chart, identify the key time points of change in the siltation rate; By integrating high-incidence areas and key time points, a time-series distribution map of high-incidence areas of siltation is generated; Based on time series distribution maps, a predictive model for areas prone to siltation is established; The initial locations and number of sensors for monitoring silt are preset based on the silt distribution characteristics and rate change trends to ensure coverage of high-incidence areas and key points, including: Acquire historical sedimentation data for the target area and extract the spatial characteristics of sedimentation distribution; Analyze time series data to determine the changing trend pattern of sedimentation rate; Based on spatial characteristics and trends, draw a spatial distribution map of areas prone to siltation; Based on the distribution characteristics of high-incidence areas, the sensor deployment locations for monitoring silt were determined. If the siltation rate changes drastically, increase the number of sensors pre-installed to monitor the silt. The spatial density calculation method was used to optimize the sensor deployment scheme for monitoring silt and ensure coverage of key points; Data collection is carried out through a pre-set plan to establish a siltation monitoring and analysis model.

8. The hydrological monitoring method according to claim 6, characterized in that: The process of calculating the monitoring coverage and error range of each preset location based on the measurement range and accuracy parameters of the sensors used to monitor silt, adjusting the deployment location and number of sensors, and optimizing the coverage effect of the monitoring network includes: The measurement range and accuracy parameters of the sensors for monitoring silt are obtained, and the monitoring coverage range is calculated using a spatial coverage algorithm for each preset location. Based on the accuracy parameters, the error range for each preset position is calculated using an error propagation model; Determine whether the monitoring coverage area meets the preset threshold; if not, adjust the deployment location and quantity. Using network optimization algorithms, the monitoring coverage and error range are recalculated for the adjusted deployment locations and quantities; Based on the recalculated results, a comprehensive evaluation model is used to determine the coverage effect of the monitoring network; If the coverage effect is not optimal, return to the step of adjusting the deployment location and quantity; Determine the final deployment plan and output the optimized monitoring network coverage effect.

9. The hydrological monitoring method according to claim 6, characterized in that: The method employs clustering analysis from machine learning algorithms to cluster historical sedimentation data, identify the main patterns and variation laws of sedimentation distribution, and assist in optimizing the sensor deployment scheme for monitoring sedimentation, including: The raw dataset is obtained from historical backlog data, and preprocessed to remove noise and standardize the data size. The K-means clustering algorithm was used to perform cluster analysis on the processed data, and Euclidean distance was selected as the distance metric. The number of cluster centers is determined based on data characteristics, and a cluster distribution map and change rate statistics are generated. If the clustering results show significant regional differences in the data distribution, then adjust the placement of the sensors monitoring the silt. By optimizing the deployment of monitoring points, the coverage of siltation data monitored by the sensors can be improved. Generate a sensor distribution map for monitoring silt based on the final deployment plan to achieve efficient monitoring; The aforementioned method utilizes regression analysis in machine learning algorithms to establish a model relating siltation rate to time, predicting future trends in siltation rate changes and providing a dynamic adjustment basis for the deployment of sensors monitoring silt. This includes: Obtain the rate values ​​on the historical timeline and use linear regression to establish the relationship between the rate values ​​and the timeline. Calculate the predicted value based on the relationship, and determine the amount of change in the rate value; Obtain the location information of the sensor deployment points for monitoring silt, and calculate the adjustment amount based on the predicted values; Determine the relationship between the change and the preset threshold. If the change exceeds the threshold, update the location information of the deployment point. Based on the updated deployment point information, the adjustment amount is recalculated to determine a new sensor deployment scheme for monitoring silt. The decision tree algorithm is used to classify the adjustment quantities and determine the dynamic indicators; Based on the dynamic indicators, the model quantity is optimized, and the final sensor deployment scheme for monitoring silt is output.

Images