Port siltation prediction system based on multi-source data fusion

Abstract

This invention relates to the field of data processing technology, and more particularly to a port siltation prediction system based on multi-source data fusion. The information analysis module confirms siltation channel information and determines whether to adjust monitoring points and their locations based on changes in the siltation channel information within a target period. It also calculates channel fluctuation parameters based on channel information within the target period to adjust the data acquisition frequency of each monitoring point. The model control module confirms the accuracy of the siltation prediction model based on measured data from each monitoring point and a three-dimensional water depth model to determine whether to adjust it before calculating port siltation. This invention considers the impact of changes in port channels and varying traffic volumes on port siltation, thereby improving the accuracy of port siltation prediction.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a port siltation prediction system based on multi-source data fusion. Background Technology

[0002] Port siltation prediction primarily relies on a variety of methods, including physical models, numerical simulations, and empirical formulas. Modern technologies increasingly employ advanced numerical simulation tools such as MIKE 21 and Delft3D for modeling. These tools comprehensively consider the interactions of factors such as waves, tidal currents, wind forces, and sediment characteristics, providing more accurate predictions. Furthermore, with the development of multi-source data fusion technologies and machine learning algorithms, data-driven models built using remote sensing, sensor networks, and historical observation data are increasingly being applied to siltation prediction, aiming to improve prediction accuracy and adapt to the specific conditions of different ports.

[0003] Chinese Patent Publication No. CN118500362B discloses a method and system for real-time monitoring of water depth in port channels. It includes: S1, selecting monitoring points within the channel and installing intelligent buoys at these points to monitor channel water data; simultaneously deploying an acoustic Doppler current profiler to monitor bottom current velocity; S2, collecting channel water data from the sonar system and acoustic Doppler current profiler; S3, constructing a channel sedimentation model and, combined with preprocessed real-time channel water data, introducing a time-dependent function to correct measurement errors caused by siltation in the real-time water depth data; S4, based on steps S1-S3, constructing a network containing multiple monitoring points to form a three-dimensional monitoring system for the channel. By establishing a preliminary channel sedimentation model based on sedimentation rate and average bottom current velocity, and introducing a time-dependent function and topographic factors, the accuracy of predicting future channel water depth changes is improved. Therefore, the existing technology has the following problems:

[0004] The impact of changes in port channels and varying traffic volumes on port siltation volume was not considered, which led to the low accuracy of port siltation volume prediction. Summary of the Invention

[0005] To address this issue, the present invention provides a port siltation prediction system based on multi-source data fusion, which overcomes the problem in the prior art that does not consider the impact of changes in port channels and different traffic volumes on port siltation, thus affecting the accuracy of port siltation prediction.

[0006] To achieve the above objectives, the present invention provides a port siltation prediction system based on multi-source data fusion, comprising:

[0007] A data acquisition module that obtains information on the siltation range and waterway of the port, collects measured data, and stores historical siltation data;

[0008] Connected to the data acquisition module, this information analysis module determines the monitoring points and their locations based on the siltation range.

[0009] A model control module connected to the data acquisition module for controlling the siltation prediction model;

[0010] The model control module is equipped with a siltation prediction model based on a three-dimensional water depth model that predicts the siltation range using historical siltation data.

[0011] The information analysis module confirms the silted-up channel information and determines whether to adjust the monitoring points and the location of the monitoring points based on the changes in the silted-up channel information within the target period.

[0012] In addition, channel fluctuation parameters are calculated based on channel information within the target period to adjust the data acquisition frequency of each monitoring point;

[0013] The model control module determines the accuracy of the siltation prediction model based on the measured data of each monitoring point and the three-dimensional water depth model, and then calculates the port siltation volume after deciding whether to control it.

[0014] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the data acquisition module includes an information acquisition unit, a data monitoring unit, and a data storage unit.

[0015] The information acquisition unit is used to acquire information on the siltation range and waterway of the port;

[0016] The data monitoring unit is used to collect measured data at each monitoring point;

[0017] The data storage unit is used to store historical siltation data of the port;

[0018] The waterway information includes the waterway and the number of navigations in each waterway, and the historical siltation data includes the historical siltation volume and the corresponding hydrological data.

[0019] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the information analysis module confirms the channel information within the siltation range as siltation channel information and determines whether to adjust the monitoring points based on the changes in each siltation channel within the target period.

[0020] The method for adjusting monitoring points is determined based on the number of silted-up channels within the target period, including:

[0021] In response to the fact that the number of different siltation channels is less than the preset number, the information analysis module adds a second monitoring point based on the different siltation channels;

[0022] In response to the number of different siltation channels being greater than or equal to a preset number, the information analysis module redetermines the monitoring points, including a first monitoring point and a second monitoring point;

[0023] The monitoring points include a first monitoring point and a second monitoring point.

[0024] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the information analysis module determines the corresponding siltation impact area according to each siltation channel and divides the siltation impact range into several monitoring areas according to the overlap between the siltation impact areas.

[0025] The overlap between adjacent monitoring areas varies, and the siltation impact range includes the location range of all the siltation impact areas.

[0026] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the information analysis module determines the number of first monitoring points within each monitoring area based on the overlap of the monitoring areas to determine the location of the first monitoring points.

[0027] The number of the first monitoring points is positively correlated with the degree of overlap of the corresponding monitoring areas.

[0028] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the information analysis module determines the turning point of each siltation channel according to the information of each siltation channel, and determines the location of the second monitoring point of the corresponding siltation channel according to the location and turning direction of the turning point.

[0029] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the information analysis module determines the total number of voyages based on channel information within the target period, and determines channel fluctuation parameters based on the total number of voyages and the preset number of voyages to adjust the data acquisition frequency of each monitoring point, including:

[0030] Based on the determination result that the channel fluctuation parameter is greater than the preset fluctuation range, the data acquisition frequency of each monitoring point is adjusted so that the adjusted data acquisition frequency is greater than the data acquisition frequency of the previous target cycle.

[0031] Based on the determination result that the channel fluctuation parameter is less than the preset fluctuation range, the data acquisition frequency of each monitoring point is adjusted so that the adjusted data acquisition frequency is less than the data acquisition frequency of the previous target cycle.

[0032] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the information analysis module determines the total number of voyages based on channel information within the target period, and determines channel fluctuation parameters based on the total number of voyages and the preset number of voyages to adjust the data acquisition frequency of each monitoring point. It also includes:

[0033] Based on the determination result that the channel fluctuation parameter is within the preset fluctuation range, it is determined that the data acquisition frequency of each monitoring point will not be adjusted so that it is the data acquisition frequency of the previous target cycle.

[0034] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the model control module determines the predicted water depth data of each monitoring point according to the three-dimensional water depth model, and determines the water depth data deviation of each monitoring point by combining the measured water depth data. Based on the water depth data deviation of each monitoring point, the module calculates its average value, standard deviation and average deviation to determine the location attributes of the corresponding monitoring point.

[0035] The location attributes include abnormal monitoring points and normal monitoring points.

[0036] As a preferred technical solution for a port siltation prediction system based on multi-source data fusion, the model control module determines whether to use the measured data of each monitoring point to retrain the siltation prediction model to re-predict the three-dimensional water depth model of the siltation range, and calculates the port siltation volume based on the re-predicted three-dimensional water depth model.

[0037] Compared with the prior art, the beneficial effects of the present invention are that the port siltation volume prediction system based on multi-source data fusion provided by the present invention effectively improves the problem of insufficient accuracy of traditional prediction in terms of dynamic monitoring, intelligent control, and precise analysis. The present invention confirms the siltation channel information through the information analysis module and adjusts the monitoring point position according to the changes in channel information within the target period, breaking through the limitation of traditional prediction not considering the dynamic changes of the channel. For situations such as channel curvature and widening, the system can adjust the monitoring layout in a timely manner to ensure that data collection covers key areas, accurately capture the fluctuations in siltation volume caused by channel changes, and significantly improve the consistency between prediction and actual situation.

[0038] In particular, this invention calculates channel fluctuation parameters based on channel information within the target period, and then adjusts the data acquisition frequency of each monitoring point, fully considering the impact of differences in the number of ships on the amount of siltation. During periods of high ship traffic, the system automatically increases the monitoring frequency to track changes in sediment movement caused by factors such as ship waves and propeller churning in real time. During periods of low traffic, the frequency is reasonably reduced to improve monitoring efficiency while ensuring data validity, making the prediction results more consistent with the actual siltation situation.

[0039] In particular, the model control module combines the measured data from each monitoring point with the three-dimensional water depth model to determine and control the accuracy of the siltation prediction model. When changes in the waterway or navigation affect the prediction, the system can use the measured data to retrain the model and optimize the parameters to ensure that the model always adapts to the dynamic changes in the port environment, effectively avoiding prediction inaccuracies caused by changes in external factors, and achieving high-precision and intelligent prediction of port siltation.

[0040] In particular, the information analysis module quantifies the complexity of the siltation environment by measuring the overlap of the siltation impact area, and dynamically adjusts the number of primary monitoring points based on this (overlap is positively correlated with the number of points). This strategy can accurately match the siltation risk level of different areas: areas with high overlap are often affected by the superposition of factors such as multi-channel water flow and ship disturbance, resulting in a more complex siltation mechanism. Dense deployment of monitoring points can capture the details of sediment movement under the coupling of multiple dynamic sources. In contrast, areas with low overlap reduce redundant monitoring and avoid resource waste. This approach breaks through the limitations of traditional fixed-density monitoring, ensuring that the layout of monitoring points is deeply aligned with the actual impact intensity of siltation, improving resource utilization efficiency while ensuring the integrity of data collection in high-risk areas.

[0041] In particular, the information analysis module delineates a fan-shaped second monitoring area based on the changes in water flow velocity and sediment deposition characteristics caused by ship turning. This area precisely covers the high-risk zone (turning side) where the flow velocity slows down due to water flow compression during ship turning, resulting in sediment deposition, while also taking into account the scouring changes on the non-turning side. At least three monitoring points are scientifically arranged within the area (one point on the circle and two points within the area). Through multi-point collaborative monitoring, it can capture the differences in siltation at different locations around the turning point (such as changes in siltation thickness near the center of the circle and the scouring effect at the circle boundary), and cross-verify the accuracy of the data, avoiding the random errors of single-point monitoring.

[0042] In particular, the processing mechanism of the model control module avoids the waste of resources or insufficient optimization caused by the traditional "one-size-fits-all" model optimization. It can quickly fix the prediction defects of the model in high-complexity areas and maintain the stability of the model in stable areas, significantly improving the accuracy and adaptability of the siltation prediction model and providing more reliable data support for port dredging decisions. Attached Figure Description

[0043] Figure 1 This is a connection diagram of the port siltation prediction system based on multi-source data fusion, as described in an embodiment of the present invention.

[0044] Figure 2 This is a flowchart illustrating the workflow of the information analysis module in an embodiment of the present invention.

[0045] Figure 3 This is a flowchart of the model control module in an embodiment of the present invention. Detailed Implementation

[0046] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0047] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0048] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0049] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0050] Please see Figure 1 The diagram shown is a connection diagram of a port siltation prediction system based on multi-source data fusion according to an embodiment of the present invention. This embodiment of the present invention provides a port siltation prediction system based on multi-source data fusion, comprising:

[0051] A data acquisition module that obtains information on the siltation range and waterway of the port, collects measured data, and stores historical siltation data;

[0052] Connected to the data acquisition module, this information analysis module determines the monitoring points and their locations based on the siltation range.

[0053] A model control module connected to the data acquisition module for controlling the siltation prediction model;

[0054] The model control module is equipped with a siltation prediction model that predicts the siltation range using a three-dimensional water depth model based on historical siltation data. It is understood that siltation prediction models are already used in the prior art, and are usually used to predict and simulate based on hydrological data and the corresponding siltation volume. This invention uses any siltation prediction model in the prior art and determines its accuracy based on measured data. When the accuracy is lower than the standard, the siltation prediction model is retrained using measured data, and a three-dimensional water depth model is generated based on the trained model to re-predict the port siltation volume.

[0055] The information analysis module confirms the siltation channel information and determines whether to adjust the monitoring points and their locations based on changes in the siltation channel information within the target period. It is understood that the target period is typically 1 to 3 months; that is, after each target period ends, the monitoring point locations and monitoring frequency for the next target period are determined based on the channel information of the current and previous target periods. In practice, channel information typically does not change significantly, therefore the target period is usually 3 months.

[0056] In addition, channel fluctuation parameters are calculated based on channel information within the target period to adjust the data acquisition frequency of each monitoring point;

[0057] The model control module determines the accuracy of the siltation prediction model based on the measured data of each monitoring point and the three-dimensional water depth model, and then calculates the port siltation volume after deciding whether to control it.

[0058] It is understood that the steps in application of this invention include: S1, predicting a three-dimensional water depth model of the siltation range based on historical siltation data; S2, determining the siltation channel information within the siltation range based on the siltation range and channel information within the target period, and determining the location of monitoring points based on the siltation channel information; S3, determining channel fluctuation parameters based on channel information within the target period to adjust the data acquisition frequency of each monitoring point, and enabling the data monitoring unit to work based on the data acquisition frequency; S4, adjusting the siltation prediction model based on the measured data of each monitoring point and the three-dimensional water depth model to calculate the port siltation volume based on the adjusted siltation prediction model.

[0059] It is understandable that existing technologies typically use a combination of multi-source data such as topography, dynamics, and sediment to accurately define the siltation range in order to provide a quantitative basis for dredging projects. However, the method of determination is not the technical problem that this application aims to solve, so it will not be elaborated upon here.

[0060] Specifically, the data acquisition module includes an information acquisition unit, a data monitoring unit, and a data storage unit;

[0061] The information acquisition unit is used to acquire information on the siltation range and waterway of the port;

[0062] The data monitoring unit is used to collect measured data at each monitoring point;

[0063] The data storage unit is used to store historical siltation data of the port;

[0064] The waterway information includes the waterway and the number of navigations in each waterway, and the historical siltation data includes the historical siltation volume and the corresponding hydrological data.

[0065] Please see Figure 2The diagram shown is a flowchart of the information analysis module in an embodiment of the present invention. Specifically, the information analysis module identifies the channel information within the siltation range as siltation channel information and determines whether to adjust the monitoring points based on the changes in each siltation channel within the target period.

[0066] The method for adjusting monitoring points is determined based on the number of silted-up channels within the target period, including:

[0067] In response to the fact that the number of different siltation channels is less than the preset number, the information analysis module adds a second monitoring point based on the different siltation channels;

[0068] In response to the number of different siltation channels being greater than or equal to a preset number, the information analysis module redetermines the monitoring points, including a first monitoring point and a second monitoring point;

[0069] The monitoring points include a first monitoring point and a second monitoring point.

[0070] It should be understood that the channel information includes the complete channel path, which spans both the siltation area and the non-siltation area. Therefore, the portion within the siltation area is defined as the siltation channel information, and dredging is only carried out within the siltation area. Thus, this invention only considers the siltation channel information.

[0071] It is understood that changes in siltation channels refer to newly added or altered siltation channels within the target period. In implementation, an altered siltation channel must have a distance greater than or equal to 0.05 nautical miles from the original siltation channel. The steps for determining the distance between the siltation channel and the original siltation channel in implementation are as follows: 1) Determine the starting and ending points and lengths of both the siltation channel and the original siltation channel; 2) Using the starting points of both siltation channels as the origin, mark points of equal proportion on both channels as corresponding points (e.g., one-tenth of the length of the siltation channel and the original siltation channel, respectively: the position starting from the origin of the siltation channel and having a length of one-tenth of the siltation channel length, and the position starting from the origin of the original siltation channel and having a length of one-tenth of the original siltation channel length); 3) Determine the straight-line distance between each corresponding point; 4) Determine whether the siltation channel has changed based on whether the average straight-line distance between each corresponding point is greater than or equal to 0.05 nautical miles.

[0072] Understandably, the information analysis module flexibly adjusts monitoring points based on changes (new additions or significant alterations) in the siltation channel within the target period, breaking through the limitations of traditional fixed monitoring modes. When new siltation channels appear or existing channels undergo significant changes, this invention can promptly capture and deploy monitoring points accordingly, ensuring comprehensive coverage of the dynamic siltation area. Differential adjustment strategies are implemented based on the number of different siltation channels to achieve efficient allocation of monitoring resources. When the number of different siltation channels is small, local monitoring density is optimized by adding a second monitoring point. When the number of different siltation channels is large, the channel conditions become more complex, requiring a complete replanning of the first and second monitoring points. Balancing the overall picture with details, this tiered approach improves monitoring efficiency in simple scenarios while ensuring accuracy in complex ones, avoiding resource waste or insufficient monitoring and ensuring a precise match between the monitoring layout and the actual complexity of siltation. Furthermore, a scientifically quantifiable method is used to determine whether the siltation channel has changed, providing an objective basis for adjusting monitoring points. A threshold of 0.05 nautical miles effectively distinguishes between normal channel fluctuations and substantial changes, avoiding resource waste caused by frequent adjustments to monitoring points due to minor fluctuations. Simultaneously, it ensures that significantly changed channels are promptly included in monitoring, achieving a balance between stability and sensitivity, and enhancing the reliability and practicality of the siltation monitoring system.

[0073] Specifically, the information analysis module determines the corresponding siltation impact area based on each siltation channel and divides the siltation impact range into several monitoring areas based on the overlap between the siltation impact areas.

[0074] The overlap between adjacent monitoring areas varies, and the siltation impact range includes the location range of all the siltation impact areas.

[0075] It is understandable that the area within a preset distance (0.05 nautical miles to 0.08 nautical miles, i.e., 93 meters to 148 meters) on both sides of the siltation channel is determined as the siltation impact area of ​​the siltation channel; preferably, the preset distance is set to 0.6 nautical miles, and the size of the preset distance is adjusted according to the angle between the channel and the tidal current during implementation. If the angle between the channel and the tidal current is greater than 30°, the preset distance is preferably 0.7 nautical miles.

[0076] Specifically, the information analysis module determines the number of first monitoring points within each monitoring area based on the overlap of the monitoring areas to determine the location of the first monitoring points;

[0077] The number of the first monitoring points is positively correlated with the degree of overlap of the corresponding monitoring areas.

[0078] Understandably, for monitoring areas with an overlap of 1 (i.e., no overlap and belonging to only one siltation influence range), there should be at least one first monitoring point per 1000 square meters; for monitoring areas with an overlap of 2 (i.e., two siltation influence ranges overlap), there should be at least two first monitoring points per 1000 square meters; for monitoring areas with an overlap of 3 (i.e., three siltation influence ranges overlap), there should be at least three first monitoring points per 1000 square meters, and so on. Therefore, in implementation, the number of first monitoring points in a monitoring area can be determined based on the overlap and area of ​​each monitoring area. The first monitoring points in a single monitoring area are evenly distributed to more completely and accurately determine the accuracy of the three-dimensional water depth model of the monitoring area through measured data.

[0079] Understandably, by adding a first monitoring point and distributing it evenly in a highly overlapping monitoring area, more densely measured data (such as water depth, flow velocity, and sediment content) can be obtained, thereby significantly improving the accuracy of the three-dimensional water depth model. In addition, evenly distributed monitoring points can cover all directions of the siltation-affected area, avoiding data blind spots, and enabling the model to accurately reflect the spatial distribution differences in siltation thickness (such as the siltation characteristics of channel intersections). The matching degree between measured data and the model is thus strengthened, thereby improving the adaptability of the siltation prediction model to complex scenarios, reducing prediction bias caused by insufficient monitoring data, and providing a more reliable quantitative basis for dredging projects.

[0080] Specifically, the information analysis module determines the turning point of each siltation channel based on the information of each siltation channel, and determines the location of the second monitoring point of the corresponding siltation channel based on the location of the turning point and the turning direction.

[0081] It is understandable that when a ship turns, the turning side of the hull will compress the water flow, slowing down the water flow speed on the turning side. According to the relationship between the sand-carrying capacity of water flow and flow velocity, the reduced flow velocity will lead to a decrease in sand-carrying capacity, making it easier for sediment to be deposited on the turning side of the hull. At the same time, the ship's turn will create a relatively low-pressure zone on the non-turning side of the hull, with a relatively faster water flow speed and enhanced sand-carrying capacity. The non-turning side may experience scouring rather than siltation.

[0082] In practice, a circle with a radius of 100 meters is drawn with the turning point as the center. The sector formed by this circle and the straight line in the direction before and after the turn is defined as the second monitoring area of ​​the turning point. At least three second monitoring points are determined within the second monitoring area, one of which is located on the circle, and the other two are located at any position inside the second monitoring area.

[0083] Understandably, the information analysis module, based on the changes in water flow velocity and sediment deposition characteristics caused by ship turning, delineates a fan-shaped second monitoring area with a radius of 100 meters centered on the turning point. This area precisely covers the high-risk zone (turning side) where water flow slows down and sediment deposition occurs due to ship turning, while also taking into account scouring changes on the non-turning side. At least three monitoring points are scientifically deployed within this area (one on the circle + two within the area). Through multi-point collaborative monitoring, it can capture differences in siltation at different locations around the turning point (such as changes in siltation thickness near the center and the scouring effect at the circle's boundary), and cross-validate data accuracy, avoiding the random errors of single-point monitoring. Compared to traditional uniform or random monitoring, this method achieves high-density, full-coverage monitoring of key siltation areas at turning points with minimal monitoring cost, significantly improving the targeting and effectiveness of siltation data collection, and providing reliable support for accurately predicting siltation volume in turning areas and optimizing dredging plans.

[0084] Specifically, the information analysis module determines the total number of voyages based on the channel information within the target period, and determines channel fluctuation parameters based on the total number of voyages and the preset number of voyages to adjust the data acquisition frequency of each monitoring point, including:

[0085] Based on the determination result that the channel fluctuation parameter is greater than the preset fluctuation range, the data acquisition frequency of each monitoring point is adjusted so that the adjusted data acquisition frequency is greater than the data acquisition frequency of the previous target cycle.

[0086] Based on the determination result that the channel fluctuation parameter is less than the preset fluctuation range, the data acquisition frequency of each monitoring point is adjusted so that the adjusted data acquisition frequency is less than the data acquisition frequency of the previous target cycle.

[0087] In practice, the preset number of voyages is the total number of voyages in the previous target period, and the channel fluctuation parameter = |total number of voyages - preset number of voyages| ÷ preset number of voyages × 100%;

[0088] In practice, the preset fluctuation range is [15%, 30%].

[0089] Specifically, the information analysis module determines the total number of voyages based on the channel information within the target period, and determines the channel fluctuation parameter based on the total number of voyages and the preset number of voyages to adjust the data acquisition frequency of each monitoring point. It also includes:

[0090] Based on the determination result that the channel fluctuation parameter is within the preset fluctuation range, it is determined that the data acquisition frequency of each monitoring point will not be adjusted so that it is the data acquisition frequency of the previous target cycle.

[0091] Understandably, the initial data collection frequency is once every 10 days. At each monitoring node, the underwater robot is controlled to move to each monitoring point to collect measured water depth data. When the channel fluctuation parameter is greater than the preset fluctuation range, the data collection frequency = channel fluctuation parameter ÷ 30% × the data collection frequency of the previous target period. When the channel fluctuation parameter is less than the preset fluctuation range, the data collection frequency = channel fluctuation parameter ÷ 15% × the data collection frequency of the previous target period.

[0092] Understandably, the method of dynamically determining the data acquisition frequency based on channel fluctuation parameters achieves dynamic adaptation between monitoring resources and the impact of ship navigation, optimizing monitoring efficiency while ensuring the effectiveness of siltation data. This invention accurately identifies the intensity of the impact of ship activities on siltation by quantitatively analyzing the difference in the number of voyages between the target period and historical periods (measured by channel fluctuation parameters). When the fluctuation parameter exceeds the preset range, the acquisition frequency is automatically increased or decreased. For example, during periods of high ship traffic (fluctuation parameter greater than 30%), the acquisition frequency will increase to capture the dynamic changes in sediment caused by factors such as ship waves and propeller churning. During periods of low traffic (fluctuation parameter less than 15%), the frequency is reasonably reduced to reduce resource consumption. This adaptive adjustment mechanism avoids the data redundancy / monitoring lag problems caused by traditional fixed-frequency monitoring, ensuring the timeliness and completeness of siltation data during high-risk periods and reducing equipment operating costs during stable periods, significantly improving the intelligence and economy of port siltation monitoring.

[0093] Please see Figure 3 The diagram shown is a flowchart of the model control module in an embodiment of the present invention. Specifically, the model control module determines the predicted water depth data for each monitoring point based on the three-dimensional water depth model, combines the measured water depth data to determine the water depth data deviation for each monitoring point, and calculates the average value, standard deviation, and average deviation based on the water depth data deviation for each monitoring point to determine the location attributes of the corresponding monitoring point.

[0094] The location attributes include abnormal monitoring points and normal monitoring points.

[0095] In implementation, the deviation fluctuation parameter is determined based on the ratio of the average deviation to the average value. If the deviation fluctuation parameter is greater than the preset fluctuation value, the location attribute of the monitoring point is determined based on the average value and the standard deviation. If the deviation fluctuation parameter is less than or equal to the preset fluctuation value, the location attribute of the monitoring point is determined based on the average value. In implementation, the preset fluctuation value is usually between 0.05 and 0.15, and is preferably set to 0.1.

[0096] Understandably, the deviation fluctuation parameter represents the difference between the measured data and the predicted data at each monitoring point. A larger deviation fluctuation parameter indicates that the data deviation varies greatly between monitoring points, meaning that the model's prediction effect is not good. Therefore, more measured data is needed to optimize and train the model. On the other hand, a smaller deviation fluctuation parameter indicates that the data deviation between monitoring points is more similar. This means that although the determined model has deviations, it fits the measured terrain better, so there is no need for a large amount of measured data to optimize and train it.

[0097] Understandably, if the deviation fluctuation parameter is greater than the preset fluctuation value, the location attribute of the monitoring point is determined based on the difference between the average value and the standard deviation: if the water depth data deviation of a single monitoring point is greater than the difference between the average value and the standard deviation, the monitoring point is determined to be an abnormal monitoring point; if the water depth data deviation of a single monitoring point is less than or equal to the difference between the average value and the standard deviation, the monitoring point is determined to be a normal monitoring point.

[0098] Understandably, if the deviation fluctuation parameter is less than or equal to the preset fluctuation value, the location attribute of the monitoring point is determined based on the average value: if the water depth data deviation of a single monitoring point is greater than the average value, the monitoring point is determined to be an abnormal monitoring point; if the water depth data deviation of a single monitoring point is less than or equal to the average value, the monitoring point is determined to be a normal monitoring point.

[0099] Understandably, the model control module scientifically and quantitatively analyzes the deviation of water depth data at monitoring points to accurately identify the inaccuracies in model predictions, thereby achieving intelligent and differentiated optimization of the siltation prediction model. This module intelligently judges the dispersion of data deviation at each monitoring point by calculating the deviation fluctuation parameter: when the deviation fluctuation parameter is large, it indicates that the model's prediction effect varies significantly in different areas. In this case, the monitoring points are loosely screened by combining the mean and standard deviation to identify more abnormal monitoring points, and targeted supplementary measured data are used for in-depth optimization training. When the deviation fluctuation parameter is small, it indicates that the overall model fit is good, and only a few monitoring points with large deviations need to be fine-tuned based on the mean. This hierarchical processing mechanism avoids the resource waste or under-optimization problems caused by the traditional "one-size-fits-all" model optimization. It can quickly correct the model's prediction defects in high-complexity areas and maintain model stability in stable areas, significantly improving the accuracy and adaptability of the siltation prediction model and providing more reliable data support for port dredging decisions.

[0100] Specifically, the model control module determines whether to use the measured data of each monitoring point to retrain the siltation prediction model based on the location attributes of each monitoring point, so as to re-predict the three-dimensional water depth model of the siltation range, and calculate the port siltation volume based on the re-predicted three-dimensional water depth model.

[0101] Understandably, for abnormal monitoring points, the coordinates and measured water depth data of the abnormal monitoring points are all input into the three-dimensional water depth model for retraining, so as to obtain a three-dimensional water depth model that better reflects the current actual situation, and the corresponding backfilling volume is determined based on the three-dimensional water depth model.

[0102] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0103] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A port siltation prediction system based on multi-source data fusion, comprising: A data acquisition module that obtains information on the siltation range and waterway of the port, collects measured data, and stores historical siltation data; Connected to the data acquisition module, this information analysis module determines the monitoring points and their locations based on the siltation range. A model control module connected to the data acquisition module for controlling the siltation prediction model; The model control module is equipped with a siltation prediction model based on a three-dimensional water depth model that predicts the siltation range using historical siltation data. Its characteristic is that it further includes: The information analysis module confirms the silted-up channel information and determines whether to adjust the monitoring points and their locations based on changes in the silted-up channel information within the target period, including: The information analysis module identifies the channel information within the siltation range as siltation channel information and determines whether to adjust the monitoring points based on the changes in each siltation channel within the target period. The method for adjusting monitoring points is determined based on the number of silted-up channels within the target period, including: In response to the fact that the number of different siltation channels is less than the preset number, the information analysis module adds a second monitoring point based on the different siltation channels; In response to the number of different siltation channels being greater than or equal to a preset number, the information analysis module redetermines the monitoring points, including a first monitoring point and a second monitoring point; The monitoring points include a first monitoring point and a second monitoring point; The information analysis module determines the corresponding siltation impact area based on each siltation channel, divides the siltation impact range into several monitoring areas based on the overlap between the siltation impact areas, and determines the number of first monitoring points in each monitoring area based on the overlap between the monitoring areas to determine the location of the first monitoring points. Among them, the degree of overlap between adjacent monitoring areas is different, the siltation impact range includes the location range of all the siltation impact areas, and the number of the first monitoring points is positively correlated with the degree of overlap of the corresponding monitoring areas; The information analysis module determines the turning point of each siltation channel based on the information of each siltation channel, and determines the location of the second monitoring point of the corresponding siltation channel based on the position and turning direction of the turning point. In addition, channel fluctuation parameters are calculated based on channel information within the target period to adjust the data acquisition frequency of each monitoring point; The model control module determines the accuracy of the siltation prediction model based on the measured data of each monitoring point and the three-dimensional water depth model, and then calculates the port siltation volume after deciding whether to control it.

2. The port siltation prediction system based on multi-source data fusion according to claim 1, characterized in that, The data acquisition module includes an information acquisition unit, a data monitoring unit, and a data storage unit; The information acquisition unit is used to acquire information on the siltation range and waterway of the port; The data monitoring unit is used to collect measured data at each monitoring point; The data storage unit is used to store historical siltation data of the port; The waterway information includes the waterway and the number of navigations in each waterway, and the historical siltation data includes the historical siltation volume and the corresponding hydrological data.

3. The port siltation prediction system based on multi-source data fusion according to claim 1, characterized in that, The information analysis module determines the total number of voyages based on the channel information within the target period, and determines channel fluctuation parameters based on the total number of voyages and the preset number of voyages to adjust the data acquisition frequency of each monitoring point, including: Based on the determination result that the channel fluctuation parameter is greater than the preset fluctuation range, the data acquisition frequency of each monitoring point is adjusted so that the adjusted data acquisition frequency is greater than the data acquisition frequency of the previous target cycle. Based on the determination result that the channel fluctuation parameter is less than the preset fluctuation range, the data acquisition frequency of each monitoring point is adjusted so that the adjusted data acquisition frequency is less than the data acquisition frequency of the previous target cycle.

4. The port siltation prediction system based on multi-source data fusion according to claim 3, characterized in that, The information analysis module determines the total number of voyages based on the channel information within the target period, and determines the channel fluctuation parameter based on the total number of voyages and the preset number of voyages to adjust the data acquisition frequency of each monitoring point. It also includes: Based on the determination result that the channel fluctuation parameter is within the preset fluctuation range, it is determined that the data acquisition frequency of each monitoring point will not be adjusted so that it is the data acquisition frequency of the previous target cycle.

5. The port siltation prediction system based on multi-source data fusion according to claim 1, characterized in that, The model control module determines the predicted water depth data of each monitoring point based on the three-dimensional water depth model, and determines the water depth data deviation of each monitoring point by combining the measured water depth data. Based on the water depth data deviation of each monitoring point, it calculates its average value, standard deviation and average deviation to determine the location attributes of the corresponding monitoring point. The location attributes include abnormal monitoring points and normal monitoring points.

6. The port siltation prediction system based on multi-source data fusion according to claim 5, characterized in that, The model control module determines whether to use the measured data of each monitoring point to retrain the siltation prediction model based on the location attributes of each monitoring point, so as to re-predict the three-dimensional water depth model of the siltation range, and calculate the port siltation volume based on the re-predicted three-dimensional water depth model.

Images