Coastal water area water level prediction and navigation capability evaluation method and system

Abstract

The invention relates to the technical field of navigation, in particular to a coastal water area water level prediction and navigation capability evaluation method and system, solves the problem that a single data source cannot comprehensively capture water level change influence factors, and forms a set of high-precision and high-comprehensiveness water level monitoring and prediction technology by integrating a GIS (Geographic Information System), ultrasonic detection and a deep learning model. The advantages of all data sources are fully utilized, fusion and optimization of water level, tide, weather and other multi-dimensional information are achieved, the applicability and reliability of the system in the complex marine environment are remarkably improved, and the safety and efficiency of navigation channel navigation are ensured.

Description

Technical Field

[0001] The present invention relates to the field of navigation technology, and in particular to a method and system for predicting the water level of coastal waters and evaluating the navigation capacity. Background Art

[0002] With the increasing frequency of economic activities in coastal areas, the monitoring and evaluation of the navigation capacity and safety of waterways have become increasingly important. In actual engineering applications, especially water transportation involving coastal shallow waters, such as during construction transportation, large ships cannot pass because the water level is shallow and it is not a conventional waterway. It is significantly affected by the tides, so the summary and prediction of the tidal water level change law is particularly critical. In order to ensure that the transportation operation can proceed smoothly, it is necessary to make full use of the characteristics of the rising water level of seawater backflow during high tide to ensure that the necessary transportation water level is reached.

[0003] The traditional waterway monitoring and management system has the following major problems: First, the traditional waterway monitoring method mainly relies on manual measurement of fixed-point water levels to estimate the water level of the entire waterway, resulting in low timeliness and accuracy of the monitoring data. Secondly, the traditional method cannot reasonably consider the time-lag properties of tidal rise and fall, and cannot comprehensively evaluate the impact of water level changes on the navigability of the waterway. Finally, the existing waterway auxiliary system lacks the function of automatic integration and processing of real-time data, data acquisition is not timely enough, it is difficult to update dynamically, and it cannot adapt to the complex situation of real-time changes in the waterway. Moreover, the existing waterway auxiliary system is also relatively limited in decision-making support for ship navigation. It usually only provides static water level information, lacks dynamic and comprehensive analysis of ship positions, water level changes and navigation paths, and cannot effectively avoid potential risks in the waterway. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for predicting water levels in coastal waters and evaluating navigability, in view of the fact that the existing waterway auxiliary systems in the prior art are also relatively limited in terms of decision support for ship navigation, usually only providing static water level information, lacking dynamic comprehensive analysis of ship positions, water level changes and navigation paths, and being unable to effectively avoid potential risks in waterways.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] In a first aspect, the present invention provides a method for predicting the water level of coastal waters and evaluating the navigability, comprising the following steps:

[0007] A plurality of measuring points are set along the waterway to obtain first information, the first information including water area information of the waterway, the water area information including geodetic coordinates and elevation of each measuring point, and water level information, and establish a three-dimensional model of underwater terrain of the waterway;

[0008] Based on the three-dimensional model, combined with the local water level historical monitoring data of each measuring point, the corresponding mapping relationship between the local point water level and the water level of each coordinate point of the waterway is established through the Splice-LSTM deep learning model, and the future prediction model of the coastal waterway is obtained;

[0009] The second information is obtained by using the virtual mouse technology. The second information includes the tide prediction information of the hydrological station and the weather forecast information. The tide prediction information includes the water level and flow velocity. The weather forecast information includes the daily rainfall.

[0010] In the Splice-LSTM model, water level, flow velocity, and daily rainfall are added as feature parameters for accuracy evaluation, and the weight factors of each feature parameter are obtained;

[0011] Based on the weight factors of each characteristic parameter, the error inverse weighted average method is used for correction, and a future prediction model for coastal waterways integrating multi-source information is established.

[0012] The future prediction model of coastal waterways that integrates multi-source information is used to evaluate the navigation capacity of the waterway, predict the navigation window period and warn of water level changes during the navigation process.

[0013] As a preferred technical solution of the present invention, the first information is obtained based on GIS+ultrasonic detection technology;

[0014] Based on the spline interpolation method, by fitting the water level data of known measurement points with multi-segment piecewise functions, continuous and smooth interpolation points are generated near the known measurement points.

[0015] The three-dimensional coordinate data of known measurement points and interpolation generated points are integrated together. The three-dimensional coordinates include geodetic coordinates and corresponding elevation information to establish the three-dimensional coordinates of the underwater terrain of the waterway.

[0016] As a preferred technical solution of the present invention, a number of observation water level gauges are arranged at the start, middle and end positions of the waterway to obtain water level information of a number of measuring points in real time, and the water level information includes time, flow rate and water level.

[0017] As a preferred technical solution of the present invention, based on the acquired multi-point historical water level information, the historical water level data is preprocessed to remove abnormal values ​​and obtain optimized data;

[0018] The feature extraction of time series data is performed based on the Splice-LSTM model architecture, which includes several submodules. The submodules include three Splice-LSTM layers and one fully connected layer. Each of the three Splice-LSTM layers contains at least 50 hidden units to capture the long-term dependencies in the time series data. The fully connected layer contains 1 neuron to output the prediction results.

[0019] Set the Splice-LSTM model parameters, including setting the training rounds to at least 500, normalizing the input data, and setting the normalization range of feature data and label data to [0,1];

[0020] The mean square error is used as the loss function to measure the error between the predicted value and the actual value of the Splice-LSTM model and optimize the prediction accuracy of the Splice-LSTM model.

[0021] After the training, the trained Splice-LSTM model was evaluated using the test dataset, and the mean square error was calculated to quantify the prediction performance of the model, and a future prediction model for coastal waterways was established.

[0022] As a preferred technical solution of the present invention, a virtual mouse mode is used to navigate to the corresponding tidal observation website and weather forecast website of the hydrological station around the waterway;

[0023] Locate specific data information on the web page, including water level, flow rate and daily rainfall;

[0024] Initialize the data storage list and set the initial mouse position;

[0025] Simulate moving the virtual mouse at the corresponding data chart position, and obtain the water level, flow rate, and daily rainfall data corresponding to each time period in the next 24 hours by continuously changing the mouse position.

[0026] As a preferred technical solution of the present invention, data is collected and preprocessed to obtain a training set, the data includes water level, flow rate, and daily rainfall, and the training set is evaluated using the Splice-LSTM model;

[0027] Based on the Splice-LSTM model, multiple single model evaluations are performed one by one based on the training set, and the prediction results of the Splice-LSTM model are analyzed. The final evaluation result label is expressed as follows:

[0028] H=α 1 w+α 2 f+α 3 p

[0029] Where H is the corresponding water level result in the full channel coordinate system output by the multi-source information complementary fusion model; α 1 , α 2 and α 3 They represent the corresponding water level results in the whole channel coordinate system obtained by taking water level w, flow velocity f and daily rainfall p as single characteristic parameters;

[0030] The weight factor corresponding to each single feature parameter:

[0031]

[0032] In the formula, ε 1 , ε 2 and ε 3 They respectively represent the accuracy of the corresponding water level results in the full channel coordinate system obtained by taking water level, flow velocity and daily rainfall as single characteristic parameters.

[0033] As a further preferred technical solution of the present invention, the weight factor α 1 , α 2 and α 3 They are 0.60, 0.31 and 0.09 respectively.

[0034] In a second aspect, the present invention further provides a coastal waters water level prediction and navigation capacity assessment system, using the coastal waters water level prediction and navigation capacity assessment method as described in any one of the above, the assessment system comprises:

[0035] The three-dimensional topographic map display module builds a three-dimensional model of the underwater terrain of the waterway through the first information, and realizes the dynamic interactive display of the underwater terrain based on the interactive visualization technology of the Plotly library;

[0036] The future water level prediction module, by setting a specified date, uses the virtual mouse technology to obtain the second information, performs data processing on the second information, and predicts the water level through the splice-LSTM model, and automatically predicts the future water level changes;

[0037] The waterway transportation window module automatically obtains the water level forecast data under the bridge from the future water level forecast module, combines the date and time information, generates the time series data of the predicted water level, analyzes the water level forecast data, selects the time period that meets the requirements based on the water level threshold conditions, uses the dynamic time segmentation algorithm to identify and calculate the continuous time window that meets the transportation requirements, automatically calculates the total duration of each continuous time period according to the time interval of the water level forecast data, and uses the preset continuous time threshold to select the effective transportation window period that meets the conditions;

[0038] The real-time water level change module is connected to the observation water level meter through the interface protocol, automatically receives the real-time water level data collected by the observation water level meter, and the real-time water level data is uploaded through the communication protocol to realize the real-time water level change correction;

[0039] The real-time navigable ship warning module includes an AIS based on an image recognition algorithm installed on the waterway, which accurately reads the size and draft parameters of ships entering the waterway, uses the satellite navigation system positioning technology configured on the ship, and combines the waterway water level data for precise spatial positioning. It uses the path optimization Dijkstra algorithm based on dynamic environment perception to calculate the best navigable route in real time, avoid areas in the waterway where the water level does not meet the ship's draft requirements, integrates water level perception and ship position data, monitors the relative position of the ship and the water level in real time, and automatically alarms when the ship's driving position approaches the critical value of the draft.

[0040] In a third aspect, the present invention further provides an electronic device, comprising:

[0041] a memory having a computer program stored thereon;

[0042] A processor is used to execute the program in the memory to implement the method for predicting the water level of coastal waters and evaluating the navigability as described in any of the above items.

[0043] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the method for predicting the water level of coastal waters and evaluating the navigability as described in any one of the above items is implemented.

[0044] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0045] The method and system for predicting the water level and assessing the navigability of coastal waters described in the present invention solve the problem that a single data source cannot fully capture the factors affecting water level changes. By integrating GIS, ultrasonic detection and deep learning models, a set of high-precision and comprehensive water level monitoring and prediction technologies is formed; and the advantages of various data sources are fully utilized to achieve the integration and optimization of multi-dimensional information such as water level, tide, and weather, which significantly improves the applicability and reliability of the system in complex marine environments and ensures the safety and efficiency of waterway navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flowchart of the method for predicting the water level and assessing the navigation capacity of coastal waters;

[0047] Figure 2 It is a three-dimensional schematic diagram of underwater terrain based on spline interpolation method;

[0048] Figure 3 It is a schematic diagram of real-time water level monitoring;

[0049] Figure 4 This is a schematic diagram of the prediction accuracy of the LSTM model;

[0050] Figure 5 This is a schematic diagram of the prediction accuracy of the Splice-LSTM model;

[0051] Figure 6 Schematic diagram of page information obtained for virtual mouse technology;

[0052] Figure 7 Schematic diagram of the display interface of the coastal waters water level prediction and navigation capacity assessment system. DETAILED DESCRIPTION

[0053] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0054] Unless otherwise specified, in the description of the specific embodiments of the present invention, the terms indicating the orientation or position relationship such as "up", "down", "left", "right", "center", "inside", "outside", etc. are all expressions based on the orientation or position relationship shown in the drawings, or are the orientation or position relationship when the invented product / equipment / device is usually used. These terms of orientation or position relationship are only for the convenience of describing the scheme of the present invention or simplifying the description in the specific embodiments, so as to facilitate the technicians to quickly understand the scheme, and do not indicate or imply that a specific device / component / element must have a specific orientation, or be constructed and operated in a specific position relationship, and therefore cannot be understood as a limitation on the present invention.

[0055] In addition, if the terms "horizontal", "vertical", "overhanging", "parallel" and the like appear, it does not mean that the corresponding devices / components / elements are required to be absolutely horizontal or vertical or overhanging or parallel, but may be slightly tilted or have deviations. For example, "horizontal" only means that its direction is more horizontal than "vertical", and does not mean that the structure must be completely horizontal, but may be slightly tilted. Alternatively, it can be simplified to mean that the corresponding devices / components / elements are set in directions such as "horizontal", "vertical", "overhanging", "parallel", etc., and can have an error / deviation of ±10% relative to the corresponding direction setting, more preferably an error / deviation within ±8%, more preferably an error / deviation within ±6%, more preferably an error / deviation within ±5%, and more preferably an error / deviation within ±4%. As long as the corresponding device / component / element is within the error / deviation range, it can still achieve its role in the scheme of the present invention.

[0056] In addition, the expressions “first”, “second”, “third”, etc., which appear in the terms, are merely used to distinguish the description of the same or similar components and should not be understood as emphasizing or implying the relative importance of specific components.

[0057] In addition, in the description of the embodiments of the present invention, "several", "plurality" and "a number" represent at least 2. It can be any number such as 2, 3, 4, 5, 6, 7, 8, 9, and even more than 9.

[0058] In addition, in the description of the technical solution of the present invention, unless otherwise clearly specified / defined / restricted, the terms "set", "install", "connect", "connected", "provided with", "laid", and "arranged" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection, and can be welding, riveting, bolting, threading, and other commonly used connection means in the field. This connection can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection, or an indirect connection through an intermediate medium, and it can be the internal connection of two elements.

[0059] In the related technology, the existing channel auxiliary system is also relatively limited in supporting the decision-making of ship navigation. It usually only provides static water level information, lacks dynamic comprehensive analysis of ship position, water level changes and navigation path, and cannot effectively avoid potential risks in the channel. Figures 1 to 7 To elaborate.

[0060] Example 1

[0061] In a canal construction project, a large-scale arch rib transport construction was carried out in the coastal shallow waters. The water transport line was 1.5 kilometers long and was located in an unconventional waterway. Due to the shallow water level, large ships could not pass directly. The construction had to make full use of the rising water level caused by the backflow of seawater during high tide to ensure that the necessary transportation water level was reached.

[0062] Since the tidal changes in the waters where the construction is located are significant, water level prediction is the key to whether the construction can proceed smoothly. It is necessary to accurately predict the changing patterns of tidal water levels so that the arch ribs can be transported and hoisted at the right time.

[0063] like Figures 1 to 6 As shown, a method for predicting the water level of coastal waters and evaluating the navigability of the coastal waters according to the present invention comprises the following steps:

[0064] Step 1: Set up several measuring points along the waterway, and obtain the first information based on GIS (Geographic Information System) + ultrasonic detection technology. The first information includes the water area information of the waterway. The water area perspective covers the entire waterway. The water area information includes the geodetic coordinates and elevation of each measuring point, as well as the water level information, as shown in Table 1.

[0065] Table 1. Channel coordinate information table

[0066]

[0067]

[0068] Based on the spline interpolation method, continuous and smooth interpolation points are generated near the known measuring points by fitting the water level data of the known measuring points with multi-segment piecewise functions.

[0069] The three-dimensional coordinate data of the known measurement points and the interpolation points are integrated together. The three-dimensional coordinates include the geodetic coordinates and the corresponding elevation information. The three-dimensional coordinates of the underwater terrain of the channel are established, and the following is formed: Figure 2 A detailed three-dimensional model of the underwater topography of the waterway is shown.

[0070] Step 2: Based on the three-dimensional model and combined with the historical monitoring data of local water levels at various measurement points on site, the Sp1ice-LSTM deep learning model is used to establish a corresponding mapping relationship between the local point water level and the water level at each coordinate point of the waterway to obtain a future prediction model for coastal waterways.

[0071] At several measuring points arranged at the beginning, middle and end of the waterway, an observation water level gauge is set up to obtain the water level information of each measuring point in real time. The water level information includes time, flow rate, water level, etc. Figure 3 shown.

[0072] According to the historical water level information of each measuring point obtained, the historical water level data is preprocessed to eliminate outliers and obtain optimized data.

[0073] The feature extraction of time series data is performed based on the Splice-LSTM model architecture, which includes several sub-modules. The sub-modules include three Sp1ice-LSTM layers and one fully connected layer.

[0074] Each of the three-layer Spiice-LSTM layers contains at least 50 hidden units. In this embodiment, each layer may contain 100 hidden units to capture long-term dependencies in time series data. The pseudo code is as follows:

[0075] model=Sequential()

[0076] model.add(Input(shape=(X_train.shape[1],X_train.shape[2])))

[0077] model.add(LSTM(100,return_sequences=True))

[0078] model.add(LSTM(100,return_sequences=True))

[0079] model.add(LSTM(100))

[0080] model.add(Dense(3))

[0081] model.compile(optimizer='adam',loss='mean_squared_error')

[0082] The fully connected layer contains 1 neuron, which is used to output the prediction results.

[0083] Set the Splice-LSTM model parameters, including setting the training rounds to at least 500. In this embodiment, the training rounds can be selected to be 1000 to ensure that the model fully learns the data features; set the batch size to 32 to balance the model training efficiency and memory usage; standardize the input data, and set the normalization range of the feature data and label data to [0,1].

[0084] The mean square error is used as the loss function to measure the error between the predicted value and the actual value of the Splice-LSTM model and to optimize the prediction accuracy of the Splice-LSTM model.

[0085] After the training, the trained Splice-LSTM model was evaluated using the test data set, and the mean square error was calculated to quantify the prediction performance of the model. The future prediction model of the coastal waterway was established, such as Figure 5 shown.

[0086] It should be noted here that the LSTM (Long Short-Term Memory) model has been widely used in time series prediction, natural language processing, and various complex time series data analysis; however, the traditional LSTM model also has some limitations in practical applications. First, the LSTM model is mainly used for the prediction of a single time series. When faced with the fusion of multi-source data of non-single time series, it is often difficult to fully capture the relationship between different data sources. Secondly, the prediction accuracy of the LSTM model may be reduced when processing complex multivariate time series data due to the complex dependencies between features. In addition, when processing multi-scale data or multi-level spatiotemporal data, the simplicity of the LSTM model structure limits the ability to extract deep-level features of the data.

[0087] By introducing multiple parallel or serial LSTM layers, Splice-LSTM can better capture and fuse the complex dependencies between multi-source data, thereby improving the prediction accuracy of the model; compared with traditional LSTM, Splice-LSTM has the ability to process multi-scale and multi-level data, and its model structure has been optimized to effectively extract deep-level features in the data. This enhanced LSTM model can more accurately predict water level changes in waterway water level monitoring and navigation capacity assessment, and better cope with the challenges brought by multivariate time series data, significantly improving the overall performance and reliability of the system. Figure 4 and Figure 5 , obviously Figure 5 The prediction accuracy of the adopted Splice-LSTM model is higher than Figure 4 Prediction accuracy of the adopted LSTM model.

[0088] Step 3: Obtain the second information through virtual mouse technology. The second information includes tide prediction information and weather forecast information of the hydrological station. The tide prediction information includes water level and flow velocity, and the weather forecast information includes daily rainfall.

[0089] Through the virtual mouse mode, navigate to the corresponding tidal observation website and weather forecast website of the hydrological station around the waterway, locate the specific data information on the web page, including water level, flow rate and daily rainfall, initialize the data storage list, and set the initial mouse position. Simulate the movement of the virtual mouse at the corresponding data chart position, and obtain the water level, flow rate, and daily rainfall data corresponding to each time period in the next 24 hours by continuously changing the mouse position. Figure 6 shown.

[0090] It should be noted here that since waterway water level prediction requires continuous acquisition of future characteristic parameters, traditional methods usually rely on manual reading or crawler technology to obtain predicted data from water level observation points. However, traditional crawler technology often needs to be applied according to complex data interaction protocols, and high-frequency crawler operations may cause excessive load on the target website server and even cause network risks. In contrast, virtual mouse technology simulates the process of human operation and obtains the required data by simulating human operations on web pages. This method can effectively reduce the pressure on website servers and reduce potential network security risks, thereby providing a safer and more efficient means of data acquisition in the water level prediction process.

[0091] Step 4: Add water level, flow velocity, and daily rainfall as feature parameters in the Sp1ice-LSTM model for accuracy evaluation, and obtain the weight factor of each feature parameter.

[0092] The data are collected and preprocessed to obtain a training set, which includes water level, flow velocity, and daily rainfall. The training set is evaluated using the Splice-LSTM model.

[0093] Based on the Sp1ice-LSTM model, multiple single model evaluations are performed one by one based on the training set, and the prediction results of the Sp1ice-LSTM model are analyzed. The final evaluation result label is expressed as follows:

[0094] H=α 1 w+α 2 f+α 3 p

[0095] Where H is the corresponding water level result in the full channel coordinate system output by the multi-source information complementary fusion model; α 1 , α 2 and α 3 They respectively represent the corresponding water level results in the whole channel coordinate system obtained by taking water level w, flow velocity f and daily rainfall p as single characteristic parameters.

[0096] The weight factor corresponding to each single feature parameter:

[0097]

[0098] In the formula, ε 1 , ε 2 and ε 3 They respectively represent the accuracy of the corresponding water level results in the full channel coordinate system obtained by taking water level, flow velocity and daily rainfall as single characteristic parameters.

[0099] In an optional embodiment, the weight factor α 1 , α 2 and α 3 They are 0.60, 0.31 and 0.09 respectively.

[0100] Step 5: Based on the weight factors of various characteristic parameters, the inverse error weighted average method is used for correction to establish a future prediction model for coastal waterways that integrates multi-source information.

[0101] Step 6: Use the future prediction model of coastal water channels that integrates multi-source information to evaluate the navigation capacity of the channel, predict the navigation window period, and warn of water level changes during the navigation process.

[0102] The method for predicting the water level and assessing the navigability of coastal waters described in the present invention solves the problem that a single data source cannot fully capture the factors affecting water level changes. By integrating GIS, ultrasonic detection and deep learning models, a set of high-precision and comprehensive water level monitoring and prediction technologies is formed. It also makes full use of the advantages of each data source to achieve the integration and optimization of multi-dimensional information such as water level, tide, and weather, significantly improving the applicability and reliability of the system in complex marine environments, and ensuring the safety and efficiency of waterway navigation.

[0103] Example 2

[0104] like Figure 7 As shown, a system for predicting the water level and evaluating the navigability of coastal waters described in the present invention utilizes the method for predicting the water level and evaluating the navigability of coastal waters described in Example 1. The evaluation system includes a three-dimensional topographic map display module, a future water level prediction module, a waterway transportation window period module, a real-time water level change module, and a real-time navigable ship warning module.

[0105] The three-dimensional topographic map display module includes a first acquisition unit and a modeling unit.

[0106] The first acquisition unit obtains the first information and transmits it to the modeling unit. The modeling unit establishes a three-dimensional model of the underwater terrain of the channel according to the first information and integrates it into the system through a program. The interactive visualization technology based on the Plotly library supports users to zoom and rotate the topographic map in real time, and click to view the coordinates and water level information of each point to realize the dynamic interactive display of the underwater terrain. The pseudo code is as follows:

[0107] fig=go.Figure(data=[go.Surface(z=z,x=x,y=y)]).

[0108] The future water level prediction module includes a second acquisition unit and a data processing unit.

[0109] The second collection unit sets a specified date (for example, the collection frequency is 7 days / time), and the system automatically obtains the second information using virtual mouse technology, that is, automatically navigates to the specified tidal data webpage and weather forecast webpage, simulates user operations, and obtains the 7-day waterway upper-middle-lower reaches of the waterway and daily rainfall data from the webpage for the future period of 24 hours, such as Figure 6 shown.

[0110] Through a preset data stream processing mechanism, the system seamlessly integrates the second information into the data processing unit for data processing, performs water level prediction through a previously trained splice-LSTM model, and automatically predicts future water level changes.

[0111] The waterway transportation window period module includes a data import unit, an analysis and screening unit, a window period calculation unit, and a result display unit.

[0112] The data import unit automatically obtains the predicted water level data under the bridge from the future water level prediction module, imports the data into the system platform, and combines the date and time information to generate time series data of the predicted water level.

[0113] The analysis and screening unit analyzes the water level prediction data through a preset algorithm, and screens out time periods that meet the requirements based on water level thresholds (such as thresholds of 1.2 meters and 1.5 meters, corresponding to the draft depths of the two workboats, respectively). The dynamic time segmentation algorithm is used to identify and calculate continuous time windows that meet the transportation requirements, ensuring the continuity and rationality of the transportation window period.

[0114] The window period calculation unit automatically calculates the total duration of each continuous time period according to the time interval of the water level prediction data, and uses the preset continuous time threshold (such as a threshold of 6 hours, indicating that the shortest continuous time is 6 hours) to screen out the valid transportation window period that meets the conditions.

[0115] The result display unit automatically displays the calculated effective transportation window period on the system interface, supporting users to interactively view the start and end times of the transportation window on a specific date.

[0116] The real-time water level change module includes several observation water level gauges arranged at the start, process and end points of the waterway. The observation water level gauges are connected through an interface protocol to automatically receive real-time water level data collected by the observation water level gauges within 24 hours. The real-time water level data is uploaded to the system platform through a standardized communication protocol to realize real-time water level change correction.

[0117] The real-time navigation ship warning module includes an AIS (Automatic Identification System) unit, a navigation path planning unit, and a critical water level warning unit.

[0118] Based on image recognition algorithms, the AIS unit is deployed at the starting point of waterway monitoring to accurately read key parameters such as the size and draft of ships entering the waterway. By combining real-time data acquisition with edge computing technology, this system can efficiently process and transmit basic ship information to ensure low latency and high accuracy of data acquisition. The AIS unit includes a high-resolution camera, which can be 1080P or higher, with waterproof, dustproof and night vision functions to ensure all-weather monitoring of ships. The camera is connected to an edge computing device that uses a GPU (graphics processing unit) or FPGA (field programmable gate array) to accelerate the processing of image data and use a deep learning algorithm to identify the feature information of the ship. The image recognition algorithm is based on a deep learning framework (such as YOLO, Faster R-CNN), which can automatically extract key parameters such as the size and draft of the ship. The AIS receiver is responsible for capturing and processing static information of the ship (such as ship ID, speed, and heading).

[0119] The navigation path planning unit is based on the positioning technology of the satellite navigation system (such as Beidou Satellite Navigation System BDS, enhanced GNSS Global Navigation Satellite System) configured on the ship, combined with the waterway water level data for accurate spatial positioning. The navigation path planning unit adopts the path optimization Dijkstra algorithm based on dynamic environment perception to calculate the best navigation route in real time, avoid areas in the waterway where the water level does not meet the ship's draft requirements, and ensure navigation safety. The Beidou satellite navigation system module and the enhanced GNSS module are installed on the ship to obtain the real-time position information of the ship. The BDS module achieves high-precision positioning by communicating with Beidou satellites. The enhanced GNSS module can simultaneously receive data from multiple satellites around the world to enhance the accuracy and stability of positioning. The navigation system is equipped with real-time dynamic differential processing software for decoding and processing multi-source satellite signals. Through the navigation control system, the ship can optimize the path based on water level prediction information and real-time channel data.

[0120] It should be noted here that the traditional waterway management system is relatively backward in data transmission and communication technology, usually relying on manual intervention or timed data transmission, and lacks fully automatic data interaction functions. Therefore, the existing system has obvious deficiencies in data transmission efficiency and real-time response capabilities. In addition, during waterway transportation, real-time ship monitoring and navigation warning functions have not been fully reflected in the existing system, resulting in ships being unable to receive timely warning information when encountering insufficient water levels or other navigation risks, increasing the safety risk of navigation. The lack of intelligent path planning and dynamic management of the navigation window period makes it difficult for the existing system to effectively guide ships to avoid dangerous waters.

[0121] In an optional implementation, the AIS unit, BDS module and GNSS module may communicate with the main system platform via RESTful API and WebSocket technology to achieve automated data transmission and real-time feedback between devices.

[0122] The critical water level warning unit integrates water level perception and ship position data to monitor the relative position of the ship and the water level in real time. When the ship's sailing position approaches the critical value of the draft depth, the critical water level warning unit will automatically send out accurate warning signals in a timely manner through text messages, on-site alarms, etc., to ensure that the ship takes measures to avoid danger.

[0123] This system adopts a distributed communication architecture. The three-dimensional topographic map display module, future water level prediction module, waterway transportation window module, real-time water level change module, and real-time navigation ship warning module interact with each other through RESTful API and WebSocket technology to ensure the real-time and reliability of data transmission. The second acquisition unit and the data processing unit realize asynchronous data transmission through message queue technology (such as Kafka or RabbitMQ) to ensure efficient transmission and processing of data between different units without manual intervention. Real-time water level change data is automatically uploaded to the system platform through the Internet of Things (IoT) communication protocol (such as MQTT or CoAP), and pushed to related modules in real time through WebSocket. This system adopts encryption-based secure transmission protocols (such as TLS / SSL) to ensure the security of data transmission, combined with distributed fault detection and recovery mechanisms to ensure that the system can still maintain high reliability and data consistency under high load conditions.

[0124] The system for predicting the water level and assessing the navigability of coastal waters described in the present invention solves the problem that a single data source cannot fully capture the factors affecting water level changes. By integrating GIS, ultrasonic detection and deep learning models, a set of high-precision and comprehensive water level monitoring and prediction technologies is formed. It also makes full use of the advantages of each data source to achieve the integration and optimization of multi-dimensional information such as water level, tide, and weather, significantly improving the applicability and reliability of the system in complex marine environments, and ensuring the safety and efficiency of waterway navigation.

[0125] Example 3

[0126] An electronic device according to the present invention comprises:

[0127] a memory having a computer program stored thereon;

[0128] A processor is used to execute the program in the memory to implement the method for predicting the water level of coastal waters and evaluating the navigability as described in Example 1.

[0129] As an optional solution of this embodiment, the electronic device may include: a processor, a memory, and the electronic device may also include one or more of a multimedia component, an input / output (I / O) interface, and a communication component.

[0130] The processor is used to control the overall operation of the electronic device to complete all or part of the steps in the above-mentioned method for predicting the water level of coastal waters and evaluating the navigability.

[0131] The memory is used to store various types of data to support the operation of the electronic device, which data may include, for example, instructions for any application or method operating on the electronic device, and application-related data; the memory can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0132] The multimedia component may include a screen and an audio component, wherein the screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals; for example, the audio component may include a microphone for receiving external audio signals, and the received audio signals may be further stored in a memory or sent via a communication component; the audio component also includes at least one speaker for outputting audio signals.

[0133] The I / O interface provides an interface between the processor and other interface modules, which may be a keyboard, a mouse, buttons, etc.; these buttons may be virtual buttons or physical buttons.

[0134] The communication component is used for wired or wireless communication between the electronic device and other devices; wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, 5G or 5.5G, or a combination of one or more of them, so the corresponding communication component may include: Wi-Fi module, Bluetooth module, Star Flash module, NFC module, mobile phone communication module.

[0135] As an optional solution of this embodiment, the electronic device can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned coastal waters water level prediction and navigation capacity assessment method.

[0136] In addition, the computer-readable storage medium provided in the embodiment of the present disclosure may be the above-mentioned memory including program instructions, and the above-mentioned program instructions may be executed by a processor of an electronic device to complete the above-mentioned method for predicting the water level of coastal waters and evaluating the navigability.

[0137] Example 4

[0138] The computer-readable storage medium described in the present invention stores a computer program thereon, characterized in that when the program is executed by a processor, the method for predicting the water level of coastal waters and evaluating the navigability as described in Example 1 is implemented.

[0139] Computer-readable storage media are used to store various types of data to support operations on the electronic device, which data may include, for example, instructions for any application or method operating on the electronic device, as well as application-related data; the memory may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0140] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for predicting water level in coastal waters and assessing navigability, characterized in that: The following steps are involved: A plurality of measuring points are set along the waterway to obtain first information, the first information including water area information of the waterway, the water area information including geodetic coordinates and elevation of each measuring point, and water level information, and establish a three-dimensional model of underwater terrain of the waterway; Based on the three-dimensional model, combined with the local water level historical monitoring data of each measuring point, the corresponding mapping relationship between the local point water level and the water level of each coordinate point of the waterway is established through the Splice-LSTM deep learning model, and the future prediction model of the coastal waterway is obtained; The second information is obtained by using the virtual mouse technology. The second information includes the tide prediction information of the hydrological station and the weather forecast information. The tide prediction information includes the water level and flow velocity. The weather forecast information includes the daily rainfall. In the Splice-LSTM model, water level, flow velocity, and daily rainfall are added as feature parameters for accuracy evaluation, and the weight factors of each feature parameter are obtained; Based on the weight factors of each characteristic parameter, the error inverse weighted average method is used for correction, and a future prediction model for coastal waterways integrating multi-source information is established. The future prediction model of coastal waterways that integrates multi-source information is used to evaluate the navigation capacity of the waterway, predict the navigation window period and warn of water level changes during the navigation process.

2. The method for predicting the water level of coastal waters and evaluating the navigability according to claim 1 is characterized in that: Obtain the first information based on GIS+ultrasonic detection technology; Based on the spline interpolation method, by fitting the water level data of known measurement points with multi-segment piecewise functions, continuous and smooth interpolation points are generated near the known measurement points. The three-dimensional coordinate data of known measurement points and interpolation generated points are integrated together. The three-dimensional coordinates include geodetic coordinates and corresponding elevation information to establish the three-dimensional coordinates of the underwater terrain of the waterway.

3. The method for predicting the water level of coastal waters and evaluating the navigability according to claim 1 is characterized in that: Several observation water level gauges are arranged at the starting, middle and end points of the waterway to obtain water level information at several measuring points in real time. The water level information includes time, flow rate and water level.

4. The method for predicting the water level of coastal waters and evaluating the navigability according to claim 1 is characterized in that: Based on the historical water level information obtained from multiple points, the historical water level data is preprocessed to remove abnormal values ​​and obtain optimized data; The feature extraction of time series data is performed based on the Splice-LSTM model architecture, which includes several submodules. The submodules include three Splice-LSTM layers and one fully connected layer. Each of the three Splice-LSTM layers contains at least 50 hidden units to capture the long-term dependencies in the time series data. The fully connected layer contains 1 neuron to output the prediction results. Set the Splice-LSTM model parameters, including setting the training rounds to at least 500, normalizing the input data, and setting the normalization range of feature data and label data to [0,1]; The mean square error is used as the loss function to measure the error between the predicted value and the actual value of the Splice-LSTM model and optimize the prediction accuracy of the Splice-LSTM model. After the training, the trained Splice-LSTM model was evaluated using the test dataset, and the mean square error was calculated to quantify the prediction performance of the model, and a future prediction model for coastal waterways was established.

5. The method for predicting the water level of coastal waters and evaluating the navigability according to claim 1 is characterized in that: Use the virtual mouse mode to navigate to the corresponding tidal observation website and weather forecast website of the hydrological station around the waterway; Locate specific data information on the web page, including water level, flow rate and daily rainfall; Initialize the data storage list and set the initial mouse position; Simulate moving the virtual mouse at the corresponding data chart position, and obtain the water level, flow rate, and daily rainfall data corresponding to each time period in the next 24 hours by continuously changing the mouse position.

6. The method for predicting the water level of coastal waters and evaluating the navigability according to any one of claims 1 to 5, characterized in that: The data was collected and preprocessed to obtain a training set, which included water level, flow velocity, and daily rainfall. The training set was evaluated using the Splice-LSTM model; Based on the Splice-LSTM model, multiple single model evaluations are performed one by one based on the training set, and the prediction results of the Splice-LSTM model are analyzed. The final evaluation result label is expressed as follows: H=α1w+α2f+α3p Where H is the corresponding water level result in the full channel coordinate system output by the multi-source information complementary fusion model; α1, α2 and α3 respectively represent the corresponding water level results in the full channel coordinate system obtained by taking water level w, flow velocity f and daily rainfall p as single characteristic parameters; The weight factor corresponding to each single feature parameter: Where ε1, ε2 and ε3 represent the accuracy of the corresponding water level results in the full channel coordinate system obtained by taking water level, flow velocity and daily rainfall as single characteristic parameters.

7. The method for predicting the water level of coastal waters and assessing the navigability according to claim 6, characterized in that: The weight factors α1, α2, and α3 are 0.60, 0.31, and 0.09, respectively.

8. A system for predicting water level in coastal waters and assessing navigation capacity, characterized in that: Using the coastal waters water level prediction and navigation capacity assessment method as described in any one of claims 1 to 7, the assessment system comprises: The three-dimensional topographic map display module builds a three-dimensional model of the underwater terrain of the waterway through the first information, and realizes the dynamic interactive display of the underwater terrain based on the interactive visualization technology of the Plotly library; The future water level prediction module, by setting a specified date, uses the virtual mouse technology to obtain the second information, performs data processing on the second information, and predicts the water level through the splice-LSTM model, and automatically predicts the future water level changes; The waterway transportation window module automatically obtains the water level forecast data under the bridge from the future water level forecast module, combines the date and time information, generates the time series data of the predicted water level, analyzes the water level forecast data, selects the time period that meets the requirements based on the water level threshold conditions, uses the dynamic time segmentation algorithm to identify and calculate the continuous time window that meets the transportation requirements, automatically calculates the total duration of each continuous time period according to the time interval of the water level forecast data, and uses the preset continuous time threshold to select the effective transportation window period that meets the conditions; The real-time water level change module is connected to the observation water level meter through the interface protocol, automatically receives the real-time water level data collected by the observation water level meter, and the real-time water level data is uploaded through the communication protocol to realize the real-time water level change correction; The real-time navigable ship warning module includes an AIS based on an image recognition algorithm installed on the waterway, which accurately reads the size and draft parameters of ships entering the waterway, uses the satellite navigation system positioning technology configured on the ship, and combines the waterway water level data for precise spatial positioning. It uses the path optimization Dijkstra algorithm based on dynamic environment perception to calculate the best navigable route in real time, avoid areas in the waterway where the water level does not meet the ship's draft requirements, integrates water level perception and ship position data, monitors the relative position of the ship and the water level in real time, and automatically alarms when the ship's driving position approaches the critical value of the draft.

9. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor is used to execute the program in the memory to implement the method for predicting the water level of coastal waters and evaluating the navigability as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, characterized in that when the program is executed by a processor, the method for predicting the water level of coastal waters and evaluating the navigability as described in any one of claims 1 to 7 is implemented.

Images