Tidal flat evolution prediction method and system based on hybrid model

Abstract

The invention provides a tidal flat evolution prediction method and system based on a hybrid model, and relates to the field of dynamic geomorphology and artificial intelligence cross technology, and the method comprises the steps: constructing a tidal flat erosion and deposition evolution database under different scenes of a target coastal zone; constructing a landform prediction model through an artificial intelligence algorithm based on the database, and taking hydrodynamic characteristic data as input and seabed erosion and deposition variation as output; hybrid simulation is executed based on the physical mechanism model and the landform prediction model, and long-duration evolution prediction is realized by updating terrain boundary conditions through loop iteration; and the configuration service application module is used for receiving user input parameters through the interactive interface and outputting a visual result. According to the method, traditional hydrodynamic force-sediment-landform coupling calculation is decoupled into a physical mechanism and AI coordinated chain process, the calculation complexity is remarkably reduced, offline rapid calculation business of tidal flat evolution is achieved, and real-time technical support is provided for coastal zone planning and disaster prevention decision making.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of dynamic geomorphology and artificial intelligence, in particular to a tidal flat evolution prediction method and system based on a hybrid model. BACKGROUND

[0002] As a key transitional zone of land-sea interaction, the tidal flat plays an irreplaceable role in maintaining the balance of the coastal ecosystem and providing disaster prevention and mitigation functions. Under the dual pressures of global climate change and high-intensity human activities, the evolution mechanism of muddy tidal flats tends to be complex, and it is urgent to establish a precise and efficient prediction system to support the sustainable development of the coastal zone. Traditional prediction methods based on dynamic geomorphology have inherent defects such as high computational complexity and large resource consumption due to the need to simulate the coupling of multiple physical fields. Especially when conducting multi-scenario simulation, the traditional numerical model is time-consuming and difficult to meet the timeliness requirements of coastal zone spatial planning and disaster prevention decision-making. In addition, factors such as storm surges caused by extreme climate events and seasonal water level fluctuations further increase the risk of systematic errors in prediction results.

[0003] Therefore, breaking through the computing power bottleneck of traditional models and building a new prediction paradigm that takes into account physical mechanisms and computational efficiency has become a core challenge for fine-grained management of the coastal zone. SUMMARY

[0004] To solve the above problems existing in the prior art, the first aspect of the present application proposes a tidal flat evolution prediction method based on a hybrid model, comprising: S1: constructing a tidal flat erosion and deposition evolution database of the target coastal zone under different scenarios; S2: constructing a geomorphology prediction model based on the tidal flat erosion and deposition evolution database through an artificial intelligence algorithm; S3: performing hybrid simulation based on the physical mechanism model and the geomorphology prediction model, comprising: generating water dynamic feature data through the physical mechanism model, inputting the water dynamic feature data into the geomorphology prediction model to generate seabed erosion and deposition changes, updating the topographic boundary conditions based on the seabed erosion and deposition changes and iterating cyclically; S4: configuring a business application module based on the hybrid simulation, the business application module receiving user input parameters, performing geomorphology speed calculation and outputting visual results.

[0005] In combination with the first aspect, in some implementations, S1 comprises: S11: establishing a dynamic geomorphology evolution model of the target coastal zone using a dynamic geomorphology model; S12: based on the dynamic geomorphology evolution model, reenacting the dynamic sedimentation process of the tidal flat, and calibrating the dynamic geomorphology evolution model in combination with historical topographic data; S13: Based on the calibrated dynamic geomorphology evolution model, the storm surge and seasonal water level fluctuation are time-scale reduced to generate an annual-scale geomorphology evolution feature dataset; S14: Based on the annual-scale geomorphology evolution feature dataset, the influence of sediment supply, sea level change, tide and wave factors on the erosion and deposition evolution is simulated to form a tidal flat erosion and deposition evolution database.

[0006] In combination with the first aspect, in some implementations, S12 includes: based on the dynamic geomorphology evolution model, long-term simulation is performed to output geomorphology data containing the influence of climate change and human activities; S13 includes: based on the geomorphology data containing the influence of climate change and human activities, characteristic parameters of storm surge and seasonal water level fluctuation are extracted for time-scale reduction; S14 includes: based on the characteristic parameters after time-scale reduction, the contribution value of coastal engineering to geomorphology evolution is quantified.

[0007] In combination with the first aspect, in some implementations, S2 includes: S21: Based on the tidal flat erosion and deposition evolution database, the inundation probability, average shear stress, flood dominance and tidal creek distance function are sorted out to construct a training dataset corresponding to the seabed erosion and deposition; S22: Based on the training dataset, a convolutional neural network model is constructed with four-channel hydrodynamic features as input and seabed erosion and deposition change as output; S23: Based on the convolutional neural network model, an encoder-decoder structure is configured and feature maps are transmitted across layers through a skip connection to output an optimized convolutional neural network model; S24: Based on the optimized convolutional neural network model and the tidal creek distance function, the loss function is regionally weighted calculated to output an accuracy-optimized geomorphology prediction model.

[0008] In combination with the first aspect, in some implementations, S23 includes: the feature maps output by each stage of the encoder are transmitted to the corresponding stage of the decoder for splicing; S24 includes: according to the tidal creek distance function value, a weight coefficient is assigned, and the weight value of the tidal creek edge area is higher than that of the flat area.

[0009] In combination with the first aspect, in some implementations, S3 includes: S31: Based on the initial terrain boundary conditions, the inundation probability, average shear stress, flood dominance and tidal creek distance function are calculated and generated by a physical mechanism model; S32: The inundation probability, average shear stress, flood dominance and tidal creek distance function are input into the geomorphology prediction model to output the seabed erosion and deposition change in the current period; S33: updating the topographic boundary condition based on the seabed erosion and deposition change amount of the current time period, taking the updated topographic boundary condition as the input of the next time period, and cyclically executing S31-S32 until the long-term simulation is completed.

[0010] In combination with the first aspect, in some implementations, S31 comprises: integrating the sea level rise parameter into the initial topographic boundary condition. S33 comprises: dividing the long-term simulation into multiple discrete time periods, and taking the output topographic boundary condition of each time period as the input topographic boundary condition of the next time period.

[0011] In combination with the first aspect, in some implementations, S4 comprises: S41: receiving the coastal zone development parameter, the climate parameter, and the hydrodynamic parameter input by the user through the interactive interface; S42: based on the coastal zone development parameter, the climate parameter, and the hydrodynamic parameter, calling the hybrid simulation to perform the geomorphology quick calculation; S43: converting the result of the geomorphology quick calculation into a visual graphic output, and providing the parameter reset, calculation trigger, and history record management functions.

[0012] In combination with the first aspect, in some implementations, S41 comprises: performing threshold range verification on the coastal zone development parameter, the climate parameter, and the hydrodynamic parameter, and triggering an alarm information when each threshold range is exceeded; S43 comprises: superimposing the seabed erosion and deposition change amount on the geographic base map in the form of a heat map, and storing the input parameter and the corresponding erosion and deposition evolution result of each calculation.

[0013] The second aspect of the present application provides a tidal flat evolution prediction system based on a hybrid model, which adopts the method provided by any of the above embodiments, and the system comprises: A first construction module for constructing a tidal flat erosion and deposition evolution database of a target coastal zone under different scenarios; A second construction module connected with the first construction module, for constructing a geomorphology prediction model based on the tidal flat erosion and deposition evolution database through an artificial intelligence algorithm; A hybrid simulation execution module connected with the second construction module, for executing hybrid simulation based on the physical mechanism model and the geomorphology prediction model, comprising: generating hydrodynamic feature data through the physical mechanism model, inputting the hydrodynamic feature data into the geomorphology prediction model to generate the seabed erosion and deposition change amount, updating the topographic boundary condition based on the seabed erosion and deposition change amount and cyclically iterating; A geomorphology quick calculation execution and output module connected with the hybrid simulation execution module, for configuring a business application module based on the hybrid simulation, the business application module receiving user input parameters, executing the geomorphology quick calculation, and outputting the visual results.

[0014] Compared with the prior art, the present application has the beneficial effects that: first, a database of tidal flat erosion and deposition evolution of the target coastal zone under different scenarios is constructed, which provides physical mechanism data covering multiple factors for subsequent model training, and ensures the completeness of the prediction system. Second, based on the database, a geomorphology prediction model is constructed through an artificial intelligence algorithm, which decouples the traditional strong coupling calculation of water dynamics-sediment-geomorphology into a chain calculation of "water dynamics (sediment)-geomorphology", significantly reducing the model complexity. This process directly establishes the mapping relationship between the feature conditions and the geomorphology evolution, which makes the computing power requirement of large-scale long-term simulation decrease by orders of magnitude.

[0015] In the hybrid simulation stage, the physical mechanism model and the geomorphology prediction model form a closed loop cooperation: the physical mechanism model generates four-channel water dynamic feature data such as flooding probability and average shear stress, which are input into the geomorphology prediction model to quickly output high-precision seabed erosion and deposition changes; based on the changes, the topographic boundary conditions are updated in real time and iterated, which not only inherits the accuracy of the dynamic response of the physical mechanism to the boundary, but also avoids the massive consumption of full coupling calculation. This "physically driven-AI fast calculation" cycle architecture realizes the stable calculation of long-period geomorphology evolution under extreme climate events for the first time.

[0016] Finally, the business application module converts the hybrid modeling capability into a business tool. After the user inputs scenario data such as coastal zone development parameters and climate parameters through the interactive interface, the system directly calls the pre-trained hybrid model chain, and outputs visual tidal flat evolution prediction results in minutes without the need for professional numerical simulation skills. This module compresses the traditional dynamic geomorphology simulation, which takes several weeks to complete, to real-time fast calculation level, completely solving the timeliness bottleneck of coastal zone management decision-making, and avoiding human operation errors through standardized operation processes.

[0017] In summary, through the four-step cooperation of "database construction-AI modeling-hybrid iteration-business integration", the algorithm and timeliness limitations of traditional models are broken through while maintaining the reliability of the physical mechanism, providing real-time technical support for coastal zone spatial planning and disaster prevention and control. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0019] Figure 1 The flowchart of a tidal flat evolution prediction method provided by an embodiment of the present application is shown.

[0020] Figure 2 Fig. 1 shows a seabed erosion and deposition map of a target sea area provided by an embodiment of the present application.

[0021] Figure 3 Fig. 4 shows a U-Net Mor speed calculation flowchart provided by an embodiment of the present application.

[0022] Figure 4 Fig. 6 shows a hybrid simulation flowchart result provided by an embodiment of the present application.

[0023] Figure 5 Fig. 7 shows an interactive interface schematic diagram provided by an embodiment of the present application.

[0024] Figure 6 Fig. 8 shows a structure schematic diagram of a tidal flat evolution prediction system based on a hybrid model provided by an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0026] The specific embodiments of the present application will be described below.

[0027] Embodiment 1 As shown in Fig. 1, the present application provides a tidal flat evolution prediction method based on a hybrid model, which comprises: Figure 1 S1: constructing a tidal flat erosion and deposition evolution database of a target coastal zone under different scenarios; S2: based on the tidal flat erosion and deposition evolution database, constructing a geomorphology prediction model through an artificial intelligence algorithm; S3: performing hybrid simulation based on the physical mechanism model and the geomorphology prediction model, comprising: generating hydrodynamic feature data through the physical mechanism model, inputting the hydrodynamic feature data into the geomorphology prediction model to generate seabed erosion and deposition change, updating the topographic boundary condition based on the seabed erosion and deposition change and iteratively looping; S4: configuring a business application module based on the hybrid simulation, the business application module receiving user input parameters, executing geomorphology speed calculation and outputting visual results.

[0028] ​Specifically, when constructing the database of tidal flat erosion and deposition under different scenarios for the target coastal zone, the dynamic geomorphology model needs to be selected first, and the numerical grid is established for the target area and the resolution of the tidal flat is improved. Taking the Caofedian sea area in Bohai Bay as an example, the outer grid is set to 0.9 km x 0.9 km, and the near-island area is encrypted to 0.11 km x 0.11 km. Through the re-evolution of the historical dynamic deposition process, the model is calibrated combined with the terrain DEM data to ensure the reliability of the physical mechanism. In order to eliminate the systematic error of extreme climate event prediction, the wave spectrum and mathematical analysis method are used to process the multi-year wave data: the wave is divided into 64 classes (8 wave heights x 8 wave directions) according to equal interval, the representative wave height is calculated by wave energy equivalence, the wave direction uses weighted value, and the duration of each wave is set to 1 day and arranged in random sequence to eliminate the influence of tidal level. At the same time, the geomorphology acceleration factor (recommended value ≤ 50) is introduced to amplify the wave flow sand lifting effect, and the storm surge and seasonal water level fluctuation are reduced to annual scale, and the mapping relationship between annual evolution process and characteristic condition is directly established. Finally, the combined effects of sediment supply, sea level rise, tidal wave and coastal engineering on erosion and deposition are simulated to form a database covering multiple scenarios.

[0029] When building the geomorphology prediction model based on the database, four types of characteristic parameters such as flooding probability, average shear stress, rising tide advantage and tidal creek distance function (SDF) are arranged as input, and the seabed erosion and deposition amount is arranged as output. The data set is divided into training set, validation set and test set according to 7:1.5:1.5. U-Net convolutional neural network framework is selected to build U-Net Mor model, the input layer is designed as four channels, the convolution kernel size is set to 5x5, ReLU activation function is used to introduce nonlinearity, and the encoder-decoder is connected through jump connection to transfer feature map to preserve details. In the loss function calculation, in addition to the mean square error to measure the overall accuracy, the tidal creek edge area is given higher weight according to the SDF value to enhance the key geomorphology prediction ability.

[0030] In the scenario simulation stage of the hybrid model, the sea level rise parameter is integrated into the water level boundary condition, and the long-term simulation is divided into N discrete time periods. The physical mechanism model (such as Delft3D) first generates four-channel hydrodynamic feature data for the i-th time period; the geomorphology prediction model (U-Net Mor) outputs the seabed erosion and deposition amount and updates the terrain accordingly; the updated terrain is used as input for the next time period, and the cycle is repeated until N simulations are completed, and the cumulative result is the long-term erosion and deposition. This process decouples the traditional coupled calculation into a chain cycle of "physical mechanism generates input-AI outputs terrain update", which significantly reduces the computational load of each time.

[0031] The business application module receives the user inputted coastal zone development parameters, climate parameters, and hydrodynamic parameters through an interactive interface, calls the pre-trained hybrid model chain to perform rapid calculation. The interface integrates parameter verification function: when the input value exceeds the threshold range, an alert is triggered; the calculation results are superimposed on the geographic base map in the form of a heat map, and parameter reset, historical record storage and clearing functions are provided. Users only need to select the scenario and click the calculation button to obtain the prediction results completed by traditional models in weeks.

[0032] In the embodiment of the present application, the tidal flat erosion and deposition evolution database provides multi-scenario data basis with complete physical mechanism for AI training; the geomorphology prediction model decouples the strong coupling process into feature mapping relationship, and the computing power demand is reduced; the hybrid simulation cycle architecture inherits the accuracy of physical mechanism while avoiding the consumption of full coupling calculation, and for the first time realizes long-period stable calculation including extreme events; the business application module converts professional numerical simulation into zero threshold operation, and the decision response time is significantly reduced. Finally, the core problems of time inefficiency and computing power bottleneck in coastal zone planning are solved.

[0033] In combination with the first aspect, in some implementations, S1 includes: S11: using a dynamic geomorphology model to establish a dynamic geomorphology evolution model of the target coastal zone; S12: based on the dynamic geomorphology evolution model, reenacting the dynamic deposition process of the tidal flat, and calibrating the dynamic geomorphology evolution model in combination with historical topographic data; S13: based on the calibrated dynamic geomorphology evolution model, performing time scale reduction on storm surge and seasonal water level fluctuation to generate annual scale geomorphology evolution feature dataset; S14: based on the annual scale geomorphology evolution feature dataset, simulating the influence of sediment supply, sea level change, tides and wave factors on erosion and deposition evolution, and forming a tidal flat erosion and deposition evolution database.

[0034] Specifically, the tidal flat erosion and deposition evolution database construction includes four sub-steps: In S11, the selection of the dynamic geomorphology evolution model needs to consider the characteristics of the target area, for example, the muddy coast should use Delft3D or XBeach with a perfect deposition module. The grid division needs to follow the principle of "coarse periphery-dense tidal flat", and in the Caofidian case, the grid near the island is encrypted to 10m x 10m to capture the tidal inlet geomorphology. The model calibration in S12 needs to combine historical topographic data, for example, comparing different years of DEM to verify the reproduction accuracy of the erosion and deposition trend. The long-term simulation length should cover the main environmental change period, for example, 50 years of simulation is used in Caofidian to include the cumulative effect of sea level rise.

[0035] The time scale reduction of S13 is the key to eliminate extreme event errors: for the intermittent characteristics of storm surge, the short-time influence is stripped through wave classification and random sequence arrangement; for seasonal water level fluctuations, the combination of spring tide and seasonal tide level simulation is used, and then the geomorphology acceleration factor is used to integrate the high-frequency fluctuations into the annual scale output.

[0036] Alternative solutions include using wavelet analysis to extract periodic components, or decomposing dominant modes through empirical orthogonal function.

[0037] The scenario simulation of S14 needs to cover the core variables of coastal zone management: the sediment supply scenario sets different sources (river sediment transport / beach erosion) and flux gradients, the sea level rise rate prediction range of low / medium / high scheme, the tidal process considers the change of tidal range and the modulation of astronomical tide period, the wave condition combines historical statistics and future wind field prediction, and the impact of coastal engineering is realized by modifying the boundary conditions (such as the coordinates of the breakwater). The database is finally integrated into a multi-dimensional parameter matrix, and each parameter combination corresponds to a set of four-channel feature data and erosion and deposition results.

[0038] In the embodiments of the present application, the calibration of dynamic geomorphological evolution model ensures the reliability of physical mechanism; the time scale reduction suppresses the divergence of AI prediction caused by extreme events; and the multi-scenario coverage enables the database to have extrapolation prediction ability. For example, the quantitative relationship between sea level rise rate and erosion and deposition in the Caofidian case can directly support the coastal protection decision under different climate policies.

[0039] In combination with the first aspect, in some implementations, S12 includes: performing long-term simulation based on a dynamic geomorphological evolution model to output geomorphological data containing the influence of climate change and human activities; S13 includes: extracting storm surge and seasonal water level fluctuation characteristic parameters for time scale reduction based on the geomorphological data containing the influence of climate change and human activities; S14 includes: quantifying the contribution value of coastal engineering to geomorphological evolution based on the characteristic parameters after time scale reduction.

[0040] Specifically, the long-term simulation of S12 needs to set a sufficient time span to capture the slow-changing process, for example, a 50-year simulation can include the linear trend of sea level rise and the stepwise influence of human activities (reclamation / sand mining). Climate change factors are realized by modifying boundary conditions: temperature rise can adjust the viscosity coefficient of water, and increased storm frequency is reflected in the energy distribution shift of wave spectrum. Human activity quantification needs to combine engineering parameters: reclamation area is converted into a grid mask, and port construction is realized by modifying local water depth.

[0041] The feature parameter extraction of S13 needs to focus on the core indicators of storm surge and seasonal fluctuations: storm surge uses three-dimensional parameters of maximum water level increase, duration, and frequency; seasonal fluctuations are represented by standard deviation of tidal level and phase angle. In addition to the wave classification method, the event response model can be used to convert discrete storm events into equivalent continuous forces.

[0042] The coastal engineering contribution value quantification of S14 needs to design a control experiment: Baseline scenario: natural evolution without engineering intervention; Engineering scenario: add boundary conditions such as embankment / siltation promoting facilities; Contribution value = (engineering scenario erosion and deposition amount - baseline scenario erosion and deposition amount) / simulation duration.

[0043] For example, in the Caofidian case, the embankment project caused the downstream annual deposition amount to increase by 0.12m, which can be directly input into the AI model training.

[0044] As shown in Figure 2 , a coastal geomorphology evolution database is constructed for the impact of sea level rise on geomorphology evolution, different 50-year scale sea level scenarios are set, and the geomorphology evolution of coastal tidal flats under different sea level rise rates is calculated.

[0045] Long-term simulation combined with engineering quantification brings three effects: revealing the cumulative geomorphic impact of human activities (such as the lag effect of reclamation leading to tidal flat shrinkage); separating the contribution proportion of natural evolution and human disturbance; providing "dose-response" data set of engineering intervention for AI model. For example, the database contains the corresponding relationship between different embankment lengths and deposition amounts, supporting the comparison and selection of planning schemes.

[0046] In combination with the first aspect, in some implementations, S2 includes: S21: Based on the tidal flat erosion and deposition evolution database, organize the submergence probability, average shear stress, rising tide advantage, and tidal ditch distance function to construct a training data set corresponding to the sea bed erosion and deposition; S22: Based on the training data set, construct a convolutional neural network model with four-channel hydrodynamic features as input and sea bed erosion and deposition change as output; S23: Based on the convolutional neural network model, configure an encoder-decoder structure and realize feature map cross-layer transmission through a jump connection to output an optimized convolutional neural network model; S24: Based on the optimized convolutional neural network model and the tidal ditch distance function, perform regional weighted calculation on the loss function to output an accuracy-optimized geomorphology prediction model.

[0047] Specifically, the training data set construction needs to strictly unify the data format: the flooding probability is a floating point matrix of the time ratio of the grid unit being flooded; the average shear stress is the average value of the bed shear force; the rising tide advantage is defined as the ratio of the rising tide duration difference to the cycle; the tidal creek distance function (SDF) is calculated by the Euclidean distance transformation to calculate the distance from the grid point to the nearest tidal creek center. The four-channel data needs to be spatially aligned and normalized to the [0, 1] interval.

[0048] As shown in Figure 3 The U-Net Mor model structure optimization focuses on three points: the encoder adopts four-level down-sampling, and the number of channels is multiplied at each level (8→16→32→64); the decoder is up-sampled by deconvolution, and the same size feature map of the encoder is connected by jumping; the final output layer adopts 1×1 convolution to compress the number of channels to 1 (scouring amount).

[0049] Alternative solutions include replacing the basic convolution block with ResNet to enhance gradient propagation, or using attention mechanism to optimize feature fusion.

[0050] The loss function weighting method divides the weight area according to the SDF value: SDF≤50m is defined as the edge of the tidal creek, 50m<SDF≤200m is the transition zone, and SDF>200m is the flat area.

[0051] In the embodiment of the present application, the four-channel input covers the core physical quantity of the water power control topography (the flooding probability controls the sedimentation range, the shear stress determines the sediment starting, the rising tide advantage affects the sediment transport direction, and the SDF represents the topography pattern); the jumping connection solves the spatial information loss caused by pooling, and the tidal creek shape prediction error is reduced; the regionally weighted loss function makes the model focus on the sensitive area of the topography, and the prediction accuracy of the beach and trough boundary is improved.

[0052] In combination with the first aspect, in some implementations, S23 includes: passing the feature maps output by each stage of the encoder to the corresponding stage of the decoder for splicing; S24 includes: assigning a weight coefficient according to the tidal creek distance function value, and assigning a higher weight value to the edge area of the tidal creek than to the flat area.

[0053] Specifically, the feature transmission from the encoder to the decoder needs to keep the spatial dimension aligned: when the size of the feature map output by the kth stage of the encoder is reduced due to the pooling operation, the input size of the corresponding stage of the decoder needs to be matched by center cropping or deconvolution up-sampling before jumping connection. In the Caofidian case, the cropping method is used to ensure that the size deviation is less than one percent during splicing. After splicing the feature map, a 3×3 convolution operation is immediately performed to compress the number of merged channels to the original number of channels of the decoder, avoiding subsequent calculation redundancy. Alternative solutions include using a channel attention mechanism to weight and select the encoder feature map, or using element-wise addition instead of channel splicing to reduce memory occupation.

[0054] The edge weight distribution of the tidal creek is based on the spatial gradient characteristics of the SDF: first, the gradient modulus of the SDF field is calculated to quantify the degree of topographic abruptness, and the greater the gradient modulus represents the closer to the tidal creek boundary. The edge region is defined as the grid cell whose gradient modulus exceeds a preset threshold, and the threshold is recommended to be more than the 90th percentile of the gradient distribution of the entire calculation domain.

[0055] This mechanism solves the core contradiction in geomorphological evolution prediction through spatially differentiated weights: the edge of the tidal creek is subjected to strong hydrodynamic action, resulting in dramatic changes in erosion and deposition, and high weights force the model to prioritize optimizing the parameters of these areas when backpropagating, avoiding key geomorphic features from being submerged in the overall error; the flat area has a slow and spatially homogeneous erosion and deposition process, and the basic weight can prevent the model from overfitting to weak noise. This method significantly improves the sensitivity of micro-topography evolution, such as the prediction accuracy of the downstream deposition hotspots caused by the embankment project.

[0056] Reference Figure 4 In combination with the first aspect, in some implementations, S3 includes: S31: based on the initial topographic boundary conditions, the submerged probability, the average shear stress, the rising tide advantage, and the tidal creek distance function are calculated and generated by the physical mechanism model; S32: input the submerged probability, the average shear stress, the rising tide advantage, and the tidal creek distance function into the geomorphological prediction model, and output the seabed erosion and deposition change in the current period; S33: update the topographic boundary conditions based on the seabed erosion and deposition change in the current period, and take the updated topographic boundary conditions as the input of the next period, and execute S31 to S32 cyclically until the long-term simulation is completed.

[0057] Specifically, the cyclic architecture of the hybrid simulation needs to clarify the division of labor between the physical mechanism model and the geomorphological prediction model: the physical mechanism model (such as Delft3D) is responsible for generating four-channel hydrodynamic feature data - the submerged probability is obtained by statistical grid cell submerged time ratio in the water level fluctuation period; the average shear stress is based on the spatial average of the flow field calculation results; the rising tide advantage is quantified by the ratio of the main tidal component and the shallow water tidal component; the tidal creek distance function (SDF) uses the Euclidean distance transformation algorithm to calculate the distance from each grid point to the nearest tidal creek center. These data need to be unified in spatial resolution and normalized before inputting into the geomorphological prediction model.

[0058] The long-term simulation strategy follows the time scale characteristics of geomorphic evolution: the total time length (e.g. 50 years) is divided into N equal or unequal time periods, which are divided according to the stability of the erosion and deposition rate, the period of external forcing change, etc. In each time period, the physical mechanism model is run based on the current topographic boundary conditions, and four-channel data are output; the geomorphic prediction model generates the sea bed erosion and deposition change amount according to this, and updates the topographic boundary conditions through the “original terrain ± erosion and deposition amount” formula. The updated topography is used as the input of the physical mechanism model of the next time period, and the iteration is repeated until all N time periods are covered. The Caofidian case divides 50 years into 10 5-year periods, and the cumulative erosion and deposition amount of each period is the long-term evolution result.

[0059] This cycle design creates a double technical effect: the physical mechanism model guarantees the physical reality of the hydrodynamic characteristics generated, especially the response to the change of the boundary conditions (such as the integration of sea level rise into the water level boundary) completely follows the principles of fluid mechanics; the geomorphic prediction model replaces the traditional time-consuming sediment transport calculation, and the time consumption of a single iteration is shortened from several hours to minutes. More importantly, the feedback mechanism after the topographic update inherits the dynamic adaptability of the coupled model, avoiding the topographic divergence problem of pure data-driven models.

[0060] In combination with the first aspect, in some implementations, S31 comprises: integrating the sea level rise parameter into the initial topographic boundary conditions; S33 comprises: dividing the long-term simulation into multiple discrete time periods, and the output topographic boundary conditions of each time period are used as the input topographic boundary conditions of the next time period.

[0061] Specifically, when the sea level rise parameter is integrated into the initial topographic boundary conditions, the relative sea level change mechanism needs to be considered: if the absolute sea level rise value is used, it is directly superimposed on the initial water depth field; if land subsidence is involved, a vertical displacement amount needs to be added.

[0062] The division of discrete time periods needs to adapt to the non-uniformity of external forcing: equal time periods are used in the stable period of erosion and deposition rate (such as the uniform rising stage of sea level); in the mutation event period (such as the year of frequent storm surges), the time period length is shortened to capture the rapid response. When the output topographic boundary conditions of each time period are passed to the next time period, data format conversion and spatial interpolation need to be performed to ensure grid compatibility. Alternative solutions include using an adaptive time step algorithm to dynamically adjust the time period length according to the change rate of the topography.

[0063] This design addresses the key difficulties of long-term simulation: dynamic integration of sea level rise parameters and boundary conditions ensures the spatio-temporal accuracy of water level driving forces, avoiding systematic bias caused by fixed boundaries; discrete time step division balances computational efficiency and evolution process resolution, especially the response accuracy of sudden events is improved compared to traditional annual-scale simulation. For example, the process of tidal gully encroachment after storm surge is completely captured in a 3-month short-term simulation, while annual simulation would miss such short-term geomorphic adjustments.

[0064] As Figure 5 shown, in combination with the first aspect, in some implementations, S4 includes: S41: receiving the coastal zone development parameters, climate parameters, and hydrodynamic parameters input by the user through the interactive interface; S42: based on the coastal zone development parameters, climate parameters, and hydrodynamic parameters, calling the hybrid simulation to perform geomorphic rapid calculation; S43: converting the results of the geomorphic rapid calculation into visual graphics output, and providing parameter reset, calculation trigger, and history record management functions.

[0065] Specifically, the interactive interface of the business application module receives three types of parameters through graphical controls: Coastal zone development parameters: reclamation range (map frame selection), port coordinates, dike length, etc. Climate parameters: sea level rise rate options (low / medium / high), storm frequency increase / decrease percentage; Hydrodynamic parameters: tide type (semi-diurnal tide / full-diurnal tide), wave dominant direction angle.

[0066] Input parameter instant trigger threshold verification: if the sea level rise rate exceeds the historical extreme value, a warning box will pop up, and if the dike length exceeds a certain proportion of the coastline, the engineering rationality will be questioned.

[0067] When the geomorphic rapid calculation is executed, the system automatically maps the user parameters to the hybrid model input: the coastal development parameters are converted into grid masks or boundary condition modifications; the climate parameters drive the database to match the closest scenario combination; the hydrodynamic parameters adjust the initial settings of the model. After calling the pre-trained hybrid model chain, the calculation core executes the loop iteration process, and finally outputs the seabed erosion and deposition change matrix.

[0068] The result visualization uses the heat map superimposed on the geographic base map method: the erosion and deposition intensity is rendered with red (deposition) -blue (erosion) color scale, and the tidal gully system is highlighted with black vector lines. The interface provides a parameter reset button (clear current input), a calculation trigger button (start rapid calculation), and a history record panel (store input parameters and result thumbnails). The user can choose a history record to reproduce the complete prediction result.

[0069] The module realizes the "zero professional threshold" business application: the planner does not need to master the principle of numerical model, and can complete the scene setting by map point selection and parameter sliding; the input verification mechanism intercepts common sense errors; the heat map output directly shows the long-term geomorphic impact of the engineering scheme.

[0070] In combination with the first aspect, in some implementations, S41 includes: verifying the threshold range of the coastal development parameter, the climate parameter and the hydrodynamic parameter, and triggering an alarm information when each threshold range is exceeded; S43 includes: superimposing the seabed erosion and deposition change amount in the form of a heat map to a geographic base map, and storing the input parameters and the corresponding erosion and deposition evolution results of each calculation.

[0071] Specifically, the parameter threshold verification is based on actual physical laws: such as the wave direction angle is limited within the historical observation range of the target sea area; the reclamation area cannot exceed a certain proportion of the total tidal flat area to prevent model distortion. The verification logic uses multi-layer conditional judgment: first, detect the legality of data type (such as numerical parameters cannot be characters), then compare the physical feasible range (such as the tidal range cannot be negative), and finally check the compatibility of parameter combination (such as the length of the jetty needs to match the coastal morphology). When the user input exceeds the limit, the interface displays a red warning icon next to the parameter and pops up a specific error message.

[0072] In the result visualization, the seabed erosion and deposition change amount heat map needs to be spatially registered with the geographic base map: a unified coordinate system is adopted using the universal transverse Mercator projection, and the erosion and deposition grid data is matched with the base map resolution through bilinear interpolation. The heat map color scale dynamically adapts to the erosion and deposition extreme value range, and a legend scale is added to improve readability. The historical records are stored in the form of a database table, each record containing a timestamp, a parameter set, a result map path and metadata. When the user clicks on a record, the system automatically restores the parameter settings and reloads the result map.

[0073] This mechanism ensures the robustness of the business application: threshold verification prevents model collapse caused by non-physical input from the source; heat map geographic registration enables decision-makers to accurately locate the engineering impact area (such as the deposition range downstream of the jetty); and the historical record function supports scheme backtracking and comparison. In actual application, a sea level rise value that exceeds the threshold is intercepted and corrected, avoiding false deposition results that violate the conservation of mass, and demonstrating the system's immunity to incorrect input.

[0074] Embodiment 2 As shown in Figure 6 The second aspect provides a tidal flat evolution prediction system based on a hybrid model, which adopts the method provided by any of the above embodiments, and the system comprises: A first construction module for constructing a tidal flat erosion and deposition evolution database of a target coastal zone under different scenarios; The second construction module is connected with the first construction module, and is configured to construct a geomorphology prediction model based on a tidal flat erosion and deposition evolution database through an artificial intelligence algorithm. The hybrid simulation execution module is connected with the second construction module, and is configured to perform hybrid simulation based on the physical mechanism model and the geomorphology prediction model, including: The physical mechanism model generates water dynamic characteristic data, the water dynamic characteristic data is input into the geomorphology prediction model to generate a seabed erosion and deposition change amount, and the seabed erosion and deposition change amount is used to update a topographic boundary condition and perform cyclic iteration. The geomorphology rapid calculation execution and output module is connected with the hybrid simulation execution module, and is configured to configure a business application module based on the hybrid simulation, the business application module receives user input parameters, executes geomorphology rapid calculation, and outputs visualized results.

[0075] The system corresponds to the method provided in the above embodiment 1, and thus will not be described here.

[0076] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the above embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present application.

Claims

1. A tidal flat evolution prediction method based on a hybrid model, characterized in that: include: S1: Construct a database of tidal flat erosion and deposition evolution in the target coastal zone under different scenarios; S2: Based on the tidal flat erosion and deposition evolution database, a landform prediction model is constructed using an artificial intelligence algorithm; S3: Performing a hybrid simulation based on the physical mechanism model and the landform prediction model, including: Generate hydrodynamic characteristic data through the physical mechanism model, input the hydrodynamic characteristic data into the landform prediction model to generate seabed scouring and silting change, update the terrain boundary conditions based on the seabed scouring and silting change, and iterate in a loop; S4: Based on the hybrid simulation, a business application module is configured, and the business application module receives user input parameters, performs landform calculations, and outputs visualization results.

2. The method for predicting tidal flat evolution based on a hybrid model according to claim 1, characterized in that: S1 includes: S11: Use the dynamic geomorphological model to establish a dynamic geomorphological evolution model of the target coastal zone; S12: Based on the dynamic landform evolution model, re-enact the tidal flat dynamic sedimentation process, and calibrate the dynamic landform evolution model in combination with historical topographic data; S13: Based on the calibrated dynamic geomorphological evolution model, the time scale of storm surges and seasonal water level fluctuations is reduced to generate an annual-scale geomorphological evolution characteristic dataset; S14: Based on the annual-scale geomorphological evolution characteristic dataset, the influence of sediment supply, sea level change, tide and wave factors on scouring and deposition evolution is simulated to form the tidal flat scouring and deposition evolution database.

3. The method for predicting tidal flat evolution based on a hybrid model according to claim 2, characterized in that: S12 includes: performing a long-term simulation based on the dynamic landform evolution model, and outputting landform data including the impact of climate change and human activities; S13 includes: extracting storm surge and seasonal water level fluctuation characteristic parameters for time scale reduction based on geomorphological data including climate change and human activity impacts; S14 includes: quantifying the contribution of coastal engineering to landform evolution based on characteristic parameters after time scale reduction.

4. The method for predicting tidal flat evolution based on a hybrid model according to claim 1, wherein S2 include: S21: Based on the tidal flat erosion and deposition evolution database, the submergence probability, mean shear stress, high tide advantage, and tidal channel distance function are sorted out to construct a training data set corresponding to seabed erosion and deposition; S22: Based on the training data set, construct a convolutional neural network model with four-channel hydrodynamic characteristics as input and seabed scouring and deposition changes as output; S23: Based on the convolutional neural network model, configure an encoder-decoder structure and implement cross-layer transfer of feature maps through skip connections, and output an optimized convolutional neural network model; S24: Based on the optimized convolutional neural network model and tidal gully distance function, a regional weighted calculation is performed on the loss function to output a landform prediction model with optimized accuracy.

5. The method for predicting tidal flat evolution based on a hybrid model according to claim 4, characterized in that: S23 includes: passing the feature maps output by each stage of the encoder to the corresponding stage of the decoder for splicing; S24 includes: allocating weight coefficients according to the tidal creek distance function value, and allocating a higher weight value to the tidal creek edge area than to the flat area.

6. The method for predicting tidal flat evolution based on a hybrid model according to claim 1, wherein S3 include: S31: Based on the initial terrain boundary conditions, the physical mechanism model is used to calculate the flooding probability, mean shear stress, high tide advantage, and tidal channel distance function; S32: Inputting the flooding probability, mean shear stress, high tide advantage, and tidal channel distance function into the landform prediction model, and outputting the seabed erosion and deposition change in the current period; S33: updating the terrain boundary conditions based on the seabed scouring and silting changes in the current period, using the updated terrain boundary conditions as input for the next period, and looping through S31 to S32 until the long-duration simulation is completed.

7. The method for predicting tidal flat evolution based on a hybrid model according to claim 6, characterized in that: S31 includes: integrating sea level rise parameters into initial terrain boundary conditions; S33 includes: dividing the long-duration simulation into a plurality of discrete time periods, and using the output terrain boundary conditions of each time period as the input terrain boundary conditions of the next time period.

8. The method for predicting tidal flat evolution based on a hybrid model according to claim 1, wherein S4 include: S41: receiving coastal zone development parameters, climate parameters and hydrodynamic parameters input by the user through an interactive interface; S42: Based on the coastal zone development parameters, climate parameters and hydrodynamic parameters, calling the hybrid simulation to perform landform rapid calculation; S43: Converts the results of rapid landform calculation into visual graphic output, and provides parameter reset, calculation triggering and history record management functions.

9. The method for predicting tidal flat evolution based on a hybrid model according to claim 8, characterized in that: S41 includes: verifying the threshold range of coastal zone development parameters, climate parameters and hydrodynamic parameters, and triggering warning information when the respective threshold ranges are exceeded; S43 includes: superimposing the seabed scouring and silting change amount on the geographic base map in the form of a heat map, and storing the input parameters of each calculation and the corresponding scouring and silting evolution results.

10. A tidal flat evolution prediction system based on a hybrid model, characterized in that: The system adopts the method according to any one of claims 1 to 9, and the system includes: The first construction module is used to construct a database of tidal flat erosion and deposition evolution in the target coastal zone under different scenarios; A second construction module, connected to the first construction module, is used to construct a landform prediction model based on the tidal flat erosion and deposition evolution database through an artificial intelligence algorithm; The hybrid simulation execution module is connected to the second construction module and is used to perform a hybrid simulation based on the physical mechanism model and the landform prediction model, including: Generate hydrodynamic characteristic data through the physical mechanism model, input the hydrodynamic characteristic data into the landform prediction model to generate seabed scouring and silting change, update the terrain boundary conditions based on the seabed scouring and silting change, and iterate in a loop; The landform quick calculation execution and output module is connected to the hybrid simulation execution module and is used to configure a business application module based on the hybrid simulation. The business application module receives user input parameters, executes landform quick calculation and outputs visual results.

Images