A method and system for predicting urban rainstorm water depth based on ConvLSTM model

Abstract

The present invention discloses a method and system for predicting urban rainstorm water depth based on a ConvLSTM model. The method comprises: collecting data from a to-be-simulated area to construct an urban rainstorm model for the to-be-simulated area; performing simulation prediction and data preprocessing on rainstorm event data information in the to-be-simulated area based on the urban rainstorm model for the to-be-simulated area to obtain normalized spatiotemporal characteristic data of rainfall in the to-be-simulated area; constructing a ConvLSTM model based on a convolutional neural network and a long short-term memory network; and performing water depth prediction on the normalized spatiotemporal characteristic data of rainfall in the to-be-simulated area based on the ConvLSTM model to obtain a rainstorm water depth prediction result for the to-be-simulated area. The embodiment of the present invention improves the accuracy of real-time waterlogging disaster forecasts by considering the correlation between adjacent spatial grids and the influence of the spatiotemporal structure of rainfall on the flood process. As a method and system for predicting urban rainstorm water depth based on a ConvLSTM model, the present invention can be widely used in the technical field of urban rainstorm waterlogging simulation.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rainstorm waterlogging simulation, and in particular to a method and system for predicting urban rainstorm water depth based on a ConvLSTM model. Background Art

[0002] Current urban flood control and drainage planning is mostly based on the assumption of uniform rainfall to simplify design and calculations. However, in reality, heavy rain often exhibits significant spatial heterogeneity, resulting in uneven spatial distribution of rainfall. Existing research indicates that this significant spatiotemporal characteristic of rainfall can exacerbate urban waterlogging. Therefore, when addressing the potential for urban waterlogging caused by heavy rain, the spatiotemporal characteristics of heavy rain should be taken into consideration. Early warning of flooding and the development of appropriate response measures can significantly mitigate its impact. However, heavy rain flooding often occurs suddenly and affects a wide area. The diverse and complex distribution of surface cover in urban areas presents challenges in early warning and prediction. Existing research is also focused on improving the timeliness and accuracy of flood forecasts. Currently, early warning and prediction of urban waterlogging are primarily based on flood models based on physical equations. These physical models are mainly categorized as hydrological models, hydrodynamic models, and coupled hydrological and hydrodynamic models. These models solve numerous complex physical equations to achieve more detailed simulations of flood inundation. However, physical model building requires high-precision terrain data and rich hydrological data, which are often difficult to obtain. At the same time, solving the physical equations takes a long time. These problems bring certain limitations to the physical model, which in turn leads to unsatisfactory results in real-time flood disaster forecasting. Summary of the Invention

[0003] In order to solve the above technical problems, the purpose of the present invention is to provide a method and system for predicting urban rainstorm water depth based on the ConvLSTM model, which improves the accuracy of real-time urban waterlogging disaster forecasts by considering the correlation between adjacent spatial grids and the impact of the spatiotemporal structure of rainfall on the flood process.

[0004] The first technical solution adopted by the present invention is: a method for predicting urban rainstorm water depth based on a ConvLSTM model, comprising the following steps:

[0005] Collect data from the area to be simulated and build an urban stormwater model for the area to be simulated;

[0006] Based on the urban rain and flood model of the area to be simulated, the rainstorm event data information of the area to be simulated is simulated and predicted and data preprocessed to obtain the normalized spatiotemporal characteristic data of rainfall in the area to be simulated;

[0007] Construct a ConvLSTM model based on convolutional neural networks and long short-term memory networks;

[0008] Based on the ConvLSTM model, the water depth prediction of the normalized spatiotemporal characteristic data of rainfall in the simulated area is performed to obtain the rainstorm water depth prediction results in the simulated area.

[0009] Furthermore, the step of collecting data from the area to be simulated and constructing an urban stormwater model for the area to be simulated specifically includes:

[0010] Obtain the land use type, elevation and drainage network data of the area to be simulated, and construct a one-dimensional hydrological model of the area to be simulated;

[0011] Obtain rainstorm data information of the area to be simulated and input it into the one-dimensional hydrological model of the area to be simulated for data deduction, and obtain the location coordinates and overflow information of the overflow pipe point under the rainfall conditions of the area to be simulated;

[0012] Obtain the elevation information of the area to be simulated and perform a two-dimensional evolution simulation based on the location coordinates and overflow information of the overflow pipe point under rainfall conditions in the area to be simulated, and construct a two-dimensional hydrodynamic model of the area to be simulated;

[0013] The one-dimensional hydrological model of the area to be simulated is coupled with the two-dimensional hydrodynamic model of the area to be simulated to construct an urban stormwater model of the area to be simulated.

[0014] Furthermore, the step of performing simulation prediction and data preprocessing on the rainstorm event data information of the simulated area based on the urban rainstorm model of the simulated area to obtain normalized spatiotemporal characteristic data of rainfall in the simulated area specifically includes:

[0015] Obtain data information on rainstorm events with different recurrence periods in the simulated area and with different moving directions and paths of rainstorm centers;

[0016] The spatially variable rainfall model based on a fixed spatial structure is used to scale the rainstorm event data information to obtain rainfall information with different spatiotemporal characteristics in the simulated area.

[0017] Inputting rainfall information with different spatiotemporal characteristics of the area to be simulated into an urban rain and flood model of the area to be simulated to perform rainfall simulation prediction, thereby obtaining predicted spatiotemporal characteristic data of rainfall in the area to be simulated, wherein the predicted spatiotemporal characteristic data of rainfall in the area to be simulated includes the inundation range, inundation depth, and hourly evolution process of inundation under different spatiotemporal rainstorm scenarios in the area to be simulated;

[0018] The predicted spatiotemporal characteristic data of rainfall in the area to be simulated are normalized to obtain normalized spatiotemporal characteristic data of rainfall in the area to be simulated.

[0019] Furthermore, the expression of the spatially variable rainfall model based on a fixed spatial structure is specifically as follows:

[0020]

[0021] R(x,y,μ1,μ2,σ,t)=γ(μ1,μ2)f(x,y,μ1,μ2,σ)r(t)

[0022]

[0023] In the above formula, R(x,y,μ1,μ2,σ,t) represents the rainfall intensity at position (x,y) at time t in the spatially variable rainfall model with a fixed spatial structure, (μ1,μ2) represents the position coordinates of the rainfall center, r(t) represents the rainfall intensity at time t under spatially uniform rainfall conditions, A represents the watershed area, f(x,y,μ1,μ2,σ) is the distribution function of the spatial rainfall intensity represented by the two-dimensional truncated Gaussian distribution function, σ represents the standard deviation of the Gaussian distribution, and γ(μ1,μ2) represents the scaling function of the two-dimensional truncated Gaussian distribution, whose purpose is to ensure that the integral of the Gaussian distribution over the domain is equal to 1.

[0024] Furthermore, the expression for normalizing the spatial and temporal characteristic data of the rainfall in the predicted area to be simulated is specifically as follows:

[0025]

[0026] In the above formula, X n represents the normalized temporal and spatial characteristic data of rainfall in the simulated area, X represents the sequence value, and X min Indicates the minimum value of the sequence value, X max Indicates the maximum value of a sequence.

[0027] Furthermore, the ConvLSTM model specifically includes a two-dimensional convolutional long short-term memory network layer, a two-dimensional convolutional neural network layer and a fully connected neural network layer, wherein the output end of the two-dimensional convolutional long short-term memory network layer is connected to the input end of the two-dimensional convolutional neural network layer, and the output end of the two-dimensional convolutional neural network layer is connected to the input end of the fully connected neural network layer.

[0028] Furthermore, the step of performing water depth prediction on the normalized spatiotemporal characteristic data of rainfall in the area to be simulated based on the ConvLSTM model to obtain a rainstorm water depth prediction result for the area to be simulated specifically includes:

[0029] The normalized spatiotemporal characteristic data of rainfall in the area to be simulated is input into the ConvLSTM model;

[0030] Based on the two-dimensional convolutional long short-term memory network layer of the ConvLSTM model, the normalized spatiotemporal characteristic data of rainfall in the simulated area are processed to obtain the spatiotemporal characteristic map of rainfall in the simulated area.

[0031] Based on the two-dimensional convolutional neural network layer of the ConvLSTM model, spatial information feature extraction is performed on the spatiotemporal feature map of rainfall in the simulated area to obtain the convolved spatiotemporal feature map of rainfall in the simulated area.

[0032] Based on the fully connected neural network layer of the ConvLSTM model, feature integration processing is performed on the convolutional spatiotemporal characteristic map of rainfall in the simulated area to obtain the rainstorm water depth prediction result of the simulated area.

[0033] Furthermore, the expression of the two-dimensional convolutional long short-term memory network layer based on the ConvLSTM model is specifically as follows:

[0034]

[0035] In the above formula, σ represents the sigmoid activation function, tanh(·) represents the hyperbolic tangent activation function, and x t represents the input of the current time step, h t-1 represents the hidden state of the previous time step, c t-1 represents the memory unit of the previous time step, f t 、i t 、c t 、o t Respectively represent the forget gate, input gate, memory unit and output gate, * represents the convolution operator, represents the Hadamard product.

[0036] Furthermore, the expression of the two-dimensional convolutional neural network layer based on the ConvLSTM model is specifically as follows:

[0037]

[0038] In the above formula, represents the input of the kth neuron in the lth convolution layer, represents the bias of the k-th neuron in the l-th convolution layer, represents the output of the i-th neuron in the (l-1)th layer of convolution, represents the convolution kernel from the i-th neuron in the (l-1)th layer to the k-th neuron in the l-th layer, and Conv2D(·) represents a two-dimensional convolution operation.

[0039] The second technical solution adopted by the present invention is: an urban rainstorm water depth prediction system based on the ConvLSTM model, comprising:

[0040] The first module is used to collect data from the area to be simulated and build an urban stormwater model for the area to be simulated;

[0041] The second module is used to simulate and predict the rainstorm event data information of the simulated area based on the urban rain and flood model of the simulated area and preprocess the data to obtain the normalized spatiotemporal characteristic data of rainfall in the simulated area;

[0042] The third module is used to build a ConvLSTM model based on convolutional neural networks and long short-term memory networks;

[0043] The fourth module is used to predict the water depth of the normalized spatiotemporal characteristic data of rainfall in the area to be simulated based on the ConvLSTM model to obtain the rainstorm water depth prediction result of the area to be simulated.

[0044] The beneficial effects of the method and system of the present invention are as follows: the present invention collects data from the area to be simulated, constructs an urban rain and flood model for the area to be simulated, further simulates and predicts and preprocesses the rainstorm event data information of the area to be simulated based on the urban rain and flood model, obtains normalized spatiotemporal characteristic data of rainfall in the area to be simulated, fully considers the impact of the spatiotemporal structure of rainfall on the flood process and the complex spatiotemporal structure of rainfall, and then constructs a ConvLSTM model based on a convolutional neural network and a long short-term memory network, predicts the water depth of the normalized spatiotemporal characteristic data of rainfall in the area to be simulated, and considers the correlation between adjacent spatial grids. It can realize rapid and accurate simulation of the evolution process of urban waterlogging under complex spatiotemporal rainfall conditions, thereby realizing rapid and accurate simulation of the urban waterlogging range, water depth and evolution process based on rainfall forecast input conditions, and ultimately improving the accuracy of real-time urban waterlogging disaster forecasts. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flowchart of the steps of a method for predicting urban rainstorm water depth based on a ConvLSTM model of the present invention;

[0046] Figure 2 This is a structural block diagram of an urban rainstorm water depth prediction system based on a ConvLSTM model of the present invention;

[0047] Figure 3 It is a schematic diagram of a one-dimensional hydrological model SWMM of a study area provided by a specific embodiment of the present invention;

[0048] Figure 4 is a schematic diagram of the movement path of the rainstorm center within the study area of the rainstorm scenario provided by a specific embodiment of the present invention;

[0049] Figure 5This is a schematic diagram of the correlation between the spatial difference map of the inundation depth and the simulated water depth in a 100-year Chicago-type flooding event in a flood-prone area provided by a specific embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram showing the correlation between a spatial difference map of flooding depth and simulated water depth in a flood-prone area under a Chicago-type flooding disaster with a return period of 80 years, provided by a specific embodiment of the present invention;

[0051] Figure 7 This is a schematic diagram of the correlation between the spatial difference map of the inundation depth and the simulated water depth in a 50-year Chicago-type flooding event in a flood-prone area provided by a specific embodiment of the present invention;

[0052] Figure 8 This is a schematic diagram of the correlation between the spatial difference map of the inundation depth and the simulated water depth in a 20-year return period Chicago-type flooding in a certain flood-prone area provided by a specific embodiment of the present invention;

[0053] Figure 9 This is a schematic diagram showing the correlation between a spatial difference map of inundation depth and simulated water depth in a flood-prone area under a Chicago-type flood with a return period of 15 years, provided by a specific embodiment of the present invention;

[0054] Figure 10 This is a schematic diagram of the correlation between the spatial difference map of the inundation depth and the simulated water depth in a 10-year Chicago-type flood in a flood-prone area provided by a specific embodiment of the present invention;

[0055] Figure 11 This is a schematic diagram showing the correlation between a spatial difference map of flooding depth and simulated water depth in a two-year Chicago-type flooding event in a flood-prone area provided by a specific embodiment of the present invention;

[0056] Figure 12 This is a schematic diagram of the correlation between the spatial difference map of the inundation depth and the simulated water depth in a Chicago-type flood in a flood-prone area with a return period of one year, provided by a specific embodiment of the present invention. DETAILED DESCRIPTION

[0057] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are provided for ease of description only and do not limit the order of the steps. The order of execution of the steps in the embodiments can be adaptively adjusted based on the understanding of those skilled in the art.

[0058] Reference Figure 1 The present invention provides a method for predicting urban rainstorm water depth based on a ConvLSTM model, which includes the following steps:

[0059] S100, collecting data for the area to be simulated and constructing an urban stormwater model for the area to be simulated;

[0060] Specifically, the land use type, elevation and drainage network data of the area to be simulated are obtained to construct a one-dimensional hydrological model of the area to be simulated; the rainstorm data information of the area to be simulated is obtained and input into the one-dimensional hydrological model of the area to be simulated for data deduction to obtain the position coordinates and overflow volume information of the overflow pipe point under rainfall conditions in the area to be simulated; the elevation information of the area to be simulated is obtained and combined with the position coordinates and overflow volume information of the overflow pipe point under rainfall conditions in the area to be simulated to perform a two-dimensional evolution simulation to construct a two-dimensional hydrodynamic model of the area to be simulated; the one-dimensional hydrological model of the area to be simulated is coupled with the two-dimensional hydrodynamic model of the area to be simulated to construct an urban stormwater model of the area to be simulated.

[0061] In this embodiment, an urban stormwater model is constructed based on the basic data of the study area. Specifically, the land use type, elevation and drainage network data of the study area are collected, and a coupled stormwater simulation based on the one-dimensional hydrological model SWMM and the two-dimensional hydrodynamic model LISFLOOD-FP is constructed. In the coupled stormwater simulation, the land use type, elevation and drainage network data are used to construct the SWMM model; the rainstorm data is input into the SWMM model, and the SWMM model derives the position coordinates and overflow information of the overflow pipe point under rainfall conditions, which are input into the LISFLOOD-FP model and combined with the elevation information to perform a two-dimensional evolution simulation of flooded water. Finally, the LISFLOOD-FP model is used to output the flooding range, flooding depth and flooding evolution process information; the constructed composite flood simulation is calibrated and verified by adjusting the model parameters until the simulation accuracy meets the requirements. Figure 3 As shown, taking a certain river basin in a certain city as an example (the area is about 116.26km 2 ), and construct an urban waterlogging model.

[0062] S200, performing simulation prediction and data preprocessing on rainstorm event data information of the simulated area based on the urban rain and flood model of the simulated area to obtain normalized spatiotemporal characteristic data of rainfall in the simulated area;

[0063] Specifically, data information of rainstorm events with different recurrence periods and different moving directions and paths of rainstorm centers in the area to be simulated are obtained; the rainstorm event data information is scaled by a spatially variable rainfall model based on a fixed spatial structure to obtain rainfall information with different spatiotemporal characteristics in the area to be simulated; the rainfall information with different spatiotemporal characteristics in the area to be simulated is input into an urban rain and flood model in the area to be simulated to perform rainfall simulation prediction to obtain predicted spatiotemporal characteristic data of rainfall in the area to be simulated, wherein the predicted spatiotemporal characteristic data of rainfall in the area to be simulated include the inundation range, inundation depth and hourly evolution process of inundation under different spatiotemporal rainstorm scenarios in the area to be simulated; the predicted spatiotemporal characteristic data of rainfall in the area to be simulated are normalized to obtain normalized spatiotemporal characteristic data of rainfall in the area to be simulated.

[0064] In this embodiment, rainfall scenarios with different spatiotemporal characteristics are designed, and a spatially variable rainfall model based on a fixed spatial structure (SVR-SS) is used to scale the rainfall in the region. The constructed coupled rainfall and flood simulation model is used to generate a spatiotemporal rainstorm-waterlogging dataset.

[0065] It should be noted that different spatiotemporal characteristics refer to rainstorm events with different recurrence periods and different moving directions and paths of the rainstorm center. Different recurrence periods include 1, 2, 5, 10, 15, 20, 30, 50, 80, and 100-year recurrence periods; different moving directions and paths of the rainstorm center refer to the rainstorm center passing through the study area horizontally from left to right or from right to left, passing through the study area vertically from top to bottom or from bottom to top, passing through the study area obliquely from upper left to lower right, from upper right to lower left, from lower left to upper right, or from lower right to upper left, such as Figure 4 shown.

[0066] The spatially variable rainfall model with fixed spatial structure (SVR-SS) is a method that uses the density function of a two-dimensional truncated Gaussian distribution to represent the spatial structure of rainfall while ensuring that the total rainfall in the basin is constant. Its governing equation is as follows:

[0067]

[0068] R(x,y,μ1,μ2,σ,t)=γ(μ1,μ2)f(x,y,μ1,μ2,σ)r(t)

[0069]

[0070] In the above formula, R(x,y,μ1,μ2,σ,t) represents the rainfall intensity at position (x,y) at time t in the spatially variable rainfall model with a fixed spatial structure, (μ1,μ2) represents the position coordinates of the rainfall center, r(t) represents the rainfall intensity at time t under spatially uniform rainfall conditions, A represents the watershed area, f(x,y,μ1,μ2,σ) is the distribution function of the spatial rainfall intensity represented by the two-dimensional truncated Gaussian distribution function, σ represents the standard deviation of the Gaussian distribution, and γ(μ1,μ2) represents the scaling function of the two-dimensional truncated Gaussian distribution, whose purpose is to ensure that the integral of the Gaussian distribution over the domain is equal to 1.

[0071] In addition, it should be noted that the spatiotemporal rainstorm-waterlogging database with different characteristics refers to inputting rainfall processes with different spatiotemporal characteristics into the constructed coupled rain-flood model to simulate the flooding range, flooding depth and hourly evolution of flooding under different spatiotemporal rainstorm scenarios.

[0072] Furthermore, the generated data set is normalized and divided into a training set, a validation set, and a test set. In this embodiment, the normalization method adopts the min-max normalization method to normalize the rainstorm waterlogging input features and the flooding depth sequence data respectively. The normalization is calculated by the following formula, which is expressed as follows:

[0073]

[0074] In the above formula, X n represents the normalized temporal and spatial characteristic data of rainfall in the simulated area, X represents the sequence value, and X min Indicates the minimum value of the sequence value, X max Indicates the maximum value of a sequence.

[0075] Finally, the normalized spatiotemporal rainstorm-flooding database was partitioned into training, validation, and test sets in a ratio of 8:1:1. To ensure the generalization of the model, the partitioned events were adjusted using a random selection and manual adjustment method to ensure that the test set contained rainfall events with different return periods and different rainstorm movement paths.

[0076] S300, based on convolutional neural network and long short-term memory network, build ConvLSTM model;

[0077] Specifically, the ConvLSTM model includes a two-dimensional convolutional long short-term memory network layer, a two-dimensional convolutional neural network layer and a fully connected neural network layer, wherein the output end of the two-dimensional convolutional long short-term memory network layer is connected to the input end of the two-dimensional convolutional neural network layer, and the output end of the two-dimensional convolutional neural network layer is connected to the input end of the fully connected neural network layer.

[0078] In this embodiment, the ConvLSTM algorithm is an integrated deep learning algorithm based on convolutional neural networks and long short-term memory networks. The rapid waterlogging prediction method of the present invention uses the spatiotemporal rainfall process as input. The ConvLSTM network used is a two-dimensional convolutional long short-term memory neural network (ConvLSTM 2D). The main structure of the constructed ConvLSTM network includes a two-dimensional convolutional long short-term memory network layer (ConvLSTM 2D), a two-dimensional convolutional neural network layer (Conv 2D), and a fully connected neural network layer.

[0079] A convolutional long short-term memory (ConvLSTM) model was constructed using the Keras Python library. The ConvLSTM model's Convolutional Long Short-Term Memory layer was designed to consist of three layers, with the tanh function selected as the activation function. The convolution kernel size was set to 7×7, and the number of kernels was 14, 28, and 56, respectively. A two-dimensional convolutional layer was connected after the Convolutional Long Short-Term Memory layer for further feature extraction. The convolution kernel size was set to 7×7, and the number of kernels was set to 1. A single fully connected layer was designed with 1024 neurons. The output layer corresponded to the water depth at each grid.

[0080] When building a ConvLSTM model, multiple parameters need to be set and optimized. In this study, the parameters of the ConvLSTM model were optimized by manual adjustment. The optimized parameter settings are shown in Table 1:

[0081]

[0082] S400: Based on the ConvLSTM model, water depth prediction is performed on the normalized spatiotemporal characteristic data of rainfall in the area to be simulated to obtain a rainstorm water depth prediction result for the area to be simulated.

[0083] Specifically, the normalized spatiotemporal characteristic data of rainfall in the area to be simulated are input into the ConvLSTM model; the two-dimensional convolutional long short-term memory network layer based on the ConvLSTM model performs spatiotemporal feature capture processing on the normalized spatiotemporal characteristic data of rainfall in the area to be simulated to obtain the spatiotemporal characteristic map of rainfall in the area to be simulated; the two-dimensional convolutional neural network layer based on the ConvLSTM model performs spatial information feature extraction processing on the spatiotemporal characteristic map of rainfall in the area to be simulated to obtain the convolved spatiotemporal characteristic map of rainfall in the area to be simulated; the fully connected neural network layer based on the ConvLSTM model performs feature integration processing on the convolved spatiotemporal characteristic map of rainfall in the area to be simulated to obtain the rainstorm water depth prediction result of the area to be simulated.

[0084] In this example, it should be noted that the independent variable of the ConvLSTM network model is the spatiotemporal sequence of rainfall (the rainfall duration is 180 minutes, with 5-minute rainfall as a feature, for a total of 36 features; the spatial resolution of the rainfall is 8 meters, and the spatial rainfall scaling factor has a total of 2340 × 1632 features). The dependent variable is the water depth of different spatial grids (a total of 345,255 grids, with a grid resolution of 8 meters × 8 meters). The spatiotemporal rainstorm-waterlogging scenario was designed, and the events were adjusted using a random selection and manual adjustment method with an 8:1:1 ratio. The dataset was divided into training, validation, and test sets. A total of 64 samples were used for model training, and 8 samples were used for validation during training.

[0085] The convolutional long short-term memory (LSTM) layer effectively captures the spatiotemporal characteristics of the input multidimensional feature grid data by combining convolution operations with the LSTM architecture. It uses the inner product of the sliding window data and the convolution kernel to extract the weights of data at different time steps and spatial locations, thereby generating a new feature map that preserves temporal dependencies.

[0086] The convolution layer focuses on feature extraction of the input data. Through the inner product operation of the sliding window and the convolution kernel, it extracts data features at different locations, forms a feature map, and then extracts spatial information.

[0087] The fully connected layer connects all the neural units in the feature map, passes through the hidden layer, and generates the final result by the output layer. This process integrates the extracted features to form the final prediction.

[0088] In the ConvLSTM network, the previous two-dimensional convolutional long short-term memory network layer is forward propagated to the neuron input of the next layer. The process is as follows:

[0089]

[0090] In the above formula, σ represents the sigmoid activation function, tanth(·) represents the hyperbolic tangent activation function, and x t represents the input of the current time step, h t-1 represents the hidden state of the previous time step, c t-1 represents the memory unit of the previous time step, f t 、i t 、c t 、o t Respectively represent the forget gate, input gate, memory unit and output gate, * represents the convolution operator, represents the Hadamard product.

[0091] A 2D convolutional layer is added after the ConvLSTM layer to further extract features. The convolution operation of the convolutional neural network layer is as follows:

[0092]

[0093] In the above formula, represents the input of the kth neuron in the lth convolution layer, represents the bias of the k-th neuron in the l-th convolution layer, represents the output of the i-th neuron in the (l-1)th layer of convolution, represents the convolution kernel from the i-th neuron in the (l-1)th layer to the k-th neuron in the l-th layer, and Conv2D(·) represents a two-dimensional convolution operation.

[0094] Finally, the output of the two-dimensional convolutional layer is used as the input of the fully connected layer to complete the construction of the model.

[0095] The spatiotemporal changing rainstorm sequence data in the test set is input into the constructed ConvLSTM model, and the changes in the flooded water depth of each spatial grid are output to achieve rapid simulation of the flooding range, flooding depth and flooding evolution process of rainstorm waterlogging.

[0096] Eight different test rainstorms were used as specific examples to test the ConvLSTM model's prediction performance on the spatiotemporal variations of water depth under different rainstorms. The eight test rainstorms included a 100-year return path type 3, an 80-year return path type 1, a 50-year return path type 2, a 20-year return path type 7, a 15-year return path type 8, a 10-year return path type 5, a 2-year return path type 4, and a 1-year return path type 6. For each rainstorm, the ConvLSTM model predicted the spatial water depth for three time steps (a total of three hours, with one prediction per hour). Because each rainstorm predicts a large number of spatial grid maps, and the flooding in the third hour (immediately after the rain ends) is the most severe, the water depth grid map from the third hour is selected here to demonstrate and illustrate the spatial prediction results. The ConvLSTM model predicted the temporal changes in the flooded water depth of 345,255 grids. The simulated values of the coupled stormwater model were compared with the predicted values of the ConvLSTM model to evaluate the water depth prediction effect of the ConvLSTM model.

[0097] Under the 100-year return path type 3 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth of the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 5 (a) and Figure 5 (b), where Figure 5 (a) Spatial difference in flood depth between the Convolutional Long Short-Term Memory Network (ConvLSTM) model prediction and the urban waterlogging model (SWMM+LISFLOOD-FP) simulation for a flood-prone area under a 100-year Chicago-type (r=0.29) storm with moving path 3 (t=3h). Figure 5(b) Correlation between the water depths predicted by the CNN model and those simulated by the urban flooding model for each grid. The spatial flooding depths predicted by the ConvLSTM model are close to those predicted by the urban flooding model, with consistent flood-prone locations. The maximum water depth difference between the two is small (generally less than 0.1 m). The water depths predicted by the ConvLSTM model are highly correlated with those simulated by the urban flooding model, showing a significant linear correlation (P < 0.001). At t = 3 hours, the correlation coefficient R² of the ConvLSTM model predictions is 0.951, the mean absolute error (MAE) is 0.051, and the mean square error (MSE) is 0.034. These results indicate that the temporal variation in water depths predicted by the ConvLSTM model and those simulated by the urban flooding model is small, indicating that the ConvLSTM model can effectively capture the temporal trends in grid water depths.

[0098] Under the 80-year return path type 1 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth by the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 6 (a) and Figure 6 As shown in (b), Figure 6 (a) Spatial difference in flood depth between the prediction of the Convolutional Long Short-Term Memory Network (ConvLSTM) model and the simulation of the urban waterlogging model (SWMM+LISFLOOD-FP) in a flood-prone area under an 80-year return period Chicago-type (r=0.29) rainstorm with moving path 1 (t=3h). Figure 6 (b) Correlation between the water depths predicted by the CNN model and those simulated by the urban flooding model for each grid. The spatial flooding depths predicted by the ConvLSTM model are close to those predicted by the urban flooding model, with consistent flood-prone locations. The maximum water depth difference between the two is small (essentially less than 0.1 m). The water depths predicted by the ConvLSTM model are highly correlated with those simulated by the urban flooding model, showing a significant linear correlation (P < 0.001). At t = 3 hours, the correlation coefficient R² of the ConvLSTM model predictions is 0.967, the mean absolute error (MAE) is 0.028, and the mean square error (MSE) is 0.015. These results indicate that the temporal variation in water depths predicted by the ConvLSTM model and those simulated by the urban flooding model is small, indicating that the ConvLSTM model can effectively capture the temporal trends in grid water depths.

[0099] Under the 50-year return path type 2 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth by the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 7 (a) and Figure 7 (b), where Figure 7(a) Spatial difference in flood depth between the Convolutional Long Short-Term Memory Network (ConvLSTM) model prediction and the urban waterlogging model (SWMM+LISFLOOD-FP) simulation for a flood-prone area under a 50-year return period Chicago-type (r=0.29) storm with moving path 2 (t=3h). Figure 7 (b) Correlation between the water depth predicted by the CNN model and the simulated water depth of the urban waterlogging model for each grid. The spatial flooding depth predicted by the ConvLSTM model is close to the result of the urban waterlogging model, and the flood-prone locations coincide with each other. The maximum water depth difference between the two is small (basically less than 0.1m). The water depth predicted by the ConvLSTM model is highly correlated with the water depth simulated by the urban waterlogging model, showing a significant linear correlation (P<0.001). At t=3h, the correlation coefficient R 2 The results show that the difference between the temporal changes in water depth predicted by the ConvLSTM model and the simulated urban flooding model is small, and the ConvLSTM model can better capture the temporal trends of grid water depth.

[0100] Under the 20-year return path type 7 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth by the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 8 (a) and Figure 8 (b), where Figure 8 (a) Spatial difference in flood depth between the Convolutional Long Short-Term Memory Network (ConvLSTM) model prediction and the urban waterlogging model (SWMM+LISFLOOD-FP) simulation in a flood-prone area under a 20-year return period Chicago-type (r=0.29) storm with moving path 7 (t=3h); Figure 8 (b) Correlation between the water depth predicted by the CNN model and the simulated water depth of the urban waterlogging model for each grid. The spatial flooding depth predicted by the ConvLSTM model is close to the result of the urban waterlogging model, and the flood-prone locations coincide with each other. The maximum water depth difference between the two is small (basically less than 0.1m). The water depth predicted by the ConvLSTM model is highly correlated with the water depth simulated by the urban waterlogging model, showing a significant linear correlation (P<0.001). At t=3h, the correlation coefficient R 2 The results show that the difference between the water depth predictions of the ConvLSTM model and the simulated water depth of the urban flooding model over time is small, and the ConvLSTM model can better capture the trend of grid water depth changes over time.

[0101] Under the 15-year return path type 8 rainstorm, the difference and correlation (t = 3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth of the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 9 (a) and Figure 9 (b), where Figure 9 (a) Spatial difference in flood depth between the Convolutional Long Short-Term Memory Network (ConvLSTM) model prediction and the urban waterlogging model (SWMM+LISFLOOD-FP) simulation in a flood-prone area under a 15-year return period Chicago-type (r=0.29) rainstorm with moving path 8 (t=3h). Figure 9 (b) Correlation between the water depth predicted by the CNN model and the simulated water depth of the urban waterlogging model for each grid. The spatial flooding depth predicted by the ConvLSTM model is close to the result of the urban waterlogging model, and the flood-prone locations coincide with each other. The maximum water depth difference between the two is small (basically less than 0.1m). The water depth predicted by the ConvLSTM model is highly correlated with the water depth simulated by the urban waterlogging model, showing a significant linear correlation (P<0.001). At t=3h, the correlation coefficient R 2 The average absolute error (MAE) is 0.964, the mean square error (MSE) is 0.026, and the mean square error (MSE) is 0.019. The results show that the difference between the water depth changes over time predicted by the ConvLSTM model and simulated by the urban flooding model is small, and the ConvLSTM model can better capture the trend of grid water depth changes over time.

[0102] Under the 10-year return path type 5 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth by the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 10 (a) and Figure 10 (b), where Figure 10 (a) Spatial difference in flood depth between the prediction of the Convolutional Long Short-Term Memory Network (ConvLSTM) model and the simulation of the Urban Waterlogging Model (SWMM+LISFLOOD-FP) in a flood-prone area under a 10-year return period Chicago-type (r=0.29) rainstorm with moving path 5 (t=3h); Figure 10 (b) Correlation between the water depth predicted by the CNN model and the simulated water depth of the urban waterlogging model for each grid. The spatial flooding depth predicted by the ConvLSTM model is close to the result of the urban waterlogging model, and the flood-prone locations coincide with each other. The maximum water depth difference between the two is small (basically less than 0.1m). The water depth predicted by the ConvLSTM model is highly correlated with the water depth simulated by the urban waterlogging model, showing a significant linear correlation (P<0.001). At t=3h, the correlation coefficient R 2The results show that the difference between the temporal changes in water depth predicted by the ConvLSTM model and the simulated water depth by the urban flooding model is small, and the ConvLSTM model can better capture the temporal trends of grid water depth.

[0103] Under the 2-year return path type 4 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth by the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 11 (a) and Figure 11 (b), where Figure 11 (a) Spatial difference in flood depth between the Convolutional Long Short-Term Memory Network (ConvLSTM) model prediction and the urban waterlogging model (SWMM+LISFLOOD-FP) simulation in a flood-prone area under a 2-year return period Chicago-type (r=0.29) rainstorm with moving path 4 (t=3h). Figure 11 (b) Correlation between the water depth predicted by the CNN model and the simulated water depth of the urban waterlogging model for each grid. The spatial flooding depth predicted by the ConvLSTM model is close to the result of the urban waterlogging model, and the flood-prone locations coincide with each other. The maximum water depth difference between the two is small (basically less than 0.1m). The water depth predicted by the ConvLSTM model is highly correlated with the water depth simulated by the urban waterlogging model, showing a significant linear correlation (P<0.001). At t=3h, the correlation coefficient R 2 The average absolute error (MAE) is 0.924, the mean square error (MSE) is 0.023, and the mean square error (MSE) is 0.010. The results show that the difference between the water depth predictions of the ConvLSTM model and the simulated water depth of the urban flooding model over time is small, and the ConvLSTM model can better capture the trend of grid water depth changes over time.

[0104] Under the 1-year return path type 6 rainstorm, the difference and correlation (t=3h) between the water depth predicted by the ConvLSTM model of each grid and the simulated water depth of the urban waterlogging model based on SWMM+LISFLOOD-FP are shown in Figure 2. Figure 12 (a) and Figure 12 As shown in (b), Figure 12 (a) shows the spatial difference in flood depth between the prediction by the Convolutional Long Short-Term Memory Network (ConvLSTM) model and the simulation by the Urban Waterlogging Model (SWMM+LISFLOOD-FP) in a flood-prone area of the embodiment under a 1-year Chicago-type (r=0.29) rainstorm with moving path 6 (t=3h); Figure 12(b) Correlation between the water depths predicted by the CNN model and those simulated by the urban flooding model for each grid. The spatial flooding depths predicted by the ConvLSTM model are close to those predicted by the urban flooding model, with consistent flood-prone locations. The maximum water depth difference between the two is small (essentially less than 0.1 m). The water depths predicted by the ConvLSTM model are highly correlated with those simulated by the urban flooding model, showing a significant linear correlation (P < 0.001). At t = 3 hours, the correlation coefficient R² of the ConvLSTM model predictions is 0.925, the mean absolute error (MAE) is 0.022, and the mean square error (MSE) is 0.005. These results indicate that the temporal variation in water depths predicted by the ConvLSTM model and those simulated by the urban flooding model is small, indicating that the ConvLSTM model can effectively capture the temporal trends in grid water depths.

[0105] From the perspective of simulation duration, the ConvLSTM model can complete the calculation of the spatial flooding depth and evolution of the study area (116.26 km2) within 2 seconds, and can output hourly urban flooding inundation maps for the next 3 hours for each rainstorm. The urban flooding simulation based on SWMM+LISFLOOD-FP basically takes 1 hour, and the computing efficiency of the former is more than 1,800 times higher than that of the latter.

[0106] Reference Figure 2 , an urban rainstorm water depth prediction system based on ConvLSTM model, including:

[0107] The first module 201 is used to collect data from the area to be simulated and build an urban stormwater model for the area to be simulated;

[0108] The second module 202 is used to simulate and predict the rainstorm event data information of the simulated area based on the urban rain and flood model of the simulated area and perform data preprocessing to obtain normalized spatiotemporal characteristic data of rainfall in the simulated area;

[0109] The third module 203 is used to build a ConvLSTM model based on a convolutional neural network and a long short-term memory network;

[0110] The fourth module 204 is used to perform water depth prediction on the normalized spatiotemporal characteristic data of rainfall in the area to be simulated based on the ConvLSTM model to obtain a rainstorm water depth prediction result for the area to be simulated.

[0111] The contents of the above method embodiments are all applicable to the present system embodiments. The functions specifically implemented by the present system embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0112] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art can make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A method for predicting urban rainstorm depth based on ConvLSTM model, characterized in that: The following steps are involved: Collect data from the area to be simulated and build an urban stormwater model for the area to be simulated; Obtain data information on rainstorm events with different recurrence periods in the simulated area and with different moving directions and paths of rainstorm centers; The different moving directions and paths of the rainstorm center refer to the rainstorm center passing through the study area horizontally from left to right or from right to left, passing through the study area vertically from top to bottom or from bottom to top, and passing through the study area obliquely from upper left to lower right, from upper right to lower left, from lower left to upper right, or from lower right to upper left. The spatially variable rainfall model based on a fixed spatial structure is used to scale the rainstorm event data information to obtain rainfall information with different spatiotemporal characteristics in the simulated area. The expression of the spatially variable rainfall model based on a fixed spatial structure is specifically as follows: ； In the above formula, A spatially variable rainfall model representing a fixed spatial structure in time Time position The rainfall intensity, represents the location coordinates of the rainfall center, Indicates spatially uniform rainfall The rainfall intensity at the time, represents the watershed area, is the distribution function of spatial rainfall intensity represented by a two-dimensional truncated Gaussian distribution function, represents the standard deviation of the Gaussian distribution, represents the scaling function of the two-dimensional truncated Gaussian distribution; Inputting rainfall information with different spatiotemporal characteristics of the area to be simulated into an urban rain and flood model of the area to be simulated to perform rainfall simulation prediction, thereby obtaining predicted spatiotemporal characteristic data of rainfall in the area to be simulated, wherein the predicted spatiotemporal characteristic data of rainfall in the area to be simulated includes the inundation range, inundation depth, and hourly evolution process of inundation under different spatiotemporal rainstorm scenarios in the area to be simulated; Performing data normalization on the predicted spatiotemporal characteristic data of rainfall in the area to be simulated to obtain normalized spatiotemporal characteristic data of rainfall in the area to be simulated; Construct a ConvLSTM model based on convolutional neural networks and long short-term memory networks; Based on the ConvLSTM model, the water depth prediction of the normalized spatiotemporal characteristic data of rainfall in the simulated area is performed to obtain the rainstorm water depth prediction results in the simulated area.

2. The urban rainstorm water depth prediction method based on the ConvLSTM model according to claim 1 is characterized in that: The step of collecting data from the area to be simulated and constructing an urban stormwater model for the area to be simulated specifically includes: Obtain the land use type, elevation and drainage network data of the area to be simulated, and construct a one-dimensional hydrological model of the area to be simulated; Obtain rainstorm data information of the area to be simulated and input it into the one-dimensional hydrological model of the area to be simulated for data deduction, and obtain the location coordinates and overflow information of the overflow pipe point under the rainfall conditions of the area to be simulated; Obtain the elevation information of the area to be simulated and perform a two-dimensional evolution simulation based on the location coordinates and overflow information of the overflow pipe point under rainfall conditions in the area to be simulated, and construct a two-dimensional hydrodynamic model of the area to be simulated; The one-dimensional hydrological model of the area to be simulated is coupled with the two-dimensional hydrodynamic model of the area to be simulated to construct an urban stormwater model of the area to be simulated.

3. The urban rainstorm water depth prediction method based on the ConvLSTM model according to claim 2 is characterized in that: The expression for normalizing the spatial and temporal characteristic data of the rainfall in the predicted area to be simulated is specifically as follows: ； In the above formula, represents the normalized temporal and spatial characteristic data of rainfall in the area to be simulated, Represents a sequence value, Indicates the minimum value of the sequence value, Indicates the maximum value of a sequence.

4. The method for predicting urban rainstorm depth based on the ConvLSTM model according to claim 3, characterized in that: The ConvLSTM model specifically includes a two-dimensional convolutional long short-term memory network layer, a two-dimensional convolutional neural network layer and a fully connected neural network layer, wherein the output end of the two-dimensional convolutional long short-term memory network layer is connected to the input end of the two-dimensional convolutional neural network layer, and the output end of the two-dimensional convolutional neural network layer is connected to the input end of the fully connected neural network layer.

5. The urban rainstorm water depth prediction method based on the ConvLSTM model according to claim 4 is characterized in that: The step of performing water depth prediction on the normalized spatiotemporal characteristic data of rainfall in the area to be simulated based on the ConvLSTM model to obtain a rainstorm water depth prediction result for the area to be simulated specifically includes: The normalized spatiotemporal characteristic data of rainfall in the area to be simulated is input into the ConvLSTM model; Based on the two-dimensional convolutional long short-term memory network layer of the ConvLSTM model, the normalized spatiotemporal characteristic data of rainfall in the simulated area are processed to obtain the spatiotemporal characteristic map of rainfall in the simulated area. Based on the two-dimensional convolutional neural network layer of the ConvLSTM model, spatial information feature extraction is performed on the spatiotemporal feature map of rainfall in the simulated area to obtain the convolved spatiotemporal feature map of rainfall in the simulated area. Based on the fully connected neural network layer of the ConvLSTM model, feature integration processing is performed on the convolved spatiotemporal characteristic map of rainfall in the simulated area to obtain the rainstorm water depth prediction result of the simulated area.

6. The urban rainstorm water depth prediction method based on the ConvLSTM model according to claim 5 is characterized in that: The expression of the two-dimensional convolutional long short-term memory network layer based on the ConvLSTM model is specifically as follows: ； ； ； ； ； In the above formula, represents the sigmoid activation function, represents the hyperbolic tangent activation function, represents the input of the current time step, represents the hidden state at the previous time step, represents the memory unit of the previous time step, Represent the forget gate, input gate, memory unit and output gate respectively. represents the convolution operator, represents the Hadamard product.

7. The urban rainstorm water depth prediction method based on the ConvLSTM model according to claim 6 is characterized in that: The expression of the two-dimensional convolutional neural network layer based on the ConvLSTM model is specifically as follows: ； In the above formula, Indicates the Convolutional layer neuron input, Indicates the Convolutional layer The bias of a neuron, Indicates the Convolutional layer The output of a neuron, express Convolutional layer neurons to the Convolutional layer The convolution kernel of neurons, Represents a two-dimensional convolution operation.

8. An urban rainstorm water depth prediction system based on ConvLSTM model, characterized by: Includes the following modules: The first module is used to collect data from the area to be simulated and build an urban stormwater model for the area to be simulated; The second module is used to obtain data information of rainstorm events with different recurrence periods in the simulated area and different moving directions and paths of the rainstorm center; The different moving directions and paths of the rainstorm center refer to the rainstorm center passing through the study area horizontally from left to right or from right to left, passing through the study area vertically from top to bottom or from bottom to top, and passing through the study area obliquely from upper left to lower right, from upper right to lower left, from lower left to upper right, or from lower right to upper left. The spatially variable rainfall model based on a fixed spatial structure is used to scale the rainstorm event data information to obtain rainfall information with different spatiotemporal characteristics in the simulated area. The expression of the spatially variable rainfall model based on a fixed spatial structure is specifically as follows: ； ； ； In the above formula, A spatially variable rainfall model representing a fixed spatial structure in time Time position The rainfall intensity, represents the location coordinates of the rainfall center, Indicates spatially uniform rainfall The rainfall intensity at the time, represents the watershed area, is the distribution function of spatial rainfall intensity represented by a two-dimensional truncated Gaussian distribution function, represents the standard deviation of the Gaussian distribution, represents the scaling function of the two-dimensional truncated Gaussian distribution; Inputting rainfall information with different spatiotemporal characteristics of the area to be simulated into an urban rain and flood model of the area to be simulated to perform rainfall simulation prediction, thereby obtaining predicted spatiotemporal characteristic data of rainfall in the area to be simulated, wherein the predicted spatiotemporal characteristic data of rainfall in the area to be simulated includes the inundation range, inundation depth, and hourly evolution process of inundation under different spatiotemporal rainstorm scenarios in the area to be simulated; Performing data normalization on the predicted spatiotemporal characteristic data of rainfall in the area to be simulated to obtain normalized spatiotemporal characteristic data of rainfall in the area to be simulated; The third module is used to build a ConvLSTM model based on convolutional neural networks and long short-term memory networks; The fourth module is used to predict the water depth of the normalized spatiotemporal characteristic data of rainfall in the area to be simulated based on the ConvLSTM model to obtain the rainstorm water depth prediction result of the area to be simulated.

Images