Rainfall prediction method, rainfall prediction device, storage medium and electronic equipment

Abstract

The invention discloses a rainfall prediction method, a rainfall prediction device, a storage medium and electronic equipment, and relates to the field of data processing. The precision of rainfall prediction can be improved. The rainfall prediction method comprises the following steps: acquiring current rainfall observation data of a target forecast area and a current environmental impact factor of the target forecast area; extracting target features from the current environmental influence factors based on a random forest; inputting the current rainfall observation data and the target features into the convolutional neural network model, the back propagation neural network model and the time convolutional network model to obtain first prediction data, second prediction data and third prediction data which are respectively output; and combining the first prediction data, the second prediction data and the third prediction data to determine continuous rainfall data of the target prediction region.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a precipitation prediction method, a precipitation prediction device, a storage medium and an electronic device. Background Art

[0002] At present, meteorological models and statistical methods are usually used to predict persistent precipitation. Meteorological models are methods that use physical principles to simulate atmospheric movement and climate change. These models usually need to consider many factors, such as temperature, humidity, wind speed, pressure, etc., as well as the interactions between these factors. By inputting initial conditions and boundary conditions, meteorological models can predict future weather conditions. However, the disadvantage of this method is that it is computationally intensive and requires high-performance computers for calculations. It is also highly sensitive to initial and boundary conditions, which may lead to unstable prediction results. Statistical methods are a method of predicting future trends by building mathematical models based on historical data. These methods usually include time series analysis, regression analysis, machine learning, etc. The advantages of these methods are that they are simple and easy to use, and have low requirements for computing resources. However, the accuracy of statistical methods may be affected by factors such as data quality and quantity, model selection, and training and testing data sets. These problems have caused inaccuracies and risks in the prediction of persistent precipitation to a certain extent. Summary of the invention

[0003] The present application provides a precipitation prediction method, a precipitation prediction device, a storage medium and an electronic device, which can improve the accuracy of precipitation prediction.

[0004] In a first aspect, the present application provides a precipitation prediction method, comprising:

[0005] Obtaining precipitation observation data during continuous precipitation in the target forecast area and environmental impact factors of the target forecast area;

[0006] Determine the importance of the environmental impact factor to the precipitation observation data based on random forest, and determine the target feature based on the importance;

[0007] A convolutional neural network model, a back propagation neural network model and a temporal convolutional network model are obtained through the precipitation observation data and the target feature training;

[0008] Inputting the current precipitation data of the target forecast area into the convolutional neural network model, the back propagation neural network model, and the time convolutional network model respectively, to obtain the first prediction data, the second prediction data, and the third prediction data as outputs;

[0009] The continuous precipitation data of the target forecast area is determined by combining the first forecast data, the second forecast data and the third forecast data.

[0010] According to the precipitation prediction method in the present embodiment, a prediction model is constructed by obtaining the precipitation observation data during continuous precipitation in the target forecast area and the environmental impact factors of the target forecast area, so that the model is more targeted to the target forecast area and continuous precipitation, and the accuracy of continuous precipitation prediction is improved. In addition, the random forest algorithm is used to extract the target features in the environmental impact factors, which helps to identify the key factors that have the greatest impact on precipitation, and then helps to improve the prediction ability of the model. By training three different models of convolutional neural network, back propagation neural network and time convolutional network, model integration is achieved, and the advantages of different models can be used to improve the robustness and accuracy of prediction. The time convolutional network model can capture the time dependence in precipitation data and improve the prediction accuracy for continuous precipitation. Combining the prediction results of three different models, a more comprehensive and reliable precipitation prediction can be obtained, the prediction error of a single model is reduced, and the overall prediction performance is improved.

[0011] In an exemplary embodiment, the combining the first prediction data, the second prediction data, and the third prediction data to determine the continuous precipitation data of the target forecast area includes:

[0012] Determining respective weights of the first prediction data, the second prediction data, and the third prediction data;

[0013] Based on the weight, the first prediction data, the second prediction data and the third prediction data are weighted averaged to obtain continuous precipitation data.

[0014] In an exemplary embodiment, determining the weights of the first prediction data, the second prediction data, and the third prediction data includes:

[0015] Acquire a test set, wherein the test set includes continuous precipitation observation data in a target forecast area;

[0016] Predicting the continuous precipitation observation data by using the convolutional neural network model, the back propagation neural network model, and the time convolutional network model to obtain a first test result, a second test result, and a third test result, respectively;

[0017] Determine the prediction pass rates of the convolutional neural network model, the back propagation neural network model, and the temporal convolutional network model based on the first test result, the second test result, and the third test result, respectively;

[0018] Based on the prediction pass rate, the weights of the convolutional neural network model, the back propagation neural network model and the temporal convolutional network model are determined.

[0019] In an exemplary embodiment, it further includes:

[0020] The first test result, the second test result and the third test result are solved by the least square method to obtain the weights of the convolutional neural network model, the back propagation neural network model and the temporal convolutional network model.

[0021] In an exemplary embodiment, the obtaining of current precipitation data of a target forecast area and current environmental impact factors of the target forecast area includes:

[0022] Obtaining current precipitation data for the target forecast area from a meteorological information center;

[0023] The meteorological forecast data output by the CFSv2 model is obtained, and the current environmental impact factors of the target forecast area are extracted from the meteorological forecast data.

[0024] In an exemplary embodiment, extracting target features from the current environmental impact factors based on random forests includes:

[0025] Obtaining a training set, wherein the training set includes daily precipitation observation data and environmental influencing factors during historical continuous precipitation;

[0026] Training a random forest with the training set enables the random forest to predict precipitation data of the next day based on the environmental influencing factors and the precipitation observation data of the previous day;

[0027] Determine the importance of the current environmental impact factor to the current precipitation observation data based on the trained random forest;

[0028] Target features are extracted from the current environmental impact factors based on the importance.

[0029] In an exemplary embodiment, it further includes:

[0030] Extracting characteristic variables from the environmental influencing factors by using the random forest;

[0031] The convolutional neural network model, the back propagation neural network model, and the time convolutional network model are obtained based on the daily precipitation observation data during the historical continuous precipitation and the characteristic variables.

[0032] In a second aspect, the present application provides a precipitation prediction device, comprising:

[0033] A data acquisition module, used to acquire current precipitation observation data of a target forecast area and current environmental impact factors of the target forecast area;

[0034] A feature extraction module, used for extracting target features from the current environmental impact factors based on random forest;

[0035] A prediction module, used for inputting the current precipitation observation data and the target features into the convolutional neural network model, the back propagation neural network model, and the time convolutional network model to obtain first prediction data, second prediction data, and third prediction data as outputs respectively;

[0036] The comprehensive prediction module is used to determine the continuous precipitation data of the target forecast area by combining the first prediction data, the second prediction data and the third prediction data.

[0037] In a third aspect, the present application provides an electronic device, the electronic device comprising a memory and one or more processors. The memory stores one or more computer programs, the computer programs comprising instructions, and when the instructions are executed by the processor, the electronic device can execute the precipitation prediction method as in the first aspect.

[0038] In a fourth aspect, the present application provides a computer-readable storage medium having instructions stored therein. When the instructions are executed on an electronic device, the electronic device executes the precipitation prediction method as in the first aspect.

[0039] In a fifth aspect, the present application provides a computer program product. When the computer program product is run on an electronic device, the electronic device executes the precipitation prediction method as described in the first aspect.

[0040] It can be understood that the beneficial effects that can be achieved by the precipitation prediction device, electronic device, computer-readable storage medium, and computer program product provided above can refer to the beneficial effects in the first aspect and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of a precipitation prediction method provided in an embodiment of the present application;

[0042] Figure 2 A line graph showing the prediction effect of the precipitation prediction method provided in the embodiment of the present application;

[0043] Figure 3 A schematic diagram of the structure of a precipitation prediction device provided in an embodiment of the present application;

[0044] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to facilitate the clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, the words "first", "second" and the like are used to distinguish the same items or similar items with substantially the same functions and effects. For example, the first chip and the second chip are only used to distinguish different chips, and their order is not limited. It can be understood by those skilled in the art that the words "first", "second" and the like do not limit the quantity and execution order, and the words "first", "second" and the like do not necessarily limit the difference. It should be noted that in the embodiments of the present application, the words "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other embodiments or design solutions. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way. In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more.

[0046] It should be noted that the “at…” in the embodiments of the present application may be the instant when a certain situation occurs, or may be a period of time after the occurrence of a certain situation, and the embodiments of the present application do not specifically limit this.

[0047] The implementation of this embodiment will be described in detail below with reference to the accompanying drawings.

[0048] This embodiment provides a precipitation prediction method that can predict continuous precipitation and improve the accuracy of the prediction. For example, the precipitation prediction method can be applied to various electronic devices such as computers (PCs), tablet computers, virtual reality / augmented reality devices, wearable devices, industrial computers, and car computers; it can also be applied to servers, cloud computing, server clusters, etc., and this embodiment does not specifically limit this.

[0049] Figure 1 A flow chart of a precipitation prediction method provided in an embodiment of the present application is shown.

[0050] like Figure 1 As shown, the precipitation prediction method may include the following steps:

[0051] Step 101: Acquire current precipitation observation data of a target forecast area and current environmental impact factors of the target forecast area.

[0052] The current precipitation observation data may include daily precipitation data up to the forecast date, such as daily precipitation. The target forecast area refers to the area where precipitation forecast is required.

[0053] Specifically, the current precipitation data of the target forecast area is obtained from the meteorological information center; the meteorological forecast data output by the CFSv2 model is obtained, and the current environmental impact factors of the target forecast area are extracted from the meteorological forecast data.

[0054] CFSv2 (Climate Forecast System Version 2) is a global climate model used for climate prediction and seasonal weather forecast. In this application, the data output by the CFSv2 model is used as the data source.

[0055] First, prepare two sets of data, one is the precipitation observation data from the National Meteorological Information Center, and the other is to prepare the initial screening factor data set. Daily meteorological observation data, that is, current precipitation data, can be obtained through the National Meteorological Information Center. Based on the CFSv2 model, according to the time of prediction, precipitation, sea level pressure, and potential heights at 200hPa, 500hPa, 700hPa, 850hPa, 925hPa, relative humidity, temperature, and UV wind speed data are extracted to construct a factor data set containing 27 characteristic variables, that is, the current environmental influencing factors.

[0056] Step 102: extracting target features from the current environmental impact factors based on random forest.

[0057] Random Forest (RF) is an ensemble learning method that improves the accuracy of predictions by constructing multiple decision trees and voting on their prediction results. In this embodiment, the random forest is pre-trained, and then the random forest can be used to evaluate the importance of features (environmental influencing factors). Specifically, it includes: obtaining a training set, which includes daily precipitation observation data and environmental influencing factors during historical continuous precipitation; training a random forest with the training set so that the random forest can predict the precipitation data of the next day based on the environmental influencing factors and daily precipitation observation data; determining the importance of the current environmental influencing factors to the current precipitation observation data based on the trained random forest; extracting target features from the current environmental influencing factors based on the importance.

[0058] This implementation method can be used to predict continuous precipitation. Therefore, daily precipitation observation data and environmental influencing factors during historical continuous precipitation are extracted, and the data is preprocessed, such as removing missing values, processing outliers, standardizing, normalizing data, etc., to obtain a training set. The acquired environmental influencing factors are used as input features to construct a decision tree. Each tree is constructed by selecting feature variables through random sampling. The features can be determined by bootstrap sampling, and the best split point can be selected by information gain or Gini impurity. The node is split until the preset stop condition is reached as a leaf node. Each tree can predict the prediction result, and finally the prediction result with the highest probability can be selected. Through training, the difference between the precipitation data predicted by the random forest for the daily precipitation observation data at the front of the time series and the actual precipitation data can be reduced to below the threshold.

[0059] Each tree in the random forest can calculate the contribution of each feature variable in the node split, which is evaluated by measuring the improvement of node purity after the feature variable is split. The purity of the feature is calculated by Gini impurity and information gain, and the contribution of each feature variable in all trees is calculated to obtain the importance score of each feature variable. Based on the importance score, 10 environmental factors with higher importance can be selected from the environmental factors as the final feature variables.

[0060] In this implementation, the current environmental impact factors can be input into the trained random forest to obtain the importance score of each current environmental impact factor, and then a certain number of environmental factors with higher importance, such as the top 10, are output as target features.

[0061] The random forest algorithm is a powerful nonlinear model that can effectively handle complex data relationships and has good generalization ability. The random forest algorithm is used to screen factors. The random forest algorithm selects the top 10 factors with the largest contribution by calculating the importance of the factors. In this way, the main factors that have the greatest impact on the continuous precipitation in Yunnan can be obtained, which can provide a scientific basis for rainstorm forecasting and disaster prevention and mitigation, so as to improve the resilience of Yunnan in extreme weather conditions.

[0062] Step 103: Input the current precipitation observation data and the target features into the convolutional neural network model, the back propagation neural network model, and the temporal convolutional network model to obtain the first prediction data, the second prediction data, and the third prediction data as outputs.

[0063] The target features output by the random forest and the current precipitation observation data are taken as input into the convolutional neural network model, the back propagation neural network model, and the temporal convolutional network model to obtain the prediction data output by these three models, which are the first prediction data, the second prediction data, and the third prediction data.

[0064] Among them, the Temporal Convolutional Network (TCN) model is a deep learning model for processing time series data. It can effectively process long sequences and control the receptive field of the network by adjusting the expansion factor, making it more suitable for processing time series data with long-term dependencies.

[0065] The Convolutional Neural Network (CNN) model is able to extract and recognize complex patterns in data, and due to the characteristics of the convolutional structure, it is data invariant to translation, scaling, and rotation.

[0066] The backpropagation neural network (BPNN) model is capable of supervised learning. The weights and biases in the network can be adjusted according to the training set through the backpropagation algorithm to minimize the prediction error.

[0067] This embodiment also includes: extracting characteristic variables from the environmental influencing factors through the random forest; training the convolutional neural network model, the back propagation neural network model, and the time convolutional network model based on the daily precipitation observation data during the historical continuous precipitation and the characteristic variables.

[0068] A training set is obtained, and a preset number of feature variables with higher importance are extracted from the environmental influencing factors in the training set through random forest. These feature variables and the precipitation observation data of the previous i days are input into the convolutional neural network model, the back propagation neural network model, and the time convolutional network model. The three models are trained separately, and the parameters of the model are adjusted by the difference between the predicted precipitation data of the i+1th day output by the model and the actual precipitation data of the i+1th day, until the difference between the predicted precipitation data output by the model and the actual precipitation data meets the preset conditions, and the training is completed.

[0069] The trained convolutional neural network model, back propagation neural network model and temporal convolutional network model are used to predict the current precipitation data.

[0070] Step 104: Determine continuous precipitation data for the target forecast area by combining the first forecast data, the second forecast data and the third forecast data.

[0071] The continuous precipitation data refers to the precipitation data after the current time, such as precipitation, etc. For example, the current precipitation observation data may be the precipitation data of the previous j days, and the continuous precipitation data may be the precipitation data of the jth day, the j+1th day, the j+2th day, and so on.

[0072] In this implementation, the final continuous precipitation data can be obtained by calculating the average value of the three prediction data output by the above three models. Alternatively, the final continuous precipitation data can be obtained by adopting a weighted method. Specifically, it includes: determining the weights of the first prediction data, the second prediction data, and the third prediction data; and performing weighted averaging on the first prediction data, the second prediction data, and the third prediction data based on the weights to obtain the continuous precipitation data.

[0073] Exemplarily, the process of determining the weight is as follows: obtaining a test set, which includes continuous precipitation observation data in the target forecast area; predicting the continuous precipitation observation data through the convolutional neural network model, the back propagation neural network model and the time convolutional network model, and obtaining a first test result, a second test result, and a third test result respectively; determining the prediction pass rates of the convolutional neural network model, the back propagation neural network model and the time convolutional network model based on the first test result, the second test result and the third test result respectively; determining the respective weights of the convolutional neural network model, the back propagation neural network model and the time convolutional network model based on the prediction pass rates.

[0074] The prediction pass rates of the convolutional neural network model, the back propagation neural network model, and the temporal convolutional network model can be counted in advance through the test set, and then weights are assigned to each model according to the prediction pass rates. For example, the pass rates of the three models in the rate period are 0.6, 0.7, and 0.8 respectively. The weights of each model are calculated according to the pass rates, which are 0.6 / (0.6+0.7+0.8)=0.286, 0.7 / (0.6+0.7+0.8)=0.334, and 0.8 / (0.6+0.7+0.8)=0.380 respectively.

[0075] In an exemplary embodiment, the least squares method is used to solve the first test result, the second test result and the third test result to obtain the respective weights of the convolutional neural network model, the back propagation neural network model and the temporal convolutional network model.

[0076] In an exemplary embodiment, the weights may be optimized by a simulated annealing algorithm, or other optimization algorithms may be used to optimize the weights, such as an Adam optimizer, a genetic algorithm, a stochastic gradient descent method, and the like.

[0077] This embodiment also provides a forecast model (Ensemble), which can implement the above steps 102 to 104. Specifically, the forecast model can be constructed using random forest, convolutional neural network model, back propagation neural network model, and time convolutional network model. Specifically, the forecast model can include three branches, the first branch is composed of random forest and convolutional neural network forecast model, the second branch is composed of random forest and back propagation neural network forecast model, and the third branch is composed of random forest and time convolutional network forecast model; the outputs of the three branches are weighted averaged by the adaptive weight module, and the continuous precipitation data is finally output. In each branch, the target feature is extracted by random forest, and the feature is predicted and processed by convolutional neural network model, back propagation neural network model or time convolutional network model to obtain future precipitation data. Then the adaptive weight module performs weighted average on the precipitation data output by each branch, and outputs the final result, i.e., continuous precipitation data. Through this scheme, the continuous precipitation in the target forecast area can be accurately predicted, thereby providing valuable reference information for related fields, which is conducive to preventing and responding to extreme weather events.

[0078] In this implementation, the forecast model constructed based on the random forest screening factor can reduce the limitations of a single model in capturing data trends, integrate the advantages of multiple models, and significantly improve the overall forecast performance, which is better than each single model. By integrating the forecast results of multiple models, the forecast can effectively reduce the deviation and uncertainty in the forecast of individual models, thereby providing more robust and accurate forecast results.

[0079] Based on the above analysis, the 10 factors selected by the random forest algorithm and the observed precipitation data were used to construct various models (including the above forecast model, a separate convolutional neural network model, a back propagation neural network model, and a time convolutional network model). The model prediction effects are as follows: Figure 2 As shown in Figure 2, OBS refers to the actual observed precipitation data. Figure 2 It can be seen that the overall trend of the predicted values ​​of each model is consistent with the measured values. The correlation coefficients of the convolutional neural network (CNN), the temporal convolutional network (TCN), the back propagation neural network (BPNN) and the ensemble forecast model (Ensemble) are 0.75, 0.80, 0.78 and 0.89, respectively, indicating that the performance of the ensemble forecast is better than that of other single models.

[0080] Furthermore, this embodiment also provides a precipitation prediction device, which can be used to execute the above precipitation prediction method. Figure 3As shown, the precipitation prediction device 300 may specifically include: a data acquisition module 301, used to obtain the current precipitation observation data of the target forecast area and the current environmental impact factors of the target forecast area; a feature extraction module 302, used to extract target features from the current environmental impact factors based on random forests; a prediction module 303, used to input the current precipitation observation data and the target features into the convolutional neural network model, the back propagation neural network model, and the time convolutional network model to obtain the first prediction data, the second prediction data, and the third prediction data outputted by each; a comprehensive prediction module 304, used to determine the continuous precipitation data of the target forecast area in combination with the first prediction data, the second prediction data, and the third prediction data.

[0081] In an exemplary embodiment, the comprehensive prediction module 304 specifically includes a weighting module for determining the weights of the first prediction data, the second prediction data, and the third prediction data; based on the weights, the first prediction data, the second prediction data, and the third prediction data are weighted averaged to obtain continuous precipitation data.

[0082] In an exemplary embodiment, the weighting module is also specifically used to: obtain a test set, which includes continuous precipitation observation data in the target forecast area; predict the continuous precipitation observation data through the convolutional neural network model, the back propagation neural network model and the time convolutional network model to obtain a first test result, a second test result and a third test result respectively; determine the prediction pass rate of the convolutional neural network model, the back propagation neural network model and the time convolutional network model based on the first test result, the second test result and the third test result respectively; determine the respective weights of the convolutional neural network model, the back propagation neural network model and the time convolutional network model based on the prediction pass rate.

[0083] In an exemplary embodiment, the weighting module is also specifically used to: use the least squares method to solve the first test result, the second test result and the third test result to obtain the respective weights of the convolutional neural network model, the back propagation neural network model and the temporal convolutional network model.

[0084] In an exemplary embodiment, the data acquisition module 301 is specifically used to obtain current precipitation data of the target forecast area from the meteorological information center; obtain meteorological forecast data output by the CFSv2 model, and extract current environmental impact factors of the target forecast area from the meteorological forecast data.

[0085] In an exemplary embodiment, the feature extraction module 302 is specifically used to: obtain a training set, the training set including daily precipitation observation data and environmental influencing factors during historical continuous precipitation; train a random forest using the training set, so that the random forest can predict the precipitation data of the next day based on the environmental influencing factors and the precipitation observation data of the previous day; determine the importance of the current environmental influencing factors to the current precipitation observation data based on the trained random forest; and extract target features from the current environmental influencing factors based on the importance.

[0086] In an exemplary embodiment, the precipitation prediction device 300 also includes a model training module for extracting characteristic variables from the environmental influencing factors through the random forest; the convolutional neural network model, the back propagation neural network model, and the time convolutional network model are obtained based on the daily precipitation observation data during the historical continuous precipitation and the characteristic variables.

[0087] The specific details of each module or unit in the above-mentioned precipitation prediction device have been described in detail in the corresponding precipitation prediction method, so they will not be repeated here.

[0088] The embodiment of the present application also provides an electronic device, Figure 4 A schematic diagram of the structure of an electronic device suitable for implementing the embodiments of the present disclosure is shown. Figure 4 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0089] like Figure 4 As shown, electronic device 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In RAM 603, various programs and data required for system operation are also stored. CPU 601, ROM 602 and RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to bus 604.

[0090] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0091] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable storage medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the embodiment of the present application are executed.

[0092] For example, when the computer program is executed by the central processing unit (CPU) 601, it can be executed as follows: obtaining current precipitation observation data of the target forecast area and the current environmental impact factors of the target forecast area; extracting target features from the current environmental impact factors based on random forests; inputting the current precipitation observation data and the target features into the convolutional neural network model, the back propagation neural network model, and the time convolutional network model to obtain the first prediction data, the second prediction data, and the third prediction data outputted by each; and determining the continuous precipitation data of the target forecast area by combining the first prediction data, the second prediction data, and the third prediction data.

[0093] It should be noted that the computer-readable medium shown in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0094] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0095] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, and the units described may also be arranged in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0096] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and the one or more programs include instructions, and when the instructions are executed by the electronic device, the electronic device implements the method described in the above embodiment.

[0097] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0098] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A precipitation prediction method, characterized in that: include: Obtaining current precipitation observation data of a target forecast area and current environmental impact factors of the target forecast area; Extracting target features from the current environmental influencing factors based on random forest; Inputting the current precipitation observation data and the target features into a convolutional neural network model, a back propagation neural network model, and a temporal convolutional network model to obtain first prediction data, second prediction data, and third prediction data as outputs; The continuous precipitation data of the target forecast area is determined by combining the first forecast data, the second forecast data and the third forecast data.

2. The precipitation prediction method according to claim 1, characterized in that: The step of combining the first prediction data, the second prediction data, and the third prediction data to determine the continuous precipitation data of the target forecast area includes: Determining respective weights of the first prediction data, the second prediction data, and the third prediction data; Based on the weight, the first prediction data, the second prediction data and the third prediction data are weighted averaged to obtain continuous precipitation data.

3. The precipitation prediction method according to claim 2, characterized in that: The determining of the weights of the first prediction data, the second prediction data, and the third prediction data includes: Acquire a test set, wherein the test set includes continuous precipitation observation data in a target forecast area; Predicting the continuous precipitation observation data by using the convolutional neural network model, the back propagation neural network model, and the time convolutional network model to obtain a first test result, a second test result, and a third test result, respectively; Determine the prediction pass rates of the convolutional neural network model, the back propagation neural network model, and the temporal convolutional network model based on the first test result, the second test result, and the third test result, respectively; Based on the prediction pass rate, the weights of the convolutional neural network model, the back propagation neural network model and the temporal convolutional network model are determined.

4. The precipitation prediction method according to claim 2, characterized in that: Also includes: The first test result, the second test result and the third test result are solved by the least square method to obtain the weights of the convolutional neural network model, the back propagation neural network model and the temporal convolutional network model.

5. The precipitation prediction method according to claim 1, characterized in that: The obtaining of current precipitation data of the target forecast area and current environmental impact factors of the target forecast area includes: Obtaining current precipitation data for the target forecast area from a meteorological information center; The meteorological forecast data output by the CFSv2 model is obtained, and the current environmental impact factors of the target forecast area are extracted from the meteorological forecast data.

6. The precipitation prediction method according to claim 3, characterized in that: The extracting target features from the current environmental impact factors based on random forests includes: Obtaining a training set, wherein the training set includes daily precipitation observation data and environmental influencing factors during historical continuous precipitation; Training a random forest with the training set enables the random forest to predict precipitation data of the next day based on the environmental influencing factors and the precipitation observation data of the previous day; Determine the importance of the current environmental impact factor to the current precipitation observation data based on the trained random forest; Target features are extracted from the current environmental impact factors based on the importance.

7. The precipitation prediction method according to claim 6, characterized in that: Also includes: Extracting characteristic variables from the environmental influencing factors by using the random forest; The convolutional neural network model, the back propagation neural network model, and the time convolutional network model are obtained based on the daily precipitation observation data during the historical continuous precipitation and the characteristic variables.

8. A precipitation prediction device, characterized in that: include: A data acquisition module, used to acquire current precipitation observation data of a target forecast area and current environmental impact factors of the target forecast area; A feature extraction module, used for extracting target features from the current environmental impact factors based on random forest; A prediction module, used for inputting the current precipitation observation data and the target features into a convolutional neural network model, a back propagation neural network model, and a temporal convolutional network model to obtain first prediction data, second prediction data, and third prediction data as outputs respectively; The comprehensive prediction module is used to determine the continuous precipitation data of the target forecast area by combining the first prediction data, the second prediction data and the third prediction data.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the precipitation prediction method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein one or more computer programs are stored in the memory, and the one or more computer programs include instructions. When the instructions are executed by the electronic device, the electronic device executes the precipitation prediction method according to any one of claims 1 to 7.

Images