An artificial intelligence-based cold region flood forecasting method and system

Abstract

The application discloses a kind of cold region flood forecasting method and system based on artificial intelligence, it is related to data processing field, wherein the method comprises: obtaining cold region flood historical disaster data information;Cold region flood historical data information is carried out index quantization processing, obtains multi-dimensional flood disaster analysis index;Based on multi-dimensional flood disaster analysis index and cold region flood historical data information carries out model fitting, verification, obtains cold region flood forecasting model;Cold region remote sensing monitoring data and cold region weather forecast data are carried out multidimensional analysis, obtain collaborative factor forecast information;Based on collaborative factor forecast information, cold region flood forecasting model is updated and optimized and flood forecasting management.It solves the technical problem that the precision of flood forecasting in cold region is insufficient in the prior art, which results in low quality of flood forecasting in cold region.The technical effect of improving the precision of cold region flood forecasting and improving the quality of cold region flood forecasting is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a cold region flood forecasting method and system based on artificial intelligence. BACKGROUND

[0002] The cold region generally refers to an area where the winter temperature is below -15 DEG C. Flood forecasting has an important influence on the cold region. Flood forecasting is an important decision basis for flood prevention and rescue and flood control in the cold region. Accurate flood forecasting can provide protection for the normal production and life of residents in the cold region and the safety of life and property. Combining artificial intelligence with cold region flood forecasting, designing a method for optimizing flood forecasting in the cold region has important practical significance.

[0003] In the prior art, there is a technical problem of insufficient accuracy of flood forecasting in the cold region, which further causes low quality of flood forecasting in the cold region. SUMMARY

[0004] The present application provides a cold region flood forecasting method and system based on artificial intelligence. The technical problem of insufficient accuracy of flood forecasting in the cold region in the prior art, which further causes low quality of flood forecasting in the cold region, is solved. The technical effect of improving the accuracy of flood forecasting in the cold region and improving the quality of flood forecasting in the cold region by intelligently and comprehensively analyzing flood forecasting in the cold region, providing reliable data reference for flood control management in the cold region is achieved.

[0005] In view of the above problems, the present application provides a cold region flood forecasting method and system based on artificial intelligence.

[0006] In the first aspect, the present application provides a cold region flood forecasting method based on artificial intelligence, wherein the method is applied to a cold region flood forecasting system based on artificial intelligence, and the method comprises: analyzing the influence factors of cold region flood disasters to obtain multi-dimensional flood disaster influence factors; based on the multi-dimensional flood disaster influence factors, data mining is performed to obtain cold region flood historical disaster data information; the cold region flood historical data information is subjected to index quantization processing to obtain multi-dimensional flood disaster analysis indexes; based on the multi-dimensional flood disaster analysis indexes and the cold region flood historical data information, model fitting and verification are performed to obtain a cold region flood forecasting model; cold region remote sensing monitoring data and cold region meteorological forecasting data are obtained; the cold region remote sensing monitoring data and the cold region meteorological forecasting data are subjected to multi-dimensional analysis to obtain collaborative factor forecasting information; and the cold region flood forecasting model is optimized and updated and flood forecasting management is performed based on the collaborative factor forecasting information.

[0007] In a second aspect, the present application also provides an artificial intelligence-based flood forecasting system for cold regions, wherein the system comprises: an influencing factor analysis module, configured to analyze influencing factors of flood disasters in cold regions and obtain multi-dimensional flood disaster influencing factors; a data mining module, configured to mine data based on the multi-dimensional flood disaster influencing factors and obtain historical flood disaster data information for cold regions; an index quantization processing module, configured to perform index quantization processing on the historical flood data information for cold regions and obtain multi-dimensional flood disaster analysis indexes; a fitting verification module, configured to perform model fitting and verification based on the multi-dimensional flood disaster analysis indexes and the historical flood data information for cold regions, and obtain a flood forecasting model for cold regions; a cold region data acquisition module, configured to acquire cold region remote sensing monitoring data and cold region meteorological forecasting data; a multi-dimensional analysis module, configured to perform multi-dimensional analysis on the cold region remote sensing monitoring data and the cold region meteorological forecasting data, and obtain collaborative factor forecasting information; and a flood forecasting management module, configured to optimize and update the flood forecasting model for cold regions and perform flood forecasting management based on the collaborative factor forecasting information.

[0008] In a third aspect, the present application also provides an electronic device, comprising: a memory configured to store executable instructions; and a processor configured to execute the executable instructions stored in the memory to implement an artificial intelligence-based flood forecasting method for cold regions provided by the present application.

[0009] In a fourth aspect, the present application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements an artificial intelligence-based flood forecasting method for cold regions provided by the present application.

[0010] One or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0011] By analyzing influencing factors of flood disasters in cold regions, multi-dimensional flood disaster influencing factors are obtained; by mining data based on the multi-dimensional flood disaster influencing factors, historical flood disaster data information for cold regions is obtained; by performing index quantization processing on the historical flood data information for cold regions, multi-dimensional flood disaster analysis indexes are obtained; by performing model fitting and verification based on the multi-dimensional flood disaster analysis indexes and the historical flood data information for cold regions, a flood forecasting model for cold regions is obtained; by performing multi-dimensional analysis on cold region remote sensing monitoring data and cold region meteorological forecasting data, collaborative factor forecasting information is obtained; and based on the collaborative factor forecasting information, the flood forecasting model for cold regions is optimized and updated, and flood forecasting management is performed. The technical effect of intelligently and comprehensively forecasting floods in cold regions is achieved, the accuracy of flood forecasting in cold regions is improved, the quality of flood forecasting in cold regions is improved, and reliable data reference is provided for flood dispatching management in cold regions.

[0012] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the following specific embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following will briefly introduce the drawings of the embodiments of the present disclosure. Obviously, the drawings described in the following description only relate to some embodiments of the present disclosure, not limit the present disclosure.

[0014] Figure 1 The flowchart of the method for forecasting flood in cold region based on artificial intelligence provided by the present application;

[0015] Figure 2 The flowchart of obtaining the flood forecasting model in cold region in the method for forecasting flood in cold region based on artificial intelligence provided by the present application;

[0016] Figure 3 The structural schematic diagram of the system for forecasting flood in cold region based on artificial intelligence provided by the present application;

[0017] Figure 4 The structural schematic diagram of the exemplary electronic device provided by the present application.

[0018] Explanation of reference signs: influence factor analysis module 11, data mining module 12, index quantization processing module 13, fitting verification module 14, cold region data acquisition module 15, multi-dimensional analysis module 16, flood forecasting management module 17, processor 31, memory 32, input device 33, output device 34. DETAILED DESCRIPTION

[0019] The present application provides a method and system for forecasting flood in cold region based on artificial intelligence. The technical problem of low accuracy of flood forecasting in cold region in the prior art is solved, and the quality of flood forecasting in cold region is not high. The technical effect of intelligent and comprehensive flood forecasting analysis in cold region, improving the accuracy of flood forecasting in cold region, improving the quality of flood forecasting in cold region, and providing reliable data reference for flood control management in cold region is achieved.

[0020] Embodiment one

[0021] Please refer to the attached Figure 1 The present application provides a method for forecasting flood in cold region based on artificial intelligence, wherein the method is applied to a system for forecasting flood in cold region based on artificial intelligence, and the method specifically comprises the following steps:

[0022] Step S100: analyzing the influence factors of flood disasters in cold regions to obtain multi-dimensional flood disaster influence factors;

[0023] Step S200: data mining based on the multi-dimensional flood disaster influence factors to obtain cold region flood history disaster data information;

[0024] Specifically, the influence factors of flood disasters in cold regions are analyzed to obtain multi-dimensional flood disaster influence factors. Based on the multi-dimensional flood disaster influence factors, historical data in cold regions is collected to obtain cold region flood history disaster data information. The multi-dimensional flood disaster influence factors include climate factors, topographic factors, location factors, river factors, river flood discharge capacity, flood storage engineering quality, and other cold region flood disaster influence factors. The cold region flood history disaster data information includes multiple cold region flood disaster history events. Each cold region flood disaster history event includes historical flood type, historical maximum flood level, historical flood occurrence process, historical flood occurrence time, historical flood volume, river confluence information before flood occurrence, and historical flood occurrence factors. Historical flood occurrence factors can be historical climate information, historical river flood discharge capacity, historical flood storage engineering quality, historical flood location and topographic information, etc. Through data mining of multi-dimensional flood disaster influence factors, cold region flood history disaster data information is obtained, which provides data support for subsequent determination of multi-dimensional flood disaster analysis indicators.

[0025] Step S300: index quantization processing of the cold region flood history data information to obtain multi-dimensional flood disaster analysis indicators;

[0026] Further, step S300 of the present application further includes:

[0027] Step S310: data cleaning and normalization processing of the cold region flood history data information to obtain standard cold region flood history data information;

[0028] Step S320: occurrence factor extraction based on the standard cold region flood history data information to obtain a set of flood analysis factor indicators;

[0029] Step S330: correlation quantization of each indicator information in the set of flood analysis factor indicators to obtain analysis indicator correlation information;

[0030] Step S340: obtaining factor indicator correlation threshold;

[0031] Step S350: screening of indicators within the factor indicator correlation threshold based on the analysis indicator correlation information to obtain the multi-dimensional flood disaster analysis indicators.

[0032] Specifically, by data cleaning and normalization processing on the historical data information of floods in cold regions, standard historical data information of floods in cold regions is obtained, and the standard historical data information of floods in cold regions is subjected to occurrence factor extraction to obtain a set of flood analysis factor indexes. The data cleaning includes format cleaning, abnormal value cleaning, logical error cleaning, and missing value cleaning of the historical data information of floods in cold regions. The normalization processing refers to dimension elimination of the historical data information of floods in cold regions after data cleaning, so that the data in the historical data information of floods in cold regions after data cleaning is converted into dimensionless pure numerical values. The standard historical data information of floods in cold regions includes a plurality of standard cold region flood disaster historical events corresponding to the historical data information of floods in cold regions after data cleaning and normalization processing. The set of flood analysis factor indexes includes a plurality of flood analysis factor indexes. The plurality of flood analysis factor indexes include a plurality of historical flood occurrence factors in the plurality of standard cold region flood disaster historical events.

[0033] Further, the plurality of flood analysis factor indexes in the set of flood analysis factor indexes are subjected to flood disaster correlation degree analysis to obtain analysis index correlation degree information. The analysis index correlation degree information includes a plurality of flood disaster correlation degree parameters corresponding to the plurality of flood analysis factor indexes. Illustratively, when the analysis index correlation degree information is obtained, the frequency of the plurality of flood analysis factor indexes in the set of flood analysis factor indexes can be counted to obtain a plurality of index frequency parameters. The plurality of index frequency parameters are set as the plurality of flood disaster correlation degree parameters in the analysis index correlation degree information.

[0034] Further, it is judged whether the plurality of flood disaster correlation degree parameters in the analysis index correlation degree information are within a factor index correlation degree threshold, and if the flood disaster correlation degree parameter is within the factor index correlation degree threshold, the flood analysis factor index corresponding to the flood disaster correlation degree parameter is added to the multi-dimensional flood disaster analysis index. The multi-dimensional flood disaster analysis index includes a plurality of flood analysis factor indexes corresponding to a plurality of flood disaster correlation degree parameters within the factor index correlation degree threshold. The technical effect of improving the reliability of cold region flood forecasting is achieved by multi-dimensional data mining on the historical data information of floods in cold regions to obtain accurate multi-dimensional flood disaster analysis indexes.

[0035] Further, the step S340 of the present application further includes:

[0036] Step S341: obtaining flood disaster parameter information from the standard historical data information of floods in cold regions, the flood disaster parameter information including flood source, generation level, duration, and scope of influence;

[0037] Step S342: obtaining index influence degree information of each index in the set of flood analysis factor indexes on the flood disaster parameter information;

[0038] Step S343: curve fitting is performed on the index influence degree information to obtain index influence degree curve information;

[0039] Step S344: correlation degree mapping is performed on the peak and trough values of the index influence degree curve information to determine the factor index correlation degree threshold.

[0040] Specifically, based on the standard cold region flood historical data information, flood disaster parameter information is set. The flood disaster parameter information includes flood source, generation level, duration, and range. Then, based on the flood disaster parameter information, influence degree evaluation is performed on the multiple flood analysis factor indexes in the flood analysis factor index set to obtain index influence degree information. The index influence degree information includes flood source index influence degree data, flood level index influence degree data, flood duration index influence degree data, and flood range index influence degree data. The flood source index influence degree data, the flood level index influence degree data, the flood duration index influence degree data, and the flood range index influence degree data respectively include multiple influence degree parameters between the flood source, the generation level, the duration, and the range and the multiple flood analysis factor indexes.

[0041] Further, average value calculation is respectively performed on the flood source index influence degree data, the flood level index influence degree data, the flood duration index influence degree data, and the flood range index influence degree data in the index influence degree information to obtain multiple-dimensional mean influence degree parameters. Then, curve fitting is performed based on the multiple-dimensional mean influence degree parameters to obtain index influence degree curve information, and peak value and trough value extraction is performed on the index influence degree curve information. Average value calculation is performed on the peak value and the trough value in the index influence degree curve information to obtain a factor index correlation degree threshold. The multiple-dimensional mean influence degree parameters include multiple average influence degree parameters corresponding to the flood source index influence degree data, the flood level index influence degree data, the flood duration index influence degree data, and the flood range index influence degree data. The index influence degree curve information includes a curve corresponding to the multiple-dimensional mean influence degree parameters. The factor index correlation degree threshold includes an average value between the peak value and the trough value in the index influence degree curve information. The technical effect of improving the accuracy of screening the multiple flood analysis factor indexes is achieved by performing influence degree evaluation on the multiple flood analysis factor indexes in the flood analysis factor index set based on the flood disaster parameter information to determine a reliable factor index correlation degree threshold.

[0042] Step S400: model fitting and verification are performed based on the multiple-dimensional flood disaster analysis index and the cold region flood historical data information to obtain a cold region flood prediction model;

[0043] Further, as shown in FIG. 6, the cold region flood prediction model is used to analyze the cold region flood historical data information to obtain a cold region flood prediction model. Figure 2As shown, the step S400 of the present application further includes:

[0044] Step S410: The index data obtained by integrating and processing the historical flood data information in cold regions according to the multi-dimensional flood disaster analysis index is identified as model sample information.

[0045] Step S420: The identified model sample information is divided according to a predetermined proportion to obtain a training data set, a verification data set, and a test data set.

[0046] Step S430: Network model supervised training is performed based on the training data set to obtain a basic cold region flood forecasting model.

[0047] Step S440: The basic cold region flood forecasting model is verified and tested based on the verification data set and the test data set until the model analysis accuracy reaches a preset accuracy to obtain the cold region flood forecasting model.

[0048] Specifically, on the basis of data cleaning, normalization processing of historical flood data information in cold regions, and obtaining standard historical flood data information in cold regions, the standard historical flood data information in cold regions is identified according to the multi-dimensional flood disaster analysis index to obtain model sample information. Further, the model sample information is divided according to a predetermined proportion to obtain a training data set, a verification data set, and a test data set. The predetermined proportion includes a sample division proportion determined by self-adaptation. For example, the predetermined proportion is 7:1:2. Then, 70% of the data information in the model sample information is divided into the training data set, 10% of the data information in the model sample information is divided into the verification data set, and 20% of the data information in the model sample information is divided into the test data set.

[0049] Further, based on the feedforward neural network, the training data set is supervised to the convergence state to obtain a basic cold region flood forecasting model. The test data set is input into the basic cold region flood forecasting model as input information, and the basic cold region flood forecasting model is updated by the test data set. The verification data set is input into the basic cold region flood forecasting model as input information, and the basic cold region flood forecasting model is verified by the verification data set. When the model analysis accuracy reaches the preset accuracy, the cold region flood forecasting model is obtained.

[0050] The feedforward neural network comprises an input layer, a plurality of neurons, and an output layer. The feedforward neural network is an artificial neural network. In the feedforward neural network, each neuron receives input from the previous stage and inputs to the next stage from the input layer to the output layer, and there is no feedback in the entire network, and the signal is unidirectionally propagated from the input layer to the output layer. The basic cold region flood forecasting model comprises an input layer, a hidden layer, and an output layer. The cold region flood forecasting model comprises a basic cold region flood forecasting model whose model analysis accuracy reaches a preset accuracy. The model analysis accuracy comprises an output parameter accuracy of the basic cold region flood forecasting model. The preset accuracy comprises an output parameter accuracy threshold of the basic cold region flood forecasting model which is set in advance. Exemplarily, when the basic cold region flood forecasting model is verified by using a verification data set, the verification data set is input into the basic cold region flood forecasting model to obtain verification flood forecasting information corresponding to the verification data set, the verification flood forecasting information is compared with historical flood information corresponding to the verification data set to obtain the model analysis accuracy, and if the model analysis accuracy meets the preset accuracy, the basic cold region flood forecasting model is set as the cold region flood forecasting model. The technical effect of improving the reliability of cold region flood forecasting is achieved by model fitting and verification on historical cold region flood data information to obtain a high-precision cold region flood forecasting model.

[0051] Step S500: acquiring cold region remote sensing monitoring data and cold region weather forecast data;

[0052] Step S600: performing multi-dimensional analysis on the cold region remote sensing monitoring data and the cold region weather forecast data to obtain collaborative factor forecasting information;

[0053] Further, the step S600 of the application further comprises:

[0054] Step S610: performing attribute integration on the cold region remote sensing monitoring data to obtain cold region remote sensing attribute data information;

[0055] Step S620: performing confluence feature analysis based on the cold region remote sensing attribute data information to obtain flood confluence influence features;

[0056] Further, the step S620 of the application further comprises:

[0057] Step S621: obtaining cold region terrain data information, water flow distribution data information, and cold region snow data information according to the cold region remote sensing attribute data information;

[0058] Step S622: performing confluence mechanism analysis based on the water flow distribution data information and the cold region snow data information to obtain snowmelt runoff flood features;

[0059] Step S623: Perform water flow loss analysis on the cold region terrain data information to determine a water flow attenuation coefficient;

[0060] Step S624: Take the ratio of the snowmelt runoff flood feature and the water flow attenuation coefficient as the flood confluence influence feature.

[0061] Specifically, historical data of the cold region is queried to obtain cold region remote sensing monitoring data and cold region weather forecast data. The cold region remote sensing monitoring data is classified according to data attributes to obtain cold region remote sensing attribute data information. The cold region remote sensing monitoring data includes historical terrain monitoring information, historical water flow distribution monitoring information, and historical snow monitoring information of the cold region. The cold region weather forecast data includes rainfall, wind speed, wind direction, temperature, humidity, and other historical weather forecast information of the cold region. The data attributes include terrain, water flow, and snow. The cold region remote sensing attribute data information includes cold region terrain data information, water flow distribution data information, and cold region snow data information in the cold region remote sensing monitoring data.

[0062] Further, the water flow distribution data information and the cold region snow data information are analyzed for confluence mechanism to obtain a snowmelt runoff flood feature. The cold region terrain data information is analyzed for water flow loss to determine a water flow attenuation coefficient. The snowmelt runoff flood feature and the water flow attenuation coefficient are calculated for a ratio to obtain a flood confluence influence feature. The snowmelt runoff flood feature includes a probability parameter of water flow converging into flood after snow melting in the cold region. The water flow attenuation coefficient is parameter information for representing the amount of water flow loss in the cold region. The greater the amount of water flow loss in the cold region, the greater the water flow attenuation coefficient. Illustratively, the ratio between the amount of water flow loss in the cold region and the historical water flow in the cold region is output as the water flow attenuation coefficient. The flood confluence influence feature includes the ratio between the snowmelt runoff flood feature and the water flow attenuation coefficient. This achieves the technical effect of obtaining accurate flood confluence influence features by analyzing the confluence features of the cold region remote sensing attribute data information, thereby improving the credibility of the subsequently obtained collaborative factor forecast information.

[0063] Step S630: Extract factors from the cold region weather forecast data to obtain weather-related factors;

[0064] Step S640: Obtain the collaborative factor forecast information based on the flood confluence influence feature and the weather-related factors.

[0065] Specifically, the meteorological correlation factor analysis is performed on the meteorological forecast data in the cold region to obtain meteorological correlation factors, and the synergistic factor prediction information is obtained by combining the flood confluence influence characteristics. The meteorological correlation factors include a plurality of meteorological flood correlation parameters corresponding to a plurality of historical meteorological prediction information in the meteorological forecast data in the cold region. The stronger the correlation between the historical meteorological prediction information and the flood disaster, the larger the corresponding meteorological flood correlation parameter. The technical effect of improving the comprehensiveness and accuracy of the flood prediction in the cold region is achieved by performing multi-dimensional analysis on the remote sensing monitoring data and the meteorological forecast data in the cold region to obtain the synergistic factor prediction information.

[0066] Step S700: updating and managing the flood prediction model in the cold region based on the synergistic factor prediction information.

[0067] Further, the step S700 of the present application further comprises:

[0068] Step S710: updating the flood prediction model in the cold region based on the synergistic factor prediction information to obtain a flood optimized prediction model in the cold region;

[0069] Step S720: obtaining the flood prediction information in the cold region based on the flood optimized prediction model in the cold region;

[0070] Step S730: obtaining the flood warning level according to the flood prediction information in the cold region;

[0071] Step S740: managing the flood warning and prediction based on the flood warning level and the flood prediction information in the cold region.

[0072] Specifically, the flood prediction model in the cold region is trained according to the synergistic factor prediction information, the parameters of the flood prediction model in the cold region are updated by the synergistic factor prediction information, and the flood optimized prediction model in the cold region is obtained. Then, the real-time data in the cold region is collected to obtain the real-time remote sensing monitoring data and the real-time meteorological forecast data in the cold region, and the real-time remote sensing monitoring data and the real-time meteorological forecast data are input into the flood optimized prediction model in the cold region as input information to obtain the flood prediction information in the cold region. The flood prediction information in the cold region is evaluated to obtain the flood warning level. The flood warning and prediction management in the cold region is performed according to the flood warning level and the flood prediction information in the cold region.

[0073] The real-time remote sensing monitoring data includes real-time topographic monitoring information, real-time water flow distribution monitoring information and real-time snow monitoring information of the cold region. The real-time weather forecast data includes rainfall, wind speed, wind direction, temperature, humidity and other real-time weather forecast information of the cold region. The cold region flood forecast information includes predicted flood duration, predicted flood range, predicted maximum flood level and predicted total flood volume and other flood forecast parameter information of the cold region. The flood warning level is parameter information for representing the degree of flood disaster corresponding to the cold region flood forecast information. For example, the greater the predicted maximum flood level and the wider the predicted flood range, the higher the corresponding flood warning level. The cold region flood prediction analysis is achieved through the cold region flood optimization prediction model, accurate and reliable cold region flood forecast information and flood warning level are obtained, and the flood prediction quality of the cold region is improved.

[0074] In summary, the cold region flood prediction method based on artificial intelligence provided in the present application has the following technical effects:

[0075] 1. Through analysis of the influencing factors of cold region flood disasters, multi-dimensional flood disaster influencing factors are obtained. Through data mining of the multi-dimensional flood disaster influencing factors, cold region flood historical disaster data information is obtained. Through index quantization processing of the cold region flood historical data information, multi-dimensional flood disaster analysis indexes are obtained. The multi-dimensional flood disaster analysis indexes and the cold region flood historical data information are fitted and verified to obtain a cold region flood prediction model. Through multi-dimensional analysis of the cold region remote sensing monitoring data and the cold region weather forecast data, collaborative factor prediction information is obtained. Based on the collaborative factor prediction information, the cold region flood prediction model is optimized and updated and flood prediction management is performed. The intelligent and comprehensive flood prediction analysis of the cold region is achieved, the accuracy of the cold region flood prediction is improved, the flood prediction quality of the cold region is improved, and reliable data reference is provided for the flood control and management of the cold region.

[0076] 2. Through multi-dimensional data mining of the cold region flood historical data information, accurate multi-dimensional flood disaster analysis indexes are obtained, thereby improving the reliability of the cold region flood prediction.

[0077] 3. Through model fitting and verification of the cold region flood historical data information, a high-precision cold region flood prediction model is obtained, thereby improving the reliability of the cold region flood prediction.

[0078] Embodiment Two

[0079] Based on the same inventive concept as the cold region flood prediction method based on artificial intelligence in the foregoing embodiments, the present application also provides a cold region flood prediction system based on artificial intelligence. Please refer to the accompanying drawings Figure 3 , the system comprises:

[0080] An influencing factor analysis module 11 is configured to analyze influencing factors of the flood disaster in the cold region, and obtain multi-dimensional flood disaster influencing factors;

[0081] A data mining module 12 is configured to mine data based on the multi-dimensional flood disaster influencing factors, and obtain cold region flood historical disaster data information;

[0082] An index quantification processing module 13 is configured to quantize the cold region flood historical data information, and obtain multi-dimensional flood disaster analysis indexes;

[0083] A fitting verification module 14 is configured to perform model fitting and verification based on the multi-dimensional flood disaster analysis indexes and the cold region flood historical data information, and obtain a cold region flood forecasting model;

[0084] A cold region data acquisition module 15 is configured to acquire cold region remote sensing monitoring data and cold region meteorological forecasting data;

[0085] A multi-dimensional analysis module 16 is configured to perform multi-dimensional analysis on the cold region remote sensing monitoring data and the cold region meteorological forecasting data, and obtain collaborative factor forecasting information;

[0086] A flood forecasting management module 17 is configured to optimize and update the cold region flood forecasting model based on the collaborative factor forecasting information, and perform flood forecasting management.

[0087] Further, the system further comprises:

[0088] A data cleaning module is configured to clean and normalize the cold region flood historical data information, and obtain standard cold region flood historical data information;

[0089] An occurrence factor extraction module is configured to extract occurrence factors based on the standard cold region flood historical data information, and obtain a set of flood analysis factor indexes;

[0090] An association quantification module is configured to quantize the association degree of each index information in the set of flood analysis factor indexes, and obtain analysis index association degree information;

[0091] An association degree threshold obtaining module is configured to obtain a factor index association degree threshold;

[0092] An index screening module is configured to screen indexes whose analysis index correlation degree information is within the factor index correlation degree threshold, to obtain the multi-dimensional flood disaster analysis index.

[0093] Further, the system further comprises:

[0094] A flood disaster parameter obtaining module is configured to obtain flood disaster parameter information according to the standard cold region flood history data information, the flood disaster parameter information including flood source, generation level, duration, and scope of influence.

[0095] An index influence degree information obtaining module is configured to obtain index influence degree information of each index in the flood analysis factor index set on the flood disaster parameter information.

[0096] A curve fitting module is configured to perform curve fitting on the index influence degree information, to obtain index influence degree curve information.

[0097] An correlation degree mapping module is configured to perform correlation degree mapping on the peak and trough values of the index influence degree curve information, to determine the factor index correlation degree threshold.

[0098] Further, the system further comprises:

[0099] A sample identification module is configured to identify index data obtained after the cold region flood history data information is integrated according to the multi-dimensional flood disaster analysis index as model sample information.

[0100] A sample division module is configured to divide the identified model sample information according to a predetermined proportion, to obtain a training data set, a verification data set, and a test data set.

[0101] A supervised training module is configured to perform network model supervised training based on the training data set, to obtain a basic cold region flood forecast model.

[0102] A first execution module is configured to verify and test the basic cold region flood forecast model based on the verification data set and the test data set, until the model analysis accuracy rate reaches a preset accuracy rate, to obtain the cold region flood forecast model.

[0103] Further, the system further comprises:

[0104] A data integration module is configured to integrate attributes of the cold region remote sensing monitoring data, to obtain cold region remote sensing attribute data information.

[0105] a confluence feature analysis module, configured to perform confluence feature analysis based on the cold region remote sensing attribute data information, and obtain a flood confluence influence feature;

[0106] a meteorological correlation factor obtaining module, configured to perform factor extraction on the cold region meteorological forecast data, and obtain a meteorological correlation factor;

[0107] a second execution module, configured to obtain the collaborative factor forecast information based on the flood confluence influence feature and the meteorological correlation factor.

[0108] Further, the system further comprises:

[0109] a third execution module, configured to obtain cold region terrain data information, water flow distribution data information, and cold region snow accumulation data information according to the cold region remote sensing attribute data information;

[0110] a confluence mechanism analysis module, configured to perform confluence mechanism analysis based on the water flow distribution data information and the cold region snow accumulation data information, and obtain snowmelt runoff flood features;

[0111] a water flow loss analysis module, configured to perform water flow loss analysis on the cold region terrain data information, and determine a water flow attenuation coefficient;

[0112] a flood confluence influence feature determination module, configured to take a ratio of the snowmelt runoff flood features and the water flow attenuation coefficient as the flood confluence influence feature.

[0113] Further, the system further comprises:

[0114] an optimization updating module, configured to perform optimization updating on the cold region flood forecast model based on the collaborative factor forecast information, and obtain a cold region flood optimization forecast model;

[0115] a cold region flood forecast information obtaining module, configured to obtain cold region flood forecast information based on the cold region flood optimization forecast model;

[0116] a flood warning level obtaining module, configured to obtain a flood warning level according to the cold region flood forecast information;

[0117] a flood warning module, configured to perform flood warning forecast management based on the flood warning level and the cold region flood forecast information.

[0118] The cold region flood forecasting system based on artificial intelligence provided by the embodiment of the present application can execute the cold region flood forecasting method based on artificial intelligence provided by any embodiment of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0119] Each module included is only divided according to functional logic, but is not limited to the above division, as long as the corresponding function can be realized; in addition, the specific name of each functional module is only for easy mutual differentiation, and is not used to limit the protection scope of the present application.

[0120] Embodiment three

[0121] Figure 4 The structural schematic diagram of the electronic device provided by the third embodiment of the present application shows the block diagram of the exemplary electronic device suitable for realizing the embodiments of the present application. Figure 4 The electronic device shown is only an example, and should not bring any limitation to the function and use range of the embodiments of the present application. As Figure 4 As shown, the electronic device includes a processor 31, a memory 32, an input device 33 and an output device 34; the number of processors 31 in the electronic device can be one or more, Figure 4 In the example, the processor 31 in the electronic device, the memory 32, the input device 33 and the output device 34 can be connected through a bus or other means, Figure 4 In the example, the connection through the bus is taken as an example.

[0122] The memory 32 as a kind of computer readable storage medium can be used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the cold region flood forecasting method based on artificial intelligence in the embodiments of the present application. The processor 31 executes the software programs, instructions and modules stored in the memory 32, thereby executing various function applications and data processing of the computer device, i.e. realizing the cold region flood forecasting method based on artificial intelligence.

[0123] The application provides a cold region flood forecasting method based on artificial intelligence, wherein the method is applied to a cold region flood forecasting system based on artificial intelligence, and the method comprises the following steps: obtaining multi-dimensional flood disaster influence factors by analyzing influence factors of cold region flood disasters; obtaining cold region flood historical disaster data information by data mining on the multi-dimensional flood disaster influence factors; obtaining multi-dimensional flood disaster analysis indexes by index quantization processing on the cold region flood historical data information; obtaining a cold region flood forecasting model by model fitting and verification on the multi-dimensional flood disaster analysis indexes and the cold region flood historical data information; obtaining collaborative factor forecasting information by multi-dimensional analysis on cold region remote sensing monitoring data and cold region meteorological forecasting data; and optimizing and updating the cold region flood forecasting model and flood forecasting management based on the collaborative factor forecasting information. The technical problem that the flood forecasting precision for the cold region is insufficient in the prior art, and thus the flood forecasting quality of the cold region is not high, is solved. The technical effects that the cold region is intelligently and comprehensively analyzed for flood forecasting, the flood forecasting precision of the cold region is improved, the flood forecasting quality of the cold region is improved, and reliable data reference is provided for the flood control scheduling management of the cold region are achieved.

[0124] It should be noted that the above only describes the preferred embodiments of the present application and the applied technical principles. Those skilled in the art should understand that the present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of the present application. Therefore, although the present application has been described in detail through the above embodiments, the present application is not limited to the above embodiments, and more other equivalent embodiments can be included without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A cold-region flood forecasting method based on artificial intelligence, characterized in that, The method includes: A multidimensional flood disaster influencing factor analysis was conducted to identify the influencing factors of flood disasters in cold regions. Data mining is performed based on the aforementioned multidimensional flood disaster influencing factors to obtain historical flood disaster data information for cold regions. The historical flood data of the cold region is subjected to index quantification to obtain multidimensional flood disaster analysis indicators; Based on the multidimensional flood disaster analysis indicators and the historical flood data information of the cold region, the model is fitted and verified to obtain the cold region flood forecasting model; Acquire remote sensing monitoring data and cold region weather forecast data; Multi-dimensional analysis was performed on the cold region remote sensing monitoring data and cold region meteorological forecast data to obtain synergistic factor forecast information; Based on the forecast information of the aforementioned synergistic factors, the cold region flood forecasting model is optimized, updated, and flood forecasting is managed. The obtained multidimensional flood disaster analysis indicators include: The historical data of floods in the cold region are cleaned and normalized to obtain standard historical data of floods in the cold region. Based on the historical data of floods in the standard cold region, the occurrence factors are extracted to obtain a set of flood analysis factor indicators; The correlation quantification of each indicator information in the flood analysis factor index set is performed to obtain the correlation degree information of the analysis indicators; Obtain the threshold for the correlation between factor indicators; The multidimensional flood disaster analysis indicators are obtained by filtering the indicators whose correlation information is within the correlation threshold of the factor indicators. The threshold for obtaining the correlation degree of the factor indicators includes: Based on the historical flood data of the standard cold region, flood disaster parameter information is obtained, including flood source, generation level, duration, and affected area; Obtain the influence information of each indicator in the flood analysis factor index set on the flood disaster parameter information; The influence information of the indicators is subjected to curve fitting to obtain the influence curve information of the indicators; The peaks and troughs of the influence curve of the indicator are mapped to determine the correlation threshold of the factor indicator.

2. The method as described in claim 1, characterized in that, The obtained cold region flood forecasting model includes: The historical flood data of the cold region is integrated and processed according to the multidimensional flood disaster analysis indicators, and the resulting indicator data is used as model sample information for identification. The labeled model sample information is divided according to a predetermined ratio to obtain a training dataset, a validation dataset, and a test dataset; Based on the training dataset, supervised training of the network model is performed to obtain a basic cold region flood forecasting model. The basic cold region flood forecasting model is validated and tested based on the validation dataset and the test dataset until the model analysis accuracy reaches the preset accuracy, thus obtaining the cold region flood forecasting model.

3. The method as described in claim 1, characterized in that, The acquisition of synergistic factor forecast information includes: The cold region remote sensing monitoring data is integrated to obtain cold region remote sensing attribute data information; Based on the aforementioned cold region remote sensing attribute data, confluence characteristic analysis is performed to obtain the flood confluence impact characteristics; Factor extraction was performed on the cold region meteorological forecast data to obtain meteorological correlation factors; Based on the characteristics of the flood confluence and the meteorological correlation factors, the forecast information of the synergistic factors is obtained.

4. The method as described in claim 3, characterized in that, The obtained flood confluence impact characteristics include: Based on the remote sensing attribute data of the cold region, obtain the topographic data, water flow distribution data, and snow cover data of the cold region. Based on the aforementioned water flow distribution data and cold region snow accumulation data, a runoff mechanism analysis was conducted to obtain the characteristics of snowmelt runoff floods. Water flow loss analysis was performed on the aforementioned cold region topographic data to determine the water flow attenuation coefficient; The ratio of the snowmelt runoff flood characteristics to the water flow attenuation coefficient is used as the flood confluence impact characteristic.

5. The method as described in claim 1, characterized in that, The optimization and updating of the cold region flood forecasting model and flood forecasting management based on the synergistic factor forecasting information include: Based on the forecast information of the synergistic factors, the cold region flood forecasting model is optimized and updated to obtain the cold region flood optimized forecasting model. Based on the optimized forecasting model for cold-region floods, flood forecasting information for cold-region floods is obtained; Based on the aforementioned cold region flood forecast information, the flood warning level is obtained; Flood warning and forecasting management is carried out based on the flood warning level and the cold region flood forecast information.

6. A cold-region flood forecasting system based on artificial intelligence, characterized in that, The system is used to implement the artificial intelligence-based cold region flood forecasting method according to any one of claims 1-5, the system comprising: The influencing factor analysis module is used to analyze the influencing factors of flood disasters in cold regions and obtain multidimensional flood disaster influencing factors. The influencing factor analysis module includes: The data cleaning module is used to clean and normalize the historical data information of cold region floods to obtain standard historical data information of cold region floods. The occurrence factor extraction module is used to extract occurrence factors based on the standard cold region flood historical data information to obtain a set of flood analysis factor indicators. The correlation quantification module is used to perform correlation quantification on the information of each indicator in the flood analysis factor indicator set to obtain the correlation degree information of the analysis indicators. A correlation threshold acquisition module, which is used to obtain the correlation threshold of factor indicators; An indicator filtering module is used to filter the correlation information of the analysis indicators within the correlation threshold of the factor indicators to obtain the multidimensional flood disaster analysis indicators. The data mining module is used to perform data mining based on the multidimensional flood disaster influencing factors to obtain historical flood disaster data information in cold regions. The data mining module includes: A flood disaster parameter acquisition module is used to obtain flood disaster parameter information based on the standard cold region flood historical data information. The flood disaster parameter information includes the flood source, generation level, duration, and affected area. The indicator impact information acquisition module is used to acquire the indicator impact information of each indicator in the flood analysis factor indicator set on the flood disaster parameter information; A curve fitting module is used to perform curve fitting on the indicator influence information to obtain indicator influence curve information. A correlation mapping module is used to map the peak and trough values ​​of the influence curve information of the indicator to determine the correlation threshold of the factor indicator. The indicator quantification processing module is used to perform indicator quantification processing on the historical data information of floods in cold regions to obtain multidimensional flood disaster analysis indicators. The fitting and verification module is used to fit and verify the model based on the multidimensional flood disaster analysis indicators and the historical data information of floods in cold regions, so as to obtain a flood forecasting model for cold regions. A cold region data acquisition module, which is used to acquire cold region remote sensing monitoring data and cold region weather forecast data; A multi-dimensional analysis module is used to perform multi-dimensional analysis on the cold region remote sensing monitoring data and cold region meteorological forecast data to obtain synergistic factor forecast information. A flood forecasting management module is used to optimize and update the cold region flood forecasting model and manage flood forecasts based on the forecasting information of the synergistic factors.

7. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the artificial intelligence-based cold region flood forecasting method according to any one of claims 1 to 5.

8. A computer-readable medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements an artificial intelligence-based cold-region flood forecasting method as described in any one of claims 1 to 5.

Images