Weather forecast detection method and device, electronic equipment and program product

Abstract

The invention provides a forecasting mode forecasting test method and device, electronic equipment and a program product. The method comprises the following steps: obtaining target parameters input by a user, wherein the target parameters comprise a target area, a target time period and at least one meteorological element; according to the at least one meteorological element, a target element set is determined, and the target element set comprises a plurality of target elements; in a target time period of the target area, determining original meteorological data corresponding to the plurality of target elements respectively; for any target element, performing anomaly detection processing on the original meteorological data corresponding to the target element, and determining target meteorological data corresponding to the target element; and according to the target meteorological data corresponding to the plurality of target elements, a visual inspection result corresponding to the meteorological forecast is generated, so that the accuracy of detecting the meteorological forecast is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of weather forecast, and in particular to a weather forecast detection method and device, an electronic device and a program product. BACKGROUND

[0002] With the continuous progress of science and technology, numerical weather prediction (NWP) has become a key tool for modern weather forecasting. Numerical weather prediction uses complex mathematical models and a large amount of observation data to predict future weather changes, covering a variety of meteorological elements such as temperature, precipitation, wind direction and wind speed.

[0003] Currently, the verification of weather forecast can be achieved in the following ways: a variety of verification methods are used in ensemble forecast evaluation, supporting a variety of verification tools for ensemble forecast, and the ensemble forecast results are evaluated by different verification methods; the frequency matching method and the probability matching method are used to correct the precipitation forecast in real time, mainly focusing on the verification of precipitation, a single meteorological element.

[0004] In the above-mentioned manner, the meteorological data to be verified usually comes from different meteorological models and different data sources, resulting in low accuracy of detecting weather forecast. SUMMARY

[0005] The present application provides a weather forecast detection method, device, electronic device and program product to solve the technical problem of low accuracy of detecting weather forecast in the prior art.

[0006] In a first aspect, the present application provides a weather forecast detection method, the method comprising:

[0007] obtaining a target parameter input by a user, the target parameter comprising a target region, a target time period and at least one meteorological element;

[0008] determining a target element set according to the at least one meteorological element, the target element set comprising a plurality of target elements;

[0009] determining original meteorological data corresponding to each of the plurality of target elements in the target region during the target time period;

[0010] performing anomaly detection processing on the original meteorological data corresponding to any one target element to determine target meteorological data corresponding to the target element;

[0011] generating a visual verification result corresponding to the weather forecast according to the target meteorological data corresponding to each of the plurality of target elements.

[0012] In this way, the target parameter input by the user is used to determine a matched target element set, and an anomaly detection process is performed on the target elements to generate a visual inspection result, thereby improving the accuracy of meteorological prediction.

[0013] Optionally, in the method described above, the anomaly detection process is performed on the original meteorological data corresponding to the target elements to determine the target meteorological data corresponding to the target elements, including:

[0014] Determining an element type corresponding to the target elements;

[0015] According to the element type, performing a weighted difference process on the original meteorological data to obtain a weighted difference value;

[0016] According to the weighted difference value and the original meteorological data, determining abnormal data in the original meteorological data by using a target algorithm;

[0017] According to the abnormal data, performing a screening process on the original meteorological data to obtain the target meteorological data.

[0018] In this way, different anomaly detection processes are performed according to different target elements, thereby improving the accuracy of meteorological prediction.

[0019] Optionally, in the method described above, the abnormal data in the original meteorological data is determined according to the weighted difference value and the original meteorological data by using a target algorithm, including:

[0020] Performing a subsampling process on the weighted difference value and the original meteorological data to obtain a plurality of sampling sets, wherein each sampling set includes a plurality of sampling data, and each sampling data is the weighted difference value or metadata, and each metadata is any data in the original meteorological data;

[0021] For any one sampling set, establishing an isolation tree of the sampling set according to the plurality of sampling data corresponding to the sampling set, and calculating a path length corresponding to each sampling data in the sampling set according to the isolation tree;

[0022] Determining the abnormal data in the original meteorological data according to a plurality of path lengths of a plurality of sampling sets.

[0023] In this way, the abnormal data is determined according to different specific conditions, thereby improving the accuracy of meteorological prediction.

[0024] Optionally, in the method described above, the abnormal data in the original meteorological data is determined according to a plurality of path lengths of a plurality of sampling sets, including:

[0025] For any one metadata, a first path length corresponding to the metadata is determined in a plurality of sampling sets;

[0026] determining a first average value of the plurality of first path lengths;

[0027] determining a dynamic threshold corresponding to the metadata;

[0028] if the first average value is greater than the dynamic threshold, determining that the metadata is in an abnormal state;

[0029] if the first average value is less than or equal to the dynamic threshold, determining that the metadata is in a normal state;

[0030] determining the metadata in the abnormal state as abnormal data in the original meteorological data.

[0031] In this way, the abnormal data is determined according to different specific conditions, and the accuracy of detecting meteorological forecasts is improved.

[0032] Optionally, the method described above, the original meteorological data includes any one of observation data, forecast data, global precipitation measurement (GPM) data and reanalysis data, according to the element type, the original meteorological data is weighted and differentially processed to obtain a weighted difference value, including:

[0033] determining whether the element type is a precipitation type;

[0034] if yes, determining a first factor corresponding to the observation data, a second factor corresponding to the reanalysis data, and a third factor corresponding to the GPM data, and determining a sum value between the first factor, the second factor and the third factor as the weighted difference value, wherein the first factor is a product of a first weight value corresponding to the observation data and a first difference value, and the first difference value is a difference value between any one of the observation data and a first weighted average value corresponding to the observation data;

[0035] if no, determining a first factor corresponding to the observation data and a second factor corresponding to the reanalysis data, and determining a sum value between the first factor and the second factor as the weighted difference value.

[0036] In this way, different abnormality detection processes are performed according to different target elements, and the accuracy of detecting meteorological forecasts is improved.

[0037] Optionally, the method described above, according to the at least one meteorological element, determining a target element set, including:

[0038] obtaining a plurality of candidate element sets, the candidate element set including a plurality of elements;

[0039] determining a plurality of similarities between the at least one meteorological element and a plurality of candidate element sets;

[0040] determine the candidate element set with the highest similarity as the target element set.

[0041] In this way, the most matched target element set is determined, and the accuracy of detecting the weather forecast is improved.

[0042] Optionally, the method described above generates a visual inspection result corresponding to the weather forecast according to the target weather data corresponding to the plurality of target elements, including:

[0043] For any one target element, the target weather data corresponding to the target element is classified to obtain a plurality of type data, and the plurality of type data includes any one of observation data, forecast data, global precipitation measurement (GPM) data, and reanalysis data.

[0044] Each type of data is standardized to obtain standard data corresponding to each type of data.

[0045] The standard data corresponding to the plurality of type data is input into a target model to determine a plurality of inspection scores corresponding to the target element.

[0046] The visual inspection result corresponding to the weather forecast is generated according to the plurality of inspection scores corresponding to the plurality of target elements.

[0047] In this way, the visual inspection result corresponding to the weather forecast is generated, and the readability of the inspection is improved.

[0048] In a second aspect, the present application provides a device for detecting a weather forecast, the device comprising:

[0049] An acquisition module is configured to acquire a target parameter input by a user, the target parameter including a target region, a target time period, and at least one weather element.

[0050] A first determination module is configured to determine a target element set according to the at least one weather element, the target element set including a plurality of target elements.

[0051] A second determination module is configured to determine original weather data corresponding to the plurality of target elements in the target time period of the target region.

[0052] An inspection module is configured to perform anomaly detection processing on the original weather data corresponding to any one target element to determine target weather data corresponding to the target element.

[0053] A display module is configured to generate a visual inspection result corresponding to the weather forecast according to the target weather data corresponding to the plurality of target elements.

[0054] Optionally, the apparatus as described above, the inspection module is specifically used for:

[0055] determining an element type corresponding to the target element;

[0056] performing weighted difference processing on the original meteorological data according to the element type to obtain a weighted difference value;

[0057] determining, according to the weighted difference value and the original meteorological data, abnormal data in the original meteorological data through a target algorithm;

[0058] performing screening processing on the original meteorological data according to the abnormal data to obtain the target meteorological data.

[0059] Optionally, the apparatus as described above, the inspection module is specifically used for:

[0060] performing a subsampling process on the weighted difference value and the original meteorological data to obtain a plurality of sampling sets, the sampling set comprising a plurality of sampling data, the sampling data being the weighted difference value or metadata, the metadata being any one data in the original meteorological data;

[0061] for any one sampling set, establishing an isolated tree of the sampling set according to a plurality of sampling data corresponding to the sampling set, and calculating path lengths corresponding to the plurality of sampling data in the sampling set according to the isolated tree;

[0062] determining abnormal data in the original meteorological data according to a plurality of path lengths of a plurality of sampling sets.

[0063] Optionally, the apparatus as described above, the inspection module is specifically used for:

[0064] for any one metadata, determining a first path length corresponding to the metadata in a plurality of sampling sets;

[0065] determining a first average value of a plurality of first path lengths;

[0066] determining a dynamic threshold value corresponding to the metadata;

[0067] if the first average value is greater than the dynamic threshold value, determining that the metadata is in an abnormal state;

[0068] if the first average value is less than or equal to the dynamic threshold value, determining that the metadata is in a normal state;

[0069] determining the metadata in the abnormal state as abnormal data in the original meteorological data.

[0070] Optionally, the apparatus as described above, the original meteorological data comprises any one of observation data, forecast data, global precipitation measurement (GPM) data and reanalysis data, the verification module is specifically used for:

[0071] determining whether the element type is a precipitation type;

[0072] if yes, determining a first factor corresponding to the observation data, a second factor corresponding to the reanalysis data and a third factor corresponding to the GPM data, and determining a sum value between the first factor, the second factor and the third factor as a weighted difference value, wherein the first factor is a product of a first weight value corresponding to the observation data and a first difference value, and the first difference value is a difference value between any one of the observation data and a first weighted average value corresponding to the observation data;

[0073] if no, determining a first factor corresponding to the observation data and a second factor corresponding to the reanalysis data, and determining a sum value between the first factor and the second factor as a weighted difference value.

[0074] Optionally, the apparatus as described above, the first determining module is specifically used for:

[0075] obtaining a plurality of candidate element sets, the candidate element sets comprising a plurality of elements;

[0076] determining a plurality of similarities between the at least one meteorological element and the plurality of candidate element sets;

[0077] determining a candidate element set with the highest similarity in the plurality of similarities as a target element set.

[0078] Optionally, the apparatus as described above, the display module is specifically used for:

[0079] for any one target element, performing classification processing on target meteorological data corresponding to the target element to obtain a plurality of type data, the plurality of type data comprising any one of observation data, forecast data, global precipitation measurement (GPM) data and reanalysis data;

[0080] performing standardization processing on each type data to obtain standard data corresponding to each type data;

[0081] inputting the standard data corresponding to the plurality of type data into a target model to determine a plurality of verification scores corresponding to the target element;

[0082] generating a visual verification result corresponding to the meteorological forecast according to the plurality of verification scores corresponding to the plurality of target elements, respectively.

[0083] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory connected with the processor in communication;

[0084] The memory stores computer-executable instructions.

[0085] The processor executes the computer-executable instructions stored in the memory to implement the method of any one of the first aspect.

[0086] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method of any one of the first aspect.

[0087] In a fifth aspect, the present application provides a computer program product, comprising a computer program, and the computer program is executed by a computer to implement the method of any one of the first aspect.

[0088] The meteorological forecast detection method, device, electronic device and program product provided by the present application obtain the target parameters input by the user, the target parameters including a target region, a target time period and at least one meteorological element; determine a target element set according to the at least one meteorological element, the target element set including a plurality of target elements; determine the original meteorological data corresponding to the plurality of target elements in the target time period of the target region; perform anomaly detection processing on the original meteorological data corresponding to any one target element to determine the target meteorological data corresponding to the target element; and generate a visual inspection result corresponding to the meteorological forecast according to the target meteorological data corresponding to the plurality of target elements. In this way, the matching target element set is determined according to the target parameters input by the user, and anomaly detection processing is performed according to the target elements to generate a visual inspection result, thereby improving the accuracy of detecting meteorological forecasts. BRIEF DESCRIPTION OF DRAWINGS

[0089] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0090] Figure 1 A structural schematic diagram of a verification system provided by an embodiment of the present application;

[0091] Figure 2 A flowchart of a verification method of a forecast mode provided by an embodiment of the present application;

[0092] Figure 3 A flowchart of another verification method of a forecast mode provided by an embodiment of the present application;

[0093] Figure 4A structure diagram of a prediction mode test device provided by an embodiment of the present application is shown in the figure.

[0094] Figure 5 A structure diagram of an electronic device provided by an embodiment of the present application is shown in the figure.

[0095] The specific embodiments of the present application have been shown in the above figures, and will be described in more detail hereinafter. These figures and the written description are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0096] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers are used to indicate the same or similar components. The embodiments described in the following exemplary embodiments are not meant to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application as detailed in the appended claims.

[0097] It should be noted that although the terms "first", "second", etc. are used to describe various information in the embodiments of the present application, the information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. Alternatively, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information without departing from the scope of the present application.

[0098] It should be understood that the terms "comprising", "including", etc. indicate the presence of the previously mentioned features, steps, operations, but do not exclude the presence, occurrence or addition of one or at least one other feature, step, operation. The terms "and / or" and the like used in the present application can be interpreted as inclusive or mean any one or any combination. Alternatively, "A and / or B" means "either A; B; A and B". In addition, the character " / " in this paper generally means that the associated objects before and after are in an "or" relationship.

[0099] With the continuous progress of science and technology, numerical weather prediction model (NWP) has become a key tool for modern weather forecasting. Numerical weather prediction model predicts the changes of future weather by complex mathematical models and a large amount of observation data, covering a variety of meteorological elements such as temperature, precipitation, wind direction and wind speed, etc.

[0100] Currently, the verification of weather forecasts can be achieved in the following ways: a variety of verification methods are used in the ensemble forecast evaluation, a variety of verification tools are supported for the ensemble forecast, and the ensemble forecast results are evaluated by different verification methods; the frequency matching method and the probability matching method are used to correct the precipitation forecast in real time, and the verification of the prediction of a single meteorological element, i.e., precipitation, is mainly focused on.

[0101] In the above manner, the meteorological data to be verified usually comes from different meteorological models and different data sources, resulting in low accuracy of detecting the accuracy of weather forecasts.

[0102] To solve the above technical problems, the embodiments of the present application provide a detection method for weather forecasts, which comprises the following steps: obtaining a target parameter input by a user, determining a target element set according to at least one meteorological element in the target parameter; determining original meteorological data corresponding to each of the target elements in a target area and a target period; performing anomaly detection processing on the original meteorological data corresponding to any one of the target elements to determine target meteorological data corresponding to the target element; and generating a visual verification result corresponding to the weather forecast according to the target meteorological data corresponding to each of the target elements. In this way, by inputting the target parameter by the user, the matching target element set is determined, and the anomaly detection processing is performed according to the target element to generate the visual verification result, thereby improving the accuracy of detecting the weather forecast.

[0103] Figure 1 A structural diagram of a detection system according to an embodiment of the present application is provided. Please refer to Figure 1 , Figure 1 The detection system can include a data layer, a quality processing layer, an analysis layer, and a visualization layer.

[0104] The data layer can be used to automatically download a plurality of meteorological data of a plurality of prediction models, and the plurality of meteorological data can include a plurality of prediction data, a plurality of observation data, Global Precipitation Measurement (GPM) data, and a plurality of reanalysis data, and support a plurality of data formats, different pressure layers, and multi-dimensional meteorological elements. For example, the plurality of data formats can include Grib format, NetCDF format, and CSV format, etc. The different pressure layers can include 200 hPa, 500 hPa, 700 hPa, 850 hPa, and 925 hPa, etc. The multi-dimensional meteorological elements can include precipitation, temperature, zonal wind, radial wind, and humidity, etc.

[0105] The data downloaded by the data layer can have different temporal and spatial resolutions.

[0106] The data layer can automatically schedule data download through a timing task scheduling mechanism, ensure to obtain real-time and accurate prediction data and observation data, significantly reduce the complexity of manual operation, and greatly improve the real-time performance of data acquisition and system processing efficiency.

[0107] The quality processing layer can be used for data quality control by a target algorithm.

[0108] The data quality control can include deleting abnormal files, checking data abnormal conditions and deleting files containing abnormal data, repairing or correcting abnormal data that can be repaired, etc.

[0109] For example, the abnormal file is a file that cannot be normally opened, a file whose data volume is lower than the normal value, etc.

[0110] After deleting the file, the data layer can resubmit the data download task for downloading, and if repeated submission fails for 3 times, the download task is no longer submitted.

[0111] The data quality control can ensure the accuracy and reliability of the input data, and avoid deviation of the test results caused by data defects.

[0112] The quality processing layer can also be used to convert the downloaded data into a unified file format, data format and spatial-temporal resolution, and align the data of the spatial grid structure.

[0113] The quality processing layer can overcome the compatibility problem of non-uniform data format in the prior art by standardizing the data of multiple prediction models from different sources, significantly improving the efficiency and accuracy of data processing. Through the data layer, the technical problem of multiple data sources and non-uniform formats is successfully solved, realizing unified management and analysis of multiple data sources.

[0114] The analysis layer can include multiple test indicators.

[0115] For example, the multiple test indicators can include root mean square error, standard deviation, correlation coefficient, threat score and reliability, etc.

[0116] The analysis layer can be used to diagnose and attribute the errors of multiple prediction models according to the values of multiple test indicators.

[0117] The visualization layer can be designed based on a lightweight B / S architecture, supporting high-performance interactive operation on the web side, with flexible layer switching, legend scaling and parameter setting capabilities, and adapting to multiple terminal browsing environments.

[0118] The visualization layer can have a "one-key skill report generation" function, automatically summarizing core evaluation results and visual graphics to form standardized or customized output reports, greatly improving the automation and intelligent level of the business evaluation process.

[0119] The technical solutions of the present application and how the technical solutions of the present application solve the above technical problems will be described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes can not be repeated in some examples. The embodiments of the present application will be described below with reference to the accompanying drawings.

[0120] The technical solutions shown in the present application will be described in detail below through specific examples. It should be noted that the following examples can exist independently or be combined with each other, and the same or similar content will not be repeated in different examples.

[0121] Figure 2 A flowchart of a detection method for weather forecasting provided by an embodiment of the present application. The execution subject of the embodiment of the present application can be a processor. The processor can be realized by software, or by the combination of software and hardware. Please refer to Figure 2 , the method comprises:

[0122] S201, obtaining a target parameter input by a user.

[0123] The target parameter includes a target area, a target period, and at least one meteorological element.

[0124] The target area can be a location area to be detected.

[0125] The meteorological element can include precipitation, temperature, zonal wind, radial wind, humidity, etc.

[0126] The target period can be a historical period to be detected.

[0127] The target parameter can be used to describe the parameter that the user wants to analyze.

[0128] The target parameter can include a detection method, which can include site detection and grid detection.

[0129] The site detection can be detection on a target area corresponding to a certain site.

[0130] The grid detection can be detection on a target area corresponding to a target grid.

[0131] An interactive interface can be displayed, which can include text boxes, selection boxes, sliding bars, etc. In response to the user's interactive operation on the target area selection control, the interactive operation on the meteorological element selection control, the sliding operation on the sliding bar corresponding to the target period, and the click operation on the submit button in the interactive interface, the control values corresponding to the multiple controls are obtained, and the target parameter input by the user is determined according to the multiple control values.

[0132] The target parameter input by the user can be obtained through a command line interface.

[0133] It should be noted that the target parameter input by the user can be obtained according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0134] S202, determining a target element set according to at least one meteorological element.

[0135] The target element set includes a plurality of target elements.

[0136] Optionally, the at least one meteorological element can be input into a preset model to output the target element set.

[0137] The preset model can be used to determine the target element set that best matches the at least one meteorological element.

[0138] Optionally, a plurality of candidate element sets can be obtained, the candidate element sets including a plurality of elements; a plurality of similarities between the at least one meteorological element and the plurality of candidate element sets are determined; and the candidate element set with the highest similarity in the plurality of similarities is determined as the target element set.

[0139] It should be noted that the target element set can be determined according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0140] S203, determining original meteorological data corresponding to the plurality of target elements respectively in a target period of a target area.

[0141] The original meteorological data includes any one or more of observation data, forecast data, global precipitation measurement (GPM) data, and reanalysis data.

[0142] The observation data can be physical quantities covering the ground surface and the near-surface atmosphere collected by ground meteorological stations, automatic weather stations, and the like.

[0143] The forecast data can be data generated by a corresponding meteorological model, a calculation algorithm, or a target algorithm according to the atmospheric state, weather phenomena, and related environmental parameters.

[0144] The global precipitation measurement (GPM) data can be multi-source fusion global precipitation observation products provided by core satellites and their carried radar and microwave radiometer instruments, and constellation satellite observation data.

[0145] The reanalysis data can be a high temporal and spatial resolution meteorological data set generated by combining observation data (such as ground observation stations, satellite observations, sounding data, etc.) with model simulation results.

[0146] The original meteorological data corresponding to the plurality of target elements respectively can be determined according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0147] The original meteorological data corresponding to the plurality of target elements respectively can be determined according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0148] The original meteorological data corresponding to the plurality of target elements respectively can be determined according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0149] S204, for any one target element, performing abnormality detection processing on the original meteorological data corresponding to the target element to determine target meteorological data corresponding to the target element.

[0150] Different target elements correspond to different abnormality detection processing modes.

[0151] The target meteorological data is meteorological data after removing abnormal data.

[0152] Optionally, a first mapping relationship can be obtained, a target algorithm can be determined according to the target element and the first mapping relationship, and abnormality detection processing can be performed on the original meteorological data corresponding to the target element according to the target algorithm to determine the target meteorological data corresponding to the target element.

[0153] The first mapping relationship can include a plurality of meteorological elements and an algorithm corresponding to each meteorological element.

[0154] Optionally, the target element can be input into an algorithm selection model to obtain a target algorithm, and abnormality detection processing can be performed on the original meteorological data corresponding to the target element according to the target algorithm to determine the target meteorological data corresponding to the target element.

[0155] Optionally, an element type corresponding to the target element can be determined, a weighted difference value can be obtained by performing weighted difference processing on the original meteorological data according to the element type, abnormal data in the original meteorological data can be determined by a target algorithm according to the weighted difference value and the original meteorological data, and the original meteorological data can be filtered to obtain the target meteorological data according to the abnormal data.

[0156] The original meteorological data corresponding to the plurality of target elements respectively can be determined according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0157] S205, according to the target meteorological data corresponding to the plurality of target elements respectively, a visual inspection result corresponding to a meteorological forecast is generated.

[0158] The visualized test result can be used to visually display the space, time and corresponding multiple test scores.

[0159] The visualized result can include line charts, column charts, scatter charts, contour charts and filled charts, etc. to meet the analysis requirements of different angles.

[0160] The visualized test result corresponding to the weather forecast can be generated according to the target meteorological data corresponding to the multiple target elements and the display mode selected by the user.

[0161] For example, the trend of the test scores (such as root mean square error, bias, Pearson correlation coefficient, etc.) corresponding to the site test indicators with the prediction length is shown by a line chart; the spatial distribution deviation of the test scores corresponding to the site test indicators is intuitively presented by a scatter chart; and the spatial field characteristics of the visualized test result are represented by a contour or filled chart.

[0162] The visualized test result corresponding to the weather forecast can be generated according to the target meteorological data corresponding to the multiple target elements by using a big data model or an artificial intelligence model.

[0163] Optionally, for any one target element, the target meteorological data corresponding to the target element can be classified to obtain multiple type data, and the multiple type data includes any multiple of observation data, prediction data, global precipitation measurement (GPM) data and reanalysis data; each type data is standardized to obtain standard data corresponding to each type data; the standard data corresponding to the multiple type data is input into a target model to determine multiple test scores corresponding to the target element; and the visualized test result corresponding to the weather forecast is generated according to the multiple test scores corresponding to the multiple target elements.

[0164] In the classification processing, the data with the same data label can be grouped into a data set according to the data label corresponding to each data in the target meteorological data to obtain a type data.

[0165] Optionally, the global and variable level labels in the target meteorological data can be read, and the labels include information such as “variable name”, “latitude and longitude coordinates”, “time dimension format”, “attribute description”, etc. to form an original label set L. At the same time, a standard field set S is defined in advance, for example, \{precipitation, temperature, latitude, longitude, time\}. Each label in the original label set L is mapped to the corresponding standard field in the standard field set S by calculating the string similarity (for example, using Levenshtein distance) or regular expression matching method, so as to generate a mapping table, and the classification processing is performed according to the mapping table.

[0166] Optionally, if the target meteorological data is grid data of different resolutions (such as 0.1°x0.1° and 0.25°x0.25°), the original field can be interpolated to the target grid using conservative interpolation, and then, in order to ensure that the total amount of information is not lost in the interpolation process, a bidirectional restoration test is performed: the target grid is mapped back to the original grid to obtain the interpolation of the original field, and the interpolation kernel parameters are iteratively optimized to minimize the following error, so as to balance the conservation and error minimization.

[0167] Optionally, the target meteorological data of different time resolutions (such as 30 minutes, 1 hour, 3 hours) is time-interpolated: if the hour resolution is required, linear or cubic spline interpolation is used for the original sequence.

[0168] For the target meteorological data of the missing interval, the neighborhood average method, the regression prediction method, and the spatial regression method can be used to determine it.

[0169] The neighborhood average method can be to fill in the average value of the adjacent grid points or the similar time period.

[0170] The regression prediction method can be to train a regression model based on historical data of the same period to predict and fill in the missing values.

[0171] The spatial regression method can be to estimate the missing values by weighted regression combined with multi-source data to ensure spatial consistency.

[0172] The standardization processing can store a single meteorological element of one pressure layer in each prediction model as a data file, and the file data format is NetCDF, the time resolution is hourly, and the spatial resolution is 0.1°x0.1°.

[0173] The target model can be a learning model optimized by historical data, which is not limited here.

[0174] Optionally, different target elements can use different test indicators.

[0175] For precipitation, the root mean square error (RMSE), bias (BIAS), threat score (TS score), equivalent skill score (ETS score), false alarm rate (FAR), missed alarm rate (MAR), and probability density function (PDF) indicators can be used for testing.

[0176] For temperature, in the station test mode, the root mean square error (RMSE), bias (BIAS), and Pearson correlation coefficient (PCC) can be used.

[0177] In the grid test mode, the RMSE and BIAS indicators can be used for testing, and the RMSE and BIAS of the daily maximum temperature and the daily minimum temperature are also tested.

[0178] For wind speed, the RMSE, BIAS and PCC indicators are used for testing.

[0179] Optionally, for each test indicator (such as TS, RMSE, BIAS, PCC, etc.), a test score can be determined by a linear mapping, which can be determined by the following formula:

[0180]

[0181] where Iraw can represent the original test score value, Imin can represent the minimum test score value, and Imax can represent the maximum test score value.

[0182] Optionally, for the score distribution of each mode, a confidence interval is calculated by Bootstrap sampling If the difference between the normalized scores of the two modes exceeds the sum of the half-widths of their confidence intervals, it is considered to be a significant difference, and a statistically supported ranking of the two modes is provided.

[0183] For example, from the original N test samples (such as N pairs of observation-data-prediction data), a new sample set is sampled with replacement The above step C is repeated times (C can be 500-1000) to obtain C groups of RMSE values Each time, the normalized score I is calculated norm ; arrange {I norm,(1) ,...,I norm,(C)} in ascending order, and take the and the values as the lower limit l and the upper limit u; if the confidence intervals of the two modes do not overlap, it is considered that there is a significant difference, that is, if l A > u B , it indicates that mode A is significantly better than mode B in this indicator;

[0184] where α can be a preset value, l A can be the lower limit of mode A, and u B can be the upper limit of mode B.

[0185] It should be noted that the visualization test result corresponding to the meteorological forecast can be generated according to any feasible implementation manner, and the embodiments of the present application do not limit this.

[0186] Optionally, after generating the visualization test result corresponding to the meteorological forecast, the method further includes: converting the visualization test result and the target meteorological data into analysis text according to natural language generation technology, and generating a picture-text report by combining the visualization test result and the analysis text.

[0187] The generated graphic reports can be edited with key points and charts rearranged through the online fine-tuning interface. The system updates the graphic reports synchronously in real time, taking into account both automation and customization needs.

[0188] The weather forecast detection method provided in this embodiment obtains target parameters input by the user, the target parameters including a target area, a target time period, and at least one meteorological element; determines a target element set based on the at least one meteorological element, the target element set including multiple target elements; determines the raw meteorological data corresponding to each of the multiple target elements within the target time period of the target area; performs anomaly detection processing on the raw meteorological data corresponding to any target element to determine the target meteorological data corresponding to the target element; and generates a visual verification result corresponding to the weather forecast based on the target meteorological data corresponding to the multiple target elements. In this way, a matching target element set is determined based on the target parameters input by the user, and anomaly detection processing is performed based on the target elements to generate a visual verification result, thereby improving the accuracy of the weather forecast detection.

[0189] Next, combine Figure 3 , for any forecast mode, anomaly detection processing is performed on the original meteorological data corresponding to the target element, and the process (S204) of determining the target meteorological data corresponding to the target element is explained.

[0190] Figure 3 This is a flow chart of another method for detecting weather forecasts provided in the embodiment of the present application. Figure 3 , the method comprising:

[0191] S301: Determine the element type corresponding to the target element.

[0192] Feature types can be precipitation, temperature, wind speed, humidity, etc.

[0193] The corresponding feature type can be determined based on the target feature.

[0194] S302: Perform weighted difference processing on the original meteorological data according to the element type to obtain a weighted difference value.

[0195] Weighted difference processing can obtain more accurate difference values ​​by performing difference calculations on the original meteorological data and assigning different weights according to the importance or reliability of the data points.

[0196] Optionally, the original meteorological data can be weighted and difference processed according to the element type to obtain a weighted difference value in the following manner: determining whether the element type is a precipitation type; if yes, determining a first factor corresponding to the observation data, a second factor corresponding to the reanalysis data, and a third factor corresponding to the GPM data, and determining a sum value between the first factor, the second factor, and the third factor as the weighted difference value, wherein the first factor is a product of a first weight value corresponding to the observation data and a first difference value, and the first difference value is a difference value between any one data in the observation data and a first weighted average value corresponding to the observation data; and if no, determining the first factor corresponding to the observation data and the second factor corresponding to the reanalysis data, and determining a sum value between the first factor and the second factor as the weighted difference value.

[0197] S303, performing a subsampling process on the weighted difference value and the original meteorological data to obtain a plurality of sampling sets.

[0198] The sampling set includes a plurality of sampling data, and the sampling data is the weighted difference value or the metadata.

[0199] The metadata is any one data in the original meteorological data.

[0200] A certain number of sampling data can be randomly extracted from the weighted difference value and the original meteorological data, and a set of the plurality of sampling data is determined as the sampling set.

[0201] S304, for any one sampling set, establishing an isolated tree of the sampling set according to a plurality of sampling data corresponding to the sampling set, and calculating path lengths corresponding to the plurality of sampling data in the sampling set according to the isolated tree.

[0202] The isolated tree can be a binary tree structure.

[0203] In the construction process, a sampling data (such as temperature data, precipitation data, etc.) in the sampling set is randomly selected, and a split point is randomly determined in the value range of the sampling data. According to the split point, the plurality of sampling data in the sampling set is divided into two parts, one part of the sampling data is less than the split point and enters the left subtree, and the other part of the sampling data is greater than or equal to the split point and enters the right subtree. This process is recursively performed until a preset stop condition is met, such as only one data point is left in the subtree, or a preset maximum depth of the tree is reached.

[0204] In the constructed isolated tree, for each sampling data in the sampling set, the number of edges from the root node to the leaf node where the data is located is the path length corresponding to the sampling data.

[0205] S305, determining abnormal data in the original meteorological data according to a plurality of path lengths of a plurality of sampling sets.

[0206] The abnormal data in the original meteorological data is determined according to the path lengths of the multiple sampling sets, and the accuracy of the abnormal data determination is improved.

[0207] Optionally, the abnormal data in the original meteorological data can be determined according to the path lengths of the multiple sampling sets in the following manner: for any one metadata, the first path lengths corresponding to the metadata are respectively determined in the multiple sampling sets; a first average value of the multiple first path lengths is determined; a dynamic threshold value corresponding to the metadata is determined; if the first average value is greater than the dynamic threshold value, the metadata is determined to be in an abnormal state; if the first average value is less than or equal to the dynamic threshold value, the metadata is determined to be in a normal state; and the metadata in the abnormal state is determined to be the abnormal data in the original meteorological data.

[0208] The dynamic threshold value can be determined according to different target elements, different target regions, or different target time periods.

[0209] In S306, the original meteorological data is filtered according to the abnormal data, and target meteorological data is obtained.

[0210] For example, there is abnormal data in the original temperature data sequence, and the abnormal data is directly deleted, so that the data sequence is more consistent with the real temperature change trend.

[0211] The implementation content of each step in the embodiments of the application can refer to the description of the corresponding steps or operations of the above method embodiments, and repeated content will not be described again.

[0212] The detection method of the meteorological forecast provided in this embodiment determines the element type corresponding to the target element, performs weighted difference processing on the original meteorological data according to the element type to obtain a weighted difference value, performs a subsampling process on the weighted difference value and the original meteorological data to obtain multiple sampling sets, establishes an isolated tree of each sampling set according to the multiple sampling data corresponding to the sampling set, calculates the path lengths corresponding to the multiple sampling data in the sampling set according to the isolated tree, determines the abnormal data in the original meteorological data according to the path lengths of the multiple sampling sets, filters the original meteorological data according to the abnormal data, and obtains target meteorological data. In this way, different abnormal detection processes are performed according to different target elements, and the accuracy of detecting meteorological forecasts is improved.

[0213] Figure 4 A structural schematic diagram of a detection device of a meteorological forecast provided in an embodiment of the application is shown in FIG. 4. Figure 4 The detection device 400 of the meteorological forecast includes an acquisition module 401, a first determination module 402, a second determination module 403, a verification module 404, and a display module 405, wherein,

[0214] The acquisition module 401 is configured to acquire a target parameter input by a user, the target parameter including a target area, a target period, and at least one meteorological element.

[0215] The first determination module 402 is configured to determine a target element set according to the at least one meteorological element, the target element set including a plurality of target elements.

[0216] The second determination module 403 is configured to determine original meteorological data corresponding to the plurality of target elements respectively within the target period in the target area.

[0217] The inspection module 404 is configured to, for any one target element, perform anomaly detection processing on the original meteorological data corresponding to the target element to determine target meteorological data corresponding to the target element.

[0218] The display module 405 is configured to generate a visual inspection result corresponding to a meteorological forecast according to the target meteorological data corresponding to the plurality of target elements respectively.

[0219] Optionally, the apparatus described above, the inspection module 404 is specifically configured to:

[0220] determine an element type corresponding to the target element;

[0221] perform weighted difference processing on the original meteorological data according to the element type to obtain a weighted difference value;

[0222] determine, according to the weighted difference value and the original meteorological data, abnormal data in the original meteorological data through a target algorithm;

[0223] perform filtering processing on the original meteorological data according to the abnormal data to obtain the target meteorological data.

[0224] Optionally, the apparatus described above, the inspection module 404 is specifically configured to:

[0225] perform a subsampling process on the weighted difference value and the original meteorological data to obtain a plurality of sampling sets, the sampling set including a plurality of sampling data, the sampling data being the weighted difference value or metadata, the metadata being any one data in the original meteorological data;

[0226] for any one sampling set, establish an isolation tree of the sampling set according to the plurality of sampling data corresponding to the sampling set, and calculate path lengths corresponding to the plurality of sampling data in the sampling set respectively according to the isolation tree;

[0227] determine the abnormal data in the original meteorological data according to the plurality of path lengths of the plurality of sampling sets.

[0228] Optionally, the apparatus as described above, the inspection module 404 is specifically configured to:

[0229] For any one metadata, a first path length corresponding to the metadata is determined in each of the plurality of sample sets;

[0230] A first average value of the plurality of first path lengths is determined;

[0231] A dynamic threshold value corresponding to the metadata is determined;

[0232] If the first average value is greater than the dynamic threshold value, it is determined that the metadata is in an abnormal state;

[0233] If the first average value is less than or equal to the dynamic threshold value, it is determined that the metadata is in a normal state;

[0234] The metadata in the abnormal state is determined as abnormal data in the original meteorological data.

[0235] Optionally, the apparatus as described above, the original meteorological data includes any one of observation data, forecast data, global precipitation measurement (GPM) data and reanalysis data, and the inspection module 404 is specifically configured to:

[0236] Judge whether the element type is a precipitation type;

[0237] If yes, a first factor corresponding to the observation data, a second factor corresponding to the reanalysis data, and a third factor corresponding to the GPM data are determined, and a sum value between the first factor, the second factor and the third factor is determined as a weighted difference value, wherein the first factor is a product of a first weight value corresponding to the observation data and a first difference value, and the first difference value is a difference between any one of the observation data and a first weighted average value corresponding to the observation data;

[0238] If no, a first factor corresponding to the observation data and a second factor corresponding to the reanalysis data are determined, and a sum value between the first factor and the second factor is determined as a weighted difference value.

[0239] Optionally, the apparatus as described above, the first determination module 402 is specifically configured to:

[0240] Obtain a plurality of candidate element sets, and the candidate element set includes a plurality of elements;

[0241] Determine a plurality of similarities between the at least one meteorological element and the plurality of candidate element sets;

[0242] The candidate element set with the highest similarity in the plurality of similarities is determined as a target element set.

[0243] Optionally, the apparatus as described above, the display module 405 is specifically configured to:

[0244] For any one target element, the target meteorological data corresponding to the target element is classified to obtain a plurality of type data, and the plurality of type data includes any one of observation data, forecast data, global precipitation measurement (GPM) data and reanalysis data;

[0245] Each type of data is standardized to obtain standard data corresponding to each type of data;

[0246] The standard data corresponding to the plurality of type data is input into the target model to determine a plurality of test scores corresponding to the target element;

[0247] According to the plurality of test scores corresponding to the plurality of target elements, a visual test result corresponding to the meteorological forecast is generated.

[0248] Figure 5 A structural schematic diagram of an electronic device provided by an embodiment of the present application is provided. Please refer to Figure 5 The electronic device 500 can include a memory 501, a processor 502, and a transceiver 503.

[0249] The memory 501 is configured to store program instructions.

[0250] The processor 502 is configured to execute the program instructions stored in the memory, so that the electronic device 500 executes the above method.

[0251] The transceiver 503 can include a transmitter and / or a receiver. The transmitter can also be referred to as a sender, a transmitter, a sending port or a sending interface, and the like. The receiver can also be referred to as a receiver, a receiving port or a receiving interface, and the like. For example, the memory 501, the processor 502 and the transceiver 503 are connected to each other through a bus 504.

[0252] The present application also provides a computer program product, which can be executed by a processor, and when the computer program product is executed, the above method can be implemented.

[0253] The meteorological forecast detection device, the electronic device, the computer readable storage medium and the computer program product provided by the embodiments of the present application can execute the technical solutions shown in the above meteorological forecast detection method embodiments, and the implementation principles and advantages are similar, which will not be repeated here.

[0254] It should be noted that, for the foregoing method embodiments, for the sake of simple description, they are all described as a combination of a series of actions, but those skilled in the art should know that the present application is not limited to the order of the actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily required by the present application.

[0255] Further, it should be noted that although each step in the flowchart is shown in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless otherwise specified herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other orders. Moreover, at least part of the steps in the flowchart can include multiple sub-steps or multiple stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily sequential, but can be executed in rotation or alternation with other steps or sub-steps or stages of other steps.

[0256] It should be understood that the above-described device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and actual implementation can have another division method. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0257] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application can be integrated in one unit / module, or each unit / module can exist physically, or two or more units / modules can be integrated together. The integrated unit / module can be realized in the form of hardware or in the form of a software program module.

[0258] If the integrated units / modules are implemented in the form of hardware, the hardware can be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. Unless otherwise specified, the storage unit can be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc.

[0259] If the integrated units / modules are implemented in the form of software program modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the essential part or all or part of the technical solutions that make contributions to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the embodiments of the present application. The aforementioned storage medium includes a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0260] In the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application

[0261] Other embodiments of the application will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. It is intended that the specification and examples be considered as exemplary only, with the true scope and spirit of the application being indicated by the following claims.

[0262] It is to be understood that the application is not limited to the precise construction herein disclosed and shown in the drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the application. The scope of the application is limited only by the claims that follow.

Claims

1. A method for detecting weather forecast, characterized in that: The method comprises: Obtaining target parameters input by a user, wherein the target parameters include a target area, a target time period, and at least one meteorological element; Determining a target element set according to the at least one meteorological element, wherein the target element set includes a plurality of target elements; Determining the original meteorological data corresponding to the plurality of target elements respectively within the target period of the target area; For any target element, perform anomaly detection processing on the original meteorological data corresponding to the target element to determine the target meteorological data corresponding to the target element; A visual inspection result corresponding to the weather forecast is generated based on the target meteorological data corresponding to the multiple target elements.

2. The method according to claim 1, characterized in that Performing anomaly detection processing on the original meteorological data corresponding to the target element to determine the target meteorological data corresponding to the target element includes: Determining the element type corresponding to the target element; performing weighted difference processing on the original meteorological data according to the element type to obtain a weighted difference value; determining abnormal data in the original meteorological data by a target algorithm based on the weighted difference value and the original meteorological data; The original meteorological data is screened and processed according to the abnormal data to obtain the target meteorological data.

3. The method according to claim 2, characterized in that Determining abnormal data in the original meteorological data using a target algorithm based on the weighted difference value and the original meteorological data includes: Performing a subsampling process on the weighted difference value and the original meteorological data to obtain a plurality of sampling sets, wherein the sampling sets include a plurality of sampling data, the sampling data being the weighted difference value or metadata, and the metadata being any one of the original meteorological data; For any sampling set, an isolation tree of the sampling set is established according to the multiple sampling data corresponding to the sampling set, and the path lengths corresponding to the multiple sampling data in the sampling set are calculated according to the isolation tree; Abnormal data in the original meteorological data is determined according to multiple path lengths of multiple sampling sets.

4. The method according to claim 3, characterized in that Determining abnormal data in the original meteorological data according to multiple path lengths of multiple sampling sets includes: For any metadata, determining the first path length corresponding to the metadata in each of the plurality of sampling sets; determining a first average of the plurality of first path lengths; determining a dynamic threshold corresponding to the metadata; If the first average value is greater than the dynamic threshold, determining that the metadata is in an abnormal state; If the first average value is less than or equal to the dynamic threshold, determining that the metadata is in a normal state; The metadata in an abnormal state is determined as abnormal data in the original meteorological data.

5. The method according to any one of claims 2 to 4, characterized in that The original meteorological data includes any one of observation data, forecast data, Global Precipitation Measurement (GPM) data, and reanalysis data. According to the element type, weighted difference processing is performed on the original meteorological data to obtain a weighted difference value, including: Determining whether the element type is a precipitation type; If yes, determine a first factor corresponding to the observation data, a second factor corresponding to the reanalysis data, and a third factor corresponding to the GPM data, and determine the sum of the first factor, the second factor, and the third factor as a weighted difference value, wherein the first factor is the product of a first weight value corresponding to the observation data and a first difference value, and the first difference value is the difference between any one of the observation data and a first weighted average value corresponding to the observation data; If not, determine a first factor corresponding to the observation data and a second factor corresponding to the reanalysis data, and determine the sum of the first factor and the second factor as a weighted difference value.

6. The method according to any one of claims 1 to 5, characterized in that Determining a target element set according to the at least one meteorological element includes: Acquire multiple candidate element sets, each of which includes multiple elements; determining a plurality of similarities between the at least one meteorological element and a plurality of candidate element sets; The candidate element set with the highest similarity among the multiple similarities is determined as the target element set.

7. The method according to claims 1-6, characterized in that Generating a visual inspection result corresponding to the weather forecast according to the target weather data corresponding to the plurality of target elements, including: For any target element, target meteorological data corresponding to the target element are classified and processed to obtain multiple types of data, where the multiple types of data include any multiple of observation data, forecast data, Global Precipitation Measurement (GPM) data, and reanalysis data; Perform standardization on each type of data to obtain the standard data corresponding to each type of data; Inputting standard data corresponding to the plurality of types of data into a target model, and determining a plurality of inspection scores corresponding to the target elements; A visual inspection result corresponding to the weather forecast is generated according to the multiple inspection scores corresponding to the multiple target elements.

8. A weather forecast detection device, characterized in that: The device comprises: An acquisition module, configured to acquire target parameters input by a user, wherein the target parameters include a target area, a target time period, and at least one meteorological element; A first determining module is configured to determine a target element set according to the at least one meteorological element, wherein the target element set includes a plurality of target elements; A second determining module is configured to determine the original meteorological data corresponding to the plurality of target elements within the target period of the target area; A verification module is used to perform anomaly detection processing on the original meteorological data corresponding to any target element, and determine the target meteorological data corresponding to the target element; The display module is used to generate a visual inspection result corresponding to the weather forecast according to the target weather data corresponding to the multiple target elements.

9. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when executed by a processor.

Images