52 results about "Rainfall estimation" patented technology

An integrated rainfall calculation method using X-band dual-polarimetric radar measurement data includes a precipitation classification step of classifying hydrometeors into four types of snow, rain / snow, rain and non-meteorological target through a fuzzy logic technique using a correlation coefficient (cross correlation coefficient, ρhv), features of a measured differential propagation phase (Ψdp(r)) or differential propagation phase (φdp) and a signal-to-noise ratio (SNR) as input variables (input feature vector); a specific differential phase calculation step of separately calculating a specific differential phase by applying a specific differential phase using a total difference of differential phase and signal-attenuation corrected reflectivity for the rain among the classified hydrometeors and applying a specific differential phase calculated using a filtering method for the other hydrometeors; and a rainfall calculation step of calculating rainfall by using a relation between the specific differential phase and the rainfall and using the separately calculated specific differential phase.

The invention provides a multi-source information fusion based rainfall estimation method. The multi-source information fusion based rainfall estimation method includes: combining ground station observation rainfall data in a study area with a multi-source satellite rainfall data body in the study area to form a multi-source data set in the study area; establishing a dynamic Bayesian theory basedBayesian rainfall prediction model; using the maximum entropy method to obtain a nonlinear optimal solution of the Bayesian rainfall prediction model, and then determining the optimal weight and uncertainty information of each satellite data source; and generating an estimation result using the multi-source information fusion rainfall in the study area. The advantages of the invention are that theresult of multi-data source fusion analysis can reduce the uncertainty of the regional rainfall estimation due to the inaccuracy of a single type of rainfall information; and more reliable data inputand richer and more refined modeling data are provided for strengthening regional high-precision disaster warning, avoiding the flood risk or small watershed rainstorm flood estimation.

The invention discloses a space statistical downscaling rainfall estimation method based on a geographical difference analysis method. The method comprises the following steps that: A, a statistical model is built according to the relationship between annual rainfall data of satellite precipitation data and a normalized differential vegetation index, and the downscaling annual rainfall is obtained; B, the geographical difference analysis method is adopted for calibrating the downscaling annual rainfall obtained in the step A; C, the calibrated downscaling annual rainfall obtained in the step B is subjected to weighted decomposition to obtain monthly rainfall data; and D, the geographical difference analysis method is adopted for calibrating the monthly rainfall data obtained in the step C, and the calibrated monthly rainfall data is obtained. Compared with a conventional method, the space statistical downscaling rainfall estimation method has the advantages of higher relevance and smaller error.

The present invention relates to a regional automatic rainfall station hourly rainfall data quality control system and method. The control system comprises a radar raw data pre-processing unit, an input unit, a quality flow main unit, an output unit, a destination file drawing unit and a destination file output parameter unit, wherein the radar raw data pre-processing unit is connected with the input unit; the input unit is connected with the quality flow main unit; the quality flow main unit comprises a rainfall estimation module, a matching module and a quality control module; and one end of the output unit is connected with the matching module while the other end is individually connected with the destination file drawing unit and the destination file output parameter unit. The control system provided by the present invention is simple and reasonable in structure design and stable and reliable in operation. According to the control method, a face state of the rainfall can be accurately captured by introducing radar files, thus to effectively help to judge uncertain identification of extreme rainfalls in a flood season.

The invention discloses a regional automatic weather station hourly rainfall data quality control method. The method comprises the steps that a discriminant model is established according to historical data; raindrop spectrometer data, weather radar data and automatic weather station rainfall data are collated in real time; the data of a number of raindrop spectrometers are used to evaluate and revise a radar echo intensity result; the relationship between the radar echo intensity and the rain intensity is fitted according to different precipitation types; the fitted relationship is substituted into a radar quantitative estimation precipitation module to realize radar quantitative rainfall estimation to acquire hourly rainfall; the set of weather radar estimated hourly rainfall and regional automatic weather station hourly rainfall is established; and comparing with the pre-established discriminant model is carried out to judge whether the regional automatic weather station hourly rainfall data are abnormal. The automatic weather station rainfall data are accurate and reliable. The observation data quality is improved. The method lays a good foundation for weather monitoring, early warning, forecasting and other businesses.

The invention provides a rainfall estimation method based on microwave rainfall attenuation, which comprises the following steps: acquiring microwave rainfall attenuation data of an estimation area and corresponding rainfall data, and obtaining a rainfall estimation result to be corrected through a microwave rainfall inversion model; acquiring multi-source non-rainfall data corresponding to the rainfall data, training a deep belief network by using the multi-source non-rainfall data and the microwave rainfall data set, constructing a correction and optimization model, and performing correctionand optimization on estimated data output by the microwave rainfall inversion model; obtaining a preliminary rainfall estimation result of the estimation area through the microwave rainfall inversionmodel according to estimation request parameters proposed by a user, and carrying out the correction and optimization of the preliminary rainfall estimation result according to the correction and optimization model, so as to obtain a final rainfall estimation result of the estimation area. The invention also provides a rainfall monitoring system based on microwave rain attenuation and a data processing device for rainfall estimation by using the rainfall estimation method.

The invention discloses a sparse region rainfall estimation fusion method based on a dual-smoothing method. On the basis of data assimilation thought, rainfall point data observed by rain gauges and TMPA grid rainfall data are integrated together to form a set of discrete points, finally, and the two integrated data sets are turned into an improved novel data set; meanwhile, smoothing is adopted for correcting discontinuity formed before TMPA fusion. According to the dual-smoothing method, rainfall values and spatial distribution are more reasonable and effective, average deviation and root-mean-square errors of a region where rain gauges are distributed sparsely change slightly along with time, heavy rainfall can be more accurately estimated, and meanwhile the rainfall visual result is smoother. The method is particularly applicable to regions where automatic meteorological stations are spatially distributed sparsely in a western region.

The invention discloses a system for satellite short-time approaching rainfall forecasting. The system is characterized by comprising a first subsystem used for receiving and processing satellite data, a second subsystem used for cloud detection and cloud classification, a third subsystem used for rainfall estimation and a fourth subsystem used for cloud motion wind deduction and rainfall forecasting. The first subsystem is connected with a second subsystem connected with a third subsystem. The fourth subsystem is respectively connected with the second subsystem and the third subsystem. A method for rainfall forecasting is further disclosed. The method includes first receiving and processing the satellite data, then conducting cloud detection and cloud classification, then conducting rainfall estimation and conducting approaching rainfall forecasting finally. The system and the method can provide abundant weather products and are good in issuing timeliness, high in issuing frequency, capable of conducting real-time issuing, capable of forecasting short-time approaching rainfall and high in forecasting accuracy.

The invention provides a drainage basin rainfall product correction method and system based on satellite-ground fusion, and the method comprises the steps: dividing all pixels in a remote sensing image to be corrected into five types, and enabling a C1 pixel to be an image grid pixel corresponding to the position of a ground in-situ sensor; wherein the C2 pixel is an image grid pixel correspondingto the rainfall grid point data position; wherein the C3 pixels and the C1 and C2 pixels are located in the same rainfall subarea, and the C3 pixels and the C1 and C2 pixels have rainfall correlation; wherein the C4 pixels and the C3 pixels are located in the same rainfall subarea and have rainfall correlation; for the C1 pixel and the C2 pixel, correcting the C1 pixel and the C2 pixel by using an improved quantile mapping method; for the C3 pixel, establishing mathematical expressions of the C1 and C2 pixel correction methods by utilizing a random forest regression method to realize correction; for a C4 pixel, realizing correction by taking a rainfall ratio relational expression before and after the C3 pixel is corrected as a reference; wherein the uncorrected grid pixels in the image are C5 pixels, and the rainfall correction value is a remote sensing image rainfall estimation value corresponding to the spatial position.

The invention discloses a foundation and satellite-borne radar rainfall data fusion method based on wavelet domain regularization. The method comprises: based on radar rainfall data wavelet domain statistical characteristics, selecting an appropriate prior model of rainfall data; and determining a regularization function of scale coefficient fusion and wavelet coefficient fusion after wavelet decomposition of the rainfall data of the foundation and the satellite-borne radar, then solving the scale coefficient of the rainfall data and the maximum posteriori estimation of the wavelet coefficientin a wavelet domain by using a gradient projection method, and finally performing wavelet inverse transformation to obtain the optimal rainfall estimation. According to the method, uncertainty of rainfall estimation of different sensors and wavelet domain statistical characteristics of rainfall data are considered in the fusion process. The fusion result reduces the uncertainty of a single sensor, and can better maintain and reconstruct detail features such as a heavy rainfall extreme value and small-scale change, thereby being more beneficial to monitoring and forecasting of heavy disaster weather such as flood monitoring and the like.

The invention relates to the technical field of information, and provides a rainfall nowcasting method and device based on deep learning. The objective of the invention is to solve the problems of insufficient attention of traditional model training on strong echo prediction and difficulty in radar echo space-time sequence prediction. According to the main scheme, the method comprises the following steps: S1, preprocessing weather radar base data; S2, dividing the preprocessed radar echo data into a training set, a verification set and a test set for deep learning network training; S3, performing space-time coding prediction network model training by using the training set, the verification set and the test set, performing radar echo extrapolation by using the trained space-time coding prediction network model to obtain predicted echo data, and comparing the predicted echo data with actual observation data; and S4, finally, carrying out radar quantitative rainfall estimation on the predicted radar echoes, and comparing the radar echoes with ground real rainfall data to carry out rainfall forecast detection.

The invention discloses a rainfall estimation method based on weather radar reflectivity feature matching. First of all, a radar reflectivity feature database is constructed, then matching rules of live radar data and the radar reflectivity feature database are determined, and based on this, an optimal parameter A and an optimal parameter b in a Z-R relation are dynamically determined. According to the invention, dynamical optimization of the Z-R relation is realized in short temporary rainfall forecast service application, and the accuracy of rainfall estimation through a radar is further improved.

The invention provides a rainfall estimation method and system and a terminal. The rainfall estimation method comprises the steps that historical rainfall information and corresponding historical environment factors of a target region are acquired; the historical rainfall information and the corresponding historical environment factors are subjected to deep learning to obtain a rainfall predictionmodel; and according to real-time environment factors of the target region, future rainfall information of the target region is evaluated by applying the rainfall prediction model. According to the technical scheme, rainfall estimation is conducted in the deep learning mode, the effectiveness of rainfall estimation is greatly improved, people can conveniently arrange affairs such as travel and work according to the effective rainfall information, and convenience is brought to daily life of people.

Based on dual-polarization radar data, raindrop spectrum data and rain gauge data, a neural network and a Logistic multiple regression method are utilized to realize the adaptive optimization inversion of the dual-polarization radar precipitation, and the precipitation inversion formula can be optimally selected to obtain high-precision radar precipitation data, so that the method can be well applied to disastrous weather monitoring and early warning forecasting. Besides, the method is good in adaptability, parameter setting does not need to be changed according to the weather process, and dual-polarization radar precipitation optimization inversion can be achieved through raindrop spectrum data, dual-polarization radar observation data and rain gauge data in the radar coverage range. Moreover, the method is less affected by rainfall system examples and data measurement precision, can be applied to any detection position of the ground-based radar, and has high rainfall estimation precision and good universality and robustness.

The invention discloses a radar-rain gauge combined rainfall estimation method based on a space-time local model. The method comprises the following steps: conducting preliminary quality control of observation data of a rain gauge; conducting radar echo quality control, gridding radar body scanning data, averaging radar echo time weight, constructing a radar echo-rain gauge sample pair, dividing radar echo, conducting quality control over sample pairs with observation values of a radar and the rain gauge through a space-time local model, establishing a precipitation echo grid point position Z-R relation through the space-time local model, realizing precipitation estimating and radar precipitation estimated value calibrating. The beneficial effects of the invention are that: the radar-raingauge combined rainfall estimation method based on a space-time local model divides rainfall echoes and non-rainfall echoes more scientifically, carries out the quality control of a Z-R sample pair more meticulously, carries out the more accurate modeling of a Z-R relation changing along with the space, and further improves the precision of radar-rain gauge combined rainfall estimation.

The invention discloses a multi-sensor rainfall estimation fusion method. the method comprises: firstly, performing radar (DPR / GR) rainfall estimation data and rain gauge data fusion based on a wavelet domain HMT model adaptive filtering method; and then carrying out multi-scale decomposition and fusion on the DPR and GR rainfall estimation data filtered by the rain gauge in a wavelet domain to obtain a high-precision high-resolution rainfall estimation result. The uncertainty of rainfall estimation of different sensors and wavelet domain statistics and geometrical characteristics of rainfall data are considered in the fusion process, and detail characteristics such as heavy rainfall extreme values and small-scale changes can be better kept and reconstructed while the uncertainty of a single sensor is reduced through the fusion result.

The invention discloses a rainfall estimation method and system coupled with dry and wet state identification and multi-source information fusion. The method comprises the following steps: obtaining various mutually independent rainfall observation information from different sources including ground station observation, remote sensing inversion and reanalysis; based on ground observation rainfall, extracting station rain / no rain state values, selecting a geographically weighted regression model, and estimating the rainfall probability of each station and a grid; preferably selecting a rainfall probability threshold value, and identifying a rain / no-rain state of each grid; fusing geography and multi-source precipitation information by adopting a geography weighted regression method to obtain grid fusion precipitation; and correcting and fusing the precipitation amount by using the rain / no-rain state identification result of each grid, and obtaining a precipitation amount estimation result. According to the method, modeling and coupling are carried out on the rainfall dry and wet state and the rainfall amount respectively, the accuracy of rain area and rain-free area identification is effectively improved, the obtained rainfall space distribution is more consistent with the reality, the false alarm rate and the false alarm error are greatly reduced, and other precision indexes are also improved.

The invention provides a method and a system for reducing short-time rainfall estimation deviation, and belongs to the field of image processing, radar rainfall models and average calibration. According to the method, the Z-R relationship is determined by utilizing radar map data information and an average calibration method, the Z-R relationship is established through actual precipitation data and radar echo data, a correction coefficient is obtained through back-stepping calculation by utilizing a related formula, and then a Z-R relationship model with the correction coefficient is obtainedto quantitatively estimate the precipitation of the observation grid point without the automatic station. The problem of short-time rainfall estimation deviation at present is solved, high-resolutionquantitative estimation rainfall data with business application value can be obtained in real time, and auxiliary support is provided for fine forecasting service and heavy rainfall waterlogging riskassessment.

The invention discloses a critical rainfall estimation method based on rainfall spatial distribution. The critical rainfall estimation method comprises the steps that firstly, estimating the confluence flow velocity in each grid unit in a hydrological response unit secondly, on the basis of the confluence flow velocity in each grid unit, obtaining a confluence time grid in the drainage basin; performing spatial interpolation on the time-interval rainfall data observed by the station to obtain a rainfall spatial distribution grid, and combining the convergence time grid to obtain a time-interval dynamic rainfall index; counting an accumulated dynamic rainfall index before a flood peak occurs in the flood process; then counting the tension water content saturation, namely the initial soil water content saturation, of the day before the flood process begins; finally, whether the flood is an over-alarm flood is judged, and making a cumulative dynamic rainfall indicator - initial soil watercontent saturation scattering point graph on the basis of multiple floods and i, marking points corresponding to the overwanted flood, and drawing a critical rainfall line in combination with the points corresponding to the overwanted flood.

The invention discloses a quantitative rainfall estimation method combining meteorological radar and rainfall bucket observation data. The method comprises steps of obtaining the original rainfall bucket accumulation data and the original meteorological radar rainfall accumulation data, and obtaining a matched rainfall bucket-meteorological radar data G / R pair on a grid at a same geographic position; calculating observation errors of original meteorological radar rainfall accumulation data by using the G / R pairs, performing abnormal data detection, generating an initial correction factor field, judging whether distance correlation exists between the initial correction factor field and the observation errors or not, if so, adjusting the initial correction factor field, and if not, adjusting the initial correction factor field; and correcting the original meteorological radar rainfall accumulation data by using the adjusted correction factor field to obtain the corrected meteorological radar rainfall accumulation data, otherwise, directly using the average field deviation to obtain the corrected meteorological radar rainfall accumulation data. According to the method, meteorological radar and rainfall bucket observation data are effectively combined, and quantitative rainfall estimation which is more accurate in space and wide in application range is formed.

The invention relates to a method for applying a comprehensive quality index to rainfall estimation by combining a radar and an automatic station, and the method comprises the following steps: obtaining a radar estimation rainfall-automatic station observation rainfall data pair and a plurality of radar data quality indexes in a target region, calculating a comprehensive quality index according to the plurality of radar data quality indexes, and calculating an average value of the comprehensive quality index; and correcting the radar estimation rainfall-automatic station observation rainfall data pair according to a first preset threshold value of the corresponding comprehensive quality index average value to obtain final rainfall estimation data. In the technical scheme of the invention, when the comprehensive quality index is applied to the process of jointly estimating rainfall by the radar and the automatic station, the rainfall estimation capability of the networking radar can be further improved.

The invention discloses a Doppler weather radar data quality control method which comprises the following steps of obtaining a meteorological echo data distribution rule and a ground clutter data distribution rule based on Doppler weather radar historical base data information; identifying ground clutter by using a skewness method based on the meteorological echo data distribution rule and the ground clutter data distribution rule; designing a radar ground clutter filter according to the data distribution rule and the skewness method, and performing filtering by using the radar ground clutter filter to obtain radar filtering data after ground clutter removal; determining a radar beam blocking rate of the radar beam blocking area based on the Doppler weather radar original base data and the radar filtering data; and performing radar filtering data correction. The method can effectively identify and remove the ground clutter in the radar base data, and fill and correct the filtered radar data to realize the quality control of the radar base data, thereby improving the accuracy of meteorological rainfall estimation.

The invention discloses an optional Kriging spatial interpolation rainfall estimation method which comprises the following steps: 1, calculating experimental variation function values of all rainfallsample point pairs through a discrete variation function formula; 2, fitting the experimental variation function values by adopting a two-scale wavelet least square support vector machine to obtain atheoretical variation function model; 3, establishing a Kriging spatial interpolation equation set, and solving a Kriging weight coefficient; 4, calculating a rainfall estimated value of the to-be-estimated position point according to the Kriging weight coefficient; and 5, quantitative precision evaluation. According to the estimation method disclosed by the invention, the defects that the shape of a traditional theoretical variation function model is fixed and the spatial variation trend and multi-scale characteristics are not considered in a common Kriging spatial interpolation rainfall estimation method are overcome, so that a rainfall estimation result conforms to the own spatial variation trend and reflects the multi-scale characteristics of spatial variation.

The invention discloses a quantitative rainfall estimation method based on multi-source data fusion. The method comprises the following steps: determining a rainfall falling area by adopting a methodbased on multi-source big data; performing interpolation on microwave link inversion rainfall to a discrete grid and accumulating the interpolation to a T-minute time scale; interpolating weather radar inversion rainfall into a discrete grid and converting the weather radar inversion rainfall into rainfall of a T-minute time scale by adopting a time weight average method; acquiring rainfall observed by a rain gauge of a T-minute time scale, establishing three types of samples of the T-minute time scale by adopting an interpolation method, performing quality control on the three types of samples by adopting corresponding space-time local models respectively, and estimating rainfall and rainfall accumulation by adopting a method of fusing multiple space-time local models. The method has thebeneficial effects: through multi-source data and multi-model fusion, the rainfall quantitative estimation precision is further improved on the basis of realizing accurate monitoring of the rainfall falling area.

The invention discloses a multi-mode rainfall estimation method integrated with machine learning, which is characterized by comprising the following steps: calculating and generating extreme climate indexes of four types of rainfall in a Chinese region by adopting newest CMIP6 climate mode simulation data and grid point data observed by a station of the China Meteorological Administration; training a plurality of machine learning models by utilizing observation and mode data in historical periods, evaluating the performance of different machine models, giving different weights based on the model performance, and forming average and extreme rainfall probability estimation results of different temperature increasing scenes and temperature increasing amplitudes in the future integrally. According to the invention, the multi-mode and multi-machine algorithm integration technology is adopted, the uncertainty introduced by different modes and different algorithms in future estimation is reduced, and the estimation result is more reliable.

The invention discloses a radar quantitative rainfall estimation method based on deep learning. The method comprises the following steps: acquiring weather radar data and ground station rainfall data; performing data screening and coordinate correction on the reflectivity data of the elevation angle of the weather radar data; performing data quality control on the rainfall data of the ground station to obtain a relatively accurate rainfall intensity label; carrying out space-time matching on the weather radar data and the ground station rainfall data after pre-data processing, cutting the weather radar data into three-dimensional radar data structures, and respectively taking the three-dimensional radar data structures as input samples and labels of a radar quantitative rainfall estimation model; and generating a complete quantitative rainfall estimation result, and superposing the result on the topographic file to generate accurate regional rainfall information. The rainfall intensity can be accurately estimated based on the radar data, and high-precision regional rainfall intensity estimation is realized.

The invention discloses a medium and short term hydrological forecasting method and system fused with multivariate information, and the method comprises the steps: collecting the related information of a whole watershed, extracting the feature information of the whole watershed, selecting a distributed hydrological model according to the feature information of the whole watershed, selecting an ARW model as an atmospheric numerical model, according to the features of a power frame, predicting an atmospheric motion state and a weather phenomenon in a certain period of time in the future; and embedding the established distributed hydrological model into a land surface model of the atmospheric numerical model, so that the atmospheric numerical model and the distributed hydrological model share the same land surface process mechanism, and constructing bidirectional coupling of the numerical weather forecast and the distributed hydrological model. Through the method for eliminating the initial error of the weather forecast numerical model, the weather forecast precision is improved, the watershed small-scale matching application of the gridding quantitative rainfall estimation model is realized, the seamless connection between the weather forecast model and the hydrological forecast model is ensured, and the hydrological forecast precision is improved.

The invention discloses a mountain small watershed radar quantitative rainfall estimation technology in the technical field of convective disaster weather monitoring. The technology comprises the steps of S1, building a membership function, and obtaining Z-R relation statistics of different rainfall types; S2, judging the echo type, and removing non-rainfall echoes; S3, carrying out precipitation type division; S4, carrying out rainfall inversion through echo observation of different rainfall types and Z-R relation statistics of different rainfall types, and generating a single-station radar quantitative rainfall estimation product; and S5, generating a radar precipitation rate puzzle product according to the precipitation puzzle and the radar reflectivity. According to the mountain small watershed radar quantitative rainfall estimation technology, through processing in five aspects, namely, reflectivity factor quality control, rainfall type classification, radar mixed elevation reflectivity calculation, Z-R conversion relation and radar puzzle, the precision of a mountain small watershed weather radar QPE product is improved, and the mountain small watershed weather radar QPE product can better serve rainfall forecast and early warning of hydrological and geological disasters.