45 results about "Empirical orthogonal functions" patented technology

The invention discloses a deep sea multibeam sound ray accurate tracking method. The method mainly comprises the steps of (1) establishing a spatial and temporal variation ocean temperature and salinity field model on the basis of the comprehensive analysis of influence factors of sea surface temperature and salinity and with the integration of marine satellites and Argo buoy multiple-source marine physical hydrological observation data, (2) calculating each beam initial incident angle through analyzing the influence of a ship instantaneous attitude on a beam initial incident angle and considering a ship attitude, (3) calculating a sound speed and an inversion sound speed based on an empirical orthogonal function through the spatial and temporal temperature and salinity field at a sound speed profile no-measured area, obtaining the average value of sound speeds obtained by two methods at a corresponding point as the sound speed of the point, and thus calculating a three-dimensional sound speed profile model, (4) constructing an efficient constant gradient sound ray accurate tracking model, and (5) providing a sound ray tracking precision evaluation method. According to the method, the influence of the ship attitude on the beam initial incident angle is considered, the precision of beam footprint coordinates can be greatly improved.

The invention discloses a multi-beam and low-glancing-angle beam homing method based on a laser-point cloud used as a constraint and belongs to the technical field of hydrographic surveying and charting. The multi-beam and low-glancing-angle beam homing method comprises the following steps: firstly, constructing a three-dimensional sound velocity field with a revised shallow layer in a region by using an experience orthogonal function analysis method according to a plurality of sound velocity sections in an offshore region and a surface layer sound velocity value along the near part of a track line, so as to finish the revising of an initial sound velocity of depth measurement data; then, taking a characteristic point extracted by the same-target laser-point cloud as an outermost sound wave beam constraint to obtain coordinate deviation of the characteristic point with the revised initial sound velocity; accordingly calculating an initial rotary angle and an initial amplification coefficient in a low-glancing layer so as to weight and interpolate the rotary angle which corresponds to each residual beam and is more than 60 degrees and the amplification coefficient according to angle linearity; finally, realizing geometrical correction of the beams. With the adoption of the method provided by the invention, the accurate homing of the sound wave beams in shallow-layer seawater is realized, and the problem of seamless splicing of upper and lower point clouds of waterlines is solved.

The invention discloses a marine multi-element medium-and-long-term statistical prediction method based on empirical orthogonal function decomposition. The method comprises the following steps of (1)based on the to-be-analyzed and predicted marine re-analysis data, constructing a marine statistical re-analysis sample matrix day by day over the years; (2) separating the seasonal signals and non-seasonal signals of various marine elements to obtain a marine re-analysis range sample matrix which is day-by-day over the years; (3) constructing a marine multi-element space-time four-dimensional orthogonal mode; (4) constructing a marine multi-element medium-and-long-term statistical prediction model; and (5) forecasting various marine elements by adopting the marine multi-element medium-and-long-term statistical prediction model constructed in the step (4). According to the invention, based on the marine re-analysis products with enough long-time sequences, the medium-and-long-term prediction method for the marine dynamic thermal environment is constructed through statistical rules, so that the marine prediction greatly improves the time efficiency of marine prediction on the basis of atraditional marine prediction mode, and a technical foundation is laid for effectively improving the safety guarantee level of the medium-and-long-term marine environment.

The invention discloses a navigation X-wave-band radar wave group detection method based on a wave theory. The navigation X-wave-band radar wave group detection method based on a wave theory includes the steps: performing empirical orthogonal function decomposition on a navigation X-wave-band radar image sequence, selecting a main modal to reconstruct a wave field, and obtaining the wave surface displacement of different positions in an observed sea area through scaling of the wave field; for the wave surface of one radial direction, utilizing change of the displacement gradient to detect the extreme value of the wave surface, and according to the difference between adjacent two maximum values and minimum values, obtaining the wave height at different positions; utilizing the sum of two sine functions to fit the wave surface between the adjacent minimum wave heights, and identifying the area as a wave group when the fitting parameter accords with the wave theory; and combining with observations at different times, obtaining the parameters of the wave group, such as amplitude, length and group speed. The navigation X-wave-band radar wave group detection method based on a wave theory utilizes the buoy to perform scaling on the main modal of the wave field so as to obtain the wave height information of a large area in the observed sea area, and then utilizes the two sine functions which are approximate in frequency based on the wave theory to determine the wave group without requirements for selecting a threshold according to the experience, and has the advantages of being simple and easy to do, and being wide in the application range.

The invention relates to an underwater temperature field reconstruction method based on a self organizing neural network and an empirical orthogonal function. A self-organizing characteristic mappinggraph including multidimensional information including an empirical orthogonal function coefficient, position information, time information, sea surface height and sea surface temperature corresponding to a temperature profile is established, an Euclidean distance between known information and a self-organizing characteristic mapping unit is used to determine an optimal matching unit, and the empirical orthogonal function coefficient to inverse is obtained. A sea surface parameter and water body temperature profile characteristic mapping network is established based on a lot of data information, and nonlinear mapping from sea surface parameters to the water body profile can be realized. The underwater temperature field reconstruction method based on the self organizing neural network and the empirical orthogonal function is excellent in performance, high in stability, needless of knowing a power process in the marine site, easy to realize, low in computing complexity, and suitable forobtaining ocean environmental parameters of the key marine site in quasi-real time by using satellite remote sensing data, and only uses correlation among the ocean environmental parameters.

The invention discloses an aerosol optical thickness inversion method of a domestic multi-angle polarization satellite sensor. The method comprises the following steps: carrying out satellite multi-angle and multi-band data matching preprocessing; identifying cloud pixels of the multi-angle polarization satellite image data; based on surface reflection contribution estimation of an empirical orthogonal function, constructing a scattering matrix under multi-angle observation to perform surface radiation item estimation; constructing an aerosol model and establishing an inversion lookup table; performing aerosol optical thickness inversion based on optimal model identification. The invention provides a global land overhead aerosol optical thickness inversion method suitable for a high-resolution No.5 satellite multi-angle polarization imager (DPC) for the first time, and particularly relates to a global land overhead aerosol optical thickness inversion method suitable for a high-resolution No.5 satellite multi-angle polarization imager. The algorithm is stable, efficient and verifiable, and the method can be applied to other satellite platforms with multi-angle detection capability in a business manner.

The invention provides a sound velocity profile inversion method based on underwater acoustic communication signals between network nodes. The method includes the following steps that: step 1) a waterregion requiring sound velocity profile inversion is selected, and a sound velocity profile empirical orthogonal function is obtained on the basis of the sound velocity profile historical data of thewater region, and the empirical orthogonal function is adopted to reconstruct a set of sound velocity profiles; 2) received signals corresponding to each sound velocity profile in the step 1) are generated through simulation by means of a sound field calculation model; and step 3) correlation operation is performed on all the generated received signals in the step 2) and the underwater acoustic communication signals transmitted between the network nodes in the water region, a sound velocity profile corresponding to the highest correlation coefficient is selected as the optimal sound velocityprofile of the water region. According to the above method, the EOF (empirical orthogonal function) is obtained on the basis of the sound velocity profile historical data; and the sound velocity profile values reconstruct by the EOF are more effective sound velocity profile samples; and therefore, the validity and accuracy of inversion results can be ensured.

The invention provides a method for X-band radar wave retrieval based on empirical orthogonal decomposition. The method includes constructing a covariance matrix by utilizing X-band radar echo data, conducting empirical orthogonal function decomposition on the covariance matrix, obtaining characteristic vector and characteristic root of main mode through determination, establishing a relation between the wave height He, the main modal characteristic root and the average signal to noise ratio SNR of radar echo, fitting by utilizing simulation or actual measurement data to obtain a relation coefficient, establishing a relation between the wave direction Di and phase of the main modal characteristic vector, fitting by utilizing simulation or actual measurement data to obtain a relation coefficient, and carrying out real-time wave parameter retrieval according to the established relation between the wave height and wave direction. The method is complementary to a conventional spectrum analysis retrieval algorithm, and improves the precision for X-band radar wave retrieval.

The invention discloses a dynamic prediction method for a flow duration curve under a climatic change condition. The method comprises steps of analyzing the correlation degree of runoff and rainfall in good and bad years by adopting a linked list inspection method; extracting one-dimensional and two-dimensional principal components of the actually measured runoff volume by adopting an empirical orthogonal function, identifying the high correlation of rainfall and runoff in magnitude through fluctuation change characteristics of the principal components in runoff and rainfall good and bad years, and further establishing the correlation of rainfall-runoff good and bad changes; and predicting a flow comprehensive duration curve under a specified weather condition by taking a future runoff good and bad division ratio speculated based on a predicted future rainfall time sequence as an adjustment coefficient and based on flow duration curves of good-diameter and bad-diameter years. Accordingto the method, rainfall information of various climate conditions can be flexibly described by introducing different climate scenes, non-stationary change characteristics of flow duration curves of different drainage basins are identified, dynamic prediction of the flow duration curves is realized, and the method has very high practicability and wide applicability.

The invention discloses a sound velocity profile inversion method based on an over-complete dictionary, and the method comprises the steps: 1), constructing a non-orthogonal and over-complete dictionary D of a sound velocity profile based on a K-SVD algorithm according to historical data; 2) linearizing the nonlinear response of the forward model by using a first-order Taylor expansion, and creating a measurement matrix Q; 3) laying a single sound source and a vertical array in a sea area to be measured, and transmitting and receiving signals; 4) based on an OMP algorithm, inverting a sparse coefficient vector X of the over-complete dictionary atoms; and 5) calculating the sound velocity profile according to the time-varying non-orthogonal dictionary atomic coefficient obtained by inversion. Compared with a method based on an empirical orthogonal function, the method is based on an over-complete dictionary, the intrinsic characteristics of the sound velocity profile can be better captured through redundancy characteristics, and the inversion precision of the sound velocity profile is greatly improved.

The invention belongs to the technical field of marine acoustics application of three-dimensional temperature-salinity-depth data, and particularly relates to a high-resolution sound velocity profile data compression method based on empirical orthogonal function decomposition. The method specifically comprises the following steps of providing data by using a high-resolution marine reanalysis product, and converting by using a sound velocity formula to obtain the seawater sound velocity profile data; according to the characteristics of the technical field of marine acoustics application, extending the high-resolution and long-time sequence sound velocity profile data under a certain space-time dimension to a seabed sedimentary layer; using a feature vector of a main mode as a primary function of the sound velocity profile empirical orthogonal function decomposition, and compressing and expressing the high-resolution and long-time sequence sound velocity profile information under the certain space-time dimension. According to the invention, the technical scheme of obtaining sound velocity profile data compression from a high-resolution ocean reanalysis product is realized; the effective compression of the sound velocity profile data under the certain time-space dimension is realized, and the beneficial effect that the compression rate of the high resolution and long-time sequence sound velocity profile data is more than 90%, is obtained.

The invention provides an industry cluster space layout simulation, optimization and screening method, which comprises the steps of firstly generating a geographic space simulation field, introducing the geographic space simulation field into a space location theoretical model and establishing an industry geographic cluster model; then performing reliability assessment to industry geographic cluster simulation by adopting a Monto Carlo spatial statistic test method; finally screening out an industry cluster space layout optimization mode from a group of simulation results by adopting an empirical orthogonal function. The industry cluster space layout simulation, optimization and screening method provided by the invention has the advantages that since the lack of geographic spaces in the space location theoretical model is defined, the randomness of the industry geographic cluster simulation is eliminated, the problem of uncertainty in the industry geographic cluster simulation is solved, the screened regional industry cluster space layout optimization mode is guaranteed to have stronger robustness and a basis of scientific decision making is provided for regional industry planning and sustainable development.

The invention provides a wind resource value obtaining method based on a mass conservation diagnosis mode. The method comprises the following steps: obtaining actual measured data of a wind field and constructing an initial wind filed; setting a model parameter, fixing the value of alpha1 as 1, and by taking a long-time wind speed sequence space component value relation based on empirical orthogonal function decomposition as a measurement standard and by taking a dominant wind direction analysis result of each wind measurement tower as a reference of a wind direction diagnosis effect, determining alpha2; and adjusting the initial wind field to enable the wind field to satisfy a mass conservation equation and a total adjusting amount to be the smallest.

The invention discloses an empirical mode decomposition-based establishment method for an hour prediction model of energetic electron flux. The method includes the following steps: step 1. simplifyinga time coefficient of the energetic electron flux by an empirical orthogonal function; step 2. performing reconstruction expansion on the empirical orthogonal function of the time coefficient of theenergetic electron flux by employing a first order primary function of the empirical orthogonal function; step 3. obtaining the time coefficient of the electron flux by using the known electron flux and primary function; step 4, selecting an input parameter; step 5. decomposing the time coefficient by using an empirical mode decomposition EMD algorithm; and step 6. performing fitting on each of components of the time coefficient. The method overcomes the prediction difficulty of the energetic electron flux caused by non-stationary, a decomposed data sequence is stronger in regularity than an original data sequence of the time coefficient, and the prediction accuracy can be significantly improved.

The invention discloses a wind power plant abandoned wind power assessment method based on a hierarchical clustering method. The method comprises the following steps: step S10, performing standardization moment flattening on actually measured wind speed data of a wind power plant; step S20, subjecting a wind speed matrix after the standardization moment flattening to empirical orthogonal function decomposition; step S30, based on the empirical orthogonal function decomposition result, calculating the variance contribution rate and the cumulative variance contribution rate of each spatial type; step S40, performing clustering analysis on the spatial matrix obtained by decomposing the empirical orthogonal function using the hierarchical clustering method; step S50, drawing a clustering dendrogram based on the result of hierarchical clustering analysis; step S60, dividing wind turbine generators based on the clustering dendrogram, selecting a mark post draught fan and determining distribution coefficients; and step S70, using the mark post draught fan method to calculate the amount of the abandoned wind power. The method of the invention can reduced errors in a mark post draught fan method, and relatively accurately reflect abandoned wind conditions in the wind power plant.

The invention is suitable for the technical field of satellite positioning, and provides a tropospheric delay model establishing and obtaining method and device and a positioning system. The establishing method comprises the steps of obtaining tropospheric delay grid data; performing elevation correction based on the tropospheric delay grid data to obtain elevation correction data; performing empirical orthogonal function decomposition on the elevation correction data to obtain a decomposition function; and performing fitting based on the decomposition function to construct a tropospheric delay model. According to the method, the tropospheric delay model is constructed by utilizing empirical orthogonal function decomposition, so that the precision of model construction can be improved.

The invention relates to a North Pacific Ocean sea surface salinity interannual modal identification method and system. The method comprises the following steps: acquiring CMIP6 climate mode data of a plurality of modes; obtaining an ORAS4 reanalysis data set of the European mid-term weather forecast center as comparison reference data; two empirical orthogonal function modes EOF1 and EOF2 of the NPSDV are calculated by comparing reference data, the EOF1 is a dipole mode, and the EOF2 is a monopole mode; two empirical orthogonal function modes EOF1'and EOF2 'of the NPSDV are calculated through the CMIP6 climate mode data of all modes, and the modes of the EOF1' and the EOF2 'are recognized according to the space correlation coefficient of the EOF1 and the EOF1', the space correlation coefficient R12 of the EOF1 and the EOF2 ', the space correlation coefficient R21 of the EOF2 and the EOF1' and the space correlation coefficient R22 of the EOF2 and the EOF2 '.

The invention relates to an offshore anemometer tower data representative year correction method based on multiple reference stations. The method comprises the steps of data acquisition, data processing, empirical orthogonal function decomposition, regression equation establishment, error checking and representative year correction completion. According to the method, the problem that the correction result of the representative year of the offshore anemometer tower data has a large error can be solved. The method has universality, is suitable for reference stations and data of various sources,and is suitable for offshore anemometer towers and data of offshore and open seas with various offshore distances.

A method and system for analyzing a seismically active field based on expansion of an empirical orthogonal function is provided. The research region of the seismic active field is gridded at equal intervals for the preset research region of a seismic active field; a seismic active field function matrix correlated with the research region of the seismic active field spatially and temporally is constructed according to the gridding of the research region of the seismic active field; and the seismic active field function matrix is expanded with an empirical orthogonal function to obtain a main typical field and a temporal factor thereof, and an anomaly on the temporal factor of the seismic active field is analyzed with a method index, a parameter index and an anomaly index.

The invention discloses a typical galloping microtopography classification method and system based on empirical orthogonal function analysis. The method comprises the following steps: collecting digital elevation data of a first resolution in a set area range and historical transmission line galloping tripping data; dividing grids by taking each galloping tripping point as a center to obtain a terrain height data matrix in the corresponding grid; performing empirical orthogonal function analysis according to the terrain height data matrix, and decomposing the terrain height data matrix into a spatial matrix representing spatial modal distribution and a time matrix representing a sequence; the front N feature vectors of the space matrix are extracted N typical microtopography modes; and extracting time coefficients corresponding to the first N feature vectors from the time matrix representing the sequence, and dividing the microtopographic region into N classes according to the time coefficients of the first to Nth feature vectors to obtain microtopographic classes of the microtopographic region. According to the method, the topographic features of the typical galloping microtopography can be extracted under the condition that the topographic types are not preset.