CN-121615364-B - Iterative deviation correction method considering satellite instrument probe element non-uniformity

CN121615364BCN 121615364 BCN121615364 BCN 121615364BCN-121615364-B

Abstract

The invention discloses an iterative deviation correcting method considering satellite instrument probe element non-uniformity, which comprises the steps of correcting deviation amplitude, constructing a cost function, initializing, constructing a deviation operator, resetting the probe element deviation, updating an atmospheric quality deviation correction BC coefficient according to an observed background deviation O-B, calculating an observed background deviation O-B residual error as the probe element deviation by adopting the updated atmospheric quality deviation correction BC coefficient, calculating the observed background deviation O-B of the next round according to the probe element deviation of the previous round, repeating the probe element deviation calculating step, iterating for a plurality of times until the atmospheric quality deviation correction BC coefficient is stable, and outputting a final atmospheric quality deviation correction BC coefficient and the probe element deviation to correct the observed data. The method leads the deviation correction coefficient of the air mass deviation and the deviation of the detector to be stable, improves the normal property of the dispersion of the initial guess observation and the simulation result, leads the peak value and the average value to be closer to zero, and can be used for solving the problem of the non-uniformity of the probe elements of a satellite instrument and an infrared hyperspectral detector.

Inventors

HAN WEI
Kong Senyi
BI LEI
Yin Ruoying

Assignees

中国气象局地球系统数值预报中心
浙江大学

Dates

Publication Date: 20260508
Application Date: 20251209

Claims (6)

1. An iterative bias correction method considering the non-uniformity of a satellite instrument probe element is characterized by comprising the following steps: S1, limiting deviation correction amplitude by adopting a constraint deviation correction CBC method, constructing a cost function and initializing the cost function; S2, constructing a deviation operator by combining the atmospheric state dependence deviation and the probe element deviation, wherein the atmospheric state dependence deviation comprises an atmospheric state prediction factor and an atmospheric quality deviation correction BC coefficient; s3, zero clearing the deviation of the probe element, calculating a preliminary guess value of the observed background deviation O-B according to the original observed bright temperature and the background field simulated bright temperature, and updating an atmospheric quality deviation correction BC coefficient according to the observed background deviation O-B; s4, calculating the atmospheric state dependence deviation by adopting the updated atmospheric quality deviation correction BC coefficient, and deducting the atmospheric state dependence deviation from the initial guess value of the observed background deviation O-B to obtain an O-B residual error as a probe element deviation; S5, calculating the observed background dispersion O-B of the next round according to the probe cell deviation of the previous round, repeating the probe cell deviation calculation step, iterating for a plurality of times until the atmospheric quality deviation correction BC coefficient is stable, and outputting the final atmospheric quality deviation correction BC coefficient and the probe cell deviation to carry out deviation correction on the observed data; the cost function is specifically: ; Wherein the method comprises the steps of As a function of the cost, In the state of the background, Correcting BC vector for atmospheric quality deviation by correcting BC coefficient for atmospheric quality deviation The composition of the composite material comprises the components, In order to observe the background dispersion O-B, For the original observation of the bright temperature of the light, The bright temperature is simulated for the background field, In order for the deviation to be an operator, In order to observe the error covariance matrix, Prior estimation for probe cell bias Is a matrix of the error covariance of (c), Is a regularization parameter; the deviation operator specifically comprises: ; ; Wherein the method comprises the steps of For the atmospheric condition dependent deviation of the gas, In order to determine the deviation of the probe element, Is a constant for a certain channel and, Is an atmospheric state predictor, derived from the background state.
2. The iterative bias correction method for considering non-uniformity of a satellite instrument probe according to claim 1, wherein said method for initializing a cost function comprises: Taking a priori estimates of probe cell bias Error covariance matrix of (a) Equal to the covariance matrix of the observed error Setting the regularization parameter to ; Performing conventional deviation correction on the instrument to obtain an observed background deviation O-B peak value, and taking a priori estimate of the probe element deviation Equal to the observed background dispersion O-B peak.
3. The method for correcting the iterative offset taking the non-uniformity of a satellite instrument probe element into consideration as set forth in claim 1, wherein the atmospheric state predictors correspond to air mass thicknesses of different heights.
4. The method for correcting iterative bias in consideration of non-uniformity of probe cells of a satellite instrument according to claim 1, wherein the condition for updating the BC coefficient based on the observed background dispersion O-B is the minimum cost function.
5. The iterative offset correction method considering the non-uniformity of the probe cells of a satellite instrument according to claim 1, wherein said method for calculating the observed background offset O-B of the next round based on the probe cell offset of the previous round comprises: Subtracting the initial guess value of the observed background dispersion O-B from the probe element deviation of the previous round to obtain the observed background dispersion O-B calculated in the next round, wherein the expression is as follows: ; Wherein the method comprises the steps of Is the first The observed background dispersion O-B of the wheel calculation, Is the first And detecting element deviation calculated by the wheel.
6. The method for correcting the iterative bias in consideration of the non-uniformity of a satellite instrument probe according to claim 1, wherein the criterion for stabilizing the BC coefficient of the atmospheric bias correction is that the norm of the variation of the BC coefficient of the atmospheric bias correction is smaller than a preset threshold.

Description

Iterative deviation correction method considering satellite instrument probe element non-uniformity Technical Field The invention relates to the technical field of satellite storage data processing, in particular to an iterative deviation correction method considering non-uniformity of satellite instrument probe elements. Background In satellite remote sensing data processing, instrument system deviation is one of key factors limiting the accuracy of observed data, and the correction effect directly influences the quality of satellite data assimilation in numerical weather forecast. The conventional correction method has a limitation in processing the deviation of the satellite instrument probe element with complex non-uniformity characteristics, and may have problems of correction or incapacity of effectively distinguishing the deviation of different sources, for example, 128 probe elements of the earth stationary orbit infrared interference type detector (GIIRS) show significant non-uniformity, which brings challenges to accurate deviation correction. Therefore, the invention provides an iterative deviation correction method considering the non-uniformity of the satellite instrument probe element, which leads the air mass deviation correction coefficient and the detector deviation to be stable through the iterative process, can effectively process the non-uniformity of the probe element and avoid excessive correction, and is beneficial to improving the application value of the observation data in downstream application of remote sensing research and satellite data assimilation research. Disclosure of Invention The invention aims to provide an iterative deviation correction method considering the non-uniformity of a satellite instrument probe. In order to achieve the above purpose, the invention is implemented according to the following technical scheme: The invention comprises the following steps: Limiting deviation correction amplitude by adopting a constraint deviation correction (CBC) method, constructing a cost function and initializing the cost function; Constructing a deviation operator by combining the atmospheric state dependence deviation and the probe element deviation, wherein the atmospheric state dependence deviation comprises an atmospheric state prediction factor and an atmospheric quality deviation correction BC coefficient; zero clearing the deviation of the probe element, calculating a preliminary guess value of the observed background deviation O-B according to the original observed brightness and the background field simulated brightness temperature, and updating an atmospheric quality deviation correction BC coefficient according to the observed background deviation O-B; calculating an atmospheric state dependent deviation by adopting the updated atmospheric quality deviation correction BC coefficient, and deducting the atmospheric state dependent deviation from the initial guess value of the observed background deviation O-B to obtain an O-B residual error as a probe element deviation; And (3) calculating the observed background dispersion O-B of the next round according to the probe cell deviation of the previous round, repeating the probe cell deviation calculation step, iterating for a plurality of times until the atmospheric quality deviation correction BC coefficient is stable, and outputting the final atmospheric quality deviation correction BC coefficient and the probe cell deviation to carry out deviation correction on the observed data. Further, the cost function is specifically: ; Wherein the method comprises the steps of As a function of the cost,In the state of the background,Correcting BC vector for atmospheric quality deviation by correcting BC coefficient for atmospheric quality deviationThe composition of the composite material comprises the components,In order to observe the background dispersion O-B,For the original observation of the bright temperature of the light,The bright temperature is simulated for the background field,In order for the deviation to be an operator,In order to observe the error covariance matrix,Prior estimation for probe cell biasIs a matrix of the error covariance of (c),Is a regularization parameter; Further, the deviation operator specifically includes: ; ; Wherein the method comprises the steps of For the atmospheric condition dependent deviation of the gas,In order to determine the deviation of the probe element,Is a constant for a certain channel and,Is an atmospheric state predictor, derived from the background state. Further, the method for initializing the cost function comprises the following steps: Taking a priori estimates of probe cell bias Error covariance matrix of (a)Equal to the covariance matrix of the observed errorSetting the regularization parameter to; Performing conventional deviation correction on the instrument to obtain an observed background deviation O-B peak value, and taking a priori estimate of the probe element