24 results about "Aerosol extinction coefficient" patented technology

The invention discloses an aerosol extinction coefficient inversion method and device, a medium and a product, and relates to the technical field of laser radar aerosol optical parameter inversion, and the method comprises the steps: carrying out the preliminary inversion based on a multi-scale feature extraction network, carrying out the supervised learning of a laser radar signal profile obtained by a Mie-scattering laser radar, and obtaining an aerosol extinction coefficient; learning a statistical mapping relationship between the Mie scattering signal and the optical parameter to obtain a hot start model; and introducing a laser radar equation to construct a physical constraint term, carrying out fine iterative optimization on the hot start model based on the physical constraint term, obtaining an extinction coefficient profile and a backscattering coefficient profile with physical credibility, and completing aerosol extinction coefficient inversion. According to the method, adaptive calibration of the laser radar constant in multiple periods and multiple conditions can be realized, inversion and physical consistency optimization of the aerosol extinction coefficient are further realized, and consistency and stability of inversion results are ensured.

The invention belongs to the technical field of aerosol extinction, and relates to a CALIOP and Mie scattering laser radar combined aerosol extinction coefficient inversion method, which comprises the following steps of: calculating a first-order gradient according to a Mie scattering laser radar echo signal, and judging whether a strong positive and negative gradient pair layer exists or not according to whether a strong gradient exists in the first-order gradient; calculating the relative increase amplitude of the strong gradient, comparing the relative increase amplitude with a threshold value to judge the type of a strong positive and negative gradient pair layer, and setting an atmospheric boundary layer height range; a discrete wavelet covariance transformation function is adopted to invert the height of the atmospheric boundary layer, and in the Gaussian process machine learning network training process, scale factor parameters are selected for the discrete wavelet covariance transformation function so as to accurately invert the height of the atmospheric boundary layer; dividing the atmospheric layer into an atmospheric boundary layer and an atmospheric boundary layer top to a troposphere top according to the inversion atmospheric boundary layer height; according to the invention, the inversion precision of the aerosol extinction coefficient can be improved.

The present application provides an iterative inversion method and system for atmospheric stratification based on the Fernald method, which involves lidar technology. The method includes preprocessing the original echo signal and identifying cloud position information based on the slope method; initializing the aerosol extinction backscattering ratio of the cloud layer, iteratively adjusting the aerosol extinction backscattering ratio of the cloud layer based on the aerosol backscattering coefficient of the cloud base area until the set conditions are met, and recording the final aerosol extinction backscattering ratio of the cloud layer; inverting the aerosol backscattering coefficient sub-profiles of the area below the cloud base and above the cloud top based on the initialized backscattering ratio, inverting the cloud layer aerosol backscattering coefficient sub-profile based on the cloud backscattering ratio, and calculating the vertical profile of the aerosol extinction coefficient at the full height. By dynamically optimizing the aerosol extinction backscattering ratio in the cloud area, negative values ​​and over-suppression at the cloud base are eliminated, which helps to improve the inversion accuracy of the vertical profile of the aerosol extinction coefficient under complex cloud conditions.

The invention discloses a dust and haze multivariable collaborative regional dynamic comprehensive evaluation method based on site observation and grid live analysis products, which combines the difference of physical and chemical properties of meteorological elements and particulate matters in dust, fog and haze weather, and introduces a Mie scattering theory to calculate an aerosol extinction coefficient to diagnose fog and haze. A scientific key characteristic index threshold value is adopted, and a multivariable cooperation and regional dynamic comprehensive evaluation method is used for diagnosis; finally, the three diagnosis results of direct diagnosis and site and grid point indirect diagnosis are comprehensively judged, the uncertainty of single data source diagnosis is reduced to the maximum extent, the evolution process, the influence range and the intensity change characteristics of sand, dust, fog and haze weather can be well represented, the temporal-spatial resolution is 5 km / 1 h, and the time efficiency lags for 55 min.

The invention discloses a method for classifying different pollution weathers based on machine learning of an aerosol laser radar, and the method comprises the steps: 1, carrying out the preliminary quality control of the observation data of the laser radar, and constructing an hour-scale extinction and depolarization feature sample set; 2, extracting aerosol optical features of different heights to form a multi-dimensional aerosol optical feature set; 3, constructing a pollution classification model by using a random forest and a gradient lifting classifier, and classifying aerosol states into four classes, namely clean days, fine particulate matter pollution, coarse particulate matter pollution and co-pollution; and 4, performing model evaluation and precision verification. According to the invention, by fusing the extinction coefficient and depolarization ratio characteristics of the aerosol, automatic classification and identification of different types of pollution processes are realized, so that the intelligent monitoring capability of foundation remote sensing on the change of the vertical structure of atmospheric pollution is improved.

This invention discloses a blind-zone-free laser remote sensing system for stable boundary layer thermal material structures. The system includes a lateral scanning Raman scattering lidar receiving system with multiple lateral receiving systems, a lidar transmitting system, and a backscattering Raman scattering lidar receiving system. The lateral receiving system receives near-surface nitrogen and water vapor Raman signals and aerosol Mie-Rayleigh scattering signals, filling the detection blind zone of the backscattering lidar. The backscattering system detects signals in the upper atmosphere. The invention also discloses a blind-zone-free laser remote sensing method for stable boundary layer thermal material structures, which utilizes the lateral and backscattering systems for collaborative observation, respectively retrieving the water vapor mixing ratio, temperature, and aerosol extinction coefficient near the ground and in the upper atmosphere, and fusing them through a data stitching algorithm to obtain a blind-zone-free profile of the entire layer. This method enables refined detection of the entire layer from the ground to the upper atmosphere, continuously, with full parameters, and high spatiotemporal resolution.

A method for detecting the atmospheric characteristics of the geometric factor region by using a scanning lidar, comprising the following steps: S1. Vertically detect the lidar, and calculate the aerosol extinction coefficients at each height within the range from the height z above the geometric factor region to the calibration distance z according to the Fernald method; g above to the calibration distance z c ; S2. Obliquely scan the lidar from the vertical direction to the horizontal direction at a certain zenith angle in sequence, that is, fan-shaped scanning. The number of scanning groups is n, the height resolution is dr, and find out the zenith angle a(n) of each scanning group and its corresponding calibration height z(n); S3. Assume that the whole atmosphere is horizontally uniform during the fan-shaped scanning process, and calculate the aerosol extinction coefficient α within the range of the geometric factor region z from the aerosol extinction coefficient profiles measured by each scanning group. g ; The present invention can measure the atmospheric characteristics within the lidar geometric factor region by using a scanning lidar, and can provide measurement data within the corresponding geometric factor for other lidar systems. a ​

A kind of space-borne wind, temperature, aerosol synchronous observation laser radar, laser emission unit emits single-frequency pulsed laser into atmosphere;Receiving telescope unit collects laser backscattering echo and sends to wave filter unit, filters out aerosol signal and retains atmospheric molecular Rayleigh scattering signal, forms echo filter light signal into interferometer unit;Interferometer unit successively carries out interference modulation to sampling single-frequency pulsed laser and echo filter light signal, generates four-way interference signals with phase difference being 0°, 90°, 180° and 270° in turn, is sent into four signal detection units, successively detects sampling laser optical signal and echo filter light signal, and after photoelectric conversion and digitization, it is sent into processing inversion unit;Processing inversion unit uses the digital signal transmitted from four signal detection units, and inverts wind speed, temperature and aerosol extinction coefficient.The present application has the functions of measuring wind speed, temperature and aerosol by one space-borne laser radar device, and has high integration and high cost performance.

The invention provides a temperature and water vapor mixing ratio inversion method in a high aerosol concentration scene, and belongs to the technical field of laser atmosphere remote sensing data inversion. According to the method, EAE and aerosol extinction coefficients at the wavelength of 355 nm, the wavelength of 532 nm and the wavelength of 1064 nm are obtained through iterative inversion; performing interpolation operation based on the aerosol extinction coefficients and EAE to obtain aerosol extinction coefficients at the wavelength of 353 nm, the wavelength of 354 nm, the wavelength of 387 nm and the wavelength of 407 nm; calculating a first spectral attenuation correction term based on the aerosol extinction coefficients at the wavelength of 387nm and the wavelength of 407nm, wherein the first spectral attenuation correction term is used for inverting a water-vapor mixing ratio; a second spectral attenuation correction term is calculated based on the aerosol extinction coefficients at the wavelength of 353 nm and the wavelength of 354 nm, the second spectral attenuation correction term being used to invert the temperature. According to the invention, through constraining the EAE, accurate calculation of the spectrum attenuation correction term is realized, and finally, the inversion precision of the temperature and water vapor mixing ratio is remarkably improved.

The disclosure provides an aerosol profile inversion method and device based on dual-wavelength Mie scattering lidar data, which comprises: obtaining an aerosol component profile initial value; constructing a mixed-mode aerosol optical property forward model; inputting the aerosol component profile initial value into the forward model to obtain an aerosol lidar ratio and a simulated aerosol extinction coefficient profile; based on the aerosol lidar ratio, using the Fernald method to obtain an inversion aerosol extinction coefficient profile; obtaining an aerosol extinction coefficient profile ratio according to the simulated aerosol extinction coefficient profile and the inversion aerosol extinction coefficient profile, and using it to adjust the aerosol total concentration; adjusting the aerosol component proportion based on the aerosol optical thickness value; obtaining a new aerosol component profile according to the adjusted aerosol total concentration and the component proportion; repeating the above steps until the difference between the simulated aerosol extinction coefficient profile and the inversion aerosol extinction coefficient profile is less than a threshold value, and completing the aerosol profile inversion.

The invention relates to an aerosol mass concentration estimation method based on a laser radar and an optical property meter. Based on collected data, the extinction coefficient, backscattering coefficient and aerosol backscattering coefficient of atmospheric molecules are calculated; based on the initial assumption of the ratio of the aerosol extinction coefficient to the backscattering coefficient, the aerosol extinction coefficient is calculated; the difference between the aerosol extinction coefficient observed by an aerosol extinction meter and the aerosol extinction coefficient calculated by a laser radar inversion is analyzed, an error convergence index is constructed, and a correction strategy for the laser radar aerosol extinction coefficient is formulated; based on the observation data of an aerosol particle size spectrometer and the corrected laser radar aerosol extinction coefficient, an addition mathematical model and a logarithmic mathematical model are combined to design a sliding window width dynamic adjustment strategy, and a relationship model between the laser radar aerosol extinction coefficient and the mass concentration that integrates the linear and nonlinear model characteristics is constructed, thereby realizing accurate estimation of the aerosol mass concentration.

An inversion method of an aerosol extinction coefficient below clouds by Light Detection and Ranging (LIDAR) detection includes: obtaining an aerosol extinction coefficient corresponding to an echo signal of LIDAR in a horizontal direction as a first calibration value EXT1; determining a calibration point altitude at a cloud, and obtaining an atmospheric extinction coefficient corresponding to the calibration point altitude at the cloud base as a second calibration value EXT0; obtaining a second extinction coefficient profile, and obtaining an extinction coefficient at a first altitude X as a second calibration value EXTX, the first altitude X being greater than a blind area altitude; comparing the first calibration value EXT1 with the second calibration value EXTX, adjusting the second calibration value EXT0 when |EXTX−EXT1|>EXT1·δ, performing the obtaining a second extinction coefficient profile based on the adjusted second calibration value EXT0 until |EXTX−EXT1|<EXT1·δ, and outputting the second calibration value EXTX, where δ is a relative error.

The invention discloses a multi-component aerosol vertical observation assimilation method based on local set assimilation. The method comprises the following steps: acquiring orbit observation data, observation profile data and multi-component aerosol extinction coefficient observation data of a target area; on the basis of latitude and longitude and time information in multi-component aerosol extinction coefficient observation data and latitude and longitude range and vertical layer structure information of a mode, converting orbit observation data to hourly mode grid points, and generating multi-dimensional data which can be directly used for assimilation; and performing multi-component aerosol vertical observation assimilation based on four-dimensional local ensemble Kalman filtering according to the generated multi-dimensional data. By establishing the relationship between the extinction coefficients of the aerosols with different components and the concentrations of the aerosols with different components, the concentration of the total aerosol and the proportion of the concentrations of the aerosols with different components can be adjusted, and assimilation with higher precision can be realized.

The invention discloses a low-altitude visibility prediction method, system and device based on wind field driving and a medium, and the method comprises the steps: inputting an obtained multivariable time sequence matrix into an air gating map attention network, and obtaining an advection space feature matrix; fusing the advection spatial feature matrix and the multivariable time sequence matrix to obtain a fused multivariable time sequence matrix; splitting the fused multivariable time sequence matrix into a plurality of independent channels to obtain a one-dimensional time sequence corresponding to each independent channel; the one-dimensional time sequence is processed through a Transform encoder, and a target time-varying burst feature vector is determined; determining a dry aerosol extinction coefficient estimation value and a moisture absorption growth parameter estimation value based on the target time-varying burst feature vector; and performing visibility prediction based on the dry aerosol extinction coefficient estimation value and the moisture absorption growth parameter estimation value to obtain a visibility prediction result. According to the invention, the accuracy of low-altitude visibility prediction can be improved.

The application provides a temperature and water vapor mixing ratio inversion method in a high-aerosol-concentration scene, and belongs to the technical field of laser atmospheric remote sensing data inversion. The application obtains EAE and aerosol extinction coefficients at 355nm wavelength, 532nm wavelength and 1064nm wavelength through iterative inversion; performs interpolation operation based on the aerosol extinction coefficients and the EAE to obtain aerosol extinction coefficients at 353nm wavelength, 354nm wavelength, 387nm wavelength and 407nm wavelength; calculates a first spectral attenuation correction term based on the aerosol extinction coefficients at 387nm wavelength and 407nm wavelength, and the first spectral attenuation correction term is used for inversion of water vapor mixing ratio; calculates a second spectral attenuation correction term based on the aerosol extinction coefficients at 353nm wavelength and 354nm wavelength, and the second spectral attenuation correction term is used for inversion of temperature. The application realizes accurate calculation of the spectral attenuation correction term by constraining the EAE, and finally realizes significant improvement in inversion accuracy of the temperature and the water vapor mixing ratio.

The invention discloses a carbon dioxide laser radar inversion method based on aerosol cooperative modulation. The method comprises the following steps: firstly, acquiring multi-wavelength laser radar echo signals of a strong absorption wavelength, a weak absorption wavelength and an aerosol reference channel; inverting the extinction coefficient and the backscattering coefficient of the aerosol and characteristic parameters representing the micro-physical form of the aerosol; constructing a collaborative modulation model based on the radiation transmission model and calculating an aerosol collaborative optical modulation factor; and the carbon dioxide concentration profile is explicitly introduced into a differential absorption inversion equation, and multiple scattering and wavelength-related scattering differences caused by the aerosol are corrected, so that a high-precision carbon dioxide concentration profile is obtained through inversion. According to the method, the influence of high-concentration aerosol and complex weather conditions on carbon dioxide detection can be effectively weakened, the inversion stability and precision of the laser radar in near-earth and polluted environments are remarkably improved, and the method has a good engineering application prospect.

The application discloses a low-altitude visibility prediction method and system based on wind field driving, equipment and a medium. The method comprises the following steps: inputting a obtained multivariate time series matrix into a wind door control graph attention network to obtain an advection space feature matrix; fusing the advection space feature matrix and the multivariate time series matrix to obtain a fused multivariate time series matrix; splitting the fused multivariate time series matrix into multiple independent channels to obtain a one-dimensional time series corresponding to each independent channel; processing the one-dimensional time series through a Transformer encoder to determine a target time-varying burst feature vector; determining a dry aerosol extinction coefficient estimate value and a hygroscopic growth parameter estimate value based on the target time-varying burst feature vector; and performing visibility prediction based on the dry aerosol extinction coefficient estimate value and the hygroscopic growth parameter estimate value to obtain a visibility prediction result. The application can improve the accuracy of low-altitude visibility prediction.

The present invention relates to the technical field of lidar detection, and particularly to a measurement method, system and medium of an LED lidar based on pseudo-random modulation. The method modulates an LED beam generated by an LED light source based on a target pseudo-random code to obtain a modulated beam, and emits the modulated beam; receives the backscattered light of the modulated beam, preprocesses the backscattered light, and converts the preprocessed backscattered light into an electrical signal; uses a photon counter to perform high-speed integration operation compensation processing on the electrical signal; demodulates the electrical signal, and calculates the aerosol extinction coefficient in the environment to be measured according to the demodulation result. In this way, the influence of optical noise can be reduced, and the LED lidar system can be adapted to daytime use.

The invention discloses a random forest-based Mie scattering laser radar aerosol extinction coefficient inversion method. The method comprises the following steps of 1, Mie scattering laser radar data set construction and standardization processing; step 2, predicting an aerosol extinction coefficient by using a random forest; and step 3, optimizing random forest model parameters by a Bayesian method. The method can improve the precision and reliability of aerosol extinction coefficient inversion.

The application discloses a single-photon laser radar horizontal direction aerosol extinction coefficient inversion method and belongs to the technical field of atmospheric monitoring. The method is as follows: obtaining horizontal direction aerosol echo signal data, performing background denoising, geometric overlap factor correction, distance square correction and smoothing processing on the original signal; setting an initial reference point near the maximum detection distance, and combining differential comparison and wavelet change to position the mutation interval of the processed aerosol echo signal data; reselecting a new reference point between adjacent mutation intervals, and using a Collis-Fernald iteration algorithm to obtain a stable solution of the extinction coefficient boundary value near the new reference point; and substituting the obtained initial reference point and the extinction coefficient boundary value at the initial reference point into a Fernald backward integral formula to inversely obtain the aerosol extinction coefficient value on the distance between the initial reference point and the starting point of the detection path. The application can improve the precision and reliability of the extinction coefficient inversion.

The present invention discloses a method and system for inverting high-resolution, three-dimensional, full-coverage aerosol spatial distribution data, relating to the field of atmospheric environment. The method comprises: collecting multi-source data and performing quality control and preprocessing; using the multi-source data as independent variables and the orbital layered aerosol extinction coefficients of cloud-aerosol lidar and infrared pathfinder satellites as dependent variables, and inputting these into the extreme gradient boosting model (XGBoost); generating prediction results, combining them with the original independent variables to form an enhanced feature set, which is then input into the lightweight gradient boosting machine learning model (LightGBM) to generate a process prediction set; and using a wavelet transform method to fuse the process prediction set with a global atmospheric reanalysis dataset, extracting the advantageous information, and obtaining an accurate prediction dataset of high-resolution, three-dimensional, full-coverage aerosol extinction coefficients. The present invention effectively improves the reliability, adaptability, and scalability of the inversion method, and the resulting dataset can comprehensively, accurately, and detailedly characterize the three-dimensional spatial distribution of aerosols.

The invention discloses a method and system for inverting high-resolution three-dimensional full-coverage aerosol spatial distribution data, and relates to the field of atmospheric environment, and the method comprises the steps: collecting multi-source data, and carrying out the quality control and preprocessing; the method comprises the following steps: taking multi-source data as an independent variable, taking an orbit layered aerosol extinction coefficient of a cloud-aerosol laser radar and an infrared path finder satellite as a dependent variable, inputting the independent variable and the dependent variable into an extreme gradient boosting model XGBoost, generating a prediction result, combining the prediction result with an original independent variable, forming an enhanced feature set, and inputting the enhanced feature set into a lightweight gradient boosting machine learning model LightGBM, generating a process prediction set; and fusing the process prediction set and the global atmosphere reanalysis data set by adopting a wavelet transform method, and extracting dominant information to obtain an accurate prediction data set of the high-resolution three-dimensional full-coverage aerosol extinction coefficient. According to the method, the reliability, the adaptability and the expansion capability of the inversion method are effectively improved, and the obtained data set can comprehensively, accurately and detailedly represent the aerosol three-dimensional space distribution condition.

The invention provides an atmosphere layering iteration inversion method and system based on a Fernald method, and relates to a laser radar technology. The method comprises the following steps: preprocessing an original echo signal, and identifying cloud layer position information according to a slope method; initializing the aerosol extinction backscattering ratio of the cloud layer, carrying out iterative adjustment on the aerosol extinction backscattering ratio of the cloud layer according to the aerosol backscattering coefficient of the cloud base region until a set condition is met, and recording the final aerosol extinction backscattering ratio of the cloud layer; and according to the initialized backscattering ratio, inverting aerosol backscattering coefficient sub-profiles of a region below the cloud bottom and a region above the cloud top, according to the backscattering ratio of the cloud, inverting a cloud layer aerosol backscattering coefficient sub-profile, and calculating an aerosol extinction coefficient vertical profile of a complete height. By dynamically optimizing the extinction backscattering ratio of the aerosol in the cloud region, the cloud bottom negative value and the excessive inhibition phenomenon are eliminated, and the inversion precision of the vertical profile of the extinction coefficient of the aerosol under the complex cloud gas condition can be improved.

The invention discloses a rainfall prediction system based on a laser radar, and the system comprises a data preprocessing module, a space-time prediction module, a full-dimension replacement mixing module and a training module, and the data preprocessing module is used for carrying out the physical calibration and spatial remodeling of aerosol extinction coefficients and humidity data collected by the laser radar. The spatio-temporal prediction module is used for extracting spatio-temporal dependence features of data to form a two-dimensional spatial feature map by using a spatio-temporal memory flow with memory decoupling, the full-dimensional replacement mixing module is used for fusing multi-channel physical data through a dynamic weighting mechanism and outputting a rainfall probability, and the training module is used for training the rainfall probability through a multi-stage joint optimization strategy. Converting the meteorological original data into high-precision rainfall prediction capability; according to the method, the space-time sequence processing capability of the space-time prediction module is fused, the full-dimensional replacement hybrid module is improved, and the dynamic weighting mechanism is combined, so that efficient fusion and decoding of the laser radar are realized, channel weights are adaptively distributed through the dynamic weighting module, and the effect of an extinction coefficient is enhanced.