12 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 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 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 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.

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 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 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.