CN108896456B - Aerosol extinction coefficient inversion method based on feedback type RBF neural network - Google Patents
Aerosol extinction coefficient inversion method based on feedback type RBF neural network Download PDFInfo
- Publication number
- CN108896456B CN108896456B CN201810397787.8A CN201810397787A CN108896456B CN 108896456 B CN108896456 B CN 108896456B CN 201810397787 A CN201810397787 A CN 201810397787A CN 108896456 B CN108896456 B CN 108896456B
- Authority
- CN
- China
- Prior art keywords
- aerosol
- extinction coefficient
- neural network
- rbf neural
- optical thickness
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000000443 aerosol Substances 0.000 title claims abstract description 122
- 230000008033 biological extinction Effects 0.000 title claims abstract description 98
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 45
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 15
- 230000003287 optical effect Effects 0.000 claims description 32
- 238000001514 detection method Methods 0.000 claims description 31
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 2
- 239000002994 raw material Substances 0.000 claims 1
- 230000007246 mechanism Effects 0.000 abstract description 3
- 239000000523 sample Substances 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 238000005259 measurement Methods 0.000 description 5
- 239000002245 particle Substances 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241000282414 Homo sapiens Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 239000005427 atmospheric aerosol Substances 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000001311 chemical methods and process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000000191 radiation effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N15/00—Investigating characteristics of particles; Investigating permeability, pore-volume or surface-area of porous materials
- G01N15/06—Investigating concentration of particle suspensions
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
Landscapes
- Chemical & Material Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Dispersion Chemistry (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses an aerosol extinction coefficient inversion method based on a feedback type RBF neural network, which comprises the following steps of 1) training the RBF neural network by utilizing input and expected output; taking the power of the historical echo signal as the input of an RBF neural network, and taking an aerosol extinction coefficient obtained according to the historical echo signal as the expected output of the RBF neural network; 2) and inverting the extinction coefficient of the aerosol based on a feedback type RBF neural network. The aerosol extinction coefficient is inverted by using the feedback type RBF neural network, and an internal mechanism between information is stored in the network through the learning of a sample mode, so that the uncertainty caused by a plurality of assumptions is effectively avoided, and the aerosol extinction coefficient inversion method has high response speed and high robustness.
Description
Technical Field
The invention relates to an aerosol extinction coefficient inversion method based on a feedback type RBF neural network, and belongs to the technical field of aerosol measurement.
Background
The atmospheric aerosol is a multi-phase system composed of various solid or liquid particles which are suspended in the atmosphere and are composed of objects in different phases, and can affect development and change of a plurality of physical and chemical processes in the atmospheric environment. The aerosol particles have a diameter ranging from 0.001 to 100 μm and can stay in the atmosphere for at least several hours or even days, so that the composition, structure and the like of the atmosphere are changed, and the original normal ecosystem is disturbed and destroyed. It is mainly distributed in the whole atmosphere, and can affect the climate effect, thereby affecting the health of human beings. Therefore, it is of great practical significance to improve the atmospheric environment by detecting and studying aerosols.
Compared with detection means such as satellites, the laser radar has the advantages of high space-time resolution, high measurement accuracy and the like, and is widely applied to the research fields such as laser atmospheric transmission, global climate detection, aerosol radiation effect, atmospheric environment and the like as an active remote sensing detection tool, so that the large-range real-time monitoring of parameters such as aerosol extinction coefficient, particle spectrum distribution, shape and the like is realized. When the laser radar is adopted for aerosol detection, the extinction coefficient or the backscattering coefficient of the aerosol is usually inverted through a radar equation, and then other characteristics of the aerosol are obtained. Then, when the extinction coefficient is inverted, because a plurality of variables exist in the radar equation, for the convenience of calculation, many variables are replaced by common empirical values or assumptions, such as an aerosol extinction coefficient boundary value, an aerosol extinction backscattering ratio and the like, so that a plurality of uncertain factors exist in the inversion of the extinction coefficient, and the fine inversion of the optical characteristics of the aerosol is obviously not facilitated.
Disclosure of Invention
In order to solve the technical problem, the invention provides an aerosol extinction coefficient inversion method based on a feedback type RBF neural network.
In order to achieve the purpose, the invention adopts the technical scheme that:
an aerosol extinction coefficient inversion method based on a feedback type RBF neural network comprises the following steps,
1) training the RBF neural network with the inputs and the desired outputs;
taking the power of the historical echo signal as the input of an RBF neural network, and taking an aerosol extinction coefficient obtained according to the historical echo signal as the expected output of the RBF neural network;
the process of obtaining the extinction coefficient of the aerosol comprises the following steps: constructing a nonlinear equation based on the principle of the optical thickness of the aerosol, iteratively calculating the extinction backscattering ratio of the aerosol by using a truncation method, and inverting the extinction coefficient of the aerosol by using a Fernald method according to the extinction backscattering ratio of the aerosol and an echo signal;
2) and inverting the extinction coefficient of the aerosol based on a feedback type RBF neural network.
And constructing a nonlinear equation by using a calculation formula for measuring the optical thickness of the aerosol by using a sunlight photometer and a calculation formula for measuring the optical thickness of the aerosol by using a radar, and iteratively calculating the extinction backscattering ratio of the aerosol by using a frustum method.
The calculation formula for measuring the optical thickness of the aerosol by the sunlight meter is as follows,
wherein S isAODOptical thickness, S, of aerosol for detection by a solar photometerALLFull thickness of atmosphere, S, for solar photometer detectionMODFor effective detection of the optical thickness, r, of atmospheric molecules in the rangeaIs the effective detection distance, σm(r) is the extinction coefficient of atmospheric molecules.
The calculation formula of the optical thickness of the aerosol measured by the radar is as follows,
wherein L R is the optical thickness, σ, of the aerosol detected by radara(r) is the extinction coefficient of the aerosol, which is a ratio S of extinction backscattering of the aerosolaAnd a function of the detection distance r.
The non-linear equation is constructed by constructing,
wherein L R is the optical thickness, σ, of the aerosol detected by radara(r) is the extinction coefficient of the aerosol, which is a ratio S of extinction backscattering of the aerosolaAnd a function of the detection distance r, SAODOptical thickness, S, of aerosol for detection by a solar photometerALLFull thickness of atmosphere, S, for solar photometer detectionMODFor effective detection of the optical thickness, r, of atmospheric molecules in the rangeaIs the effective detection distance, σm(r) is the extinction coefficient of atmospheric molecules。
Training the RBF neural network by using input and expected output, inputting the current echo signal into the trained RBF neural network to obtain the current aerosol extinction coefficient, and calculating the AOD (optical thickness of aerosol) according to the current aerosol extinction coefficientlidarThe current aerosol optical thickness AOD measured by a solar photometersunAnd comparing, if an error exists, correcting the current aerosol extinction coefficient, taking the corrected aerosol extinction coefficient and the corresponding echo signal as a new sample, and performing secondary training on the feedback RBF neural network.
The formula for correcting the extinction coefficient of the aerosol is as follows,
AECcorrected＝Networkoutput×(1+ω)
wherein AECcorrectedFor corrected aerosol extinction coefficient, NetworkoutputFor the aerosol extinction coefficient before correction, ω is the error.
The invention achieves the following beneficial effects: 1. according to the method, the feedback type RBF neural network is used for inverting the aerosol extinction coefficient, an internal mechanism between information is stored in the network through learning of a sample mode, uncertainty caused by a plurality of assumptions is effectively avoided, meanwhile, due to the addition of a dynamic feedback adjustment process approaching instrument measurement, the reliability of an inversion result is greatly increased, and quick and accurate inversion of the aerosol extinction coefficient is realized; 2. according to the invention, an equation about the extinction backscattering ratio is constructed by combining the echo signal and the detection data of the sunshine photometer, the solution is carried out by using a chord section method, and the aerosol extinction coefficient is calculated by the extinction backscattering ratio, so that the error caused by the extinction backscattering ratio is avoided to a certain extent, and the aerosol detection precision is improved.
Drawings
FIG. 1 is a structural schematic diagram for inverting aerosol extinction coefficients based on a feedback type RBF neural network;
FIG. 2 is a flow chart of a method for determining an extinction backscattering ratio for an aerosol;
FIG. 3 is a flow chart of the acquisition of training samples;
FIG. 4 is a graph of the inversion results of a feedback-type RBF neural network;
FIG. 5 is a diagram of the inversion result of a cloudy-day feedback type RBF neural network;
fig. 6 is a graph of inversion results of a feedback type RBF neural network in sunny days.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
As shown in fig. 1, an aerosol extinction coefficient inversion method based on a feedback RBF neural network includes the following steps:
The training sample comprises an input and an expected output, the historical echo signal power is used as the input of the RBF neural network, and the aerosol extinction coefficient obtained according to the historical echo signal is used as the expected output of the RBF neural network.
In order to ensure that the network finally obtains the extinction coefficient precision, the expected output should be as accurate as possible, so when the Fernald method is adopted to obtain the expected output, the aerosol extinction backscattering ratio with high precision needs to be determined, and the aerosol extinction coefficient with high precision is obtained.
In order to obtain the aerosol extinction backscattering ratio with high precision, a nonlinear equation can be constructed, and the aerosol extinction backscattering ratio is iteratively calculated by utilizing a frustum method, wherein the specific process comprises the following steps: a nonlinear equation is constructed by using a calculation formula for measuring the optical thickness of the aerosol by a sunlight photometer and a calculation formula for measuring the optical thickness of the aerosol by a radar (here, a laser radar), and the extinction backscattering ratio of the aerosol is iteratively calculated by using a chord section method.
The calculation formula for measuring the optical thickness of the aerosol by the sunlight meter is as follows:
wherein S isAODOptical thickness, S, of aerosol for detection by a solar photometerALLWhole-layer atmospheric optics for solar photometer detectionThickness, SMODFor effective detection of the optical thickness, r, of atmospheric molecules in the rangeaIs the effective detection distance, σm(r) is the extinction coefficient of atmospheric molecules, is a function of the probe distance r, and can be obtained according to the standard atmospheric molecule extinction mode in the United states.
The calculation formula of the optical thickness of the aerosol measured by the radar is as follows,
wherein L R is the optical thickness, σ, of the aerosol detected by radara(r) is the extinction coefficient of the aerosol, which is a ratio S of extinction backscattering of the aerosolaAnd a function of the detection distance r.
The two methods for measuring the optical thickness of the aerosol are independent from each other, so that joint inversion can be formed by the two methods, and a nonlinear equation about extinction backscattering ratio is constructed:
the aerosol extinction backscattering ratio is calculated by iteration through a truncation method, for example, as shown in fig. 2, two points are selected as an iteration initial value in a value interval, then an iteration sequence is generated according to an iteration formula of the truncation method, and judgment is performed according to an iteration stop condition, so that the aerosol extinction backscattering ratio is obtained finally.
In summary, the process of obtaining training samples as shown in fig. 3: a nonlinear equation is constructed based on the principle of the optical thickness of the aerosol, the extinction backscattering ratio of the aerosol is iteratively calculated by utilizing a truncation method, and the extinction coefficient of the aerosol is inverted by adopting a Fernald method according to the extinction backscattering ratio of the aerosol and an echo signal.
And 2, training the RBF neural network by using the input and the expected output.
Definition of X ═ (X)1,x2,…,xn)TFor the network input vector, Y ═ Y1,y2,…,ys)TIs output by the network, [ phi ]iIsThe radial basis function of the ith hidden layer node. The distribution function of the RBF neural network is:
where m is the number of hidden layer neuron nodes, i.e. the number of radial basis function centers, and the coefficient wiIs the connection weight.
Wherein φ (#) is a radial basis function, | | x-ciI is the Euclidean norm, ciIs the ith center of RBF, ξiFor the ith radius of the RBF, the available network outputs are:
thus, the matrix expression for an RBF network can be expressed as:
D＝HW+E
wherein the desired output vector is D ═ D (D)1,d2,…,dp)TThe error vector between the desired output and the network output is E ═ E (E)1,e2,…,ep)TWeight vector W ═ W1,w2,…,wm)TThe regression matrix H ═ H1,h2,…,hm)T。
Taking into account the influence of all training samples, ci、ξiAnd wiThe adjustment amounts of (a) and (b) are:
in the formula, phii(xj) For the ith implicit node pair xjη1，η2，η3Respectively corresponding learning rates, ci(t) and ci(t +1) c at the t-th and t + 1-th iterations, respectivelyi，ξi(t) and ξi(t +1) ξ at the t and t +1 iterations, respectivelyi，wi(t) and wi(t +1) w at the t-th and t + 1-th iterations, respectivelyi. And obtaining the mean square error according to the cost function E, thereby finishing the training condition. When the mean square error of the actual output and the expected output is less than the set threshold, the network is considered to be trained.
And 3, inputting the current echo signal into the trained RBF neural network to obtain the current aerosol extinction coefficient.
AECcorrected＝Networkoutput×(1+ω)
wherein AECcorrectedFor corrected aerosol extinction coefficient, NetworkoutputThe aerosol extinction coefficient before correction is represented by omega as an error;
and taking the corrected aerosol extinction coefficient and the corresponding echo signal as a new sample, and carrying out secondary training on the feedback neural network, so that the output of the feedback neural network gradually approaches to the measurement result of the instrument.
Fig. 4 is a graph of the inversion result of the feedback-type RBF neural network, and the network performance is detected by using a test sample, where the test sample also includes an input and an expected output, and it can be seen that the output after correction has higher consistency with the expected output. In order to adapt the network to inversion of different weathers, inversion tests of cloudy days and sunny days are carried out, as shown in fig. 5 and 6, high consistency is kept between the two weather conditions and expected output, and the feasibility of the method is verified.
According to the method, the feedback type RBF neural network is used for inverting the aerosol extinction coefficient, an internal mechanism between information is stored in the network through learning of a sample mode, uncertainty caused by a plurality of assumptions is effectively avoided, meanwhile, due to the addition of a dynamic feedback adjustment process approaching instrument measurement, the reliability of an inversion result is greatly increased, and quick and accurate inversion of the aerosol extinction coefficient is realized; meanwhile, the method combines the echo signal and the detection data of the sunshine photometer to construct an equation about the extinction backscattering ratio, a chord section method is used for solving, and the aerosol extinction coefficient is calculated according to the extinction backscattering ratio, so that errors caused by the extinction backscattering ratio are avoided to a certain extent, and the aerosol detection precision is improved.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (5)
1. An aerosol extinction coefficient inversion method based on a feedback type RBF neural network is characterized by comprising the following steps: comprises the steps of (a) preparing a mixture of a plurality of raw materials,
1) training the RBF neural network with the inputs and the desired outputs;
taking the power of the historical echo signal as the input of an RBF neural network, and taking an aerosol extinction coefficient obtained according to the historical echo signal as the expected output of the RBF neural network;
the process of obtaining the extinction coefficient of the aerosol comprises the following steps: constructing a nonlinear equation by using a calculation formula for measuring the optical thickness of the aerosol by using a sunshine photometer and a calculation formula for measuring the optical thickness of the aerosol by using a radar, iteratively calculating an aerosol extinction backscattering ratio by using a truncation method, and inverting the extinction coefficient of the aerosol by using a Fernald method according to the aerosol extinction backscattering ratio and an echo signal;
the non-linear equation is constructed by constructing,
wherein L R is the optical thickness, σ, of the aerosol detected by radara(r) is the extinction coefficient of the aerosol, which is a ratio S of extinction backscattering of the aerosolaAnd a function of the detection distance r, SAODOptical thickness, S, of aerosol for detection by a solar photometerALLFull layer atmospheric optical thickness, r, for solar photometer detectionaIs the effective detection distance, σm(r) is the extinction coefficient of atmospheric molecules;
2) and inverting the extinction coefficient of the aerosol based on a feedback type RBF neural network.
2. The feedback-type RBF neural network-based aerosol extinction coefficient inversion method of claim 1, wherein: the calculation formula for measuring the optical thickness of the aerosol by the sunlight meter is as follows,
wherein S isAODOptical thickness, S, of aerosol for detection by a solar photometerALLFull thickness of atmosphere, S, for solar photometer detectionMODFor effective detection of the optical thickness, r, of atmospheric molecules in the rangeaIs the effective detection distance, σm(r) is the extinction coefficient of atmospheric molecules.
3. The feedback-type RBF neural network-based aerosol extinction coefficient inversion method of claim 1, wherein: the calculation formula of the optical thickness of the aerosol measured by the radar is as follows,
wherein L R is the optical thickness, σ, of the aerosol detected by radara(r) is the extinction coefficient of the aerosol, which is a ratio S of extinction backscattering of the aerosolaAnd a function of the detection distance r.
4. The feedback-type RBF neural network-based aerosol extinction coefficient inversion method of claim 1, wherein: training the RBF neural network by using input and expected output, inputting the current echo signal into the trained RBF neural network to obtain the current aerosol extinction coefficient, and calculating the AOD (optical thickness of aerosol) according to the current aerosol extinction coefficientlidarThe current aerosol optical thickness AOD measured by a solar photometersunAnd comparing, if an error exists, correcting the current aerosol extinction coefficient, taking the corrected aerosol extinction coefficient and the corresponding echo signal as a new sample, and performing secondary training on the feedback RBF neural network.
5. The feedback-type RBF neural network-based aerosol extinction coefficient inversion method of claim 4, wherein: the formula for correcting the extinction coefficient of the aerosol is as follows,
AECcorrected＝Networkoutput×(1+ω)
wherein AECcorrectedFor corrected aerosol extinction coefficient, NetworkoutputFor the aerosol extinction coefficient before correction, ω is the error.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810397787.8A CN108896456B (en) | 2018-04-28 | 2018-04-28 | Aerosol extinction coefficient inversion method based on feedback type RBF neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810397787.8A CN108896456B (en) | 2018-04-28 | 2018-04-28 | Aerosol extinction coefficient inversion method based on feedback type RBF neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108896456A CN108896456A (en) | 2018-11-27 |
| CN108896456B true CN108896456B (en) | 2020-07-31 |
Family
ID=64342521
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810397787.8A Active CN108896456B (en) | 2018-04-28 | 2018-04-28 | Aerosol extinction coefficient inversion method based on feedback type RBF neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108896456B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110108672B (en) * | 2019-04-12 | 2021-07-06 | 南京信息工程大学 | Aerosol extinction coefficient inversion method based on deep belief network |
| CN110109149B (en) * | 2019-05-08 | 2022-11-25 | 南京信息工程大学 | Laser radar low-layer extinction coefficient profile correction method |
| CN113009437B (en) * | 2021-03-03 | 2023-10-03 | 云南大学 | Method for solving dynamic radar ratio of laser radar according to real-time aerosol distribution |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944503A (en) * | 2012-10-11 | 2013-02-27 | 中国科学院安徽光学精密机械研究所 | PM2.5 mass concentration value automatic inversion algorithm based on sun photometer and lidar |
| CN103175759A (en) * | 2013-02-25 | 2013-06-26 | 中国科学院安徽光学精密机械研究所 | Method for acquiring complex refractive index of urban aerosol on basis of various ground-based remote sensing technologies |
| CN103278479A (en) * | 2013-04-23 | 2013-09-04 | 中国科学院安徽光学精密机械研究所 | Atmospheric radiation transmission correction system and correction method |
| CN107144829A (en) * | 2017-06-29 | 2017-09-08 | 南京信息工程大学 | A kind of efficient laser radar echo signal antinoise method |
| CN107421917A (en) * | 2017-05-17 | 2017-12-01 | 南京信息工程大学 | A kind of multifunction high-precision atmosphere visibility meter and visibility measurement method |
| CN107728163A (en) * | 2017-09-05 | 2018-02-23 | 兰州大学 | Atmospheric Characteristics layer detection method and device |
| CN108845306A (en) * | 2018-07-05 | 2018-11-20 | 南京信息工程大学 | Laser radar echo signal antinoise method based on variation mode decomposition |
| CN109696412A (en) * | 2018-12-20 | 2019-04-30 | 南京信息工程大学 | Infrared gas sensor and atmospheric pressure compensating method based on AGNES Optimized BP Neural Network |
-
2018
- 2018-04-28 CN CN201810397787.8A patent/CN108896456B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102944503A (en) * | 2012-10-11 | 2013-02-27 | 中国科学院安徽光学精密机械研究所 | PM2.5 mass concentration value automatic inversion algorithm based on sun photometer and lidar |
| CN103175759A (en) * | 2013-02-25 | 2013-06-26 | 中国科学院安徽光学精密机械研究所 | Method for acquiring complex refractive index of urban aerosol on basis of various ground-based remote sensing technologies |
| CN103278479A (en) * | 2013-04-23 | 2013-09-04 | 中国科学院安徽光学精密机械研究所 | Atmospheric radiation transmission correction system and correction method |
| CN107421917A (en) * | 2017-05-17 | 2017-12-01 | 南京信息工程大学 | A kind of multifunction high-precision atmosphere visibility meter and visibility measurement method |
| CN107144829A (en) * | 2017-06-29 | 2017-09-08 | 南京信息工程大学 | A kind of efficient laser radar echo signal antinoise method |
| CN107728163A (en) * | 2017-09-05 | 2018-02-23 | 兰州大学 | Atmospheric Characteristics layer detection method and device |
| CN108845306A (en) * | 2018-07-05 | 2018-11-20 | 南京信息工程大学 | Laser radar echo signal antinoise method based on variation mode decomposition |
| CN109696412A (en) * | 2018-12-20 | 2019-04-30 | 南京信息工程大学 | Infrared gas sensor and atmospheric pressure compensating method based on AGNES Optimized BP Neural Network |
Non-Patent Citations (1)
| Title |
|---|
| 基于Madaline网络的气溶胶消光系数反演算法;韩道文等;《光学学报》;20070331;第27卷(第3期);第384-390页第1-6节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108896456A (en) | 2018-11-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fan et al. | Empirical and machine learning models for predicting daily global solar radiation from sunshine duration: A review and case study in China | |
| CN108896456B (en) | Aerosol extinction coefficient inversion method based on feedback type RBF neural network | |
| CN103593538A (en) | Fiber optic gyroscope temperature drift modeling method by optimizing dynamic recurrent neural network through genetic algorithm | |
| CN110908364B (en) | Fault detection method based on robust interval estimation | |
| CN110807554B (en) | Generation method and system based on wind power/photovoltaic classical scene set | |
| CN110441374B (en) | Dissolved oxygen sensor compensation method based on SA-BP neural network | |
| CN112232554A (en) | Construction method of local short-term rainfall forecast model based on BP neural network | |
| CN111366156A (en) | Transformer substation inspection robot navigation method and system based on neural network assistance | |
| CN113344149B (en) | PM2.5 hourly prediction method based on neural network | |
| CN110108672B (en) | Aerosol extinction coefficient inversion method based on deep belief network | |
| CN111896029A (en) | MEMS gyroscope random error compensation method based on combined algorithm | |
| CN114117886A (en) | Water depth inversion method for multispectral remote sensing | |
| CN111190211A (en) | GPS failure position prediction positioning method | |
| CN113468799A (en) | Method and system for acquiring near-ground PM2.5 concentration in static meteorological satellite observation | |
| Guermoui et al. | Daily global solar radiation modelling using multi-layer perceptron neural networks in semi-arid region | |
| CN107656905B (en) | Air quality data real-time calibration method using error transfer | |
| CN114239796A (en) | Power system state estimation method based on extended Kalman filtering | |
| CN108154273B (en) | Rainfall data estimation method and device based on neural network | |
| CN109101759A (en) | A kind of parameter identification method based on forward and reverse response phase method | |
| CN114757096A (en) | Bridge temperature prediction method, device, equipment and medium based on NARX neural network | |
| Zaiani et al. | New approach to establish a clear sky global solar irradiance model | |
| CN111044489B (en) | Method for obtaining atmosphere refractive index height distribution profile based on multi-wavelength measurement | |
| CN114001759A (en) | Array type MEMS sensor control method and system | |
| CN113077110A (en) | GRU-based harmonic residual segmented tide level prediction method | |
| Cui et al. | The atmospheric model of neural networks based on the improved Levenberg-Marquardt algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |