CN113189014B - Ozone concentration estimation method integrating satellite remote sensing and ground monitoring data - Google Patents
Ozone concentration estimation method integrating satellite remote sensing and ground monitoring data Download PDFInfo
- Publication number
- CN113189014B CN113189014B CN202110401303.4A CN202110401303A CN113189014B CN 113189014 B CN113189014 B CN 113189014B CN 202110401303 A CN202110401303 A CN 202110401303A CN 113189014 B CN113189014 B CN 113189014B
- Authority
- CN
- China
- Prior art keywords
- ozone concentration
- layer
- remote sensing
- concentration estimation
- basic model
- 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
Images
Classifications
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/89—Lidar systems specially adapted for specific applications for mapping or imaging
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/95—Lidar systems specially adapted for specific applications for meteorological use
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- G01N2021/1765—Method using an image detector and processing of image signal
-
- 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
- G01N2021/1793—Remote sensing
- G01N2021/1795—Atmospheric mapping of gases
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Radar, Positioning & Navigation (AREA)
- Electromagnetism (AREA)
- Remote Sensing (AREA)
- Software Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data, and belongs to the technical field of environmental monitoring. Comprising the following steps: step one, collecting, preprocessing and fusing multi-source sample data to obtain input parameters; step two, establishing an ozone concentration estimation operation basic model based on the multi-layer mapping back propagation neural network; and thirdly, searching an optimized input parameter combination of the obtained ozone concentration estimation operation basic model based on influence factors, forward tracing time and a spatial range, accurately estimating the ground ozone concentration according to the obtained optimized input parameter combination, obtaining a spatial continuous distribution condition of the ozone concentration, and realizing an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data. The method has the advantages of high accuracy, strong reliability and simple operation, and the used multi-source sample data are free and open-source, so that the universality is enhanced, and the ozone concentration can be rapidly estimated and a continuous distribution diagram of the ozone concentration in a target area can be drawn.
Description
Technical Field
The invention belongs to the technical field of environmental monitoring, and relates to an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data.
Background
With the acceleration of the urban and industrial processes, the problem of atmospheric pollution is becoming serious, ozone has become the primary pollutant affecting the quality of the ambient air and directly affecting the physical and mental health of human beings. Therefore, the rule of continuous distribution of ozone in time and space is monitored and revealed, and the method has important significance for preventing ozone pollution and preventing the harm to health.
At present, common ozone monitoring methods include ground monitoring and remote sensing monitoring. The ground monitoring is based on all-weather continuous observation by a monitoring station, so that the ozone concentration in the ground surface space and the accurate information of the ozone concentration along with the time change can be directly obtained. However, the construction cost of the monitoring stations is high, the quantity is limited, the distribution is uneven, and continuous and accurate ozone concentration monitoring in a large-scale space is difficult to realize. The research of monitoring ozone by using satellite remote sensing image data starts in the 80 s of the 20 th century and mainly comprises a mode scale factor method, a semi-experimental method based on a physical mechanism, a statistical model method and the like. In the past few decades, the above ozone estimation method, although having a wide range of applications, has the following problems: 1) The model structure and the simulation process are very complex, and the calculation cost is high; 2) The requirements on basic data are high, and a pollutant emission list often has large uncertainty, so that the estimation accuracy is limited; 3) The estimation result is greatly influenced by parameter setting, and the parameters are very complicated in calculation process and have obvious differences among different regions.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide the ozone concentration estimation method integrating satellite remote sensing and ground monitoring data, which has the advantages of high accuracy, high reliability and simple operation, and solves the problems of high uncertainty and complex calculation of a conclusion obtained by the existing ozone concentration estimation method.
In order to achieve the above purpose, the invention is realized by adopting the following technical scheme:
the invention discloses an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data, which comprises the following steps:
step one, collecting, preprocessing and fusing multi-source sample data to obtain input parameters;
step two, establishing an ozone concentration estimation operation basic model based on the multi-layer mapping back propagation neural network;
and thirdly, based on three dimensions of influencing factors, forward tracing time and a spatial range, combining the input parameters obtained in the first step, exploring an optimized input parameter combination of an ozone concentration estimation operation basic model obtained in the second step, estimating the ground ozone concentration according to the obtained optimized input parameter combination, obtaining a spatial continuous distribution result of the ozone concentration, and realizing an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data.
Preferably, in step one, the multi-source sample data includes: satellite remote sensing image data, air quality monitoring data and meteorological data.
Preferably, in the first step, the preprocessing of the multi-source sample data includes cloud removal processing and obtaining of normalized vegetation index NDVI: cloud layer identification and classification based on satellite remote sensing image data, a projection influence range of cloud layers on the ground and cloud layer coverage rate, and cloud removal processing of the remote sensing images according to different cloud layer characteristics; and carrying out orthographic correction and spatial position registration on the satellite remote sensing image data, and extracting the wave band reflectivity of each wave band in the satellite remote sensing image data to obtain a normalized vegetation index NDVI.
Preferably, in the first step, the fusion of the multi-source sample data includes the following operations: acquiring a meteorological site closest to an environment monitoring site through a proximity analysis algorithm, and taking meteorological data monitored by the meteorological site closest to the environment monitoring site as meteorological information of the environment monitoring site to realize fusion of multi-source sample data; wherein an index for traceability analysis is established for the multi-source sample data.
Preferably, in the second step, based on a multi-layer mapping back propagation neural network, an obtained ozone concentration estimation operation basic model is established, wherein the model comprises an input layer, an hidden layer and an output layer; the neurons in the same layer are not connected, and the neurons in each layer can receive signals of the neurons in the previous layer and generate signals to be output to the next layer.
Further preferably, one input layer, one output layer and L hidden layers are included, wherein L.gtoreq.1.
Further preferably, the node number of the hidden layer is obtained by: continuously training and comparing the multi-layer mapping back propagation neural network by gradually expanding the number of nodes in the hidden layer; and when the predicted result and the real result tend to be consistent, obtaining the node number of the hidden layer.
Preferably, in the second step, the input parameters of the obtained ozone concentration estimation operation basic model include: band reflectivity of different bands in satellite remote sensing image data, normalized vegetation index and meteorological data; the output data of the ozone concentration estimation operation basic model is the ozone concentration value of the monitoring station during remote sensing image imaging.
Preferably, in the third step, the method searches the optimized input parameter combination of the ozone concentration estimation operation basic model obtained in the second step based on the influence factors, and the method comprises the following operations:
firstly, adopting a statistical method to analyze the correlation between the input parameters obtained in the first step and the ozone concentration; then, classifying and grouping according to the intensity of the obtained correlation and the class characteristics of the input parameters obtained in the step one, and inputting each group of input parameter data obtained by grouping into the ozone concentration estimation operation basic model obtained in the step two for training and verification; and finally, determining the optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model obtained in the step two in the dimension of the influencing factors based on the determination coefficient, the average error and the root mean square error.
Preferably, in the third step, the method searches the optimized input parameter combination of the ozone concentration estimation operation basic model obtained in the second step based on the time tracing, and the method comprises the following operations: by adopting a cyclic exploration mode, t is as follows s The hour is the step length to gradually increase the range of the front tracing period; every increment of the trace-ahead period t s The input parameters corresponding to the time period are added to the input parameters of the ozone concentration estimation operation basic model obtained in the step two; based on the determination coefficient, the average error and the root mean square error, determining an optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model obtained in the step two in the forward tracing time dimension; wherein, 5 is greater than or equal to t s 1 or more; and when the current tracing period exceeds a preset threshold value and the estimation result of the ozone concentration estimation operation basic model obtained in the step two is kept unchanged or continuously worsened, the exploration is stopped.
Preferably, in the third step, the operation of exploring the optimized input parameter combination of the ozone concentration estimation operation basic model obtained in the second step based on the space scope includes: training and verifying an ozone concentration estimation operation basic model, and recording a decision coefficient, an average error and a root mean square error; gradually expanding a training and verifying area according to a step length set in a research area, and determining an optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model obtained in the second step in a space range dimension through a decision coefficient, an average error and a root mean square error; and when the training and verification area of the ozone concentration estimation operation basic model exceeds a preset threshold value, and the ozone concentration estimation result is kept unchanged or continuously deteriorated, the exploration is stopped.
Compared with the prior art, the invention has the following beneficial effects:
the invention discloses an ozone concentration estimation method for fusing satellite remote sensing and ground monitoring data, which can obtain input parameters through collection, pretreatment and fusion of multi-source sample data, can establish an ozone concentration estimation operation basic model through multi-layer mapping back propagation neural network and combining machine learning, and can explore the optimal combination of the input parameters through three dimensions based on influence factors, forward tracing time and space range so as to realize accurate estimation of ground ozone concentration. In addition, the optimizing process in three dimensions of influencing factors, forward tracing time and space range is not mutually separated and independently operated, but comprehensively considered, and the optimal input parameter combination can be more comprehensively and accurately found. Therefore, the ozone concentration estimation method based on machine learning has the advantages of high accuracy, strong reliability, simple operation, capability of estimating the ozone concentration, capability of indirectly obtaining the ozone concentration in the environment monitoring work and strong popularization and application value.
Furthermore, the multisource sample data are free and open-source, so that the universality of the model is enhanced, and the ozone concentration can be rapidly estimated and a continuous distribution diagram of the ozone concentration in a target area can be drawn.
Further, through cloud removal processing and obtaining of the normalized vegetation index NDVI, the influence of the cloud on the estimation result can be removed, and the accuracy of the model on estimating the ground ozone concentration is improved.
Furthermore, the fusion of the multi-source sample data is adopted, and the index of the traceability analysis is established, so that the related data (such as satellite remote sensing image data, meteorological data and air quality data) of different sources can be summarized and fused, and the ozone estimation model is convenient to use.
Furthermore, the multi-layer mapping back propagation neural network can be adapted to the very complex nonlinear relation between the distribution of the ground ozone concentration and a plurality of factors such as temperature, relative humidity, atmospheric pressure, wind speed, wind direction and the like through the establishment of the multi-layer mapping back propagation neural network, and has relatively great advantages in the process and the solution of the nonlinear mapping problem by the unique structure of the multi-layer mapping back propagation neural network.
Further, from the dimension of influencing factors, the optimized input parameter combination of the ozone concentration estimation operation basic model is determined, and the accuracy of ground ozone concentration estimation is improved.
Further, in the course of the ozone concentration estimation operation, the influence on the estimated value includes not only the time of satellite imaging (denoted as T 0 ) Meteorological parameters at the same time should also include the parameter consisting of T 0 The time of tracing to a certain time (denoted as T 1 ) During the period (i.e.: t (T) 0 -T 1 ) Is a meteorological condition of (a); therefore, the method well solves the problem based on the method of searching the forward tracing time, determines the optimal forward tracing time and improves the accuracy of estimating the ground ozone concentration. Wherein, setting the step length t by a cyclic searching mode s ，5≥t s 1 or more; if t s Not only is sample data difficult to acquire, but also the excessive data volume can complicate the prediction process; if t s And if the error is more than 5, the error of the optimal forward tracing period is increased, so that the accuracy of the model is greatly reduced.
Further, when there are fewer monitoring sites within the research area, but a certain number of ground monitoring sites are deployed around the research area, the optimal spatial range for model training and verification will most likely be larger than the research area; therefore, the method overcomes the influence of the space range on the estimation result through space range exploration, determines the optimal space range and improves the accuracy of estimating the ground ozone concentration.
Drawings
FIG. 1 is a schematic flow chart of an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data according to the invention;
FIG. 2 is a schematic diagram of an ozone concentration estimation operation basic model established by the multi-layer mapping back propagation neural network according to the embodiment of the invention;
FIG. 3 is a schematic diagram of an optimal spatial range for training, verifying and testing an ozone concentration estimation calculation basic model according to an embodiment of the present invention; wherein (a) the optimal spatial extent is equal to the investigation region; (b) the optimal spatial extent is greater than the investigation region;
fig. 4 is a fitting chart of correlation between estimated values and monitored values of ground ozone concentration in beijing city according to the embodiment of the invention;
FIG. 5 is a graph showing the ground ozone concentration estimated by Beijing city model and the monitored value of the monitoring station according to the embodiment of the present invention;
fig. 6 is a spatial distribution diagram of Beijing ozone concentration at different times according to an embodiment of the present invention: (a) 2018, 10/1 UTC 2:53; (b) month 3 of 2019, 26 days UTC 2:53.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The invention is described in further detail below with reference to the attached drawing figures:
referring to fig. 1, it can be known that the method for estimating the ozone concentration based on machine learning and integrating satellite remote sensing and ground monitoring data comprises the following steps:
step one, collecting, preprocessing and fusing multi-source sample data to obtain input parameters.
Wherein, the multisource sample data mainly comprises: satellite remote sensing image data (Landsat 8, MODIS, etc.), air quality monitoring data (O) 3 ) And meteorological data (wind speed, wind direction, humidity, air temperature, air pressure, etc.)
Specifically, in one embodiment of the present invention, satellite remote sensing image data is collected from Google Earth Engine (GEE), air quality monitoring data is collected from the chinese environmental monitoring headquarters (CNEMC), and weather data is collected from the united states national marine and atmospheric administration (NOAA).
Wherein the preprocessing of the multi-source sample data comprises: based on an API (application program interface) provided by Google Earth Engine (GEE) and other related open source programs, cloud layer identification and classification of satellite remote sensing image data in a research area, calculation of a projection influence range of cloud layers on the ground and cloud layer coverage rate are realized, and cloud removal processing is carried out on the remote sensing images by adopting corresponding algorithms according to different cloud layer characteristics; carrying out orthographic correction and spatial position registration on satellite remote sensing image data in a research area, extracting wave band reflectivity of each wave band in the remote sensing image, and calculating to obtain a normalized vegetation index NDVI according to the following formula:
ndvi= (NIR-R)/(nir+r); wherein, NIR is the reflection value of near infrared band, R is the reflection value of red light band.
Wherein, the fusion of the multisource sample data includes: and establishing an index of the sample data, so that the estimation result can be conveniently subjected to traceability analysis. And acquiring a meteorological site closest to the environment monitoring site through a proximity analysis algorithm, and taking the monitored meteorological data as meteorological information of the environment monitoring site to realize fusion of multi-source sample data.
And step two, establishing an ozone concentration estimation operation basic model based on the multi-layer mapping back propagation neural network.
The ozone concentration estimation operation basic model constructed by the invention consists of an input layer, L hidden layers (L is more than or equal to 1) and an output layer. Wherein the neurons in the same layer are not connected. The neurons of each layer may receive signals from neurons of a previous layer and generate signals for output to a next layer. When a set of sample data is provided to the ozone concentration estimation operation base model, an input signal is propagated backward from the input layer through the hidden layer by layer until the output layer; if the output layer cannot obtain the expected output result, the connection weight of the network is corrected from the output layer to the front layer by layer through each middle layer along the direction of reducing the error until the connection weight reaches the input layer; the forward computing process and the backward propagation process are repeatedly performed, and weights and thresholds of all layers are continuously adjusted, so that the predicted output of the backward propagation neural network is continuously approximate to the expected output. The ozone concentration estimation operation basic model adopts a tan sig function as a transfer function between different hidden layers, a purelin function as a transfer function between the last hidden layer and an output layer, and a trainlm function of a Levenberg-Marquardt (LM) algorithm is adopted for calculation in a network training process. For determining the number of nodes in the hidden layer of the neural network, the invention gradually expands the number of nodes in the hidden layer according to the Kolmogorov theorem and continuously trains and compares the network, thereby selecting the most suitable network structure. The input parameters of the ozone concentration estimation operation basic model mainly comprise: the output data of the ozone concentration estimation operation basic model is the ozone concentration value of the ground monitoring station when the remote sensing image is imaged.
Specifically, in a specific embodiment of the present invention, the ozone concentration estimation operation basic model constructed by the present invention is composed of one input layer, L hidden layers (l=2) and one output layer.
The invention selects the average error (ME), root Mean Square Error (RMSE) and the determination coefficient (R) 2 ) And comprehensively and objectively evaluating the ozone concentration estimation operation basic model.
And thirdly, searching an optimized input parameter combination of an ozone concentration estimation operation basic model from three different dimensions of influence factors, forward tracing time and space range, accurately estimating the ground ozone concentration according to the obtained optimized input parameter combination, obtaining a space continuous distribution result of the ozone concentration, rapidly estimating the ozone concentration and drawing a continuous distribution diagram of the ozone concentration in a target area, and realizing the ozone concentration estimation method based on machine learning.
Searching for an optimized combination of influencing factors, wherein the operation comprises the following steps: firstly, adopting a statistical method to analyze the correlation between the input parameters obtained in the first step and the ozone concentration; and then, classifying and grouping all possible input parameters according to the intensity of the correlation and the class characteristics of the input parameters obtained in the step one, wherein each group of input parameter data obtained by grouping is input into an ozone concentration estimation operation basic model for training and verification. By reacting R 2 And (3) comprehensively judging the ME and the RMSE, and determining the optimized combination of the input data of the ozone concentration estimation operation basic model in the dimension of the influencing factors.
Searching for an optimized combination of time tracing before searching, and the operation comprises the following steps: adopts a cyclic exploration mode, takes 1-5 hours as a step length (t s ) The range of the forward trace period is gradually increased. Every increment of the trace-ahead periodFor 1-5 hours, the input parameters corresponding to the period are added to the input parameters of the ozone concentration estimation calculation basic model, and R is calculated by 2 And (3) comprehensively judging the three parameters of ME and RMSE to determine the optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model in the forward time dimension. The termination of the discovery process can be controlled according to two principles: 1) The pre-tracing period is long enough, namely the pre-tracing period exceeds a preset threshold value; 2) The estimation result of the ozone concentration estimation operation basic model is kept unchanged or continuously worsened.
Specifically, in a specific embodiment of the present invention, the range of the forward trace period is gradually increased in steps of 3 hours. Searching for an optimal combination of spatial ranges: firstly, training and verifying an ozone concentration estimation operation basic model in a research area, and recording R of the model 2 ME and RMSE. The training and validation area is then expanded gradually until one or more new monitoring sites are present in the study area, by R 2 And comprehensively judging the advantages and disadvantages of the corresponding estimation results by using the three characteristic values of ME and RMSE, and determining the optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model in the space range dimension. The termination of the discovery process can be controlled according to two principles: 1) The training and verification area of the ozone concentration estimation operation basic model is large enough to exceed a preset threshold; 2) The ozone concentration estimation remains unchanged or continuously worsens.
The optimizing process in three dimensions of influencing factors, forward tracing time and space range is not mutually separated and independently operated, but is combined and comprehensively considered to search the optimized combination of the input parameters of the ozone concentration estimation operation basic model, so that the optimal estimation result under the existing sample data support is obtained.
The ozone concentration estimation operation basic model created based on the back propagation neural network is high in accuracy and reliability, can estimate the ozone concentration to a certain extent, and has strong application potential. Meanwhile, the data used in the model estimation process are free and open-source, so that the universality of the model is enhanced. In addition, the back propagation neural network model selected by the invention is very suitable for a processing method adopting distributed parallel calculation, which can greatly improve the operation efficiency, thereby enabling rapid estimation of continuous distribution of ozone concentration in space.
In order to illustrate the technical scheme of the invention, the following description uses Beijing as a specific embodiment.
Step one, collecting, preprocessing and fusing multi-source sample data to obtain input parameters.
In this embodiment, landsat 8 remote sensing image data of Beijing, from 5.month 1.year 2014 to 10.month 1.year 2019 is collected by Google Earth Engine (GEE, https:// eartenginine. Google. Com /); collecting ground ozone concentration data of Beijing city at the same time through a China environmental monitoring total station (CNEMC, http:// www.cnemc.cn /); meteorological data of the Beijing city at the same time was collected by the national ocean and atmosphere administration (NOAA, https:// gis. Ncdc. NOAA. Gov/maps/ncei/cdo/horly), including: wind speed, wind direction, humidity, air temperature, air pressure, etc.
Based on an API (application program interface) and other related open source programs provided by Google Earth Engine (GEE), cloud layer identification and classification of satellite remote sensing image data in a research area, calculation of the projection influence range of cloud layers on the ground and cloud layer coverage rate are realized, and cloud removal processing is carried out on the remote sensing images by adopting corresponding algorithms according to different cloud layer characteristics. Carrying out orthographic correction and spatial position registration on satellite remote sensing image data in a research area, extracting wave band reflectivity of each wave band in the remote sensing image, and calculating to obtain a normalized vegetation index NDVI according to the following formula:
NDVI＝(band5–band4)/(band5+band4)
wherein band5 is the reflectivity of the band5 of the Landsat 8 remote sensing image, and band4 is the reflectivity of the band4 of the Landsat 8 remote sensing image.
And an index of sample data is established, so that the ozone concentration estimation result can be conveniently subjected to traceability analysis. And extracting the band reflectivity and NDVI data in the 15-meter buffer zone of the square and round of the Chinese environment monitoring station, and giving the average value to the corresponding Chinese environment monitoring station to realize the fusion of Landsat 8 remote sensing image data and air quality data. And secondly, acquiring a weather monitoring site closest to the China environment monitoring site through a proximity analysis algorithm, and taking the monitored weather data as weather information of the China environment monitoring site.
And step two, establishing a basic model of ozone concentration estimation operation based on the multi-layer mapping back propagation neural network.
As shown in fig. 2, the ozone concentration estimation operation basic model constructed in this embodiment is composed of one input layer, two hidden layers and one output layer. Wherein the neurons in the same layer are not connected. The neurons of each layer may receive signals from neurons of a previous layer and generate signals for output to a next layer. When a set of sample data is provided to the model, the input signal propagates back from the input layer, layer by layer, through the hidden layer, to the output layer. If the output layer does not obtain the desired output result, the connection weights of the network are modified layer by layer from the output layer through the intermediate layers in the direction of error reduction until the input layer is reached. The forward calculation process and the backward propagation process are repeatedly performed, and the weight and the threshold value of each layer are continuously adjusted, so that the predicted output of the ozone concentration estimation operation basic model is continuously approximate to the expected output.
In FIG. 2, X 1 ，X 2 ，…，X m The input parameters of the ozone concentration estimation operation basic model mainly comprise band reflectivities of different bands in Landsat 8 remote sensing image data, NDVI and related meteorological data, such as: wind speed, wind direction, humidity, air temperature, air pressure, etc. Y is an estimated value of the model, namely the ozone concentration value of the ground monitoring station during Landsat 8 remote sensing image imaging.
Representing the weight on the connection from the jth neuron of layer l-1 to the ith neuron of layer l,/h>
Indicating the bias of the ith neuron at layer l,/->
Representing the activation value of the i-th neuron of layer i. The ozone concentration estimation operation basic model adopts a tan sig function as a transfer function between different hidden layers, a purelin function as a transfer function between the last hidden layer and an output layer, and a trainlm function of a Levenberg-Marquardt (LM) algorithm is adopted for calculation in a network training process.
In addition, for determining the number of nodes in the hidden layer of the neural network, the invention gradually expands the number of nodes in the hidden layer according to the Kolmogorov theorem, and continuously trains and compares the multi-layer mapping counter-propagation neural network, thereby selecting the most suitable network structure. And finally determining [15,15] as the optimal node number of the multi-layer mapping back propagation neural network in the embodiment.
In this embodiment, a set-aside method commonly used in machine learning model evaluation is adopted, and a training data set and a verification test data set are selected through a mode of multiple random sampling, wherein the training set accounts for 80% and the verification test set accounts for 20%. According to the determined neural network structure and transfer function, the maximum training frequency is set to 500, the network training precision is set to 0.001, and the learning rate is set to 0.1. Each set of experiments was repeated 300 times to average as the final result of the model evaluation. Finally, the average error (ME), the Root Mean Square Error (RMSE) and the decision coefficient (R) are selected 2 ) And comprehensively and objectively evaluating the ozone concentration estimation operation basic model.
And thirdly, searching an optimized input parameter combination of the ozone concentration estimation operation basic model from three different dimensions of influence factors, forward tracing time and space range.
Searching for an optimized combination of influencing factors: and adopting a statistical method to analyze the correlation between various data and ozone concentration. And classifying and grouping all possible input data according to the intensity of the correlation and the category characteristics of the data. In this embodiment, the 17 candidate input influencing factors are divided into the following three groups according to the intensity of the correlation with ozone and different data sources and characteristics, namely:
(1) The reflectivity of the wave band 1, the wave band 2 and the wave band 3 in Landsat 8OLI/TIRS have strong correlation with the concentration of ozone;
(2) Reflectivity of other bands, NDVI calculated from band4 and band5 of Landsat 8 OLI;
(3) Meteorological parameters, wind speed, wind direction, humidity, air temperature and air pressure.
Because the parameters in the group (1) have strong correlation with the ozone concentration, the reflectivities of the wave band 1, the wave band 2 and the wave band 3 are used as the basis of ozone estimation, so that the parameters are always adopted in the whole process to explore the optimal combination of the alternative input influencing factors. The three different parameters are input into the model step by step for training, learning and verification, and according to ME, RMSE and R 2 The training results generated by the different alternative input parameter combinations may be compared to achieve an optimized input parameter combination of the input parameters with the various influencing factors.
Search for optimized input parameter combinations of trace-ahead time: since the weather conditions vary significantly over time, both the weather parameters during satellite imaging and the weather conditions before can have a significant impact on the accuracy of the ozone concentration estimation. For a particular area of investigation (e.g., a city), the process of exploring the optimal forward trace time can be described by the following six steps:
the first step: by T 0 Representing satellite imaging time, T 1 Represents the time of tracing ahead, t s Representing a time step.
And a second step of: let T be 1 ＝T 0 And evaluating O of the ozone concentration estimation operation basic model 3 Estimating performance, denoted as P 0 。
And a third step of: since the meteorological parameters are collected by the ground monitoring station every 3 hours, the meteorological parameters are collected by the ground monitoring station every 3 hoursLet t be s ＝3h，T 1 ＝T 0 -n×t s Wherein n is an integer greater than zero.
Fourth step: will [ T ] 0 ,T 1 ]All meteorological parameters in the time period are input into an ozone concentration estimation operation basic model, and O of the ozone concentration estimation operation basic model is estimated 3 Estimating performance, recorded as
Fifth step: if the current performance
Is superior to the previous P 0 Is provided with->
n＝n+1。
Sixth step: repeating the third to fifth steps repeatedly until
Continuous ratio P 0 Difference, or [ T ] 0 ,T 1 ]The time period becomes sufficiently long, for example, greater than an empirically determined threshold.
[ T ] explored by the iterative procedure 0 ,T 1 ]The optimal time-to-trace will be trained as a multi-layer mapped back propagation neural network.
Searching for an optimal combination of spatial ranges: empirically, the optimal spatial extent of the multi-layer mapped back propagation neural network training is not necessarily the same as the minimum bounding rectangle of the region of interest. In practice, it is relevant to the distribution of monitoring stations, and may be larger than the minimum bounding rectangle of the investigation region, see fig. 3. For a particular region of investigation (e.g., a city), the process of exploring the optimal spatial range can be described by the following six steps:
the first step: by S 0 And
respectively represent the research areasThe minimum bounding rectangle of the domain and the spatial extent of the neural network training.
And a second step of: is provided with
And evaluating O of the ozone concentration estimation operation basic model 3 Estimating performance, denoted as P 0 。
And a third step of: gradually expanding in both longitudinal and latitudinal directions
Until one or more new monitoring stations are located +.>
And (3) inner part.
Fourth step: estimating O of ozone concentration estimation operation basic model 3 Estimating performance, recorded as
Fifth step: if the current performance
Is superior to the previous P 0 Is provided with->
Sixth step: repeating the third to fifth steps repeatedly until the area
Become large enough, e.g.>
Greater than an empirically determined threshold.
As found by the iterative process
Optimal spatial paradigm to be trained as neural networkAnd (5) enclosing.
In the embodiment, the optimizing process in three dimensions of influencing factors, forward tracing time and space range is not mutually separated and independently operated, but is combined and comprehensively considered to search the optimized combination of the input parameters of the ozone concentration estimation operation basic model, so that the optimal estimation result under the existing sample data support is obtained.
FIG. 4 shows the correlation between estimated and monitored values of ground ozone concentration in Beijing, wherein R 2 0.91 and ME 1.2. Mu.g/m 3 RMSE was 18.4. Mu.g/m 3 . The slope of the fitted line approaches 1 and the correlation is quite significant. Fig. 5 compares the estimated ground ozone concentration of the beijing city model with the monitored value of the monitoring station, wherein the red line represents the actual observed ozone concentration value and the blue line represents the estimated ozone concentration value. In most cases, the estimated data is substantially consistent with the trend of the monitored data. Therefore, the ozone concentration estimation operation basic model established by the invention has the capability of accurately estimating the ground ozone concentration, and is expected to become a new important means for monitoring the atmospheric pollution change and analyzing the area.
Fig. 6 (a) and 6 (b) show spatial distributions of ozone concentration in Beijing area at times UTC 2:53 at 10.1 and UTC 2:53 at 3.26 and 2019, respectively, with spatial resolution up to 30m. The ground ozone concentration at both moments shows a gradually decreasing trend from southwest to northwest, and is consistent with the features of high northwest, low southwest and Beijing as well as industrialization and population density.
The foregoing is merely illustrative of specific embodiments of this invention and is not intended to limit the scope of the invention, but it is to be construed as limited to only those equivalent changes and modifications which will occur to those skilled in the art without departing from the spirit and principles of the invention as defined in the appended claims.
Claims (6)
1. The ozone concentration estimation method integrating satellite remote sensing and ground monitoring data is characterized by comprising the following steps of:
step one, collecting, preprocessing and fusing multi-source sample data to obtain input parameters;
the multi-source sample data includes: satellite remote sensing image data, air quality monitoring data and meteorological data;
pretreatment of multi-source sample data, including cloud removal and obtaining of normalized vegetation index NDVI: cloud layer identification and classification based on satellite remote sensing image data, a projection influence range of cloud layers on the ground and cloud layer coverage rate, and cloud removal processing of the remote sensing images according to different cloud layer characteristics; carrying out orthographic correction and spatial position registration on satellite remote sensing image data, and extracting the wave band reflectivity of each wave band in the satellite remote sensing image data to obtain a normalized vegetation index NDVI;
fusion of multi-source sample data, comprising the operations of: acquiring a meteorological site closest to an environment monitoring site through a proximity analysis algorithm, and taking meteorological data monitored by the meteorological site closest to the environment monitoring site as meteorological information of the environment monitoring site to realize fusion of multi-source sample data;
the method comprises the steps of establishing an index for tracing analysis aiming at multi-source sample data;
the input parameters include: band reflectivities of different bands in satellite remote sensing image data, normalized vegetation index NDVI and related meteorological data; the meteorological data comprise wind speed, wind direction, humidity, air temperature and air pressure;
step two, establishing an ozone concentration estimation operation basic model based on the multi-layer mapping back propagation neural network;
thirdly, based on three dimensions of influencing factors, forward tracing time and a spatial range, combining the input parameters obtained in the first step, exploring an optimized input parameter combination of an ozone concentration estimation operation basic model obtained in the second step, estimating the ground ozone concentration according to the obtained optimized input parameter combination, obtaining a spatial continuous distribution result of the ozone concentration, and realizing an ozone concentration estimation method integrating satellite remote sensing and ground monitoring data;
in the third step, the optimal input parameter combination of the basic model is calculated based on the ozone concentration estimation obtained in the second step of space range exploration, and the operation comprises the following steps:
firstly, training and verifying an ozone concentration estimation operation basic model in a research area, and recording a decision coefficient, an average error and a root mean square error; then expanding the training and verifying area step by step according to the set step length until one or more new monitoring stations appear in the research area, comprehensively judging the quality of the corresponding estimation result by determining three characteristic values of the coefficient, the average error and the root mean square error, and determining the optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model in the space range dimension;
and when the training and verification area of the ozone concentration estimation operation basic model exceeds a preset threshold value, and the ozone concentration estimation result is kept unchanged or continuously deteriorated, the exploration is stopped.
2. The method for estimating the ozone concentration by fusing satellite remote sensing and ground monitoring data according to claim 1, wherein in the second step, an obtained ozone concentration estimation operation basic model is built based on a multi-layer mapping back propagation neural network, and the model comprises an input layer, an implicit layer and an output layer; the neurons in the same layer are not connected, and the neurons in each layer can receive signals of the neurons in the previous layer and generate signals to be output to the next layer.
3. The method for estimating the concentration of ozone by fusing satellite remote sensing and ground monitoring data according to claim 2, comprising an input layer, an output layer and L hidden layers, wherein L is more than or equal to 1.
4. The method for estimating the concentration of ozone by fusing satellite remote sensing and ground monitoring data according to claim 2, wherein the number of nodes of the hidden layer is obtained by: continuously training and comparing the multi-layer mapping back propagation neural network by gradually expanding the number of nodes in the hidden layer; and when the predicted result and the real result tend to be consistent, obtaining the node number of the hidden layer.
5. The method for estimating ozone concentration by fusing satellite remote sensing and ground monitoring data according to claim 1, wherein in the third step, the optimum input parameter combination of the ozone concentration estimation operation basic model obtained in the second step is explored based on influence factors, and the operation comprises the following steps:
firstly, adopting a statistical method to analyze the correlation between the input parameters obtained in the first step and the ozone concentration;
then, classifying and grouping according to the intensity of the obtained correlation and the class characteristics of the input parameters obtained in the step one, and inputting each group of input parameter data obtained by grouping into the ozone concentration estimation operation basic model obtained in the step two for training and verification;
and finally, determining the optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model obtained in the step two in the dimension of the influencing factors based on the determination coefficient, the average error and the root mean square error.
6. The method for estimating ozone concentration by fusing satellite remote sensing and ground monitoring data according to claim 1, wherein in the third step, the operation of exploring the optimized input parameter combination of the basic model for estimating and calculating ozone concentration obtained in the second step based on the forward tracing time comprises the following steps:
by adopting a cyclic exploration mode, t is as follows s The hour is the step length to gradually increase the range of the front tracing period; every increment of the trace-ahead period t s The input parameters corresponding to the time period are added to the input parameters of the ozone concentration estimation operation basic model obtained in the step two; based on the determination coefficient, the average error and the root mean square error, determining an optimized input parameter combination of the input parameters of the ozone concentration estimation operation basic model obtained in the step two in the forward tracing time dimension;
wherein, 5 is greater than or equal to t s ≥1；
And when the current tracing period exceeds a preset threshold value and the estimation result of the ozone concentration estimation operation basic model obtained in the step two is kept unchanged or continuously worsened, the exploration is stopped.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110401303.4A CN113189014B (en) | 2021-04-14 | 2021-04-14 | Ozone concentration estimation method integrating satellite remote sensing and ground monitoring data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110401303.4A CN113189014B (en) | 2021-04-14 | 2021-04-14 | Ozone concentration estimation method integrating satellite remote sensing and ground monitoring data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113189014A CN113189014A (en) | 2021-07-30 |
| CN113189014B true CN113189014B (en) | 2023-05-02 |
Family
ID=76975794
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110401303.4A Active CN113189014B (en) | 2021-04-14 | 2021-04-14 | Ozone concentration estimation method integrating satellite remote sensing and ground monitoring data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113189014B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113821987B (en) * | 2021-11-19 | 2022-03-25 | 浙江甲骨文超级码科技股份有限公司 | Complex terrain meteorological data prediction method, system and device |
| CN115855761B (en) * | 2022-11-03 | 2023-09-26 | 生态环境部卫星环境应用中心 | Five-base collaborative sky-ground integrated atmospheric environment three-dimensional remote sensing monitoring system and method |
| CN115825338A (en) * | 2022-11-03 | 2023-03-21 | 浙江时空智子大数据有限公司 | Ozone monitoring downscaling method, device, equipment and computer readable storage medium |
| CN116486278B (en) * | 2023-04-24 | 2023-11-21 | 河南大学 | Hour-level ozone estimation method based on space-time information mosaic |
| CN117216480B (en) * | 2023-09-18 | 2024-06-28 | 宁波大学 | Near-surface ozone remote sensing estimation method for deep coupling geographic space-time information |
| CN117592005B (en) * | 2024-01-19 | 2024-04-26 | 中国科学院空天信息创新研究院 | PM2.5 concentration satellite remote sensing estimation method, device, equipment and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909722A (en) * | 2017-02-10 | 2017-06-30 | 广西壮族自治区气象减灾研究所 | A kind of large area accurate inversion method of temperature near the ground |
| CN110595968A (en) * | 2019-08-02 | 2019-12-20 | 中国科学院遥感与数字地球研究所 | PM2.5 concentration estimation method based on geostationary orbit satellite |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102011081197A1 (en) * | 2011-08-18 | 2013-02-21 | Siemens Aktiengesellschaft | Method for the computer-aided modeling of a technical system |
| CN110595960B (en) * | 2019-08-02 | 2021-05-14 | 中国科学院遥感与数字地球研究所 | PM2.5 concentration remote sensing estimation method based on machine learning |
| CN111310386B (en) * | 2020-02-13 | 2023-04-21 | 北京中科锐景科技有限公司 | Near-ground ozone concentration estimation method |
| AU2020101095A4 (en) * | 2020-06-23 | 2020-07-23 | Tarim University | A Satellite Remote Sensing Monitoring Method for the Content of Chlorophyll A in the Canopy of Jujube Trees |
| CN112052627A (en) * | 2020-08-21 | 2020-12-08 | 海南星瞰信息咨询中心(有限合伙) | Method, device, medium and equipment for estimating near-surface ozone space distribution |
| CN112163375B (en) * | 2020-09-28 | 2024-05-10 | 中国科学院空天信息创新研究院 | Long-time sequence near-ground ozone inversion method based on neural network |
-
2021
- 2021-04-14 CN CN202110401303.4A patent/CN113189014B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909722A (en) * | 2017-02-10 | 2017-06-30 | 广西壮族自治区气象减灾研究所 | A kind of large area accurate inversion method of temperature near the ground |
| CN110595968A (en) * | 2019-08-02 | 2019-12-20 | 中国科学院遥感与数字地球研究所 | PM2.5 concentration estimation method based on geostationary orbit satellite |
Non-Patent Citations (1)
| Title |
|---|
| Sarah Safieddine等.Tropospheric ozone variability during the East Asian summer monsoon as observed by satellite (IASI), aircraft (MOZAIC) and ground stations.《Atmospheric Chemistry and Physics》.2016,10489-10500. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113189014A (en) | 2021-07-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113189014B (en) | Ozone concentration estimation method integrating satellite remote sensing and ground monitoring data | |
| CN111639747B (en) | GNSS-R sea surface wind speed inversion method and system based on BP neural network | |
| CN110174359B (en) | Aviation hyperspectral image soil heavy metal concentration assessment method based on Gaussian process regression | |
| CN110865040A (en) | Sky-ground integrated hyperspectral water quality monitoring and analyzing method | |
| CN111639746B (en) | GNSS-R sea surface wind speed inversion method and system based on CNN neural network | |
| CN109782274A (en) | A kind of Moisture Damage recognition methods based on Gpr Signal time-frequency statistical nature | |
| CN107230197B (en) | Tropical cyclone objective strength determination method based on satellite cloud image and RVM | |
| CN110595968B (en) | PM2.5 concentration estimation method based on geostationary orbit satellite | |
| CN114460013B (en) | Coastal wetland vegetation overground biomass GAN model self-learning remote sensing inversion method | |
| CN105787184A (en) | Atmospheric aerosol optical depth estimation method based on PM2.5 | |
| CN116879297B (en) | Soil moisture collaborative inversion method, device, equipment and medium | |
| CN115690632A (en) | Water environment monitoring method for inland river water body | |
| CN110110025B (en) | Regional population density simulation method based on feature vector space filtering value | |
| Barkataki et al. | Classification of soil types from GPR B scans using deep learning techniques | |
| CN110321528B (en) | Hyperspectral image soil heavy metal concentration assessment method based on semi-supervised geospatial regression analysis | |
| Machap et al. | IoT audio sensor networks and decision trees for enhanced rain sound classification | |
| CN114254692B (en) | Multi-scale intelligent thunderstorm classifying and identifying method based on multi-source lightning data | |
| CN114371519B (en) | Foundation microwave radiometer atmospheric temperature and humidity profile inversion method based on unbiased bright temperature | |
| CN118396482B (en) | Climate change-oriented dynamic monitoring method for cultivated quality | |
| CN114972984A (en) | Accumulated snow time-space analysis and prediction method based on random forest | |
| CN114005027A (en) | Urban single tree detection system and method based on unmanned aerial vehicle image | |
| CN116466368B (en) | Dust extinction coefficient profile estimation method based on laser radar and satellite data | |
| CN115840908B (en) | Method for constructing PM2.5 three-dimensional dynamic monitoring field based on microwave link of LSTM model | |
| CN117456342A (en) | Traffic accident risk assessment method and system based on street view image | |
| Qian et al. | Deep Learning-Augmented Stand-off Radar Scheme for Rapidly Detecting Tree Defects |
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 |