Airport dominant visibility measurement method and system based on multi-source data fusion

Abstract

The invention belongs to the technical field of airport meteorological monitoring, and relates to an airport dominant visibility measurement method and system based on multi-source data fusion, and the method comprises the steps: 1) employing the backscattering coefficient, PM2.5 concentration and relative humidity of a laser radar as feature vectors, and carrying out the weather classification through a support vector machine classifier; 2) under different weather types, using a laser radar to obtain echo signals of the whole airport area, and using a point-type visibility meter to obtain point-type visibility; 3) preprocessing the echo signal and obtaining laser visibility based on the echo signal; 4) performing dynamic weight iterative optimization on the point-mode visibility, the laser visibility and the near visibility under different weather types through a dynamic weight multilayer sensor to obtain final visibility; and 5) performing dominant visibility judgment based on the final visibility, and determining the dominant visibility. Through multi-source data fusion and an intelligent algorithm, the problems of insufficient precision, poor environmental adaptability and low automation degree in the prior art can be solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of aviation meteorological monitoring, and relates to an airport dominant visibility measurement method and system, in particular to an airport dominant visibility measurement method and system based on multi-source data fusion. BACKGROUND

[0002] In aviation meteorological guarantee, the dominant visibility is a core parameter for the safe take-off and landing of an aircraft. The current dominant visibility measurement technology mainly faces the following challenges:

[0003] (1) Limitations of manual observation: dependent on the experience of observers, significantly affected by subjective factors such as light and vision, data update is lagging (usually once per hour), unable to capture the dynamic changes of low-visibility weather (such as advection fog, heavy rain) in real time, and high labor cost.

[0004] (2) Serious defects of traditional instruments: forward scatter meters, transmission meters and other devices only provide single-point data, which cannot reflect the non-uniformity of the atmosphere in a large range of the airport. For example, when the rainfall or aerosol distribution is uneven, the single-point data may deviate from the actual value by up to 49%. The traditional laser radar has incomplete signals in the near-field blind area (≤60m), and is significantly affected by weather. For example, under the condition of rainfall, the measurement error of the laser radar can reach -18.88%, which needs to be manually assisted to correct.

[0005] (3) Fragmentation of multi-source data: the existing system lacks deep fusion of laser radar data and meteorological elements (such as PM2.5, humidity), cannot achieve multi-dimensional cross verification, and is difficult to distinguish the differentiated influence of fog, haze, rain and other weather on visibility.

[0006] Therefore, in view of the defects in the prior art, it is necessary to develop a new airport dominant visibility measurement method and system. SUMMARY

[0007] In order to overcome the defects of the prior art, the application provides an airport dominant visibility measurement method and system based on multi-source data fusion, which can solve the problems of insufficient precision, poor environmental adaptability and low automation level in traditional technology through multi-source data fusion and intelligent algorithm.

[0008] In order to achieve the above purpose, the application provides the following technical scheme:

[0009] An airport dominant visibility measurement method based on multi-source data fusion, characterized in that it comprises the following steps:

[0010] 1) The backscattering coefficient of the laser radar, the PM2.5 concentration and the relative humidity are taken as feature vectors, and a support vector machine classifier is used for weather classification to obtain different weather types;

[0011] 2) Under different weather conditions, use lidar to obtain echo signals of the entire airport area and use point visibility meters to obtain point visibility;

[0012] 3) Preprocess the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient and obtain the laser visibility based on the preprocessed extinction coefficient;

[0013] 4) Dynamic weighted multilayer perceptrons are used to dynamically iterate and optimize point visibility, laser visibility, and near-visibility under different weather types to obtain the final visibility.

[0014] 5) Based on the final visibility, determine the dominant visibility to ascertain the dominant visibility.

[0015] Preferably, step 3) involves preprocessing the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient, specifically including:

[0016] 31) Geometric overlap factor self-calibration: An adaptive piecewise fitting algorithm is used to process the echo signal and outliers in the echo signal are removed by box plot;

[0017] 32) Geometric factor fitting: Fit the effective region of the echo signal using the least squares method and remove echo signals that are not within the effective region;

[0018] 33) Perform wavelength conversion: Use a solar photometer or particle spectrometer to obtain the wavelength index α of the lidar, and convert the wavelength λ of the lidar into the equivalent extinction coefficient α of the 550nm standard wavelength. lidar : α(550nm) is the extinction coefficient at the standard wavelength of 550nm.

[0019] Preferably, when the weather type is rain, step 3) of preprocessing the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient further includes:

[0020] 34) Calculate the rainfall intensity R = 0.01Z using the reflectivity factor Z of millimeter-wave radar. 0.714 The extinction coefficient α of the lidar under rainfall conditions is calculated based on the rainfall intensity. rain α rain =α lidar ·(1+k R ·R), k R This is the rainfall attenuation coefficient.

[0021] Preferably, the dynamic weight iterative optimization in step 4) to obtain the final visibility specifically involves: Among them, V finalis the final visibility, V i is the i-th visibility among the point visibility, the laser visibility and the proximate visibility under different weather types, n is the total number of the point visibility, the laser visibility and the proximate visibility under different weather types, b is an offset constant, w i is the weight of the i-th visibility and is the dynamic weight output by the i-th layer of the dynamic weight multi-layer perceptron, w i = w i-1 + Δw i · η, wherein w i-1 is the dynamic weight output by the i-1-th layer of the dynamic weight multi-layer perceptron, Δw i is the weight adjustment amount of the i-1-th layer of the dynamic weight multi-layer perceptron, and η is a learning rate.

[0022] Preferably, the step 5) is specifically: when the final visibility V final > 1500 m, the dominant visibility is V final ; when the final visibility V final ≤ 1500 m, the dominant visibility is max(V final , V manualthreshold ), wherein V manualthreshold is a visibility threshold of artificial observation.

[0023] Preferably, the method further comprises:

[0024] 6) acquiring echo signals of the entire area of the airport in real time by using the laser radar, and pre-processing the echo signals acquired by the laser radar to obtain pre-processed radar extinction coefficients; acquiring point extinction coefficients in real time by using the point visibility meter, and fusing the pre-processed radar extinction coefficients and the point extinction coefficients to obtain fused extinction coefficients.

[0025] Preferably, the method further comprises:

[0026] 7) acquiring echo signals of the entire area of the airport within a period of time before the current time by using the laser radar, and pre-processing the echo signals acquired by the laser radar to obtain a pre-processed radar extinction coefficient series; predicting future extinction coefficients α pred based on the pre-processed radar extinction coefficient series by using a recurrent neural network, and obtaining a future dominant visibility V: pred based on the predicted future extinction coefficients α

[0027] In addition, the application also provides an airport dominant visibility measurement system based on multi-source data fusion, characterized by comprising:

[0028] a weather classification module, configured to classify weather by a support vector machine classifier using a backscattering coefficient of a laser radar, a PM2.5 concentration and a relative humidity as feature vectors to obtain different weather types;

[0029] a data acquisition module, configured to acquire echo signals of a full area of an airport and point visibility by a point visibility meter under different weather types by a laser radar;

[0030] an echo signal preprocessing module, configured to preprocess the echo signals acquired by the laser radar to obtain a preprocessed extinction coefficient and laser visibility based on the preprocessed extinction coefficient;

[0031] a final visibility calculation module, configured to perform dynamic weight iterative optimization on the point visibility, the laser visibility and the near-approximate visibility under different weather types by a dynamic weight multi-layer perceptron to obtain final visibility;

[0032] a dominant visibility judgment module, configured to perform dominant visibility judgment based on the final visibility to determine dominant visibility.

[0033] Furthermore, the application also provides an airport dominant visibility measurement device based on multi-source data fusion, characterized by comprising:

[0034] one or more processors;

[0035] a memory configured to store one or more programs;

[0036] when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the airport dominant visibility measurement method based on multi-source data fusion as described above.

[0037] Finally, the application also provides a computer readable storage medium having a computer program stored thereon, characterized by that the program is executed by a processor to implement the steps of the airport dominant visibility measurement method based on multi-source data fusion as described above.

[0038] Compared with the prior art, the airport dominant visibility measurement method and system based on multi-source data fusion has one or more of the following beneficial technical effects:

[0039] 1、The application can solve the problems of insufficient precision, poor environmental adaptability and low automation in traditional airport dominant visibility measurement technology by multi-source data fusion and intelligent algorithms.

[0040] 2、The application adopts advanced extinction coefficient inversion algorithm and multi-source data fusion technology, can continuously output high-precision visibility data under complex weather conditions, and effectively makes up for the short board of artificial and point visibility instrument timeliness and continuity. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 is a flow chart of the airport dominant visibility measurement method based on multi-source data fusion of the application;

[0042] Figure 2 is a structural schematic diagram of the airport dominant visibility measurement system based on multi-source data fusion of the application. DETAILED DESCRIPTION

[0043] Before any embodiments of the application are explained in detail, it is to be understood that the application is not limited in its application to the details of construction and the arrangement of components set forth in the following description or illustrated in the following drawings. The application is capable of other embodiments and of being practiced or of being carried out in various ways. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of "including," "comprising," or "having" and variations thereof herein is meant to encompass the items listed thereafter and equivalents thereof as well as additional items. Unless specified or limited otherwise, the terms "mounted," "connected," "supported," and "coupled" and variations thereof are used broadly and encompass both direct and indirect mountings, connections, supports, and couplings. Further, "connected" and "coupled" are not restricted to physical or mechanical connections or couplings.

[0044] Also, in the disclosure of the application, the terms "longitudinal", "transverse", "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like indicate the orientation or positional relationship shown in the drawings, which are only for the convenience of describing the application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the above terms cannot be understood as limiting the application; secondly, the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of one element can be one, and in another embodiment, the number of the element can be multiple, the term "one" cannot be understood as limiting the number.

[0045] In view of the problems existing in the existing airport dominant visibility measurement, the application proposes a high-precision and high-robustness airport dominant visibility measurement method, which can solve the problems of insufficient precision, poor environmental adaptability and low automation degree in traditional airport dominant visibility measurement technology through multi-source data fusion and intelligent algorithm.

[0046] Figure 1 A flow chart of the airport dominant visibility measurement method based on multi-source data fusion of the present application is shown. As shown in the figure, the airport dominant visibility measurement method based on multi-source data fusion of the present application comprises the following steps: Figure 1

[0047] I. Weather classification

[0048] The backscattering coefficient of the laser radar, the PM2.5 concentration and the relative humidity are taken as the feature vectors, and the weather classification is performed by the support vector machine classifier to obtain different weather types.

[0049] In the present application, a laser radar, a multi-point point-type visibility meter, a meteorological element sensor and a millimeter wave radar can be deployed at the airport. Among them, the laser radar can adopt a three-dimensional scanning visibility laser radar (such as 1064nm wavelength, scanning range 0.06-15km, spatial resolution ≤15m) for obtaining the echo signal of the whole area of the airport and obtaining the three-dimensional distribution of the atmospheric extinction coefficient based on the echo signal. The multi-point point-type visibility meter is arranged at both ends and the middle of the runway (such as a combination of a transmission meter and a forward scattering meter) to obtain the point-type visibility and the point-type extinction coefficient, thereby providing local calibration data. The meteorological element sensor collects meteorological data such as PM2.5 and relative humidity in real time for environmental feature analysis. The millimeter wave radar is used to provide backscattering data in thick fog and rainfall scenarios to improve the signal reliability in complex weather.

[0050] Moreover, in the present application, a weather classification support vector machine (SVM) is constructed, i.e. a support vector machine classifier. Among them, the backscattering coefficient β of the laser radar, the PM2.5 concentration C and the relative humidity RH are extracted as the feature vectors [β, C, RH], the SVM classifier is trained by the radial basis kernel function, the weather types such as fog (category 1), haze (category 2) and rainfall (category 3) are distinguished, and the classification accuracy is ≥90%.

[0051] II. Data acquisition

[0052] In the present application, the echo signal of the whole area of the airport is obtained by the laser radar and the point-type visibility is obtained by the point-type visibility meter under different weather types (fog, haze, rainfall). Thus, it is convenient to perform weighted value fusion on the laser radar data and the point-type visibility meter data under different weather types.

[0053] III. Echo signal preprocessing

[0054] After obtaining the echo signal of the laser radar, the echo signal obtained by the laser radar can be preprocessed to obtain a preprocessed extinction coefficient and obtain the laser visibility based on the preprocessed extinction coefficient. In the present application, the echo signal preprocessing specifically comprises: ​

[0055] 1. Geometric overlap factor self-calibration.

[0056] For the problem of incomplete signal in the near-field transition zone of the laser radar, the adaptive piecewise fitting algorithm is used for processing.

[0057] Firstly, the atmospheric uniformity is determined: the distance square correction is performed on the echo signal P(z) to obtain S(z0=P(z)·z 2 , and Y(z)=ln(S(z)) is obtained after taking the logarithm. Wherein, z is the one-way distance of laser from the emission end, through atmospheric propagation, to the position of a voxel (i.e. a single sampling point) in the atmosphere, and then back to the receiving end after scattering.

[0058] Then, the first-order difference SA=Y(z+1-Y(z) is calculated, and the abnormal value is removed through the box plot. The specific operation of removing the abnormal value through the box plot is: the calculated first-order difference data is sorted from small to large, and the first quartile Q1, the third quartile Q3 and the interquartile range IQR are calculated. The first quartile Q1 refers to the data value at the 25% position after the first-order difference data is arranged from small to large. The third quartile Q3 refers to the data value at the 75% position after the first-order difference data is arranged from small to large. The interquartile range IQR refers to the range of the middle 50% of the data after the first-order difference data is arranged from small to large. Based on the first quartile Q1, the third quartile Q3 and the interquartile range IQR, the upper and lower limits of the abnormal value are set. The lower limit is Q1-1.5×IQR, and the upper limit is Q3+1.5×IQR. Finally, the data less than the lower limit or greater than the upper limit is regarded as an abnormal value and removed.

[0059] At the same time, if |SA| is less than the empirical threshold (such as 0.1), it is determined that the atmosphere is uniform, otherwise it is non-uniform. The echo signal under the non-uniform atmosphere is also deleted.

[0060] 2. Geometric factor fitting.

[0061] Under the uniform atmosphere, the radar equation of the laser radar is simplified as Y(z)=kz+b (k is the extinction coefficient slope, and b is the offset), and the fitting curve F(z)=kz+b is obtained by fitting the effective area (SNR>1) through the least square method. The geometric overlap factor OL(z)=S(z) / exp(F(z)). That is, the sequence echo signal is fitted into multiple radar equations, and only the effective area is retained by fitting the effective area of the multiple radar equations through the least square method, and the others are removed.

[0062] 3. Perform wavelength conversion algorithm.

[0063] The extinction coefficient α of 550 nm standard wavelength is converted from the wavelength λ (1064 nm) of the laser radar to the equivalent extinction coefficient α of 550 nm standard wavelength by using a sun photometer or a particle spectrometer lidar : α(550nm) is the extinction coefficient of 550 nm standard wavelength.

[0064] 4. Rainfall compensation.

[0065] In the present application, the laser radar is compensated for rainfall attenuation when the weather is classified as rainfall. Specifically, the rainfall intensity R = 0.01Z is calculated using the reflectivity factor Z of the millimeter wave radar 0.714 , and the extinction coefficient α of the laser radar in the case of rainfall is calculated based on the rainfall intensity rain : α rain = α lidar ·(1+k R ·R). Wherein, k R is the rainfall attenuation coefficient, which can be calibrated by actual measurement.

[0066] Four, final visibility calculation.

[0067] The point visibility, laser visibility and near-approximate visibility under different weather types are dynamically weighted and iteratively optimized by a dynamic weight multi-layer perceptron (DW-MLP) to obtain the final visibility.

[0068] In the present application, the final visibility wherein, V final is the final visibility, V i is the i-th visibility among the point visibility, laser visibility and near-approximate visibility under different weather types, n is the total number of the point visibility, laser visibility and near-approximate visibility under different weather types, b is the offset constant, and w i is the weight of the i-th visibility.

[0069] The w i is the dynamic weight output by the i-th layer of the dynamic weight multi-layer perceptron. Wherein, w i = w i-1 + Δw i · η, wherein, w i-1 is the dynamic weight output by the i-1-th layer of the dynamic weight multi-layer perceptron, Δw i is the weight adjustment amount of the i-1-th layer of the dynamic weight multi-layer perceptron, and η is the learning rate.

[0070] At the same time, in the present application, the near-approximate visibility V approx = -3L / lnt (L is the scene depth, and t is the transmittance).

[0071] V. Dominant visibility judgment.

[0072] Based on the final visibility, a dominant visibility judgment is made to determine the dominant visibility.

[0073] When the final visibility V final >1500m, the dominant visibility is V final ; when the final visibility V final ≤1500m, the dominant visibility is max(V final ,V manualthreshold ), wherein V manualthreshold is a manually observed visibility threshold.

[0074] Thus, the present application can solve the problems of insufficient precision, poor environmental adaptability and low automation level in traditional dominant visibility measurement technology through multi-source data fusion and intelligent algorithms.

[0075] Of course, in the present application, spatio-temporal weighted fusion can also be performed on the laser radar and point visibility meter to perform real-time measurement of visibility based on the weighted fusion result.

[0076] Specifically, the echo signals of the entire airport area are acquired in real time by the laser radar, and the echo signals acquired by the laser radar are preprocessed by the echo signal processing method of step three to obtain preprocessed radar extinction coefficients. At the same time, the point extinction coefficient is acquired in real time by the point visibility meter. Then, the preprocessed radar extinction coefficients and the point extinction coefficient are fused to obtain a fused extinction coefficient.

[0077] In the present application, the method of spatio-temporal weighted fusion is used for fusion. That is, the atmospheric uniformity index U = σ / μ is defined (σ is the standard deviation of the point data, and μ is the mean value); the weights of the laser radar and the point data are dynamically adjusted: The fused extinction coefficient is α fusion =w lidar ·α lidar +w point ·α point . Wherein w lidar and w point are the weight coefficients of the laser radar and the weight coefficients of the point visibility meter, respectively, and α lidar and α point are the extinction coefficients of the laser radar and the point visibility meter, respectively. With the fused extinction coefficient, the visibility can be converted.

[0078] In addition, in the present application, the future dominant visibility can also be predicted based on the echo signal of the laser radar. Specifically, the echo signal of the entire area of the airport within a period of time before the current time (for example, 10 minutes before the current time) is obtained by using the laser radar, and the echo signal obtained by the laser radar is preprocessed by using the echo signal processing method described in step three to obtain a preprocessed radar extinction coefficient series ([a1, a2, …, a 60 ], wherein 6 extinction coefficients are obtained within each minute). Based on the preprocessed radar extinction coefficient series, the future extinction coefficient a pred is predicted by a recurrent neural network, and the future dominant visibility V is obtained based on the predicted future extinction coefficient a pred . Thus, the present application can predict the future dominant visibility.

[0079] Figure 2 The present application is shown in the schematic diagram of the airport dominant visibility measurement system based on multi-source data fusion. As shown in the figure, Figure 2 the airport dominant visibility measurement system based on multi-source data fusion of the present application comprises:

[0080] 1. Weather classification module.

[0081] The weather classification module is used to take the backscattering coefficient of the laser radar, the PM2.5 concentration and the relative humidity as the feature vector, and perform weather classification by using the support vector machine classifier to obtain different weather types.

[0082] 2. Data acquisition module.

[0083] The data acquisition module is used to obtain the echo signal of the entire area of the airport by using the laser radar under different weather types and obtain the point visibility by using the point visibility meter.

[0084] 3. Echo signal preprocessing module.

[0085] The echo signal preprocessing module is used to preprocess the echo signal obtained by the laser radar to obtain the preprocessed extinction coefficient and obtain the laser visibility based on the preprocessed extinction coefficient.

[0086] 4. Final visibility calculation module.

[0087] The final visibility calculation module is used to perform dynamic weight iterative optimization on the point visibility, the laser visibility and the near-approximate visibility under different weather types by using the dynamic weight multilayer perceptron to obtain the final visibility.

[0088] 5. Dominant visibility judgment module.

[0089] The active visibility judgment module is configured to perform a dominant visibility judgment based on the final visibility to determine a dominant visibility.

[0090] In addition, the present application also provides an airport dominant visibility measurement device based on multi-source data fusion, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors are caused to implement the airport dominant visibility measurement method based on multi-source data fusion as described above.

[0091] Finally, the present application also provides a computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the airport dominant visibility measurement method based on multi-source data fusion as described above.

[0092] Finally, it should be explained that the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the protection scope of the present application. Those skilled in the art can modify or equivalently replace the technical solutions of the present application according to the idea of the present application, without departing from the essence and scope of the technical solutions of the present application.

Claims

1. A method for measuring airport-dominant visibility based on multi-source data fusion, characterized in that, Includes the following steps: 1) Use the backscattering coefficient, PM2.5 concentration and relative humidity of the lidar as feature vectors, and use a support vector machine classifier to classify the weather to obtain different weather types; 2) Under different weather conditions, use lidar to obtain echo signals of the entire airport area and use point visibility meters to obtain point visibility; 3) Preprocess the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient and obtain the laser visibility based on the preprocessed extinction coefficient; 4) Dynamic weighted multilayer perceptrons are used to dynamically iterate and optimize point visibility, laser visibility, and near-visibility under different weather types to obtain the final visibility. 5) Based on the final visibility, determine the dominant visibility to ascertain the dominant visibility.

2. The airport-dominant visibility measurement method based on multi-source data fusion according to claim 1, characterized in that, Step 3) involves preprocessing the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient, specifically including: 31) Geometric overlap factor self-calibration: An adaptive piecewise fitting algorithm is used to process the echo signal and outliers in the echo signal are removed by box plot; 32) Geometric factor fitting: Fit the effective region of the echo signal using the least squares method and remove echo signals that are not within the effective region; 33) Perform wavelength conversion: Use a solar photometer or particle spectrometer to obtain the wavelength index α of the lidar, and convert the wavelength λ of the lidar into the equivalent extinction coefficient α of the 550nm standard wavelength. lidar : α(550nm) is the extinction coefficient at the standard wavelength of 550nm.

3. The airport-dominant visibility measurement method based on multi-source data fusion according to claim 2, characterized in that, When the weather type is rain, step 3) of preprocessing the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient further includes: 34) Calculate the rainfall intensity R = 0.01Z using the reflectivity factor Z of millimeter-wave radar. 0.714 The extinction coefficient α of the lidar under rainfall conditions is calculated based on the rainfall intensity. rain α rain =α lidar ·(1+k R ·R), k R This is the rainfall attenuation coefficient.

4. The airport-dominant visibility measurement method based on multi-source data fusion according to claim 1, characterized in that, The dynamic weight iterative optimization in step 4) to obtain the final visibility specifically involves: Among them, V final It is the final visibility, V i is the i-th visibility among point visibility, laser visibility, and near-visibility under different weather types, n is the total number of point visibility, laser visibility, and near-visibility under different weather types, b is the offset constant, and w i w is the weight of the i-th visibility and it is the dynamic weight output by the i-th layer of the dynamic weight multilayer perceptron. i =w i-1 +Δw i ·η, where w i-1 It is the dynamic weights output by the (i-1)th layer of the dynamic weight multilayer perceptron, Δw i η represents the weight adjustment amount of the (i-1)th layer of the dynamic weighted multilayer perceptron, and η is the learning rate.

5. The airport-dominant visibility measurement method based on multi-source data fusion according to claim 1, characterized in that, Step 5) specifically involves: when the final visibility V... final At a depth of >1500m, the dominant visibility is V. final When the final visibility V final When the visibility is ≤1500m, the dominant visibility is max(V final V manualthreshold ), where V manualthreshold This is the visibility threshold for manual observation.

6. The airport-dominant visibility measurement method based on multi-source data fusion according to any one of claims 1-5, characterized in that, Further includes: 6) Use lidar to acquire echo signals of the entire airport area in real time and preprocess the echo signals acquired by lidar to obtain the preprocessed radar extinction coefficient. Use a point visibility meter to acquire the point extinction coefficient in real time, and fuse the preprocessed radar extinction coefficient and the point extinction coefficient to obtain the fused extinction coefficient.

7. The airport-dominant visibility measurement method based on multi-source data fusion according to any one of claims 1-5, characterized in that, Further includes: 7) Use lidar to acquire echo signals of the entire airport area over a period of time prior to the current moment, and preprocess the acquired echo signals to obtain a preprocessed radar extinction coefficient series. Based on the preprocessed radar extinction coefficient series, predict the future extinction coefficient α using a recurrent neural network. pred And based on the predicted future extinction coefficient α pred Obtain future dominant visibility V:

8. An airport-dominant visibility measurement system based on multi-source data fusion, characterized in that, include: The weather classification module uses the backscattering coefficient of the lidar, PM2.5 concentration, and relative humidity as feature vectors, and performs weather classification through a support vector machine classifier to obtain different weather types. The data acquisition module is used to acquire echo signals of the entire airport area using lidar and to acquire point visibility using a point visibility meter under different weather conditions. The echo signal preprocessing module is used to preprocess the echo signal acquired by the lidar to obtain the preprocessed extinction coefficient and obtain the laser visibility based on the preprocessed extinction coefficient. The final visibility calculation module is used to perform dynamic weight iterative optimization of point visibility, laser visibility and near-visibility under different weather types through a dynamic weight multilayer sensor to obtain the final visibility. A dominant visibility determination module is used to determine the dominant visibility based on the final visibility.

9. An airport-dominant visibility measurement device based on multi-source data fusion, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the airport-dominant visibility measurement method based on multi-source data fusion as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the airport-dominated visibility measurement method based on multi-source data fusion as described in any one of claims 1-7.

Images