Meteorological monitoring method and system for low-altitude navigation and Internet of Things

Abstract

The invention discloses a meteorological monitoring method and system for low-altitude navigation and the Internet of Things. The meteorological monitoring method comprises the steps that a historical time sequence of meteorological monitoring data and historical observation records of meteorological events are obtained; preprocessing the historical time sequence of the meteorological monitoring data to generate standard sequence data; carrying out correlation analysis between the standard sequence data and meteorological event historical observation records to generate a meteorological event correlation; inputting the standard sequence data and the meteorological event association into an improved meteorological monitoring model, and executing model training; obtaining meteorological monitoring data in real time, inputting the meteorological monitoring data into the improved meteorological monitoring model, and generating real-time meteorological event prediction; and the real-time meteorological event prediction is sent to a low-altitude meteorological monitoring guarantee platform for real-time publishing, and if the real-time meteorological event prediction shows an extreme meteorological event, an early warning signal is broadcasted. The problem that a meteorological monitoring method and system specially oriented to low-altitude navigation and low-altitude internet-of-things scenes are lacked at the present stage is solved.

Description

Technical Field

[0001] The present invention relates to the field of meteorological monitoring, and in particular to a meteorological monitoring method and system for low-altitude navigation and the Internet of Things. Background Art

[0002] General aviation (GA) covers a variety of flight activities such as short-distance transportation, production operations, emergency rescue, etc., involving many fields, and has been positioned as a national strategic emerging industry. The rapid increase in low-altitude flight activities urgently requires the enhancement of low-altitude airspace surveillance capabilities. In light of my country's general aviation development needs and the actual situation of the pilot reform of low-altitude airspace management, the Civil Aviation Administration of China has clearly proposed to improve the meteorological service capabilities of low-altitude general aviation, promote low-altitude meteorological services based on the Internet, and enrich the meteorological information data platform.

[0003] The existing meteorological monitoring systems include those that can conduct real-time monitoring of fog, dust, visibility parameters, wind direction, wind speed, and rain and snow for highway traffic, as well as meteorological monitoring devices based on the Internet of Things for agricultural production. However, there is a lack of meteorological monitoring systems for low-altitude aviation and the Internet of Things.

[0004] The existing general aviation low-altitude surveillance system includes a ground-to-air communication data link, which is used to transmit data information from the ground to the air. In addition to control command information, it also includes meteorological data information. However, there is a lack of a meteorological monitoring system specifically for low-altitude general aviation and the Internet of Things. Summary of the invention

[0005] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a meteorological monitoring method and system for low-altitude general aviation and the Internet of Things, aiming to solve the problem of the current lack of meteorological monitoring methods and systems specifically for low-altitude general aviation and low-altitude Internet of Things scenarios, and to realize accurate meteorological event prediction for low-altitude general aviation and low-altitude Internet of Things based on artificial intelligence technology.

[0006] In order to solve the above problems, the present invention provides a meteorological monitoring method and system for low-altitude aviation and the Internet of Things.

[0007] The first aspect disclosed in the present application provides a meteorological monitoring method for low-altitude general aviation and the Internet of Things, the method comprising: obtaining a historical time series of meteorological monitoring data and a historical observation record of meteorological events; preprocessing the historical time series of meteorological monitoring data to generate standard sequence data; performing a correlation analysis between the standard sequence data and the historical observation record of meteorological events to generate meteorological event associations; dividing the standard sequence data and the meteorological event associations into a training set, a validation set and a test set in a ratio of 8:1:1, respectively, inputting the data into an improved meteorological monitoring model, performing model training and model parameter optimization; obtaining meteorological monitoring data in real time, inputting the data into the improved meteorological monitoring model, and generating real-time meteorological event prediction results; sending the real-time meteorological event prediction results to a low-altitude meteorological monitoring support platform for real-time publication, and if an extreme meteorological event occurs in the real-time meteorological event prediction results, a warning signal is broadcast.

[0008] Preferably, the meteorological monitoring data include PWV, temperature, air pressure, cloud base height, cloud cover, relative humidity, dew point temperature, wet bulb temperature, wind speed, prevailing wind direction, cloud top observation and solar radiation, and the meteorological events include precipitation, low-altitude wind speed, low-altitude wind direction, thunderstorms, strong winds, atmospheric turbulence, downbursts and low-altitude wind shear.

[0009] Preferably, the preprocessing includes: using a data spatiotemporal interpolation method to fill in missing data in the historical time series, and using a data consistency repair method to construct a historical time series in a unified format.

[0010] Preferably, the correlation analysis includes: using correlation coefficient analysis and regression analysis to perform correlation analysis between various meteorological elements in the standard sequence data, and using multiple correlation coefficients, principal component analysis, and change trend analysis to perform correlation analysis between various meteorological elements in the standard sequence data and historical observation records of meteorological events.

[0011] Preferably, the improved meteorological monitoring model comprises the following steps: The meteorological monitoring data at time t is mapped into a feature vector through the embedding layer , where the numerical data in the meteorological monitoring data is directly mapped to the corresponding components of the feature vector, the descriptive data is converted into an integer value that can be uniquely represented in the form of a hashmap, and then the integer value is mapped to the corresponding component of the feature vector; Will Input LSTM network, and LSTM network output at time t-1 Formula (1) is used to calculate the forget gate: Formula (1) Among them, in formula (1) is the calculation result of the forget gate, is the weight matrix of the forget gate, and then Use equations (2) and (3) to calculate the input gate: Formula (2) Formula (3) Among them, in formula (2) and formula (3), and is the calculation result of the input gate, is the weight matrix of the input gate, and then based on the calculation results of formula (1), formula (2) and formula (3) , and The state control parameters are calculated using formula (4): Formula (4) Among them, * in formula (4) is the Hadamard product, is the state control parameter at time t-1, is the state control parameter at time t, and then , The calculation result of formula (4) Use formula (5) to calculate the output gate: Formula (5) Among them, in formula (5) is the weight matrix of the output gate, is the output of the long short-term memory network at time t, and Calculate the forget gate, input gate, and output gate of the long short-term memory network at time t+1 together; Will Input a two-layer back-propagation neural network and reversely map the output vector to obtain the meteorological event prediction result at time t.

[0012] Preferably, the extreme meteorological events include: thunderstorms, strong winds, atmospheric turbulence, downbursts and low-level wind shear.

[0013] The second aspect disclosed in the present application provides a meteorological monitoring system for low-altitude general aviation and the Internet of Things, the system is used to implement the above-mentioned meteorological monitoring method for low-altitude general aviation and the Internet of Things, and the system includes: a collection module, the collection module is used to obtain the historical time series of meteorological monitoring data, and the historical observation records of meteorological events; a preprocessing module, the preprocessing module is used to preprocess the historical time series of meteorological monitoring data to generate standard sequence data; a correlation analysis module, the correlation analysis module is used to perform correlation analysis between the standard sequence data and the historical observation records of meteorological events to generate meteorological event associations; a model training module, the model training module is used to divide the standard sequence data and the meteorological event associations into a training set, a validation set and a test set according to a ratio of 8:1:1, input an improved meteorological monitoring model, perform model training, and optimize model parameters; a meteorological event prediction module, the meteorological event prediction module is used to obtain meteorological monitoring data in real time, input the improved meteorological monitoring model, and generate real-time meteorological event prediction results; a publishing module, the publishing module is used to send the real-time meteorological event prediction results to the low-altitude meteorological monitoring support platform for real-time publishing, and if an extreme meteorological event occurs in the real-time meteorological event prediction results, a warning signal is broadcast and issued.

[0014] The beneficial effects of the present invention are: (1) it solves the problem of the lack of meteorological monitoring methods and systems specifically for low-altitude general aviation and low-altitude Internet of Things scenarios at this stage; (2) it realizes accurate meteorological event prediction for low-altitude general aviation and low-altitude Internet of Things based on artificial intelligence technology; (3) it can combine historical meteorological event associations and real-time meteorological monitoring data, and perform large-scale parameter fitting through the forgetting gate, input gate and output gate of the long short-term memory network, so that the prediction results are accurate and the error rate is low; (4) by mapping and reverse mapping meteorological monitoring data and meteorological events, they are projected into Euclidean space in the form of feature vectors, so that meteorological monitoring data and meteorological events have mathematical meanings, are easy to calculate, and are accurately expressed. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 This is a flow chart of the meteorological monitoring method for low-altitude general aviation and the Internet of Things.

[0017] Figure 2 This is the structural diagram of the meteorological monitoring system for low-altitude aviation and the Internet of Things. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] like Figure 1 As shown, the embodiment of the present application provides a meteorological monitoring method for low-altitude general aviation and the Internet of Things, and the scheme includes: Obtain historical time series of meteorological monitoring data and historical observation records of meteorological events; Specifically, the historical meteorological monitoring data are arranged in chronological order in a time series manner, and the observation records of meteorological events in the corresponding time period are obtained.

[0020] Preprocess the historical time series of meteorological monitoring data to generate standard series data; Performing correlation analysis between the standard sequence data and historical observation records of meteorological events to generate meteorological event associations; The standard sequence data and meteorological event associations were divided into training set, validation set and test set in a ratio of 8:1:1, respectively, and input into the improved meteorological monitoring model to perform model training and model parameter optimization; Specifically, the standard sequence data and the corresponding meteorological events are associated to form a key-value pair, that is, the meteorological monitoring data at time t in the standard sequence data corresponds to the meteorological events that occurred in the time period after time t. Every 10 moments are taken as a sample sequence, and there are N / 10 sample sequences in total, where N is the length of the standard sequence data. The training set, validation set, and test set are divided in a ratio of 8:1:1 in the form of leave-one-out. The improved meteorological monitoring model is constructed using the pytorch framework, and the Adam optimizer is used for model training. The grid search method is used to adjust the optimal hyperparameters.

[0021] Obtaining meteorological monitoring data in real time, inputting it into the improved meteorological monitoring model, and generating real-time meteorological event prediction results; The real-time meteorological event forecast results are sent to the low-altitude meteorological monitoring and support platform for real-time publication. If an extreme meteorological event occurs in the real-time meteorological event forecast results, a warning signal will be broadcast.

[0022] Furthermore, the meteorological monitoring data include PWV, temperature, air pressure, cloud base height, cloud cover, relative humidity, dew point temperature, wet bulb temperature, wind speed, prevailing wind direction, cloud top observation and solar radiation, and the meteorological events include precipitation, low-altitude wind speed, low-altitude wind direction, thunderstorms, strong winds, atmospheric turbulence, downbursts and low-altitude wind shear.

[0023] Furthermore, the preprocessing includes: using a data spatiotemporal interpolation method to fill in missing data in the historical time series, and using a data consistency repair method to construct a historical time series in a unified format.

[0024] Furthermore, the correlation analysis includes: using correlation coefficient analysis and regression analysis to perform correlation analysis between various meteorological elements in standard sequence data, and using multiple correlation coefficients, principal component analysis, and change trend analysis to perform correlation analysis between various meteorological elements in standard sequence data and historical observation records of meteorological events.

[0025] Furthermore, the improved meteorological monitoring model comprises the following steps: The meteorological monitoring data at time t is mapped into a feature vector through the embedding layer , where the numerical data in the meteorological monitoring data is directly mapped to the corresponding components of the feature vector, the descriptive data is converted into an integer value that can be uniquely represented in the form of a hashmap, and then the integer value is mapped to the corresponding component of the feature vector; Will Input LSTM network, and LSTM network output at time t-1 Formula (1) is used to calculate the forget gate: Formula (1) Among them, in formula (1) is the calculation result of the forget gate, is the weight matrix of the forget gate, and then Use equations (2) and (3) to calculate the input gate: Formula (2) Formula (3) Among them, in formula (2) and formula (3), and is the calculation result of the input gate, is the weight matrix of the input gate, and then based on the calculation results of formula (1), formula (2) and formula (3) , and The state control parameters are calculated using formula (4): Formula (4) Among them, * in formula (4) is the Hadamard product, is the state control parameter at time t-1, is the state control parameter at time t, and then , The calculation result of formula (4) Use formula (5) to calculate the output gate: Formula (5) Among them, in formula (5) is the weight matrix of the output gate, is the output of the long short-term memory network at time t, and Calculate the forget gate, input gate, and output gate of the long short-term memory network at time t+1 together; Specifically, sigmod is an S-curve activation function with a domain of , the range is [0,1], tanh is the hyperbolic tangent function, the domain is , the value range is [-1,1]. For the weight matrix of the forget gate, input gate and output gate , by selecting from a Gaussian distribution with standard deviation 0.01 and mean 0 N (0,0.01) is randomly sampled to initialize the parameters in each weight matrix. In addition, the weight matrices of the forget gate, input gate, and output gate are the targets of model training and model parameter optimization, that is, through back propagation and gradient descent, the parameters in the weight matrix are iteratively optimized using the divided training set and validation set, so that the parameters in the weight matrix can be fitted more accurately.

[0026] Will Input a two-layer back-propagation neural network and reversely map the output vector to obtain the meteorological event prediction result at time t.

[0027] Exemplary: For example, temperature 25.3°C, air pressure 1004hPa and humidity 77.3%, the numerical data can be directly mapped to the corresponding components of the vector , and , Northeast wind can map 45 to the corresponding component of the vector through hashmap{[East wind, 0],[Northeast wind, 45],......,[South wind, 270]......} For example, the weight The value is 0.7, the component The value is 0.1, through a The corresponding mapping table between each component and each meteorological event, The reverse mapping is that the probability of light rain is 0.7, and the component The value of is reversely mapped to a moderate rain probability of 0.1.

[0028] Furthermore, the extreme meteorological events include: thunderstorms, strong winds, atmospheric turbulence, downbursts and low-level wind shear.

[0029] In summary, the meteorological monitoring method for low-altitude general aviation and the Internet of Things provided by the embodiment of the present application has the following technical effects: (1) It solves the problem of the lack of meteorological monitoring methods and systems specifically for low-altitude general aviation and low-altitude Internet of Things scenarios at this stage; (2) Realized accurate weather event prediction for low-altitude general aviation and low-altitude Internet of Things based on artificial intelligence technology; (3) It can combine past meteorological event associations with real-time meteorological monitoring data, and perform large-scale parameter fitting through the forget gate, input gate, and output gate of the long short-term memory network, with accurate prediction results and low error rate; (4) By mapping and reverse mapping meteorological monitoring data and meteorological events, they are projected into Euclidean space in the form of feature vectors, so that meteorological monitoring data and meteorological events have mathematical meaning, are easy to calculate, and are accurately expressed.

[0030] Based on the same inventive concept as the meteorological monitoring method for low-altitude general aviation and the Internet of Things in the aforementioned embodiment, Figure 2 As shown, the present application provides a meteorological monitoring system for low-altitude general aviation and the Internet of Things, the system comprising: A collection module, the collection module is used to obtain the historical time series of meteorological monitoring data and the historical observation records of meteorological events; A preprocessing module, which is used to preprocess the historical time series of meteorological monitoring data to generate standard sequence data; A correlation analysis module, the correlation analysis module is used to perform correlation analysis between the standard sequence data and historical observation records of meteorological events to generate meteorological event associations; Model training module: The model training module is used to associate standard sequence data with meteorological events and divide them into training set, validation set and test set in the ratio of 8:1:1, input the improved meteorological monitoring model, perform model training and optimize model parameters; A meteorological event prediction module, which is used to obtain meteorological monitoring data in real time, input the improved meteorological monitoring model, and generate real-time meteorological event prediction results; The publishing module is used to send the real-time meteorological event prediction results to the low-altitude meteorological monitoring and support platform for real-time publishing. If an extreme meteorological event occurs in the real-time meteorological event prediction results, an early warning signal is broadcast.

[0031] Through the above-mentioned detailed description of a meteorological monitoring method for low-altitude aviation and the Internet of Things in this specification, those skilled in the art can clearly know a meteorological monitoring system for low-altitude aviation and the Internet of Things in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0032] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0033] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A meteorological monitoring method for low-altitude air navigation and the Internet of Things, characterized in that: The method comprises: Obtain historical time series of meteorological monitoring data and historical observation records of meteorological events; Preprocess the historical time series of meteorological monitoring data to generate standard series data; Performing correlation analysis between the standard sequence data and historical observation records of meteorological events to generate meteorological event associations; The standard sequence data and meteorological event associations were divided into training set, validation set and test set in a ratio of 8:1:1, respectively, and input into the improved meteorological monitoring model to perform model training and model parameter optimization; Obtaining meteorological monitoring data in real time, inputting it into the improved meteorological monitoring model, and generating real-time meteorological event prediction results; The real-time meteorological event forecast results are sent to the low-altitude meteorological monitoring and support platform for real-time publication. If an extreme meteorological event occurs in the real-time meteorological event forecast results, a warning signal will be broadcast.

2. A meteorological monitoring method for low-altitude general aviation and the Internet of Things as claimed in claim 1, characterized in that: The meteorological monitoring data include PWV, temperature, air pressure, cloud base height, cloud amount, relative humidity, dew point temperature, wet bulb temperature, wind speed, prevailing wind direction, cloud top observation and solar radiation; the meteorological events include precipitation, low-altitude wind speed, low-altitude wind direction, thunderstorms, strong winds, atmospheric turbulence, downbursts and low-altitude wind shear.

3. A meteorological monitoring method for low-altitude air navigation and the Internet of Things as claimed in claim 1, characterized in that: The preprocessing includes: using a data spatiotemporal interpolation method to fill in missing data in the historical time series, and using a data consistency repair method to construct a historical time series in a unified format.

4. A meteorological monitoring method for low-altitude air navigation and the Internet of Things as claimed in claim 1, characterized in that: The correlation analysis includes: using correlation coefficient analysis and regression analysis to analyze the correlation relationship between various meteorological elements in the standard sequence data, and using multiple correlation coefficients, principal component analysis, and change trend analysis to analyze the correlation relationship between various meteorological elements in the standard sequence data and historical observation records of meteorological events.

5. A meteorological monitoring method for low-altitude general aviation and the Internet of Things as claimed in claim 1, characterized in that: The improved meteorological monitoring model comprises the following steps: The meteorological monitoring data at time t is mapped into a feature vector through the embedding layer , where the numerical data in the meteorological monitoring data is directly mapped to the corresponding components of the feature vector, the descriptive data is converted into an integer value that can be uniquely represented in the form of a hashmap, and then the integer value is mapped to the corresponding component of the feature vector; Will Input LSTM network, and LSTM network output at time t-1 Formula (1) is used to calculate the forget gate: Formula (1) Among them, in formula (1) is the calculation result of the forget gate, is the weight matrix of the forget gate, and then Use equations (2) and (3) to calculate the input gate: Formula (2) Formula (3) Among them, in formula (2) and formula (3), and is the calculation result of the input gate, is the weight matrix of the input gate, and then based on the calculation results of formula (1), formula (2) and formula (3) , and The state control parameters are calculated using formula (4): Formula (4) Among them, * in formula (4) is the Hadamard product, is the state control parameter at time t-1, is the state control parameter at time t, and then , The calculation result of formula (4) Use formula (5) to calculate the output gate: Formula (5) Among them, in formula (5) is the weight matrix of the output gate, is the output of the long short-term memory network at time t, and Calculate the forget gate, input gate, and output gate of the long short-term memory network at time t+1 together; Will Input a two-layer back-propagation neural network and reversely map the output vector to obtain the meteorological event prediction result at time t.

6. A meteorological monitoring method for low-altitude general aviation and the Internet of Things as claimed in claim 1, characterized in that: The extreme meteorological events include: thunderstorms, high winds, atmospheric turbulence, downbursts and low-level wind shear.

7. A meteorological monitoring system for low-altitude air navigation and the Internet of Things, characterized in that: The system comprises: A collection module, the collection module is used to obtain the historical time series of meteorological monitoring data and the historical observation records of meteorological events; A preprocessing module, which is used to preprocess the historical time series of meteorological monitoring data to generate standard sequence data; A correlation analysis module, the correlation analysis module is used to perform correlation analysis between the standard sequence data and historical observation records of meteorological events to generate meteorological event associations; Model training module: The model training module is used to associate standard sequence data with meteorological events and divide them into training set, validation set and test set in the ratio of 8:1:1, input the improved meteorological monitoring model, perform model training and optimize model parameters; A meteorological event prediction module, which is used to obtain meteorological monitoring data in real time, input the improved meteorological monitoring model, and generate real-time meteorological event prediction results; The publishing module is used to send the real-time meteorological event prediction results to the low-altitude meteorological monitoring and support platform for real-time publishing. If an extreme meteorological event occurs in the real-time meteorological event prediction results, an early warning signal is broadcast.

Images