A wind field inversion method and system based on miniaturized laser wind radar

Abstract

The present invention belongs to the technical field of wind field inversion, and provides a wind field inversion method and system based on a miniaturized laser wind radar. The method comprises: determining a first prediction duration according to the size of a monitoring area, determining a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and using the correction coefficient to correct the first prediction duration to a second prediction duration; obtaining wind measurement data of each miniaturized laser wind radar arranged in the monitoring area, using a CNN‑LSTM model to perform wind field inversion on each wind measurement data, and obtaining wind field evolution data matching the second prediction duration; and outputting the wind field evolution data. The present invention adopts a miniaturized laser wind radar for more flexible wind measurement, and takes into account the influence of severe weather on the wireless communication stability of the wind radar, thereby improving the accuracy of the wind field evolution data.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind field inversion, and in particular relates to a wind field inversion method and system based on a miniaturized laser wind measuring radar. Background Art

[0002] Wind field inversion is a method of using Doppler radar to infer the distribution of wind speed and direction in the atmosphere. It is crucial to understanding weather systems, especially the evolution of severe convective weather phenomena such as typhoons and tornadoes. Through inversion, three-dimensional wind field data with high temporal and spatial resolution can be obtained, thus providing support for weather forecasts and warnings.

[0003] Existing wind field inversion schemes are all based on fixed Doppler radars. In order to adapt to their fixed monitoring sites and stronger detection capabilities, Doppler radars in the field of meteorological detection are usually large in size, which limits the increase in their deployment.

[0004] The use of miniaturized wind measuring radars is a relatively good solution for wind field inversion. However, miniaturized wind measuring radars are usually connected by wireless communication, which has the problem of unstable communication in severe weather. As a result, it may be impossible to stably obtain wind speed data for wind field inversion, resulting in blank spots or interruptions in the wind field inversion data, which is not conducive to assisting relevant personnel in weather forecasting and early warning. Summary of the invention

[0005] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a wind field inversion method, system, electronic equipment, computer storage medium and computer program product based on a miniaturized laser wind measurement radar.

[0006] The present invention provides a wind field inversion method based on a miniaturized laser wind measuring radar, comprising the following steps:

[0007] Determine a first prediction duration according to the size of the monitoring area, determine a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and use the correction coefficient to correct the first prediction duration to a second prediction duration;

[0008] Acquire wind measurement data of each miniaturized laser wind measurement radar deployed in the monitoring area, the wind measurement data including wind speed data and wind direction data, use the CNN-LSTM model to perform wind field inversion on each wind measurement data, and obtain wind field evolution data matching the second prediction duration;

[0009] The wind field evolution data is output.

[0010] In some embodiments, determining the first predicted duration according to the size of the monitoring area includes:

[0011] Determine the predicted overlap area between the monitored area and the target observed meteorological conditions, and calculate the area ratio of the predicted overlap area to the monitored area;

[0012] Determining a wind intensity assessment value of the target observed meteorology based on observation data of the target observed meteorology in adjacent monitoring areas;

[0013] The first prediction duration is calculated based on the area ratio and the wind intensity assessment value.

[0014] In some embodiments, the calculating the first prediction duration according to the area ratio and the wind severity assessment value includes:

[0015] ;

[0016] In the above formula, is the first prediction duration, is a fixed duration corresponding to the first predicted duration, This fixed duration is a historical experience value; is the area ratio, is the wind intensity assessment value; , is a constant value used to adjust the calculation.

[0017] In some embodiments, the determining of the correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area includes:

[0018] Determine a ratio of a first number of miniaturized laser wind radars in the monitoring area to a second number of all wind radars, and a third number of miniaturized laser wind radars in the predicted overlap area;

[0019] The correction coefficient is calculated based on the ratio and the third quantity.

[0020] In some embodiments, calculating the correction coefficient according to the ratio and the third quantity includes:

[0021] ;

[0022] In the above formula, is the correction factor, is the ratio, is the third quantity, is a constant value used to adjust the calculation, , is the weighted weight, and .

[0023] In some embodiments, the CNN-LSTM model is trained in the following manner:

[0024] Collecting historical observation data, each piece of the historical observation data includes a regional wind field data group that conforms to the fixed time length;

[0025] The CNN-LSTM model is self-supervised learned using each of the regional wind field data groups to obtain a qualified CNN-LSTM model.

[0026] The present invention also provides a wind field inversion system based on a miniaturized laser wind measurement radar, the system comprising a first processing module, a second processing module, and an output module;

[0027] The first processing module is used to determine a first prediction duration according to the size of the monitoring area, determine a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and use the correction coefficient to correct the first prediction duration to a second prediction duration;

[0028] The second processing module is used to obtain wind measurement data of each miniaturized laser wind measurement radar deployed in the monitoring area, the wind measurement data including wind speed data and wind direction data, and use the CNN-LSTM model to perform wind field inversion on each of the wind measurement data to obtain wind field evolution data matching the second prediction duration;

[0029] The output module is used to output the wind field evolution data.

[0030] The beneficial effects of the present invention are:

[0031] 1) The present invention uses a miniaturized laser wind measuring radar that is more convenient to deploy to measure the wind speed and wind direction in various places in the monitoring area, which effectively overcomes the defects of conventional Doppler radars for meteorological detection that are large in size and can only be designed as ground-based equipment due to their large size and are difficult to flexibly deploy, and helps to measure more detailed wind speed data in the monitoring area;

[0032] 2) The present invention comprehensively considers the size of the monitoring area and the distribution data of the miniaturized laser wind radar in the monitoring area to obtain the optimal prediction duration, and evolves the wind field accordingly, so that the obtained wind field evolution data can effectively address the defect of unstable wireless communication of the miniaturized laser wind radar in severe weather. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0034] Figure 1 It is a flow chart of a wind field inversion method based on a miniaturized laser wind measuring radar disclosed in an embodiment of the present invention;

[0035] Figure 2 is a schematic diagram of a process for determining a first predicted duration disclosed in an embodiment of the present invention;

[0036] Figure 3 is a schematic diagram of a process for determining correction coefficient points disclosed in an embodiment of the present invention;

[0037] Figure 4 It is a structural schematic diagram of a wind field inversion system based on a miniaturized laser wind measuring radar disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0039] like Figure 1 As shown, the embodiment of the present invention discloses a wind field inversion method based on a miniaturized laser wind measuring radar, comprising the following steps:

[0040] Determine a first prediction duration according to the size of the monitoring area, determine a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and use the correction coefficient to correct the first prediction duration to a second prediction duration;

[0041] Acquire wind measurement data of each miniaturized laser wind measurement radar deployed in the monitoring area, the wind measurement data including wind speed data and wind direction data, use the CNN-LSTM model to perform wind field inversion on each wind measurement data, and obtain wind field evolution data matching the second prediction duration;

[0042] The wind field evolution data is output.

[0043] Laser wind radar is a remote sensing device that uses the Doppler effect to measure wind speed and direction. It emits laser pulses and receives light signals scattered by atmospheric particles (such as dust, salt crystals, cloud mist, water vapor, pollution particles and other aerosols), analyzes the Doppler frequency shift of these signals to determine wind speed and direction, and can provide high temporal and spatial resolution and high-precision wind field data. The present invention uses a miniaturized laser wind radar to measure the wind speed and direction in the monitoring area in real time, and then feeds it back to the monitoring terminal for the monitoring terminal to invert the wind field evolution data, and to assist relevant personnel in weather forecasting and early warning.

[0044] Specifically, first, determine the monitoring area that needs to be inverted for the wind field, and then determine the first prediction duration according to the size of the monitoring area; and then determine the distribution data of the miniaturized laser wind radar in the monitoring area, determine the corresponding correction coefficient based on the distribution data, and use the correction coefficient to correct the first prediction duration obtained above to the second prediction duration. The above-mentioned first prediction duration and second prediction duration are both used to determine the time span of the wind field evolution data finally obtained, such as 1 hour, 3 hours, etc. Then, obtain the wind measurement data of all miniaturized laser wind radars deployed in various parts of the monitoring area, the wind measurement data includes wind speed data and wind direction data, and use the pre-built CNN-LSTM model to perform wind field inversion on these wind measurement data, that is, obtain the wind field evolution data of the monitoring area in the next second prediction duration, and the wind field evolution data can be output to relevant personnel in the form of dynamic graphics to assist them in weather forecasting and early warning.

[0045] The above scheme of the present invention, on the one hand, uses a more conveniently deployed miniaturized laser wind radar to measure the wind speed and wind direction at various locations in the monitoring area, effectively overcoming the defect that the conventional Doppler radar for meteorological detection can only be designed as a ground-based device due to its large size and is difficult to flexibly deploy, and helps to measure more detailed wind data and wind speed in the monitoring area. On the other hand, the present invention comprehensively considers the size of the monitoring area and the distribution data of the miniaturized laser wind radar in the monitoring area to obtain the best prediction time, and evolves the wind field accordingly, so that the obtained wind field evolution data can effectively deal with the defect of unstable wireless communication of the miniaturized laser wind radar in bad weather.

[0046] In some embodiments, Figure 2 As shown, the determining of the first prediction duration according to the area size of the monitoring area includes:

[0047] Determine the predicted overlap area between the monitored area and the target observed meteorological conditions, and calculate the area ratio of the predicted overlap area to the monitored area;

[0048] Determining a wind intensity assessment value of the target observed meteorology based on observation data of the target observed meteorology in adjacent monitoring areas;

[0049] The first prediction duration is calculated based on the area ratio and the wind intensity assessment value.

[0050] In this embodiment, the size of the monitoring area in the present invention does not refer to the size of the area of ​​the monitoring area itself, but is based on the degree of influence of the target observation meteorology in the monitoring area. Specifically, the target observation meteorology that is about to pass through the monitoring area is determined, such as typhoons, tornadoes, etc. The predicted path and radius of the target observation meteorology can be predicted by the existing meteorological forecasting system. According to the geographical area information of the monitoring area and the predicted path and radius of the target observation meteorology, the overlap area of ​​the two, i.e., the predicted overlap area, can be calculated, and then the area ratio of the predicted overlap area to the monitoring area can be calculated. At the same time, the actual observation data of the target observation meteorology when it passes through the adjacent monitoring area can also be obtained, and the wind intensity of the target observation meteorology is evaluated based on this, and the corresponding wind intensity evaluation value can be obtained. The higher the wind intensity evaluation value, the higher the wind speed of the target observation meteorology, the more frequent the wind direction changes, and the larger the impact radius; and when the wind intensity evaluation value is lower, the wind speed of the target observation meteorology is lower, the wind direction is more stable, and the impact radius is smaller. Finally, the first forecast duration is calculated based on the obtained area ratio and wind intensity assessment value.

[0051] In some embodiments, the calculating the first prediction duration according to the area ratio and the wind severity assessment value includes:

[0052] ;

[0053] In the above formula, is the first prediction duration, is a fixed duration corresponding to the first predicted duration, This fixed duration is a historical experience value; is the area ratio, is the wind intensity assessment value; , is a constant value used to adjust the calculation.

[0054] Based on the above formula, the first prediction duration is determined based on the fixed duration determined based on the historical experience value and the corresponding coefficient, and the coefficient is determined by the area ratio and wind intensity assessment value obtained above. Obviously, the functional relationship between the area ratio, the wind intensity assessment value and the first prediction duration conforms to a positive correlation: when the area of ​​the target observation meteorology passing through the monitoring area is smaller and the intensity of the target observation meteorology is lower, the impact of the target observation meteorology on the monitoring area (mainly the economic impact, such as the impact on transportation safety, planting, breeding, etc.) is smaller, and the first prediction duration can be appropriately reduced at this time, that is, there is no need to perform a longer time span wind field inversion to reduce the allocation of wind field inversion computing power for the target observation meteorology, and other computing power can be used to allocate to other observation meteorology; and when the area of ​​the target observation meteorology passing through the monitoring area is larger and the intensity of the target observation meteorology is higher, the impact of the target observation meteorology on the monitoring area is greater, and the first prediction duration can be appropriately increased at this time, that is, a longer time span wind field inversion is required, which helps to assist relevant personnel in making more reasonable decisions related to the target observation meteorology.

[0055] In some embodiments, Figure 3 As shown, the correction coefficient is determined based on the distribution data of the miniaturized laser wind radar in the monitoring area, including:

[0056] Determine a ratio of a first number of miniaturized laser wind radars in the monitoring area to a second number of all wind radars, and a third number of miniaturized laser wind radars in the predicted overlap area;

[0057] The correction coefficient is calculated based on the ratio and the third quantity.

[0058] In this embodiment, the setting of the correction coefficient is mainly to consider the insufficient communication stability of the miniaturized laser wind radar in the monitoring area due to the influence of the target observation meteorology (especially the target observation meteorology belonging to the severe weather). That is, when the intensity of the target observation meteorology is higher, the wireless communication of the miniaturized laser wind radar in the passing area is more easily interfered, resulting in the more serious loss of wind measurement data (the number of missing points, the missing duration, etc.) obtained by the monitoring terminal for wind field evolution, which is not conducive to determining accurate wind field evolution data, especially wind field evolution data with a long span; and when the intensity of the target observation meteorology is lower, the wireless communication of the miniaturized laser wind radar in the passing area is less likely to be interfered, and the less serious loss of wind measurement data (the number of missing points, the missing duration, etc.) obtained by the monitoring terminal for wind field evolution, which is conducive to determining accurate wind field evolution data, especially wind field evolution data with a long span;

[0059] In view of the above technical problems, the present invention counts the number of miniaturized laser wind radars deployed in the monitoring area, calculates the ratio of the first number of miniaturized laser wind radars to the second number of all wind radars, and determines the third number of miniaturized laser wind radars located in the predicted overlapping area of ​​the target meteorological observation and the monitoring area. Finally, a correction coefficient is calculated based on the above ratio and the third number. Then, the correction coefficient is multiplied by the first prediction duration obtained above to obtain the corrected second prediction duration.

[0060] It should be noted that in addition to mobile miniaturized laser wind measuring radars, conventional fixed ground-based wind measuring radars are also installed in the monitoring area, that is, the second number includes these fixed ground-based wind measuring radars. Since the fixed ground-based wind measuring radars use wired communication, the stability of their communication transmission will basically not be affected by the target observed meteorology.

[0061] In some embodiments, calculating the correction coefficient according to the ratio and the third quantity includes:

[0062] ;

[0063] In the above formula, is the correction factor, is the ratio, is the third quantity, is a constant value used to adjust the calculation, , is the weighted weight, and .

[0064] In this embodiment, the correction coefficient is determined by combining the above-mentioned ratio and the third quantity. Since the wireless communication stability of the miniaturized laser wind measurement radar located in the predicted overlapping area is more easily affected by the target observed meteorology, its weighted weight is set larger, and the corresponding weighted weight of the ratio is set smaller.

[0065] In some embodiments, the CNN-LSTM model is trained in the following manner:

[0066] Collecting historical observation data, each piece of the historical observation data includes a regional wind field data group that conforms to the fixed time length;

[0067] The CNN-LSTM model is self-supervised learned using each of the regional wind field data groups to obtain a qualified CNN-LSTM model.

[0068] In this embodiment, based on the above fixed time length, multiple groups of regional wind field data sets are collected and obtained, each of which contains the real wind evolution data of each region within the fixed time span, and these regional wind field data sets are constructed as unlabeled training data. Then, using the algorithm based on self-supervised learning, these training data are input one by one into the pre-built CNN-LSTM model until the training goal is reached, that is, convergence.

[0069] In addition to using the above CNN-LSTM model for wind field inversion, the following methods can also be used for wind field inversion:

[0070] 1. Simplified VVP inversion algorithm: A volume velocity processing method for Doppler radar data that can invert the three-dimensional structure of the wind field. This method selects the three main parameters with the largest magnitude for inversion and introduces random observation errors to determine the scope of application of the inversion algorithm. Studies have shown that the simplified VVP algorithm is insensitive to observation errors, and the root mean square error at different elevation angles is basically unchanged, especially when the wind speed is high, the inversion accuracy is higher.

[0071] 2. Satellite remote sensing wind field inversion: Using satellite observation and remote sensing technology to invert the wind field is of great significance for supplementing the wind field information in areas with sparse observation stations such as oceans, plateaus and deserts. This method can be used for numerical forecasting of initial field structure, typhoons, high-altitude jet streams and water vapor transport research.

[0072] 3. Doppler lidar wind field inversion: Doppler lidar has important application value in wind field detection in clear sky weather due to its high-precision measurement and high-altitude resolution. It is preferred to use assimilation methods (such as variational assimilation).

[0073] 4. Nonlinear wind field inversion method based on modified VAD technology: Traditional VAD (velocity azimuth display) technology assumes that the horizontal wind field is uniformly distributed in space. However, when facing strong convective weather systems such as typhoons and tornadoes, this method does not hold true due to the complex spatial distribution of the internal wind field, resulting in low inversion accuracy. The nonlinear wind field inversion method based on modified VAD technology establishes a geometric model of the viewing cone scanning under the mobile platform and a nonlinear spatial distribution model of the complex wind field, which can better reflect the real complex wind field structure and has a better inversion effect.

[0074] like Figure 4 As shown, an embodiment of the present invention discloses a wind field inversion system based on a miniaturized laser wind measurement radar, the system comprising a first processing module, a second processing module, and an output module;

[0075] The first processing module is used to determine a first prediction duration according to the size of the monitoring area, determine a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and use the correction coefficient to correct the first prediction duration to a second prediction duration;

[0076] The second processing module is used to obtain wind measurement data of each miniaturized laser wind measurement radar deployed in the monitoring area, the wind measurement data including wind speed data and wind direction data, and use the CNN-LSTM model to perform wind field inversion on each of the wind measurement data to obtain wind field evolution data matching the second prediction duration;

[0077] The output module is used to output the wind field evolution data.

[0078] An embodiment of the present invention further discloses an electronic device, comprising: a memory storing executable program code; a processor coupled to the memory; the processor calls the executable program code stored in the memory to execute the method as described in any of the preceding items.

[0079] The embodiment of the present invention further discloses a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described in any of the above items is executed.

[0080] An embodiment of the present invention further discloses a computer program product, which includes a computer program stored in a computer storage medium, and when the computer program is executed by a processor of an electronic device, any of the above methods is implemented.

[0081] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0082] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0083] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0084] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0085] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0086] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0087] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of this disclosure can be achieved, and this document is not limited here.

[0088] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A wind field inversion method based on a miniaturized laser wind radar, characterized in that: The steps include: Determine a first prediction duration according to the size of the monitoring area, determine a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and use the correction coefficient to correct the first prediction duration to a second prediction duration; Acquire wind measurement data of each miniaturized laser wind measurement radar deployed in the monitoring area, the wind measurement data including wind speed data and wind direction data, use the CNN-LSTM model to perform wind field inversion on each wind measurement data, and obtain wind field evolution data matching the second prediction duration; Outputting the wind field evolution data; The determining the first prediction duration according to the size of the monitoring area includes: Determine the predicted overlap area between the monitored area and the target observed meteorological conditions, and calculate the area ratio of the predicted overlap area to the monitored area; Determining a wind intensity assessment value of the target observed meteorology based on observation data of the target observed meteorology in adjacent monitoring areas; The first prediction duration is calculated according to the area ratio and the wind intensity assessment value; The method of determining the correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area includes: Determine a ratio of a first number of miniaturized laser wind radars in the monitoring area to a second number of all wind radars, and a third number of miniaturized laser wind radars in the predicted overlap area; The correction coefficient is calculated based on the ratio and the third quantity.

2. The wind field inversion method based on a miniaturized laser wind radar according to claim 1 is characterized in that: The first prediction duration is calculated according to the area ratio and the wind intensity assessment value, comprising: ； In the above formula, is the first prediction duration, is a fixed duration corresponding to the first predicted duration. is the historical experience value; is the area ratio, is the wind intensity assessment value; , is a constant value used to adjust the calculation.

3. The wind field inversion method based on a miniaturized laser wind radar according to claim 2 is characterized in that: Calculating the correction coefficient according to the ratio and the third quantity includes: ； In the above formula, is the correction factor, is the ratio, is the third quantity, is a constant value used to adjust the calculation, , is the weighted weight, and .

4. The wind field inversion method based on a miniaturized laser wind radar according to claim 2 is characterized in that: The CNN-LSTM model is trained as follows: Collecting historical observation data, each piece of the historical observation data includes a regional wind field data group that conforms to the fixed time length; The CNN-LSTM model is self-supervised learned using each of the regional wind field data groups to obtain a qualified CNN-LSTM model.

5. A wind field inversion system based on a miniaturized laser wind radar, characterized in that: The system comprises a first processing module, a second processing module, and an output module; The first processing module is used to determine a first prediction duration according to the size of the monitoring area, determine a correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area, and use the correction coefficient to correct the first prediction duration to a second prediction duration; The second processing module is used to obtain wind measurement data of each miniaturized laser wind measurement radar deployed in the monitoring area, the wind measurement data including wind speed data and wind direction data, and use the CNN-LSTM model to perform wind field inversion on each of the wind measurement data to obtain wind field evolution data matching the second prediction duration; The output module is used to output the wind field evolution data; The step of determining the first predicted duration according to the size of the monitoring area includes: Determine the predicted overlap area between the monitored area and the target observed meteorological conditions, and calculate the area ratio of the predicted overlap area to the monitored area; Determining a wind intensity assessment value of the target observed meteorology based on observation data of the target observed meteorology in adjacent monitoring areas; The first prediction duration is calculated according to the area ratio and the wind intensity assessment value; The method of determining the correction coefficient according to the distribution data of the miniaturized laser wind radar in the monitoring area includes: Determine a ratio of a first number of miniaturized laser wind radars in the monitoring area to a second number of all wind radars, and a third number of miniaturized laser wind radars in the predicted overlap area; The correction coefficient is calculated based on the ratio and the third quantity.

Images