A method and system for improving the accuracy of laser interferometry information in atmospheric turbulence

Abstract

The present invention provides a method and system for improving the accuracy of laser interferometry information in atmospheric turbulence, wherein the method comprises: obtaining a first temperature set and a first humidity set; training a neural network model to obtain a first laser echo spot prediction model; obtaining a first model parameter, obtaining a first air pressure set and a first wind speed set; training a neural network model to obtain a second laser echo spot prediction model, obtaining a second model parameter; updating the neural network model to obtain a third laser echo spot prediction model; inputting the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain the first predicted spot information. The method solves the technical problem in the prior art that the spot obtained by the pulse laser propagating in the atmospheric turbulent medium is affected by the weather conditions, the light intensity distribution of the spot drifts, and the accuracy is low.

Description

Technical Field

[0001] The present invention relates to the field of laser echo interference, and in particular to a method and system for improving the accuracy of laser interference information in atmospheric turbulence. Background Art

[0002] Laser echo interferometry is an important means of target detection and has been widely used in radar systems, target tracking, regional reconnaissance, satellite remote sensing and other fields. The pulsed laser transmission characteristics in atmospheric turbulence are important theories of atmospheric optics and laser technology. The refractive index fluctuations of atmospheric turbulence will lead to unstable output of laser transmission characteristics, causing statistical fluctuations in amplitude, phase and polarization degree, which will have a direct impact on the laser echo interference.

[0003] However, the inventors of this application found that the above technology has at least the following technical problems in the process of implementing the technical solution of the invention in the embodiment of this application:

[0004] In the prior art, the light spot obtained by the pulsed laser propagating in the atmospheric turbulent medium is affected by weather conditions, and the light intensity distribution of the light spot drifts, resulting in a technical problem of low accuracy. Summary of the invention

[0005] The embodiments of the present application provide a method and system for improving the accuracy of laser interferometric information in atmospheric turbulence, thereby solving the technical problem in the prior art that the light spot obtained by the propagation of a pulsed laser in an atmospheric turbulent medium is affected by weather conditions, the light spot intensity distribution drifts, and the accuracy is low. The technical effect of reducing the interference of weather conditions, improving the accuracy of laser echo light spot prediction, and enhancing the ability to predict the drift of light spot intensity distribution is achieved.

[0006] In view of the above problems, the embodiments of the present application provide a method and system for improving the accuracy of laser interferometry information in atmospheric turbulence.

[0007] In a first aspect, an embodiment of the present application provides a method for improving the accuracy of laser interference information in atmospheric turbulence, wherein the method includes: obtaining a first predetermined area, the first predetermined area being an atmospheric turbulent medium; obtaining a first temperature set and a first humidity set in the first predetermined area within a predetermined time period; training a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; obtaining a first model parameter of the first laser echo spot prediction model; obtaining a first air pressure set and a first wind speed set in the first predetermined area within a predetermined time period; training a neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; obtaining a second model parameter of the second laser echo spot prediction model; updating the neural network model according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model; obtaining a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed in the first predetermined area; inputting the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain first predicted spot information.

[0008] On the other hand, an embodiment of the present application provides a system for improving the accuracy of laser interference information in atmospheric turbulence, wherein the system includes: a first acquisition unit, the first acquisition unit is used to obtain a first predetermined area, the first predetermined area is an atmospheric turbulent medium; a second acquisition unit, the second acquisition unit is used to obtain a first temperature set and a first humidity set of the first predetermined area within a predetermined time period; a third acquisition unit, the third acquisition unit is used to train a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; a fourth acquisition unit, the fourth acquisition unit is used to obtain a first model parameter of the first laser echo spot prediction model; a fifth acquisition unit, the fifth acquisition unit is used to obtain a first air pressure set and a first wind speed set of the first predetermined area within a predetermined time period; a sixth acquisition unit , the sixth acquisition unit is used to train the neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; the seventh acquisition unit, the seventh acquisition unit is used to obtain the second model parameters of the second laser echo spot prediction model; the eighth acquisition unit, the eighth acquisition unit is used to update the neural network model according to the first model parameters and the second model parameters to obtain a third laser echo spot prediction model; the ninth acquisition unit, the ninth acquisition unit is used to obtain the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed of the first predetermined area; the tenth acquisition unit, the tenth acquisition unit is used to input the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain the first predicted spot information.

[0009] In a third aspect, an embodiment of the present application provides a system for improving the accuracy of laser interferometry information in atmospheric turbulence, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the methods described in the first aspect when executing the program.

[0010] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0011] The first predetermined area is obtained, which is an atmospheric turbulent medium; a first temperature set and a first humidity set of the first predetermined area within a predetermined time period are obtained; a neural network model is trained through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; a first model parameter of the first laser echo spot prediction model is obtained; a first air pressure set and a first wind speed set of the first predetermined area within a predetermined time period are obtained; a neural network model is trained through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; a second model parameter of the second laser echo spot prediction model is obtained; based on The neural network model is updated according to the first model parameters and the second model parameters to obtain a third laser echo spot prediction model; the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed of the first predetermined area are obtained; the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed are input into the third laser echo spot prediction model to obtain a technical solution for first predicted spot information. The embodiment of the present application provides a method and system for improving the accuracy of laser interference information in atmospheric turbulence, thereby achieving the technical effect of reducing interference from weather conditions, improving the accuracy of laser echo spot prediction, and enhancing the ability to predict the drift of spot intensity distribution.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A schematic diagram of a process for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0014] Figure 2 A schematic diagram of a process of adjusting first predicted spot information in a method for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0015] Figure 3 A schematic diagram of a flow chart of obtaining a weather change regularity coefficient of a first predetermined area in a method for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0016] Figure 4 A schematic diagram of a process for performing incremental learning of a method for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0017] Figure 5A schematic diagram of a flow chart of data loss analysis for a method of improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0018] Figure 6 A schematic diagram of a process for obtaining a first laser echo spot prediction model in a method for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0019] Figure 7 A schematic diagram of a process for dimensionality reduction processing of a method for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0020] Figure 8 This is a schematic diagram of the structure of a system for improving the accuracy of laser interferometry information in atmospheric turbulence according to an embodiment of the present application;

[0021] Fig. 9 It is a schematic diagram of the structure of an exemplary electronic device according to an embodiment of the present application.

[0022] Explanation of the figure marks: first obtaining unit 11, second obtaining unit 12, third obtaining unit 13, fourth obtaining unit 14, fifth obtaining unit 15, sixth obtaining unit 16, seventh obtaining unit 17, eighth obtaining unit 18, ninth obtaining unit 19, tenth obtaining unit 20, electronic device 300, memory 301, processor 302, communication interface 303, bus architecture 304. DETAILED DESCRIPTION

[0023] The embodiments of the present application provide a method and system for improving the accuracy of laser interferometric information in atmospheric turbulence, thereby solving the technical problem in the prior art that the light spot obtained by the propagation of a pulsed laser in an atmospheric turbulent medium is affected by weather conditions, the light spot intensity distribution drifts, and the accuracy is low. The technical effect of reducing the interference of weather conditions, improving the accuracy of laser echo light spot prediction, and enhancing the ability to predict the drift of light spot intensity distribution is achieved.

[0024] Application Overview

[0025] Laser echo interferometry is an important means of target detection and has been widely used in radar systems, target tracking, regional reconnaissance, satellite remote sensing and other fields. The pulsed laser transmission characteristics in atmospheric turbulence are an important theory of atmospheric optics and laser technology. The refractive index fluctuations of atmospheric turbulence will lead to unstable output of laser transmission characteristics, causing its amplitude, phase and polarization degree to undergo statistical fluctuations, which will have a direct impact on the laser echo interference. In the prior art, the light spot obtained by the pulsed laser propagating in the atmospheric turbulent medium is affected by weather conditions, and the light intensity distribution of the light spot drifts, resulting in low accuracy.

[0026] In response to the above technical problems, the overall idea of ​​the technical solution provided by this application is as follows:

[0027] The embodiment of the present application provides a method for improving the accuracy of laser interference information in atmospheric turbulence, wherein the method includes: obtaining a first predetermined area, the first predetermined area being an atmospheric turbulent medium; obtaining a first temperature set and a first humidity set of the first predetermined area within a predetermined time period; training a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; obtaining a first model parameter of the first laser echo spot prediction model; obtaining a first air pressure set and a first wind speed set of the first predetermined area within a predetermined time period; training a neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; obtaining a second model parameter of the second laser echo spot prediction model; updating the neural network model according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model; obtaining a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed of the first predetermined area; inputting the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain first predicted spot information.

[0028] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0029] Embodiment 1

[0030] like Figure 1 As shown, an embodiment of the present application provides a method for improving the accuracy of laser interferometry information in atmospheric turbulence, wherein the method comprises:

[0031] S100: Obtain a first predetermined area, where the first predetermined area is an atmospheric turbulent medium;

[0032] Specifically, atmospheric turbulence is an important form of motion in the atmosphere. Its existence significantly enhances the vertical and horizontal exchange of momentum, heat, water vapor and suspended pollutants in the atmosphere, which is much greater than the exchange intensity of molecular motion. The intensity of turbulence has a two-way restrictive relationship with atmospheric pressure, wind speed and temperature. Atmospheric turbulence not only affects the stability of high-altitude operations such as aircraft, but also brings certain interference to the reception of optical signals. The laser propagates in the first predetermined area, and the first predetermined area is the medium for light propagation in the atmosphere, that is, the atmospheric turbulence medium in any area. The selection of the first predetermined area can lay the foundation for the reception and analysis of the subsequent detection light path echo spot.

[0033] S200: Obtaining a first temperature set and a first humidity set of the first predetermined area within a predetermined period of time;

[0034] S300: training a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model;

[0035] S400: Obtaining first model parameters of the first laser echo spot prediction model;

[0036] Specifically, the light spot obtained by the propagation of the pulsed laser in its medium has a certain randomness and will be affected by weather conditions, such as temperature, humidity, air pressure and wind speed. The first temperature set and the first humidity set of the first predetermined area in the predetermined time period are obtained. In order to make the acquired pulsed laser echo light spot have a certain universality, the predetermined time period can be a period of sunny weather, high visibility, and conducive to the formation of atmospheric turbulence. Obtaining the first temperature set and the first humidity set can take into account the instability of the weather. The first temperature set and the first humidity set are input into the neural network model for training to obtain the first laser echo light spot prediction model, which can predict the laser echo light spot properties corresponding to different temperature and humidity conditions, and obtain the first model parameters of the first laser echo light spot prediction model, which can reduce the interference of weather conditions and improve the accuracy of laser echo light spot prediction.

[0037] S500: Obtaining a first air pressure set and a first wind speed set in the first predetermined area within a predetermined period of time;

[0038] S600: training a neural network model using the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model;

[0039] S700: Obtaining second model parameters of the second laser echo spot prediction model;

[0040] Specifically, air pressure and wind speed are also important weather factors that affect laser echoes and can affect the intensity of the echo spot. The first air pressure set and the first wind speed set of the first predetermined area within a predetermined period are obtained, and the first air pressure set and the first wind speed set are input into the neural network model for training to obtain the second laser echo spot prediction model. The difference in laser echo spot properties caused by the difference in air pressure and wind speed can be predicted, and the second model parameters of the first laser echo spot prediction model can be obtained, which can improve the accuracy and reliability of the laser echo spot prediction.

[0041] S800: updating the neural network model according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model;

[0042] S900: Obtaining a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed of the first predetermined area;

[0043] S1000: Input the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain first predicted spot information.

[0044] Specifically, in traditional machine learning modeling, the data required for model training is usually collected in a data center, and then the model is trained and predicted. Federated learning can achieve joint modeling and improve the effect of the model on the basis of ensuring data privacy, security, legality and compliance. By combining temperature, humidity, air pressure and wind speed samples, sample-based distributed model training is carried out, and all data is distributed to different machines. Each machine downloads the model from the server, and then uses local data to train the model, and then returns the parameters that need to be updated to the server. The server aggregates the returned parameters on each machine, updates the model, and then feeds the latest model back to each machine. In this process, each machine has the same and complete model, and the machines do not communicate or rely on each other. Each machine can also make independent predictions during prediction. Finally, through data update, the third laser echo spot prediction model is obtained. The third laser echo spot prediction model does not need to collect original data, which reduces communication costs and local storage and computing costs. Furthermore, the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed of the first predetermined area are collected, that is, data of any one or more indicators of the first temperature, the first humidity, the first air pressure and the first wind speed are collected, and the data are input into the third laser echo spot prediction model to perform laser echo spot prediction to obtain the first predicted spot information. This improves the efficiency of machine learning, makes the output prediction information more accurate, and enhances the prediction capability of spot intensity distribution drift.

[0045] Further, such as Figure 2 As shown, the embodiment of the present application also includes:

[0046] S1010: Obtaining a weather change regularity coefficient for the first predetermined area;

[0047] S1020: Obtaining a loss function of the first predicted light spot information under a weather change law state in the first predetermined area according to a weather change law coefficient of the first predetermined area and the first predicted light spot information;

[0048] S1030: Obtain a first adjustment parameter according to the loss function;

[0049] S1040: Adjust the first predicted light spot information according to the first adjustment parameter to obtain second predicted light spot information.

[0050] Specifically, the propagation of pulsed laser in its medium is open and easily affected by different weather conditions. Based on the historical data of weather changes, the weather change law coefficient of the first area is obtained. The first predicted spot information refers to the spot information of the laser echo spot obtained by the propagation of the pulsed laser in the medium. The first state loss function is used to represent the state loss function of the initial weather state data to the current weather state data of the weather change law coefficient of the first predicted spot information in the first predetermined area. Commonly used loss functions include 0-1 loss function, absolute value loss function, log loss function, etc. The use of different loss functions has different effects on the model. By minimizing the loss function, the model reaches a convergence state and reduces the error of the model prediction value. Based on the introduction of the loss function, the data loss analysis is completed to obtain the first adjustment parameter, and the first predicted spot information is adjusted to obtain the second predicted spot information. The accuracy of spot information prediction can be improved by introducing the loss function.

[0051] Further, such as Figure 3 As shown, the step S1010 of obtaining the weather change regularity coefficient of the first predetermined area further includes:

[0052] S1011: Obtaining a value threshold of a weather change regularity coefficient of the first predetermined area;

[0053] S1012: Randomly obtain M weather change regularity coefficients from the value threshold of the weather change regularity coefficient of the first predetermined area;

[0054] S1013: Calculating the M weather change law coefficients according to a genetic algorithm to obtain M predicted light spot morphology curves, wherein the M predicted light spot morphology curves correspond to the M weather change law coefficients one by one;

[0055] S1014: Obtaining an actual spot shape curve of the pulsed laser;

[0056] S1015: Compare the M predicted light spot morphology curves with the actual light spot morphology curve to obtain the weather change law coefficient of the first predetermined area, wherein the predicted light spot morphology curve corresponding to the weather change law coefficient of the first predetermined area has the greatest similarity with the actual light spot morphology curve.

[0057] Specifically, the essence of the genetic algorithm is to continuously perform random searches in the solution space, continuously generate new solutions in the search process, and retain the algorithm of the better solution. The implementation difficulty is relatively low, and a relatively satisfactory result can be obtained in a short time. When using, the genetic algorithm directly operates on the structural object, there is no limitation on the derivation and function continuity, and it has inherent implicit parallelism and better global optimization ability. It adopts a probabilistic optimization method, and can automatically obtain and guide the optimized search space without determining the rules, and adjust the search direction adaptively. Therefore, it is widely used in various fields as a search algorithm. The threshold value of the weather change law coefficient of the first predetermined area is obtained, that is, the threshold range of the weather change law coefficient is obtained according to historical data, and the M weather change law coefficients are calculated according to the genetic algorithm to obtain M predicted light spot morphology curves. The M weather change law coefficients are randomly obtained from the threshold value of the weather change law coefficient of the first predetermined area, and the M predicted light spot morphology curves correspond to the M weather change law coefficients one by one. The actual spot shape curve of the pulsed laser is the recorded data of the spot shape of the pulsed laser echo actually received by the pulsed laser. By comparing the M predicted spot shape curves with the actual spot shape curve, the predicted value with the closest similarity is obtained, and the corresponding weather change law coefficient is the weather change law coefficient of the first predetermined area. A better weather change law coefficient can be obtained with low implementation difficulty, achieving the technical effect of obtaining a more satisfactory result in a short time.

[0058] Further, such as Figure 4 As shown, the embodiment of the present application also includes:

[0059] S1050: Obtain a second predetermined area;

[0060] S1060: Obtaining a second temperature and / or a second humidity and / or a second air pressure and / or a second wind speed of the second predetermined area according to the second predetermined area;

[0061] S1070: performing incremental learning on the third laser echo spot prediction model according to the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area to obtain a fourth laser echo spot prediction model;

[0062] S1080: Obtain fourth predicted spot information according to the fourth laser echo spot prediction model.

[0063] Specifically, a second predetermined area different from the first predetermined area is obtained, and at least one of a second temperature, a second humidity, a second air pressure and a second wind speed of the second predetermined area is obtained; the third laser echo spot prediction model is incrementally learned according to the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area to obtain a fourth laser echo spot prediction model; the third laser echo spot prediction model is a corresponding evaluation model obtained by machine learning the neural network model based on the first model parameters and the second model parameters, and the first model parameters and the second model parameter information need to be learned twice incrementally. Since the second temperature and / or the second humidity and / or the second wind speed are The second air pressure and / or the second wind speed need to be combined with the old training data of the third laser echo spot prediction model to complete the comprehensive incremental learning results. Therefore, after the incremental learning of the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed, the basic performance of the third laser echo spot prediction model can be retained, and the corresponding incremental learning can be completed, and then based on the fourth laser echo spot prediction model, the corresponding output information is obtained, that is, the fourth predicted spot information is obtained, wherein the fourth laser echo spot prediction model is an updated model after incremental learning, thereby achieving the technical effect of incremental learning of delay features to improve the updating performance of laser echo spot prediction results.

[0064] Further, such as Figure 5 As shown, the third laser echo spot prediction model is incrementally learned according to the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area to obtain a fourth laser echo spot prediction model, and step S1070 includes:

[0065] S1071: inputting the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area into the third laser echo spot prediction model to obtain third predicted spot information;

[0066] S1072: Performing data loss analysis on the third predicted light spot information to obtain light spot loss data;

[0067] S1073: Input the spot loss data into the third laser echo spot prediction model for training to obtain the fourth laser echo spot prediction model.

[0068] Specifically, since the third laser echo spot prediction model is obtained by data training based on the first model parameters and the second model parameters, the first loss data is obtained by introducing a loss function to complete the data loss analysis, wherein the first loss data represents the relevant loss data of the third laser echo spot prediction model for the first model parameters and the second model parameters, and then the incremental learning of the third laser echo spot prediction model is completed based on the first loss data. Since the third laser echo spot prediction model is obtained by a neural network composed of multiple neurons connected to each other, the fourth laser echo spot prediction model retains the basic functions of the third laser echo spot prediction model through loss data training, and maintains the performance of continuous updating of the model, thereby improving the prediction update performance and ensuring the technical effect of the accuracy of the updated prediction results.

[0069] Further, such as Figure 6 As shown, after obtaining the first temperature set and the first humidity set of the first predetermined area within the predetermined period, step S200 further includes:

[0070] S210: performing dimensionality reduction processing on the first temperature set and the first humidity set according to a principal component analysis method to obtain a first temperature reduced dimensionality set and a first humidity reduced dimensionality set;

[0071] S220: Training a neural network model using the first temperature set and the first humidity set to obtain a first laser echo spot prediction model.

[0072] Specifically, the principal component analysis algorithm (PCA) is the most commonly used linear dimensionality reduction method. Its goal is to map high-dimensional data to low-dimensional space through some kind of linear projection, and it is expected that the amount of information of the data on the projected dimension is the largest (the variance is the largest), so as to use fewer data dimensions while retaining more characteristics of the original data points. The purpose of PCA dimensionality reduction is to reduce the dimensions of the original features while ensuring that "the amount of information is not lost", that is, to project the original features onto the dimension with the largest amount of projected information as much as possible. The original features are projected onto these dimensions to minimize the loss of information after dimensionality reduction. Since the temperature and humidity data are constantly changing and the amount of data is large, the first temperature set and the first humidity set are reduced in dimension by the principal component analysis method. Under the premise of ensuring the amount of information, redundant data are eliminated to obtain the first temperature reduced dimension set and the first humidity reduced dimension set. The reduced dimension data set is further used as the first temperature set and the first humidity set to train the neural network model to obtain the first laser echo spot prediction model. It can reduce the sample size of feature data in the database and minimize the loss of information after dimensionality reduction, thereby speeding up the calculation speed of the training model on the data.

[0073] Further, such as Figure 7 As shown, the first temperature set and the first humidity set are subjected to dimensionality reduction processing according to the principal component analysis method to obtain a first temperature dimensionality reduction set and a first humidity dimensionality reduction set, and step S210 further includes:

[0074] S211: Decentralize the first temperature set and the first humidity set to obtain a second temperature set and a second humidity set;

[0075] S212: Obtain a first covariance matrix and a second covariance matrix of the second temperature set and the second humidity set respectively;

[0076] S213: Obtain a first eigenvector and a second eigenvector of the first covariance matrix and the second covariance matrix respectively;

[0077] S214: Project the first temperature set and the first humidity set onto the first eigenvector and the second eigenvector respectively to obtain a first temperature reduced-dimensionality set and a first humidity reduced-dimensionality set.

[0078] Specifically, the extracted feature data is numerically processed, and a feature data set matrix is ​​constructed to obtain the first temperature set and the first humidity set. Then, each feature data in the first temperature set and the first humidity set is centrally processed. First, the average value of each feature in the first temperature set and the first humidity set is solved. Then, for all samples, each feature is subtracted from its own mean value to obtain a new feature value. The new feature value constitutes the second temperature set and the second humidity set. The second temperature set and the second humidity set are two data matrices. Through the covariance formula:

[0079] The second temperature set and the second humidity set are respectively operated to obtain the first covariance matrix and the second covariance matrix of the second temperature set and the second humidity set. The characteristic data of is the average value of the feature data; M is the total amount of sample data. Then, through matrix operations, the eigenvalues ​​and eigenvectors of the first covariance matrix and the second covariance matrix are respectively obtained, and each eigenvalue corresponds to an eigenvector. The eigenvalues ​​of the first covariance matrix correspond to the first eigenvector, and the eigenvalues ​​of the second covariance matrix correspond to the second eigenvector. Among the first eigenvector and the second eigenvector obtained, the largest first K eigenvalues ​​and their corresponding eigenvectors are respectively selected, and the original features in the first temperature set and the first humidity set are projected onto the selected eigenvectors to obtain the first temperature set and the first humidity set after dimensionality reduction. The feature data in the database are reduced in dimension by principal component analysis. Under the premise of ensuring the amount of information, redundant data is eliminated, so that the sample size of the feature data in the database is reduced, and the information loss after dimensionality reduction is minimized, thereby speeding up the training model's operation speed on the data.

[0080] In summary, the method and system for improving the accuracy of laser interferometry information in atmospheric turbulence provided by the embodiments of the present application have the following technical effects:

[0081] 1. Due to the adoption of obtaining a first predetermined area, the first predetermined area is an atmospheric turbulent medium; obtaining a first temperature set and a first humidity set of the first predetermined area within a predetermined period; training a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; obtaining a first model parameter of the first laser echo spot prediction model; obtaining a first air pressure set and a first wind speed set of the first predetermined area within a predetermined period; training a neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; obtaining a second model parameter of the second laser echo spot prediction model; The neural network model is updated according to the first model parameters and the second model parameters to obtain a third laser echo spot prediction model; the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed of the first predetermined area are obtained; the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed are input into the third laser echo spot prediction model to obtain a technical solution for first predicted spot information. The embodiment of the present application provides a method and system for improving the accuracy of laser interference information in atmospheric turbulence, thereby achieving the technical effect of reducing interference from weather conditions, improving the accuracy of laser echo spot prediction, and enhancing the ability to predict the drift of spot intensity distribution.

[0082] 2. Since the loss data is used to train the third laser echo spot prediction model, the fourth laser echo spot prediction model retains the basic functions of the third laser echo spot prediction model and maintains the performance of the model being continuously updated, thereby improving the prediction update performance and ensuring the technical effect of the accuracy of the updated prediction results.

[0083] Embodiment 2

[0084] Based on the same inventive concept as the method for improving the accuracy of laser interferometry information in atmospheric turbulence in the aforementioned embodiment, Figure 8 As shown, an embodiment of the present application provides a system for improving the accuracy of laser interferometry information in atmospheric turbulence, wherein the system comprises:

[0085] A first obtaining unit 11, wherein the first obtaining unit 11 is used to obtain a first predetermined area, wherein the first predetermined area is an atmospheric turbulent medium;

[0086] A second obtaining unit 12, the second obtaining unit 12 is used to obtain a first temperature set and a first humidity set of the first predetermined area within a predetermined period of time;

[0087] A third obtaining unit 13, wherein the third obtaining unit 13 is used to train a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model;

[0088] A fourth obtaining unit 14, the fourth obtaining unit 14 is used to obtain a first model parameter of the first laser echo spot prediction model;

[0089] A fifth obtaining unit 15, the fifth obtaining unit 15 is used to obtain a first air pressure set and a first wind speed set in the first predetermined area within a predetermined time period;

[0090] A sixth obtaining unit 16, wherein the sixth obtaining unit 16 is used to train a neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model;

[0091] A seventh obtaining unit 17, wherein the seventh obtaining unit 17 is used to obtain a second model parameter of the second laser echo spot prediction model;

[0092] An eighth obtaining unit 18, wherein the eighth obtaining unit 18 is used to update the neural network model according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model;

[0093] A ninth obtaining unit 19, wherein the ninth obtaining unit 19 is used to obtain a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed of the first predetermined area;

[0094] The tenth obtaining unit 20 is used to input the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain first predicted spot information.

[0095] Furthermore, the system comprises:

[0096] an eleventh obtaining unit, the eleventh obtaining unit being used to obtain a weather change regularity coefficient of the first predetermined area;

[0097] a twelfth obtaining unit, the twelfth obtaining unit being used to obtain a loss function of the first predicted light spot information under a weather change law state of the first predetermined area according to a weather change law coefficient of the first predetermined area and the first predicted light spot information;

[0098] a thirteenth obtaining unit, the thirteenth obtaining unit being used to obtain a first adjustment parameter according to the loss function;

[0099] A fourteenth obtaining unit, wherein the fourteenth obtaining unit is used to adjust the first predicted light spot information according to the first adjustment parameter to obtain second predicted light spot information.

[0100] Furthermore, the system comprises:

[0101] a fifteenth obtaining unit, the fifteenth obtaining unit being used to obtain a value threshold of a weather change regularity coefficient of the first predetermined area;

[0102] A sixteenth obtaining unit, the sixteenth obtaining unit is used to randomly obtain M weather change regularity coefficients from a value threshold of the weather change regularity coefficient of the first predetermined area;

[0103] A seventeenth obtaining unit, the seventeenth obtaining unit is used to calculate the M weather change law coefficients according to a genetic algorithm to obtain M predicted light spot morphology curves, wherein the M predicted light spot morphology curves correspond to the M weather change law coefficients one by one;

[0104] An eighteenth obtaining unit, the eighteenth obtaining unit is used to obtain an actual spot morphology curve of the pulsed laser;

[0105] A nineteenth obtaining unit is used to compare the M predicted light spot morphology curves with the actual light spot morphology curve to obtain the weather change law coefficient of the first predetermined area, wherein the predicted light spot morphology curve corresponding to the weather change law coefficient of the first predetermined area has the greatest similarity with the actual light spot morphology curve.

[0106] Furthermore, the system comprises:

[0107] a twentieth obtaining unit, the twentieth obtaining unit being used to obtain a second predetermined area;

[0108] a twenty-first obtaining unit, the twenty-first obtaining unit being configured to obtain, according to the second predetermined area, a second temperature and / or a second humidity and / or a second air pressure and / or a second wind speed of the second predetermined area;

[0109] a twenty-second obtaining unit, the twenty-second obtaining unit being configured to perform incremental learning on the third laser echo spot prediction model according to the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area to obtain a fourth laser echo spot prediction model;

[0110] The twenty-third obtaining unit is used to obtain fourth predicted spot information according to the fourth laser echo spot prediction model.

[0111] Furthermore, the system comprises:

[0112] A twenty-fourth obtaining unit, the twenty-fourth obtaining unit is used to input the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area into the third laser echo spot prediction model to obtain third predicted spot information;

[0113] A twenty-fifth obtaining unit, the twenty-fifth obtaining unit is used to perform data loss analysis on the third predicted light spot information to obtain light spot loss data;

[0114] The twenty-sixth obtaining unit is used to input the spot loss data into the third laser echo spot prediction model for training to obtain the fourth laser echo spot prediction model.

[0115] Furthermore, the system comprises:

[0116] A twenty-seventh obtaining unit, the twenty-seventh obtaining unit is used to perform dimensionality reduction processing on the first temperature set and the first humidity set according to a principal component analysis method to obtain a first temperature reduced dimension set and a first humidity reduced dimension set;

[0117] The twenty-eighth obtaining unit is used to train the neural network model through the first temperature set and the first humidity set to obtain the first laser echo spot prediction model.

[0118] Furthermore, the system comprises:

[0119] a twenty-ninth obtaining unit, the twenty-ninth obtaining unit being configured to perform decentralized processing on the first temperature set and the first humidity set to obtain a second temperature set and a second humidity set;

[0120] a thirtieth obtaining unit, configured to obtain a first covariance matrix and a second covariance matrix of the second temperature set and the second humidity set, respectively;

[0121] A thirty-first obtaining unit, the thirty-first obtaining unit being used to obtain a first eigenvector and a second eigenvector of the first covariance matrix and the second covariance matrix, respectively;

[0122] A thirty-second obtaining unit is used to project the first temperature set and the first humidity set onto the first eigenvector and the second eigenvector respectively to obtain a first temperature reduced-dimensionality set and a first humidity reduced-dimensionality set.

[0123] Exemplary Electronic Devices

[0124] Reference below Fig. 9 To describe the electronic device of the embodiment of the present application,

[0125] Based on the same inventive concept as a method for improving the accuracy of laser interference information in atmospheric turbulence in the aforementioned embodiment, an embodiment of the present application also provides a system for improving the accuracy of laser interference information in atmospheric turbulence, including: a processor, the processor is coupled to a memory, the memory is used to store a program, when the program is executed by the processor, the system executes any one of the methods described in the first aspect.

[0126] The electronic device 300 includes: a processor 302, a communication interface 303, and a memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. The communication interface 303, the processor 302, and the memory 301 may be interconnected via the bus architecture 304; the bus architecture 304 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig. 9 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0127] The processor 302 may be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application.

[0128] The communication interface 303 uses any transceiver-like system for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), wired access networks, etc.

[0129] The memory 301 may be a ROM or other type of static storage device that can store static information and instructions, a RAM or other type of dynamic storage device that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor via a bus architecture 304. The memory may also be integrated with the processor.

[0130] The memory 301 is used to store computer-executable instructions for executing the solution of the present application, and the execution is controlled by the processor 302. The processor 302 is used to execute the computer-executable instructions stored in the memory 301, thereby realizing a method for improving the accuracy of laser interferometry information in atmospheric turbulence provided in the above embodiment of the present application.

[0131] Optionally, the computer-executable instructions in the embodiments of the present application may also be referred to as application code, which is not specifically limited in the embodiments of the present application.

[0132] The embodiment of the present application provides a method for improving the accuracy of laser interference information in atmospheric turbulence, wherein the method includes: obtaining a first predetermined area, the first predetermined area being an atmospheric turbulent medium; obtaining a first temperature set and a first humidity set of the first predetermined area within a predetermined time period; training a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; obtaining a first model parameter of the first laser echo spot prediction model; obtaining a first air pressure set and a first wind speed set of the first predetermined area within a predetermined time period; training a neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; obtaining a second model parameter of the second laser echo spot prediction model; updating the neural network model according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model; obtaining a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed of the first predetermined area; inputting the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain first predicted spot information.

[0133] Those of ordinary skill in the art will appreciate that the various digital numbers such as the first and second involved in the present application are only for the convenience of description, and are not used to limit the scope of the embodiments of the present application, nor do they represent the order of precedence. "And / or" describes the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one" refers to one or more. At least two refers to two or more. "At least one", "any one" or similar expressions refer to any combination of these items, including any combination of single items (individuals) or plural items (individuals). For example, at least one item (individual, kind) of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0134] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable systems. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server, data center, etc. that includes one or more available media integrated. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0135] The various illustrative logic units and circuits described in the embodiments of the present application can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic system, a discrete gate or transistor logic, a discrete hardware component, or the design of any combination of the above functions. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any traditional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of a computing system, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration to implement.

[0136] The steps of the method or algorithm described in the embodiments of the present application can be directly embedded in hardware, a software unit executed by a processor, or a combination of the two. The software unit can be stored in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and the storage medium can be arranged in an ASIC, and the ASIC can be arranged in a terminal. Optionally, the processor and the storage medium can also be arranged in different components in the terminal. These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0137] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined by the appended claims, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for improving the accuracy of laser interferometry information in atmospheric turbulence, wherein: The method comprises: Obtaining a first predetermined area, wherein the first predetermined area is an atmospheric turbulent medium; Obtaining a first temperature set and a first humidity set of the first predetermined area within a predetermined period of time; The neural network model is trained by using the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; Obtaining first model parameters of the first laser echo spot prediction model; Obtaining a first air pressure set and a first wind speed set for the first predetermined area within a predetermined period of time; The neural network model is trained by using the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; Obtaining second model parameters of the second laser echo spot prediction model; The neural network model is updated according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model; Obtaining a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed in the first predetermined area; The first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed are input into the third laser echo spot prediction model to obtain first predicted spot information.

2. The method of claim 1, wherein: The method comprises: Obtaining a weather change regularity coefficient for the first predetermined area; According to the weather change law coefficient of the first predetermined area and the first predicted light spot information, a loss function of the first predicted light spot information under the weather change law state of the first predetermined area is obtained; According to the loss function, obtaining a first adjustment parameter; The first predicted light spot information is adjusted according to the first adjustment parameter to obtain second predicted light spot information.

3. The method of claim 2, wherein: The step of obtaining the weather change regularity coefficient of the first predetermined area includes: Obtaining a value threshold of a weather change regularity coefficient of the first predetermined area; Randomly obtain M weather change regularity coefficients from the value threshold of the weather change regularity coefficient of the first predetermined area; Calculating the M weather change law coefficients according to a genetic algorithm to obtain M predicted light spot morphology curves, wherein the M predicted light spot morphology curves correspond one to one to the M weather change law coefficients; Obtain the actual spot shape curve of the pulse laser; The M predicted light spot morphology curves are compared with the actual light spot morphology curve to obtain the weather change law coefficient of the first predetermined area, wherein the predicted light spot morphology curve corresponding to the weather change law coefficient of the first predetermined area has the greatest similarity with the actual light spot morphology curve.

4. The method of claim 1, wherein: The method further comprises: obtaining a second predetermined area; According to the second predetermined area, obtaining a second temperature and / or a second humidity and / or a second air pressure and / or a second wind speed of the second predetermined area; performing incremental learning on the third laser echo spot prediction model according to the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area to obtain a fourth laser echo spot prediction model; According to the fourth laser echo spot prediction model, fourth predicted spot information is obtained.

5. The method of claim 4, wherein: The step of incrementally learning the third laser echo spot prediction model according to the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area to obtain a fourth laser echo spot prediction model includes: Inputting the second temperature and / or the second humidity and / or the second air pressure and / or the second wind speed of the second predetermined area into the third laser echo spot prediction model to obtain third predicted spot information; Performing data loss analysis on the third predicted light spot information to obtain light spot loss data; The spot loss data is input into the third laser echo spot prediction model for training to obtain the fourth laser echo spot prediction model.

6. The method of claim 1, wherein: After obtaining the first temperature set and the first humidity set of the first predetermined area within a predetermined period of time, the method further includes: Performing dimensionality reduction processing on the first temperature set and the first humidity set according to a principal component analysis method to obtain a first temperature reduced dimensionality set and a first humidity reduced dimensionality set; The neural network model is trained by using the first temperature set and the first humidity set to obtain a first laser echo spot prediction model.

7. The method of claim 6, wherein: The performing dimensionality reduction processing on the first temperature set and the first humidity set according to the principal component analysis method to obtain a first temperature dimensionality reduction set and a first humidity dimensionality reduction set includes: Decentralizing the first temperature set and the first humidity set to obtain a second temperature set and a second humidity set; Obtaining a first covariance matrix and a second covariance matrix of the second temperature set and the second humidity set respectively; Obtaining first eigenvectors and second eigenvectors of the first covariance matrix and the second covariance matrix respectively; The first temperature set and the first humidity set are projected onto the first eigenvector and the second eigenvector respectively to obtain a first temperature reduced-dimensionality set and a first humidity reduced-dimensionality set.

8. A system for improving the accuracy of laser interferometry information in atmospheric turbulence, wherein: The system comprises: A first obtaining unit, the first obtaining unit is used to obtain a first predetermined area, the first predetermined area is an atmospheric turbulent medium; a second obtaining unit, the second obtaining unit being used to obtain a first temperature set and a first humidity set of the first predetermined area within a predetermined period of time; a third obtaining unit, the third obtaining unit being used to train a neural network model through the first temperature set and the first humidity set to obtain a first laser echo spot prediction model; A fourth obtaining unit, the fourth obtaining unit is used to obtain a first model parameter of the first laser echo spot prediction model; a fifth obtaining unit, the fifth obtaining unit being used to obtain a first air pressure set and a first wind speed set of the first predetermined area within a predetermined period of time; a sixth obtaining unit, the sixth obtaining unit being configured to train a neural network model through the first air pressure set and the first wind speed set to obtain a second laser echo spot prediction model; A seventh obtaining unit, the seventh obtaining unit is used to obtain a second model parameter of the second laser echo spot prediction model; an eighth obtaining unit, the eighth obtaining unit being used to update the neural network model according to the first model parameter and the second model parameter to obtain a third laser echo spot prediction model; a ninth obtaining unit, configured to obtain a first temperature and / or a first humidity and / or a first air pressure and / or a first wind speed of the first predetermined area; A tenth obtaining unit, wherein the tenth obtaining unit is used to input the first temperature and / or the first humidity and / or the first air pressure and / or the first wind speed into the third laser echo spot prediction model to obtain first predicted spot information.

9. A system for improving the accuracy of laser interferometry information in atmospheric turbulence, comprising: A processor, the processor is coupled to a memory, the memory is used to store a program, when the program is executed by the processor, the system is enabled to execute the method according to any one of claims 1 to 7.

Images