Wind disturbance compensation control method, system, device and medium for large-aperture reflector antenna

Abstract

The present invention relates to the field of antenna control technology, and discloses a method, system, device and medium for controlling wind disturbance compensation of a large-aperture reflector antenna. The method comprises: obtaining an optimal gain obtained by a Kalman filter for wind disturbance compensation of a large-aperture reflector antenna; inputting the optimal gain into a trained error prediction model, converting it into a two-dimensional matrix, mining the correlation features between adjacent elements in different directions of the two-dimensional matrix, and the correlation features between non-adjacent elements in different directions; optimizing the optimal gain obtained by the Kalman filter through the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements; using the optimized optimal gain to compensate for wind disturbance of the large-aperture reflector antenna, the wind disturbance compensation control effect of the large-aperture reflector antenna can be improved.

Description

Technical Field

[0001] The present invention relates to the field of antenna control technology, and in particular to a method, system, device and medium for wind disturbance compensation control of a large-caliber reflector antenna. Background Art

[0002] Large-aperture reflector antennas are key devices in the field of radio astronomy and deep space exploration. Stable and high-precision pointing is the basic prerequisite for carrying out high-quality missions. Gusts during mission operation can cause serious interference to the antenna, so it is necessary to design a reasonable method for wind disturbance control compensation.

[0003] The influence of wind disturbance can be reduced by establishing a reasonable wind disturbance control model. At present, the linear quadratic Gaussian (LOG) control method is usually used to estimate the various states of the system based on the incomplete state measurement value of the system, and then obtain the gain caused by wind disturbance, so as to achieve optimal control. Figure 1 As shown, the controlled object is a large-aperture reflector antenna. Various parameters of the controlled object are input into a Kalman filter, and the optimal gain of the controlled object is obtained through the Kalman filter. Then, the large-aperture reflector antenna is controlled for wind disturbance compensation through the optimal gain.

[0004] However, the Kalman filter does not work well in scenarios where external factors vary greatly, resulting in poor wind disturbance control compensation effect. Summary of the invention

[0005] The object of the present invention is to provide a method, system, device and medium for controlling wind disturbance compensation of a large-aperture reflector antenna, which can realize optimal wind disturbance compensation control of the large-aperture reflector antenna.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a large-aperture reflector antenna wind disturbance compensation control method, comprising the following steps:

[0007] The state parameters of the large-aperture reflector antenna are input into the Kalman filter, and the state of the large-aperture reflector antenna is estimated by the Kalman filter to obtain the optimal gain for wind disturbance compensation of the large-aperture reflector antenna;

[0008] Inputting the optimal gain into a trained error prediction model, wherein the error prediction model includes a data conversion module, a cross scan module and a dilated convolution module;

[0009] The feature information in the optimal gain is extracted through the data conversion module, the feature information is stored in a one-dimensional vector, and the one-dimensional vector is converted into a two-dimensional matrix through a preset projection matrix; the correlation features between adjacent elements in different directions of the two-dimensional matrix are mined through the cross scan module, and the correlation features between non-adjacent elements in different directions of the two-dimensional matrix are mined through the expansion convolution module;

[0010] The optimal gain obtained by the Kalman filter is optimized through the correlation characteristics between adjacent elements in different directions of the two-dimensional matrix and the correlation characteristics between non-adjacent elements;

[0011] The optimized optimal gain is used to compensate for wind disturbance of large-aperture reflector antennas.

[0012] In some optional embodiments, mining the association features between adjacent elements in different directions of the two-dimensional matrix by the Cross scan module includes:

[0013] The Cross scan module is used to scan the two-dimensional matrix horizontally, vertically, anti-horizontally and anti-vertically respectively. During the horizontal, vertical, anti-horizontally and anti-vertical scanning processes, the correlation relationship between each element in the two-dimensional matrix and adjacent elements in different directions is extracted to obtain the correlation features between adjacent elements in different directions of the two-dimensional matrix.

[0014] In some optional embodiments, the dilated convolution module is composed of dilated convolution operators of different scales, and mining the association features between non-adjacent elements in different directions of the two-dimensional matrix through the dilated convolution module includes:

[0015] In the process of the Cross scan module scanning the two-dimensional matrix horizontally, vertically, reversely horizontally and reversely vertically, the dilated convolution operators of different scales in the dilated convolution module are used to increase the step size spacing of the Cross scan module, so as to extract the correlation relationship between each element in the two-dimensional matrix and non-adjacent elements in different directions, and obtain the correlation features between non-adjacent elements in different directions of the two-dimensional matrix.

[0016] In some optional embodiments, the error prediction model further includes a fully connected layer, and the optimization of the optimal gain of the Kalman filter output by the association features between adjacent elements in different directions of the two-dimensional matrix and the association features between non-adjacent elements includes:

[0017] The fully connected layer is made to output the correction amount corresponding to the optimal gain according to the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements;

[0018] The correction amount output by the error prediction model is added to the optimal gain to obtain the optimized optimal gain.

[0019] In some optional embodiments, the error prediction model is trained in the following manner:

[0020] Obtain the optimal gain obtained by the Kalman filter as sample data;

[0021] Gaussian noise is added to the sample data, and the error prediction model is trained using the sample data after the noise is added. During the training process, the intensity of the Gaussian noise is gradually increased so that the error prediction model is trained using sample data with different noise intensities.

[0022] In some optional embodiments, the state parameters of the large-aperture reflector antenna include:

[0023] The speed output command of the large-aperture reflector antenna, the current azimuth angle, pitch angle, antenna pointing deviation, equivalent force area, and the wind speed, wind direction, temperature, humidity, and air pressure of the large-aperture reflector antenna's environment.

[0024] In some optional embodiments, the optimal gain includes the difference between the state prediction value and the state estimation value, the difference between the observation value and the estimated observation value, and the filter gain of the Kalman filter for the large-aperture reflector antenna.

[0025] An embodiment of the present invention further provides a large-aperture reflector antenna wind disturbance compensation control system, comprising:

[0026] A gain acquisition module is used to input the state parameters of the large-aperture reflector antenna into a Kalman filter, estimate the state of the large-aperture reflector antenna through the Kalman filter, and obtain the optimal gain for wind disturbance compensation of the large-aperture reflector antenna;

[0027] A model solving module, used for inputting the optimal gain into the trained error prediction model, wherein the error prediction model includes a data conversion module, a cross scan module and a dilated convolution module;

[0028] The feature information in the optimal gain is extracted through the data conversion module, the feature information is stored in a one-dimensional vector, and the one-dimensional vector is converted into a two-dimensional matrix through a preset projection matrix; the correlation features between adjacent elements in different directions of the two-dimensional matrix are mined through the cross scan module, and the correlation features between non-adjacent elements in different directions of the two-dimensional matrix are mined through the expansion convolution module;

[0029] A gain optimization module, used for optimizing the optimal gain obtained by the Kalman filter through the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements;

[0030] The wind disturbance compensation module is used to compensate for the wind disturbance of the large-aperture reflector antenna by using the optimized optimal gain.

[0031] An embodiment of the present invention also provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned large-aperture reflector antenna wind disturbance compensation control method.

[0032] An embodiment of the present invention further provides a computer-readable storage medium storing a computer program, which implements the above-mentioned large-aperture reflector antenna wind disturbance compensation control method when executed by a processor.

[0033] The wind disturbance compensation control method for a large-aperture reflector antenna provided by the present invention has at least the following beneficial effects:

[0034] First, the optimal gain estimated by the Kalman filter for wind disturbance compensation of large-aperture reflector antennas is obtained, that is, the output of the Kalman filter in traditional technology, which is used as the model input data. It is processed using the trained error prediction model and converted into a two-dimensional matrix. By mining the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements in different directions of the two-dimensional matrix, deep fusion extraction of the input data can be achieved. Based on this, the output of the traditional Kalman filter is optimized, which can achieve a more accurate estimation of the antenna state and achieve the goal of improving the accuracy of antenna wind disturbance compensation. Then, under the influence of complex external factors, the antenna position can be better adjusted to the correct position. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] One or more embodiments are exemplarily described by the pictures in the corresponding drawings, and these exemplary descriptions do not constitute limitations on the embodiments.

[0036] Figure 1 It is a schematic diagram of wind disturbance compensation control of an antenna;

[0037] Figure 2 is a flow chart of a wind disturbance compensation control method for a large-aperture reflector antenna provided according to an embodiment of the present invention;

[0038] Figure 3A schematic diagram of a network structure provided according to an embodiment of the present invention Figure 1 ;

[0039] Figure 4 is a schematic diagram of a wind disturbance compensation control method for a large-aperture reflector antenna provided according to an embodiment of the present invention;

[0040] Figure 5 is a schematic diagram of a denoising process of a diffusion model provided according to an embodiment of the present invention;

[0041] Figure 6 A schematic diagram of a network structure provided according to an embodiment of the present invention Figure 2 ;

[0042] Figure 7 is a schematic diagram of a wind disturbance compensation control system for a large-aperture reflector antenna provided according to an embodiment of the present invention;

[0043] Figure 8 It is a structural schematic diagram of a computer device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in the embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present invention can be implemented. The division of the following embodiments is for the convenience of description and should not constitute any limitation on the specific implementation of the present invention. The various embodiments can be combined and referenced with each other without contradiction.

[0045] An embodiment of the present invention relates to a large-aperture reflector antenna wind disturbance compensation control method. The specific process of the large-aperture reflector antenna wind disturbance compensation control method of this embodiment can be as follows: Figure 2 As shown, including:

[0046] Step 201: input the state parameters of the large-aperture reflector antenna into a Kalman filter, perform state estimation on the large-aperture reflector antenna through the Kalman filter, and obtain an optimal gain for wind disturbance compensation of the large-aperture reflector antenna.

[0047] Step 202: input the optimal gain into the trained error prediction model, where the error prediction model includes a data conversion module, a cross scan module and a dilated convolution module.

[0048] Step 203, extracting feature information in the optimal gain through a data conversion module, storing the feature information in a one-dimensional vector, and converting the one-dimensional vector into a two-dimensional matrix through a preset projection matrix; mining the association features between adjacent elements in different directions of the two-dimensional matrix through a cross scan module, and mining the association features between non-adjacent elements in different directions of the two-dimensional matrix through a dilated convolution module.

[0049] Step 204 , optimizing the optimal gain obtained by the Kalman filter through the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements.

[0050] Step 205: Use the optimized optimal gain to perform wind disturbance compensation on the large-aperture reflector antenna.

[0051] In this embodiment, the optimal gain for wind disturbance compensation of a large-aperture reflector antenna estimated by a Kalman filter, that is, the output of the Kalman filter in traditional technology, is used as model input data, and is processed using a trained error prediction model. After converting it into a two-dimensional matrix, deep fusion extraction of the input data can be achieved by mining the correlation features between adjacent elements in different directions of the two-dimensional matrix, as well as the correlation features between non-adjacent elements in different directions of the two-dimensional matrix. Based on this, the output of the traditional Kalman filter is optimized, and a more accurate estimation of the antenna state can be achieved, thereby achieving the goal of improving the accuracy of antenna wind disturbance compensation, and then better adjusting the antenna to the correct position through adjustment under the influence of complex external factors.

[0052] The implementation details of the wind disturbance compensation control method for a large-aperture reflector antenna of this embodiment are specifically described below. The following content is only the implementation details provided for easy understanding and is not necessary for implementing this solution.

[0053] First of all, it should be noted that the antennas described below are all large-aperture reflector antennas.

[0054] In step 201, the influence of wind disturbance on antenna pointing can be mainly divided into two parts according to the disturbance mechanism: one part is the axis angle error caused by the disturbance torque acting on the antenna shaft, and the other part is the pointing error caused by the wind pressure acting on the reflection surface and the deformation of the antenna structure. Therefore, it is very necessary to establish a reasonable wind disturbance control model to reduce the influence of wind disturbance, for example, see Figure 1The traditional wind disturbance control model shown in FIG. This model is actually an output feedback control problem, which is a comprehensive control of the linear quadratic regulator (LQR) control and the Kalman filter based on state feedback. This embodiment actually obtains the optimal gain for wind disturbance compensation of the large-aperture reflector antenna estimated by the Kalman filter, that is, Figure 1 The output of the Kalman filter is shown.

[0055] First, let me introduce the working principle of the Kalman filter:

[0056] The state equation of the Kalman filter is:

[0057] ;

[0058] In the formula, x k represents the state vector of the target, A represents the state transfer matrix, B represents the control matrix that transforms input into state, u k-1 represents the control vector, w k-1 Represents the predicted noise, which follows a Gaussian distribution.

[0059] The observation equation is:

[0060] ;

[0061] In the formula, z k represents the observation vector of the target, v k is the observed noise, which follows a Gaussian distribution.

[0062] The implementation of Kalman filter is divided into two steps: prediction and update. Prediction: According to the previous moment, that is, The posterior estimate of the moment is used to estimate the current moment, that is, The state of the moment, The prior estimate at the moment; the update is to use the measurement at the current moment to correct the estimate at the prediction stage to obtain the posterior estimate at the current moment. The following time update equation (i.e., the prediction stage) is used to infer the prior estimate of the state variable and the prior estimate of the error covariance at the current moment based on the state estimate at the previous moment, and the state update equation (i.e., the update stage) is responsible for combining the prior estimate and the new measurement variable to construct an improved posterior estimate.

[0063] Time update equation:

[0064] ;

[0065] ;

[0066] In the formula, and Respectively Moment and The posterior state estimate at time , express The prior state estimate at time , and Respectively Moment and The posterior estimated covariance at time , A T express A The transposed matrix, Q, represents the covariance of the excitation noise.

[0067] The state update equation is as follows:

[0068] ;

[0069] ;

[0070] ;

[0071] In the formula, represents the filter gain, represents the transformation matrix from state variables to measurements (observations), H T express H The transposed matrix, represents the covariance of the observation noise, I Represents the identity matrix.

[0072] Since the linear system shown in the above state estimation formula is only valid in theory, the state estimation in practical applications is often not a linear system, and filtering using this method will produce a large error. Therefore, this embodiment needs to correct the error, and at this time, it is first necessary to obtain the parameters that will affect the Kalman filter estimation error, that is, the optimal gain for wind disturbance compensation of the large-aperture reflector antenna obtained by Kalman filter estimation in this embodiment.

[0073] In a specific implementation, the optimal gain for wind disturbance compensation of the large-aperture reflector antenna estimated by the Kalman filter specifically includes the difference between the state prediction value and the state estimation value of the large-aperture reflector antenna by the Kalman filter. , the difference between the observed value and the estimated observed value And the filter gain The Kalman filter estimates the optimal gain for wind disturbance compensation of the large-aperture reflector antenna through the following data: the speed output command of the large-aperture reflector antenna, the current azimuth angle, the pitch angle, the antenna pointing deviation, the equivalent force area, and the wind speed, wind direction, temperature, humidity, and air pressure of the large-aperture reflector antenna. These data are the input of the Kalman filter (the state parameters of the large-aperture reflector antenna).

[0074] In step 202 and step 203, the network structure used by the error prediction model described in this embodiment is established on the basis of the classic three-layer fully connected neural network, combined with the local receptive field and shared weight concepts of the convolutional network. The network structure used by the error prediction model can be found in Figure 3 The error prediction model mainly consists of three parts. The first part is the data conversion module, which is used to flatten the control parameters into a one-dimensional vector and then convert it into a two-dimensional matrix; the second part is the Cross scan module, which is used to scan from different directions of the two-dimensional matrix and mine the correlation features between adjacent elements in different directions of the two-dimensional matrix; the third part is the dilated convolution module, which is used to mine the correlation features between non-adjacent elements in different directions of the two-dimensional matrix.

[0075] Regarding the data conversion module, for the data of each model (i.e., the optimal gain), if the optimal gain is multiple dimensions, the module first extracts the feature information of each dimension of the optimal gain according to the order of the multiple dimensions, and then stores all the feature information in a one-dimensional vector to flatten the control parameters into a one-dimensional vector, and finally projects the one-dimensional vector into a two-dimensional matrix through a preset projection matrix.

[0076] In the specific implementation, before projecting the one-dimensional vector into a two-dimensional matrix through the preset projection matrix, the features of each dimension in the one-dimensional vector must be normalized first, so that the value of the feature of each dimension in the normalized one-dimensional vector is in [0,1]. In order to complete the vector normalization operation, batch collection Each feature One by one Dimension information Perform a normalization operation, in which all Moment The maximum value of and minimum value , after normalization ; In subsequent new sample operations, if there is an The value of the range is first constrained to the range , and then normalize.

[0077] The error prediction model converts the three input quantities (i.e. control parameters) into one-dimensional vectors through the above operations, and then connects them in series to obtain an integrated input vector Then, through the preset projection matrix Function, get the mapped two-dimensional matrix . The function of is to project a one-dimensional vector into a two-dimensional matrix, which is convenient for subsequent convolution model operations. and The relationship cannot be described directly, so the The parameters need to be obtained through a large number of sample training. Here, As part of the model of this solution, it is trained and learned together with the parameters of subsequent modules. The size of and The dimension of is determined by the matrix, and the elements in the matrix are determined in the training phase like other parameters in the network. In other words, the mapping described here is to flatten the previous output to form a vector, and then construct a linear mapping relationship to obtain .

[0078] Regarding the Cross scan module, it specifically scans the two-dimensional matrix horizontally, vertically, inversely and vertically, and reorders the adjacent elements in the two-dimensional matrix during the horizontal, vertical, inversely and vertical scanning process to extract the association between each element in the two-dimensional matrix and the adjacent elements in different directions, thereby obtaining the association features between adjacent elements in different directions of the two-dimensional matrix. Different from the indiscriminate convolution operation of conventional convolution within the local receptive field, its approach is to weight and sum the elements in all rows and columns in the region at the same time, where the weight is the value of the convolution kernel matrix.

[0079] By introducing a new convolution operator Cross Scan scanning module, the feature fusion method in different directions is used to reduce the calculation amount of a single scan, and the correlation between adjacent information in different directions is mined by scanning in different directions. The reason why the feature fusion method is used here is that the matrix after projection is There may be special correlations between features connected in different directions in the feature information. The CrossScan module can be used to better mine relevant information and extract more valuable features. At the same time, the CrossScan module can improve computational efficiency by scanning in one direction compared to traditional convolution operations.

[0080] Regarding the dilated convolution module, it is specifically composed of dilated convolution operators of different scales. In the process of the Cross scan module scanning the two-dimensional matrix horizontally, vertically, reverse horizontally and reverse vertically, the dilated convolution operators of different scales in the dilated convolution module are used to increase the step spacing of the Cross scan module, so as to extract the correlation relationship between each element in the two-dimensional matrix and non-adjacent elements in different directions, and obtain the correlation features between non-adjacent elements in different directions of the two-dimensional matrix.

[0081] Since the Cross Scan module operates by reordering adjacent elements in different directions, the relationship between non-adjacent positions has not been fully explored. Therefore, this embodiment adds dilated convolution modules of different scales at this position to achieve feature fusion between non-adjacent elements by increasing the step size of operator operations during feature fusion.

[0082] Based on the above, the network structure used in the error prediction model of this embodiment is based on a convolutional neural network, wherein the convolution operator is a basic operator of the convolutional neural network, and the convolution kernel is the matrix value in the sliding box (matrix form) in the convolution operation. The convolutional neural network forms local perception ability through local connection, and at the same time completes a significant reduction in the parameters to be trained, reduces the difficulty of network training, and further reduces the number of parameters to be trained by weight sharing while keeping the amount of calculation unchanged. The traditional convolution structure operates on all points in the neighborhood for each pixel position, while the Cross scan scanning module only fuses in one direction for each operation, has high operating efficiency, and can freely combine specific directions as needed. The difference between the dilated convolution and the traditional convolution is that the pixel position processed by the discrete receptive field can be feature fused, which increases the receptive field range without increasing the complexity of the convolution operation. Therefore, the model can perform multiple error associations on information of different dimensions, and then mine the potential relationship between information of different dimensions, thereby further improving the processing ability of the network model.

[0083] In step 204, the error prediction model in the above steps also includes a fully connected layer in actual implementation. The fully connected layer of the error prediction model can output a correction amount corresponding to the optimal gain (i.e., model input data) according to the correlation characteristics between adjacent elements in different directions of the two-dimensional matrix and the correlation characteristics between non-adjacent elements. In this step, when optimizing the optimal gain output by the Kalman filter through the correlation characteristics between adjacent elements in different directions of the two-dimensional matrix and the correlation characteristics between non-adjacent elements, the correction amount output by the error prediction model is specifically added to the optimal gain to obtain the optimized optimal gain.

[0084] like Figure 4As shown, the input parameters are the state parameters of the large-aperture reflector antenna. The Kalman filter outputs the optimal gain based on it and uses it as the input of the error prediction model. The output of the error prediction model is , where is the theoretical value of the target state vector. After the model is trained, the model output (i.e. the correction amount corresponding to the optimal gain) and the result obtained by Kalman filtering Adding them together, we can get the optimized optimal gain.

[0085] In some embodiments, the above-mentioned error prediction model is specifically trained through the following strategy: first, the optimal gain obtained by the Kalman filter is obtained as sample data, then Gaussian noise is added to the sample data, and the error prediction model is trained using the sample data with added noise, and during the training process, the intensity of the Gaussian noise is gradually increased so that the error prediction model is trained using sample data with different noise intensities.

[0086] Since the collected data related to wind disturbance intensity, control torque, etc. have certain errors, using such data as training samples will inevitably affect the network parameters (mapping matrix And the parameters of the three dilated convolution modules) have a certain impact on the training. Moreover, factors such as wind in the environment where large antennas are located are changeable, and there is no fixed pattern for its direction and strength. Therefore, it is difficult to reasonably analyze and predict external disturbance factors such as wind. Therefore, the adaptability of the network model to disturbances plays a great role in the prediction ability of the network. This embodiment uses the idea of ​​diffusion model to actively add noise and reverse learning to the network model training to improve its anti-interference performance, thereby improving the prediction accuracy of the network in complex scenarios.

[0087] The schematic diagram of denoising process based on diffusion model is shown in Figure 5 , during the training process, through a fixed forward diffusion process Gradually add Gaussian noise to the sample features until pure noise is finally obtained. Then use the trainable reverse denoising diffusion process , train a neural network, starting from pure noise and gradually denoising until a true The number of steps forward and backward is represented by the subscript Defined, and has a predefined total number of steps .

[0088] And given that Figure 3 The difference between the error prediction model and the generative problem shown in the figure cannot directly use the diffusion model to improve the training effect. Therefore, when using the diffusion model, Figure 3 After optimization, the structure is Figure 6 As shown in Figure 1, the three inputs (i.e., control parameters) and one output (i.e., the output of the error prediction model) are taken as a whole, and a unified denoising and reverse denoising process based on the diffusion model is performed. The process adopted is still a fixed forward diffusion process. Gradually add Gaussian noise to the image until you end up with pure noise, then Figure 6 The error prediction model in is used as a trainable Starting from pure noise, we gradually remove the noise until we get a real four-component system. Through active noise training, we can improve the error prediction model's ability to deal with disturbances caused by noise in different parameters and enhance the network's ability to resist disturbances. Figure 6 The new structure Figure 3 The structure in is further optimized, turning the prediction and estimation process into a cyclic iterative generation process.

[0089] In step 205, after the optimized optimal gain is obtained, the wind disturbance compensation is performed on the large-aperture reflector antenna, so that more accurate antenna pointing can be achieved.

[0090] The step division of the various methods above is only for clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the protection scope of the present invention. Adding insignificant modifications or introducing insignificant designs to the algorithm or process without changing the core design of the algorithm and process are all within the protection scope of the invention.

[0091] Another embodiment of the present invention relates to a wind disturbance compensation control system for a large-aperture reflector antenna. The implementation details of the wind disturbance compensation control system for the large-aperture reflector antenna of this embodiment are specifically described below. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The schematic diagram of the wind disturbance compensation control system for the large-aperture reflector antenna of this embodiment can be as follows Figure 7 As shown, it includes: a gain acquisition module 701, a model solving module 702, a gain optimization module 703 and a wind disturbance compensation module 704.

[0092] Specifically, the gain acquisition module 701 is used to input the state parameters of the large-aperture reflector antenna into the Kalman filter, perform state estimation on the large-aperture reflector antenna through the Kalman filter, and obtain the optimal gain for wind disturbance compensation of the large-aperture reflector antenna.

[0093] A model solving module 702, used for inputting the optimal gain into the trained error prediction model, wherein the error prediction model includes a data conversion module, a cross scan module and a dilated convolution module;

[0094] The feature information in the optimal gain is extracted through the data conversion module, the feature information is stored in a one-dimensional vector, and the one-dimensional vector is converted into a two-dimensional matrix through a preset projection matrix; the correlation features between adjacent elements in different directions of the two-dimensional matrix are mined through the Cross scan module, and the correlation features between non-adjacent elements in different directions of the two-dimensional matrix are mined through the expansion convolution module.

[0095] The gain optimization module 703 is used to optimize the optimal gain obtained by the Kalman filter through the correlation characteristics between adjacent elements in different directions of the two-dimensional matrix and the correlation characteristics between non-adjacent elements.

[0096] The wind disturbance compensation module 704 is used to perform wind disturbance compensation on the large-aperture reflector antenna using the optimized optimal gain.

[0097] It is not difficult to find that this embodiment is a system embodiment corresponding to the above method embodiment, and this embodiment can be implemented in conjunction with the above method embodiment. The relevant technical details and technical effects mentioned in the above embodiment are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the above embodiment.

[0098] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by the present invention, but this does not mean that there are no other units in this embodiment.

[0099] Another embodiment of the present invention relates to a computer device, such as Figure 8 As shown, it includes: at least one processor 801; and a memory 802 that is communicatively connected to the at least one processor 801; wherein the memory 802 stores instructions that can be executed by the at least one processor 801, and the instructions are executed by the at least one processor 801 so that the at least one processor 801 can execute the large-aperture reflector antenna wind disturbance compensation control method in the above-mentioned embodiments.

[0100] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0101] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0102] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0103] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as: ROM), random access memory (Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.

[0104] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present invention, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A method for controlling wind disturbance compensation of a large-aperture reflector antenna, characterized in that: include: The state parameters of the large-aperture reflector antenna are input into the Kalman filter, and the state of the large-aperture reflector antenna is estimated by the Kalman filter to obtain the optimal gain for wind disturbance compensation of the large-aperture reflector antenna; Inputting the optimal gain into a trained error prediction model, wherein the error prediction model includes a data conversion module, a cross scan module and a dilated convolution module; The feature information in the optimal gain is extracted through the data conversion module, the feature information is stored in a one-dimensional vector, and the one-dimensional vector is converted into a two-dimensional matrix through a preset projection matrix; the correlation features between adjacent elements in different directions of the two-dimensional matrix are mined through the cross scan module, and the correlation features between non-adjacent elements in different directions of the two-dimensional matrix are mined through the expansion convolution module; The optimal gain obtained by the Kalman filter is optimized through the correlation characteristics between adjacent elements in different directions of the two-dimensional matrix and the correlation characteristics between non-adjacent elements; The optimized optimal gain is used to compensate for wind disturbance of large-aperture reflector antennas.

2. The large-aperture reflector antenna wind disturbance compensation control method according to claim 1, characterized in that: The cross scan module is used to mine the association features between adjacent elements in different directions of the two-dimensional matrix, including: The Cross scan module is used to scan the two-dimensional matrix horizontally, vertically, anti-horizontally and anti-vertically respectively. During the horizontal, vertical, anti-horizontally and anti-vertical scanning processes, the correlation relationship between each element in the two-dimensional matrix and adjacent elements in different directions is extracted to obtain the correlation features between adjacent elements in different directions of the two-dimensional matrix.

3. The large-aperture reflector antenna wind disturbance compensation control method according to claim 1, characterized in that: The dilated convolution module is composed of dilated convolution operators of different scales, and the dilated convolution module is used to mine the correlation features between non-adjacent elements in different directions of the two-dimensional matrix, including: The step size of the dilated convolution module is increased by using dilated convolution operators of different scales in the dilated convolution module to extract the correlation relationship between each element in the two-dimensional matrix and non-adjacent elements in different directions, thereby obtaining the correlation features between non-adjacent elements in different directions of the two-dimensional matrix.

4. The large-aperture reflector antenna wind disturbance compensation control method according to claim 1, characterized in that: The error prediction model also includes a fully connected layer, and the optimal gain of the Kalman filter output is optimized through the association features between adjacent elements in different directions of the two-dimensional matrix and the association features between non-adjacent elements, including: The fully connected layer is made to output the correction amount corresponding to the optimal gain according to the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements; The correction amount output by the error prediction model is added to the optimal gain to obtain the optimized optimal gain.

5. The large-aperture reflector antenna wind disturbance compensation control method according to claim 1, characterized in that: The error prediction model is trained in the following way: Obtain the optimal gain obtained by the Kalman filter as sample data; Gaussian noise is added to the sample data, and the error prediction model is trained using the sample data after the noise is added. During the training process, the intensity of the Gaussian noise is gradually increased so that the error prediction model is trained using sample data with different noise intensities.

6. The method for controlling wind disturbance compensation of a large-aperture reflector antenna according to any one of claims 1 to 5, characterized in that: The state parameters of the large-aperture reflector antenna include: The speed output command of the large-aperture reflector antenna, the current azimuth angle, pitch angle, antenna pointing deviation, equivalent force area, and the wind speed, wind direction, temperature, humidity, and air pressure of the large-aperture reflector antenna's environment.

7. The large-aperture reflector antenna wind disturbance compensation control method according to claim 6, characterized in that: The optimal gain includes the difference between the state prediction value and the state estimation value, the difference between the observation value and the estimated observation value, and the filter gain of the Kalman filter for the large-aperture reflector antenna.

8. A wind disturbance compensation control system for a large-aperture reflector antenna, characterized in that: include: A gain acquisition module is used to input the state parameters of the large-aperture reflector antenna into a Kalman filter, estimate the state of the large-aperture reflector antenna through the Kalman filter, and obtain the optimal gain for wind disturbance compensation of the large-aperture reflector antenna; A model solving module, used for inputting the optimal gain into the trained error prediction model, wherein the error prediction model includes a data conversion module, a cross scan module and a dilated convolution module; The feature information in the optimal gain is extracted through the data conversion module, the feature information is stored in a one-dimensional vector, and the one-dimensional vector is converted into a two-dimensional matrix through a preset projection matrix; the correlation features between adjacent elements in different directions of the two-dimensional matrix are mined through the cross scan module, and the correlation features between non-adjacent elements in different directions of the two-dimensional matrix are mined through the expansion convolution module; A gain optimization module, used for optimizing the optimal gain obtained by the Kalman filter through the correlation features between adjacent elements in different directions of the two-dimensional matrix and the correlation features between non-adjacent elements; The wind disturbance compensation module is used to compensate for the wind disturbance of the large-aperture reflector antenna by using the optimized optimal gain.

9. A computer device, characterized in that: include: at least one processor; And, a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the large-aperture reflector antenna wind disturbance compensation control method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the wind disturbance compensation control method for a large-aperture reflector antenna according to any one of claims 1 to 7 is implemented.

Images