A satellite fault early warning method and ground station
By obtaining satellite attitude information and using neural network to determine the attitude type and outputting corresponding early warning information, the problem of accurate prediction difficulties caused by fixed early warning thresholds in the prior art is solved, and flexible and accurate early warning of satellite failures is achieved.
Patent Information
- Application Number
- CN202410882348.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-03
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-07-03
AI Technical Summary
The prior art uses a fixed warning threshold in satellite attitude information warning, which makes it impossible to accurately predict satellite failures and cannot adapt to changes in different scenarios.
By obtaining multiple sets of attitude information of satellites at multiple moments, determining their attitude type, using neural network to convert attitude information into feature maps, and outputting corresponding early warning information based on the feature maps, avoiding the use of fixed early warning thresholds, and flexibly adjusting the early warning method .
Accurate prediction of satellite failures is achieved, and the early warning method is flexibly changed according to different scenarios, which improves the accuracy and adaptability of the early warning.
Smart Images

Figure CN118419285B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of satellite engineering technology, and in particular to a satellite fault early warning method and a ground station. Background Art
[0002] Satellite attitude information can characterize the attitude change process of the satellite during its in-orbit operation. Accurate and reliable attitude information is crucial to the normal operation of the satellite. During the operation of the satellite, the attitude information can be processed and then transmitted to the ground.
[0003] The current early warning method for telemetry data is to analyze the satellite attitude information generated during the satellite's in-orbit operation, predict possible satellite failures, and issue a warning when any parameter in the satellite attitude information exceeds the set warning threshold. However, using a fixed warning threshold regardless of the scenario may make it impossible to accurately predict satellite failures. Summary of the invention
[0004] In order to solve the above-mentioned problems in the prior art, the present application example provides a satellite fault warning method and a ground station, which can use different methods to warn the satellite attitude information according to the satellite attitude corresponding to the satellite attitude information.
[0005] In a first aspect, an embodiment of the present application provides a satellite fault warning method, which can be applied to a ground station, and the method includes:
[0006] Obtain multiple sets of satellite attitude information at multiple times;
[0007] Determining the attitude type of the satellite according to the multiple sets of satellite attitude information;
[0008] If the attitude type of the satellite is the first attitude type, outputting warning information for the satellite.
[0009] In a possible implementation, determining the attitude type of the satellite according to the multiple sets of satellite attitude information includes:
[0010] Determine the deviation between each set of satellite attitude information and set standard attitude information in the multiple sets of satellite attitude information;
[0011] The deviation between each set of satellite attitude information and the standard attitude information is expressed as a state characteristic value to obtain a plurality of state characteristic values;
[0012] Converting the plurality of state feature values into a feature map;
[0013] The feature map is input into a trained neural network to obtain the posture type output by the neural network.
[0014] In one possible implementation, each set of satellite attitude information includes an attitude angle and an angular velocity; the standard attitude information includes a standard attitude angle and a standard angular velocity; the deviation between each set of satellite attitude information and the standard attitude information includes a deviation between the attitude angle in each set of satellite attitude information and the standard attitude angle, and a deviation between the angular velocity in each set of satellite attitude information and the standard angular velocity.
[0015] In a possible implementation, expressing the deviation between each set of satellite attitude information and the standard attitude information as a state characteristic value includes:
[0016] For each set of satellite attitude information, perform the following operations respectively:
[0017] The state characteristic value corresponding to the satellite attitude information is determined according to the weighted sum of the squares of each deviation component in the multiple deviation components corresponding to the satellite attitude information; the multiple deviation components include the deviation between the attitude angle in the satellite attitude information and the standard attitude angle, including the attitude angle deviation component of the X-axis, the attitude angle deviation component of the Y-axis, and the attitude angle deviation component of the Z-axis; and the deviation between the angular velocity in the satellite attitude information and the standard angular velocity includes the angular velocity deviation component of the X-axis, the angular velocity deviation component of the Y-axis, and the angular velocity deviation component of the Z-axis.
[0018] In a possible implementation, the neural network includes a plurality of pooling layers, each of which is connected to a random dropout layer; the random dropout layer is used to randomly discard output results of the pooling layer connected to the random dropout layer.
[0019] In a possible implementation, the acquiring multiple sets of satellite attitude information of the satellite at multiple times includes:
[0020] Receiving multiple groups of satellite attitude information sent by the satellite;
[0021] If the multiple sets of satellite attitude information are compressed multiple sets of satellite attitude information, the compressed multiple sets of satellite attitude information are decompressed.
[0022] In a possible implementation, if the attitude type of the satellite is the first attitude type, after outputting the warning information for the satellite, the method further includes:
[0023] receiving a control instruction for the satellite input by a user;
[0024] The control instruction is sent to the satellite; the control instruction is used to guide the satellite to adjust its attitude.
[0025] In a possible implementation, after determining the attitude type of the satellite according to the multiple sets of satellite attitude information, the method further includes:
[0026] If the attitude type of the satellite is the second attitude type, information indicating that the attitude of the satellite is normal is output.
[0027] In a possible implementation, after determining the attitude type of the satellite according to the multiple sets of satellite attitude information, the method further includes:
[0028] If the attitude type of the satellite is the third attitude type, outputting prompt information for evaluating the satellite;
[0029] The evaluation information for the satellite input by a user is received.
[0030] In a possible implementation manner, after receiving the evaluation information for the satellite input by the user, the method further includes:
[0031] If the evaluation information indicates that the satellite is in a healthy state, sending an instruction to the satellite to continue to perform the mission;
[0032] If the evaluation information indicates that the satellite is in an unhealthy state, an instruction to stop executing the mission is sent to the satellite.
[0033] In a possible implementation, if the attitude type of the satellite is the first attitude type, after outputting the warning information for the satellite, the method further includes:
[0034] Taking the Nth group of satellite attitude information from the multiple groups of satellite attitude information as a starting point, selecting the same number of satellite attitude information as the multiple groups of satellite attitude information; wherein N is a set value;
[0035] The attitude type of the satellite is determined according to the satellite attitude information having the same number as the multiple groups of satellite attitude information.
[0036] In a second aspect, an embodiment of the present application provides a satellite failure warning device, the device comprising:
[0037] An acquisition unit, used to acquire multiple sets of satellite attitude information of the satellite at multiple times;
[0038] A determination unit, configured to determine the attitude type of the satellite according to the multiple sets of satellite attitude information;
[0039] The early warning unit is used to output early warning information for the satellite if the attitude type of the satellite is the first attitude type.
[0040] In a third aspect, an embodiment of the present application provides a ground station, including a memory and a processor, wherein the memory stores a computer program running on the processor, and when the computer program is executed by the processor, the above-mentioned satellite fault warning method can be implemented.
[0041] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the above-mentioned satellite fault warning method can be implemented.
[0042] The embodiment of the present application provides a satellite fault warning method and ground station. After obtaining multiple sets of satellite attitude information of the satellite at multiple times, the attitude type of the satellite is determined according to the multiple sets of satellite attitude information. If the attitude type of the satellite is a first attitude type, the warning information corresponding to the first attitude type is displayed to warn the user of possible problems with the satellite. By determining the corresponding warning information displayed for different satellite attitude types, there is no need to adopt a fixed warning threshold, and the warning method can be flexibly changed according to different scenarios, and the possible faults of the satellite can be accurately predicted. BRIEF DESCRIPTION OF THE DRAWINGS
[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0044] Figure 1 An application scenario diagram of a satellite fault warning method provided in an embodiment of the present application;
[0045] Figure 2 A flowchart of a method for executing satellite failure warning provided in an embodiment of the present application;
[0046] Figure 3 A schematic diagram of satellite attitude angle values provided in an embodiment of the present application;
[0047] Figure 4 A schematic diagram of the value of a satellite angular velocity provided in an embodiment of the present application;
[0048] Figure 5 A schematic diagram of multiple state characteristic values determined according to a satellite attitude angle and a satellite angular velocity provided in an embodiment of the present application;
[0049] Figure 6 A schematic diagram of the correspondence between a Gram angle and a field diagram and a satellite attitude type provided in an embodiment of the present application;
[0050] Figure 7 A schematic diagram of generating a data set provided in an embodiment of the present application;
[0051] Figure 8 A specific flow chart of a satellite fault warning method performed by a satellite according to an embodiment of the present application;
[0052] Fig. 9 A structural block diagram of a satellite fault warning device provided in an embodiment of the present application;
[0053] Fig.10 A structural block diagram of a ground station provided in an embodiment of the present application. DETAILED DESCRIPTION
[0054] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.
[0055] It should be noted that the terms "including" and "having" and their variations involved in the documents of this application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0056] The terms "first" and "second" in this document are used for descriptive purposes only and should not be understood as explicitly or implicitly indicating relative importance or implicitly indicating the number of technical features indicated. Therefore, features defined as "first" and "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.
[0057] The word “exemplary” is used hereinafter to mean “serving as an example, example, or illustration.” Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.
[0058] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0059] Figure 1 FIG. 1 shows an application scenario diagram of a satellite fault warning method provided by an embodiment of the present application, see Figure 1 As shown,
[0060] The application scenario includes: a satellite 101 and a ground station (102a, 102b), and the satellite 101 and the ground station (102a, 102b) can exchange information through a wireless communication network. The ground station (102a, 102b) can also be called a ground base station (102a, 102b). The following description is based on the ground station.
[0061] Exemplarily, the satellite 101 can access the network through cellular mobile communication technology to communicate with the ground station (102a, 102b), wherein the cellular mobile communication technology, for example, includes the fifth generation mobile communication (5th Generation Mobile Networks, 5G) technology.
[0062] The embodiment of the present application does not impose any restriction on the number of communication devices involved in the above application scenario. For example, there may be multiple ground stations, or no ground station, or other network / communication devices may be included, such as Figure 1 As shown, only the satellite 101 and the ground station (102a, 102b) are taken as an example for description; and, a data sending device 103 is deployed on the satellite 101, and a data receiving device, namely, a data receiving device 104a and a data receiving device 104b, is deployed on the ground station (102a, 102b).
[0063] The data transmission device 103 deployed on the satellite 101 can be used to transmit telemetry data to the data receiving device (104a, 104b) deployed on the ground station (102a, 102b). It should be noted that the data transmission device 103 can transmit uncompressed telemetry data or compressed telemetry data, and the present application does not limit whether the telemetry data is compressed.
[0064] The data receiving device (104a, 104b) deployed on the ground station (102a, 102b) is used to receive the telemetry data sent from the data sending device 103. If the data sending device 103 compresses the telemetry data before sending it, the data receiving device (104a, 104b) deployed on the ground station (102a, 102b) decodes and decompresses the compressed data to obtain corresponding telemetry data; wherein the compressed data includes: video stream data after video compression encoding and / or single frame image data after image compression encoding.
[0065] It should be noted that single-frame image data is the information contained in a single-frame image. Therefore, in the embodiments of the present application, single-frame image data and single-frame image refer to the same thing and there is no need to distinguish them.
[0066] Exemplarily, assuming that the compressed data is video stream data after video compression encoding, the video stream data after video compression encoding can be decoded and decompressed to obtain corresponding telemetry data; similarly, assuming that the compressed data is single-frame image data after image compression encoding, the single-frame image data after image compression encoding can be decoded and decompressed to obtain corresponding telemetry data.
[0067] After receiving the telemetry data sent from the data sending device 103, the satellite fault warning device deployed on the ground station (102a, 102b) can determine the attitude type of the satellite according to the multiple sets of satellite attitude information. If the attitude type of the satellite is the first attitude type, the warning information corresponding to the first attitude type is displayed to warn the user of possible problems with the satellite. By determining the corresponding warning information for different satellite attitude types, there is no need to adopt a fixed warning threshold, and the warning method can be flexibly changed according to different scenarios, and the possible faults of the satellite can be accurately predicted.
[0068] It should also be noted that in the above application scenarios, there may be no ground station (102a, 102b), that is, the data receiving device (104a, 104b) and the satellite fault warning device can be widely used in terminal devices that can receive data sent by spacecraft.
[0069] The satellite fault warning method provided by the exemplary embodiment of the present application is described below in combination with the above-mentioned application scenarios and with reference to the accompanying drawings. It should be noted that the above-mentioned application scenarios are only shown to facilitate understanding of the spirit and principles of the present application, and the implementation methods of the present application are not limited in this regard.
[0070] Figure 2 A flowchart of a satellite fault warning method provided by an embodiment of the present application is shown. Figure 2 As shown, the satellite failure early warning method may include the following steps:
[0071] Step S201, obtaining multiple sets of satellite attitude information of the satellite at multiple times.
[0072] In a possible embodiment, the satellite usually includes various sensors that can measure the direction, angle or position of the satellite relative to reference objects such as the earth, the sun, and stars. For example, the satellite may include a gyroscope, which can calculate the angle and direction changes of the satellite by integrating the measurement values, and can also measure the angular velocity and angular acceleration of the satellite; the satellite may also include a solar sensor, which can detect the position and direction of the sun to determine the satellite's attitude relative to the sun; it may also include sensors such as earth magnetic field sensors and inertial measurement units.
[0073] In a possible embodiment, the satellite attitude information may include an attitude angle describing the orientation of the satellite in three-dimensional space, an angular velocity describing the rotation speed of the satellite around each axis, an angular acceleration describing the rate of change of the satellite angular velocity, and other values that can describe the satellite attitude. Such attitude information is essential for the satellite's operation, communication, navigation, earth observation, and other tasks. Specifically, the satellite can obtain the satellite attitude information through various sensors included on the satellite.
[0074] It should be noted that the satellite does not acquire satellite attitude information all the time, but has a fixed sampling frequency. In general, the satellite attitude control system that can control the attitude of the satellite is limited by factors such as the output capacity of the actuator on the satellite and the satellite inertia characteristics. The satellite control system does not frequently send attitude change signals to the actuator on the satellite, and the frequency of the satellite control system sending attitude change signals to the actuator on the satellite is also different for different satellite models. Therefore, in this application, it is recommended to set the sampling frequency to a value that is one order of magnitude higher than the frequency at which the satellite attitude control system sends attitude change signals to the actuator on the satellite. For example, if the frequency of the satellite attitude control system is 0.016~0.4Hz, the sampling frequency for acquiring satellite attitude information can be set to 4Hz, that is, the satellite attitude information is acquired every 0.25 seconds. In this application, the sampling frequency for acquiring satellite attitude information is not limited, and can be 5Hz or 3Hz or other values. The following description will be based on the example of setting the sampling frequency for acquiring satellite attitude information to 4Hz.
[0075] In a possible embodiment, after the satellite acquires multiple sets of satellite attitude information at multiple times, it can send the multiple sets of satellite attitude information at multiple times to a ground station. The ground station can receive the multiple sets of satellite attitude information sent by the satellite. If the multiple sets of satellite attitude information received by the ground station are compressed sets of satellite attitude information, the compressed sets of satellite attitude information are decompressed, and then the multiple sets of satellite attitude information can be obtained.
[0076] Step S202: determining the attitude type of the satellite according to multiple sets of satellite attitude information.
[0077] In a possible embodiment, the deviation between the attitude angle in each group of satellite attitudes and the standard attitude angle, and the deviation between the angular velocity in each group of satellite attitude information and the standard angular velocity can be determined first in multiple groups of satellite attitude information, and then multiple characteristic values are used to respectively characterize the deviation between the attitude angle in each group of satellite attitudes and the standard attitude angle, and the deviation between the angular velocity in each group of satellite attitude information and the standard angular velocity. Since the attitude angle characterizes the orientation of the satellite in three-dimensional space, the attitude angle can include the attitude angles of the X-axis, Y-axis, and Z-axis respectively; similarly, the angular velocity also includes the angular velocities of the X-axis, Y-axis, and Z-axis respectively.
[0078] That is to say, a state characteristic value can be used to indicate whether the attitude angles of the X-axis, Y-axis and Z-axis and the angular velocities of the X-axis, Y-axis and Z-axis in a set of satellite attitude information deviate from their corresponding standard values. If any one or more of the six values deviate from the corresponding standard values, the state characteristic value can reflect the deviation. Among them, the standard attitude is the optimal operating attitude of the satellite in orbit, and the satellite can achieve the highest operating efficiency under the optimal operating attitude; the standard attitude angle is the attitude angle of the X-axis, Y-axis and Z-axis of the satellite under the optimal operating attitude; the standard angular velocity is the angular velocity of the X-axis, Y-axis and Z-axis of the satellite under the optimal operating attitude.
[0079] Specifically, the satellite attitude generalized energy function V(x) can be selected as the determination function of the state eigenvalue, and the expression of V(x) is as follows:
[0080]
[0081] Among them, E x 、E y 、E z They respectively represent the dimensionless satellite X-axis attitude angle, which is the ratio of the actually measured X-axis attitude angle to the X-axis standard attitude angle index; the dimensionless satellite Y-axis attitude angle, which is the ratio of the actually measured Y-axis attitude angle to the Y-axis standard attitude angle index; the dimensionless satellite Z-axis attitude angle, which is the ratio of the actually measured Z-axis attitude angle to the Z-axis standard attitude angle index; They represent the dimensionless satellite three-axis deviation attitude angular velocity, which is the ratio between the actual measured angular velocity and the standard angular velocity; It represents the motion state coefficient. The motion state coefficient can be selected as the product of the proportional coefficient of the satellite's moment of inertia and the proportional coefficient of the criticality of the mission. That is to say, the easier it is to change, the smaller the corresponding motion state coefficient will be. In addition, the sum of the motion state coefficients meets the following conditions: .
[0082] When the satellite attitude gradually converges with the participation of the control system, it is also through E x 、Ey 、E z , The convergence of oscillation is reflected in the convergence characteristics of the energy function, which is also reflected in the state eigenvalue; the failure and performance degradation of the control system, external environmental interference, and movable components on the satellite will also affect the satellite attitude E. x 、E y 、E z and angular velocity The different state fluctuations of , which are reflected in the oscillation, deviation and divergence characteristics of the energy function curve. It should be noted that the method of determining the state characteristic value V(x) in this application is not limited to the method of determining the state characteristic value V(x) by E x 、E y 、E z , The state characteristic value can also be determined directly based on the attitude angle and angular velocity, or it can be determined based on other different values included in the satellite attitude information, which is not limited in this application.
[0083] For example, Figure 3 A schematic diagram of satellite attitude angle values provided in an embodiment of the present application is shown. Figure 3 As shown, the horizontal axis is time, and the vertical axis is the value that the attitude angle can take. Figure 3 In , all the values taken according to the frequency are connected; Figure 4 A schematic diagram of the value of a satellite angular velocity provided in an embodiment of the present application is shown, such as Figure 4 As shown, the horizontal axis is time, and the vertical axis is the value that the satellite angular velocity can take. Figure 4 In , all the values taken according to the frequency are connected; Figure 5 This embodiment of the present application provides a method according to Figure 3 and Figure 4 A schematic diagram of multiple state characteristic values determined, such as Figure 5 As shown, the horizontal axis is time, and the vertical axis is the value that the state characteristic value can take. Figure 5 In , all state eigenvalues are connected.
[0084] In a possible embodiment, a feature map corresponding to a plurality of state feature values may be determined, wherein the plurality of state feature values are one-dimensional time series signals, and the feature map is a two-dimensional image signal. After obtaining the plurality of state feature values, the one-dimensional time series signal may be converted into a two-dimensional image signal using algorithms such as a Gramian Angular Difference Field (GADF), a Gramian Angular Summation Field (GASF), a Markov Transition Field (MTF), and a Recurrence Plot (RP), that is, the plurality of state feature values may be converted into corresponding feature maps.
[0085] For example, the process of using GASF to convert multiple state eigenvalues into corresponding feature maps is roughly as follows: The Gram matrix is a matrix composed of the inner products of the eigenvectors of a one-dimensional signal at each moment. For a one-dimensional time series signal, it can be regarded as the relationship between time and amplitude. By calculating the inner product between the eigenvectors at each moment, a matrix representing the frequency information of the signal can be obtained. This matrix is usually called the Gram matrix. After obtaining the Gram matrix corresponding to multiple state eigenvalues, the values of the Gram matrix can be used as pixel intensities and then rearranged into a rectangular image.
[0086] In a possible embodiment, the feature map can be input into the trained neural network to obtain the attitude type corresponding to the feature map. Among them, the attitude type of the satellite can include seven types: Category 1 "attitude flip state" mainly includes two situations: the satellite detaches from the rocket to adjust the attitude and the attitude flip in the event of a failure of the satellite attitude control system, which requires the attention of operation and maintenance personnel or attitude control experts; Category 2 "attitude maneuvering state", which mainly exists in the process of satellite attitude adjustment, and belongs to the normal attitude state; Category 3 "attitude stable state", refers to the attitude angle and angular velocity are in a stable state, and will hardly change for a period of time, but are not completely zero or close to zero, which belongs to the normal attitude state; Category 4 "attitude jitter state", the satellite may have a failure or be suddenly disturbed by the onboard body and the space environment, Attitude jitter occurs, which is an unexpected attitude change state and requires the attention of operation and control personnel; Category 5 "attitude fluctuation state" may be caused by a fault or continuous disturbance by the satellite body and the space environment, which is an unexpected attitude change state and requires the attention of operation and control personnel; Category 6 "attitude deviation state" means that the satellite attitude has a large pointing deviation, and in severe cases there may be a risk of loss of control. This is an unhealthy attitude change state and requires the attention of operation and maintenance personnel or attitude control experts; Category 7 "satellite attitude failure state" means that the satellite fails, resulting in the inability to obtain satellite attitude data and accurate attitude classification. This is an unhealthy attitude change state and requires the attention of operation and maintenance personnel or attitude control experts. Figure 6 A schematic diagram showing the correspondence between a Gram angle and a field diagram and a satellite attitude type provided in an embodiment of the present application is shown, such as Figure 6 As shown, there is a number corresponding to each Gram angle and field diagram, and the numbers 0 to 6 correspond to the satellite attitude types 1 to 7 respectively.
[0087] It should be noted that the neural network is not trained on the satellite, but offline on the ground. When training the neural network offline on the ground, it is impossible to obtain multiple sets of satellite attitude information at multiple times from the satellite's sensors. However, the data required for training can be obtained from the satellite attitude information saved during the previous operation of the satellite. In the following description, the data required for training obtained from the satellite attitude information saved during the previous operation is called offline data. Similar to the above steps, during the training process of the neural network, multiple state feature values can be used to represent the deviation between each group of satellite attitudes and the standard attitude in the offline data, and then the multiple state feature values can be converted into corresponding feature maps using algorithms such as GADF, GASF, MTF or RP. Finally, the feature maps can be classified and labeled using expert experience. Similarly, the attitude fusion feature curves can be divided into 7 categories according to features such as category 1 "attitude flip state", category 2 "attitude maneuvering state", category 3 "attitude stable state", category 4 "attitude jitter state", category 5 "attitude fluctuation state", category 6 "attitude offset state", and category 7 "satellite attitude fault state", and the corresponding feature images can be saved in the corresponding category folders, and the data can be divided into training set and test set according to the ratio of 98:2 or 8:2.
[0088] Among them, since offline data is not necessarily sufficient, the data set can be generated by data sliding. Figure 7 A schematic diagram of generating a data set provided in an embodiment of the present application is shown, such as Figure 7 As shown, in order to ensure that each segmented data contains enough features and considering the size of the neural network, the length of the data segment can be comprehensively considered. If the sampling frequency of the satellite attitude information is 4Hz, the data of every 0.25 seconds can be selected as the starting point, and the data of 125 seconds backward can be used as the original feature data point corresponding to a single feature image. It should be noted that in some cases, the end of the offline data set can also be used as the starting point, and the data of 125 seconds from the back to the front can be used as the original feature data point corresponding to a single feature image, and the selection of the sampling data frequency and time should be combined with the frequency characteristics of the satellite attitude change. In addition to the above examples, the data of every 0.5 seconds can be selected as the starting point, or the data of every 0.75 seconds can be selected as the starting point, and the data of 100 seconds backward can be selected as the original feature data point corresponding to a single feature image, or the data of 150 seconds backward can be selected as the original feature data point corresponding to a single feature image. This application does not limit this.
[0089] After the preparation of the data set is completed, the data set can be input into a convolutional neural network (CNN) to achieve rapid classification of different satellite attitudes. CNN has the characteristics of weight sharing, local connection and multi-level structure, which can realize feature extraction of nonlinear data. It is particularly suitable for detecting attitude signals with strong nonlinear characteristics. Using CNN as a detector for multi-classification of attitude data, its adaptability and accuracy can be significantly improved. In this application, the training process of the neural network uses cross entropy as the loss function, and the selection of optimizers includes but is not limited to SGD (gradient descent), Adagrad, Adam and other methods. The appropriate optimizer can be selected according to the actual task needs.
[0090] Unlike common CNNs, the neural network in the present application may include multiple convolutional layers, multiple average pooling layers, maximum pooling layers, and multiple matrix multiplication layers, wherein the multiple convolutional layers and the multiple pooling layers may add up to 9 layers. In this way, a certain depth can be added to the traditional CNN model, which can reduce the rate of descent of the gradient of the neural network of the present application and ensure the accuracy of the neural network. However, the present application does not limit the total number of multiple convolutional layers and multiple pooling layers, and the number may be more than 9 layers, or less than 9 layers, which is not limited in the present application.
[0091] In addition, in order to speed up the training time of the neural network in this application and avoid overfitting, a random inactivation Dropout layer can be added after each pooling layer, and the hyperparameters in the Dropout layer are used as the discard probability to randomly discard the data passing through the neural network of this application. Although this may cause the convergence speed of the neural network to slow down, it can reduce the dependency between the neurons in the neural network and reduce the probability of noise and outliers in the neural network training data. Since the data redundancy near the input end is high and the data redundancy near the output end is low, the value of the hyperparameter from the first Dropout layer to the last Dropout layer should be gradually decreasing. A hyperparameter combination of 0.6, 0.3, and 0.1 can be used. After improvement, the classification accuracy of the neural network can reach 98.9% after multiple trainings. It should be noted that this application does not limit the specific values of the hyperparameters in the Dropout layer.
[0092] Therefore, during the training process, the feature map can be input into the convolution layer in the neural network, and the feature map is extracted through multiple convolution layers to obtain a convolution feature map; the convolution feature map is then input into the average pooling layer in the neural network, and the convolution feature map is divided into regions through multiple average pooling layers to determine the average eigenvalue of each region to obtain an average feature map, wherein the convolution layer is connected to the average pooling layer, and the average pooling layer can continue to be connected to the convolution layer until the neural network no longer contains other average pooling layers and convolution layers, and then the average feature map is input into the maximum pooling layer in the neural network, and the average feature map is divided into regions through the maximum pooling layer to determine the maximum eigenvalue of each region to obtain a maximum feature map, wherein each pooling layer is connected to a random inactivation layer, which can be used to randomly discard the output results of the pooling layer connected to the random inactivation layer, and finally, the maximum feature map can be input into multiple matrix multiplication layers in the neural network, and each feature in the maximum feature map is multiplied by a weight through multiple matrix multiplication layers to determine the posture type corresponding to the feature map. By setting multiple convolutional layers, the lower-level convolutional layers usually learn some simple features, such as edges and textures, while the higher-level convolutional layers can learn more complex and abstract features, such as the parts and the whole of an object; setting multiple average pooling layers makes the neural network robust to a certain degree of position changes in the input data, so it can better handle translation and local changes.
[0093] After the training of the neural network is completed by the above method, the attitude type of the satellite can be determined through steps S201 and S202, and then step S203 can be executed.
[0094] Step S203: if the attitude type of the satellite is the first attitude type, output warning information for the satellite.
[0095] In a possible embodiment, the first attitude type may include five attitude types: category 1, category 4, category 5, category 6, and category 7. If the attitude type is category 1 "attitude flip state", category 6 "attitude offset state", and category 7 "satellite attitude fault state", the corresponding satellite attitude information reflects the state of rapid change of satellite attitude, which belongs to abnormal attitude health status, and requires attitude control experts to intervene in time, comprehensively evaluate satellite-related data, and execute corresponding disposal measures to restore the satellite to a normal state as much as possible; if the attitude type is the third attitude type, including: category 4 "attitude jitter state", category 5 "attitude fluctuation state", it means that the satellite attitude data changes in a long period or fixed interval, which belongs to the decline of satellite attitude health status, and the discrimination network will give warning information, and the attitude control expert is required to evaluate the satellite status; if the attitude type is the second attitude type, including: category 2 "attitude maneuvering state" or category 3 "attitude stable state", the satellite attitude data is relatively stable, which belongs to the normal attitude change state, and no warning will be issued for the satellite.
[0096] The embodiment of the present application provides a satellite fault warning method and ground station. After obtaining multiple sets of satellite attitude information of the satellite at multiple times, the attitude type of the satellite is determined according to the multiple sets of satellite attitude information. If the attitude type of the satellite is a first attitude type, the warning information corresponding to the first attitude type is displayed to warn the user of possible problems with the satellite. By determining the corresponding warning information displayed for different satellite attitude types, there is no need to adopt a fixed warning threshold, and the warning method can be flexibly changed according to different scenarios, and the possible faults of the satellite can be accurately predicted.
[0097] In a specific embodiment, the process of a satellite failure early warning method executed by a ground station can be as follows: Figure 8 As shown, the following steps are included:
[0098] Step S801, obtaining multiple sets of satellite attitude information of the satellite at multiple times.
[0099] Step S802: determining the deviation between each set of satellite attitude information and the set standard attitude information in the multiple sets of satellite attitude information.
[0100] Step S803: Express the deviation between each set of satellite attitude information and the standard attitude information as a state characteristic value to obtain multiple state characteristic values.
[0101] Step S804: convert multiple state feature values into a feature graph.
[0102] Step S805, input the feature map into the trained neural network to obtain the posture type output by the neural network.
[0103] Step S806: If the attitude type of the satellite is the first attitude type, output warning information for the satellite.
[0104] Based on the same inventive concept, the present application also provides a satellite fault warning device, such as Fig. 9 As shown, the satellite failure warning device 900 may include:
[0105] An acquisition unit 901 is used to acquire multiple sets of satellite attitude information of a satellite at multiple times;
[0106] A determination unit 902 is used to determine the attitude type of the satellite according to the multiple sets of satellite attitude information;
[0107] The early warning unit 903 is configured to output early warning information for the satellite if the attitude type of the satellite is the first attitude type.
[0108] In a possible implementation, the determination unit 902 is specifically configured to determine a deviation between each set of satellite attitude information and set standard attitude information in the multiple sets of satellite attitude information;
[0109] The deviation between each set of satellite attitude information and the standard attitude information is expressed as a state characteristic value to obtain a plurality of state characteristic values;
[0110] Converting the plurality of state feature values into a feature map;
[0111] The feature map is input into a trained neural network to obtain the posture type output by the neural network.
[0112] In a possible implementation, the determining unit 902 is specifically configured to perform the following operations for each set of satellite attitude information:
[0113] The state characteristic value corresponding to the satellite attitude information is determined according to the weighted sum of the squares of each deviation component in the multiple deviation components corresponding to the satellite attitude information; the multiple deviation components include the deviation between the attitude angle in the satellite attitude information and the standard attitude angle, including the attitude angle deviation component of the X-axis, the attitude angle deviation component of the Y-axis, and the attitude angle deviation component of the Z-axis; and the deviation between the angular velocity in the satellite attitude information and the standard angular velocity includes the angular velocity deviation component of the X-axis, the angular velocity deviation component of the Y-axis, and the angular velocity deviation component of the Z-axis.
[0114] In a possible implementation, the compression unit 903 is further configured to, after compressing the multiple groups of satellite attitude information at the target compression ratio corresponding to the first attitude type, if the first attitude type is a specified type, take the Nth group of satellite attitude information in the multiple groups of satellite attitude information as a starting point, and select the same number of satellite attitude information as the multiple groups of satellite attitude information; N is a set value;
[0115] Determine a first attitude type of the satellite according to the satellite attitude information having the same number as the multiple groups of satellite attitude information.
[0116] In a possible implementation, the acquiring unit 901 is specifically configured to receive multiple sets of satellite attitude information sent by the satellite;
[0117] If the multiple sets of satellite attitude information are compressed multiple sets of satellite attitude information, the compressed multiple sets of satellite attitude information are decompressed.
[0118] In a possible implementation, the early warning unit 903 may also be used to receive a control instruction for the satellite input by a user;
[0119] The control instruction is sent to the satellite; the control instruction is used to guide the satellite to adjust its attitude.
[0120] In a possible implementation, the determination unit 902 may also be configured to output information that the satellite attitude is normal if the attitude type of the satellite is the second attitude type.
[0121] In a possible implementation, the determination unit 902 may also be configured to output prompt information for evaluating the satellite if the attitude type of the satellite is the third attitude type;
[0122] The evaluation information for the satellite input by a user is received.
[0123] In a possible implementation, the determination unit 902 may also be configured to send an instruction to the satellite to continue executing the mission if the evaluation information indicates that the satellite is in a healthy state;
[0124] If the evaluation information indicates that the satellite is in an unhealthy state, an instruction to stop executing the mission is sent to the satellite.
[0125] In a possible implementation, the early warning unit 903 may also be used to select the same number of satellite attitude information as the multiple groups of satellite attitude information, starting from the Nth group of satellite attitude information in the multiple groups of satellite attitude information; N is a set value;
[0126] The attitude type of the satellite is determined according to the satellite attitude information having the same number as the multiple groups of satellite attitude information.
[0127] Based on the same inventive concept, the present application embodiment provides a ground station 1000, which can implement the motor automatic calibration method discussed above, please refer to Fig.10 The ground station 1000 includes a memory 1001, a processor 1002 and a bus 1003.
[0128] The memory 1001 is used to store computer programs executed by the processor 1002. The memory 1001 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and programs required for running the instant messaging function, etc.; the data storage area may store various instant messaging information and operation instruction sets, etc.
[0129] The memory 1001 may be a volatile memory, such as a random-access memory (RAM); the memory 1001 may also be a non-volatile memory, such as a read-only memory, a flash memory, a hard disk drive (HDD) or a solid-state drive (SSD), or the memory 1001 is any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 1001 may be a combination of the above memories.
[0130] The processor 1002 may include one or more central processing units (CPU) or a digital processing unit, etc. The processor 1002 is configured to implement the resource checking method in the above embodiment when calling the computer program stored in the memory 1001 .
[0131] The specific connection medium between the memory 1001 and the processor 1002 is not limited in the embodiment of the present application. Fig.10 In the embodiment, the memory 1001 and the processor 1002 are connected via a bus 1003. The bus 1003 is connected to the processor 1002 via a bus 1003. Fig.10 The connections between other components are shown in bold lines, and are not intended to be limiting. Bus 1003 can be divided into address bus, data bus, control bus, etc. For ease of representation, Fig.10 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.
[0132] Based on the same inventive concept, the embodiment of the present application provides a computer-readable storage medium, and the computer program product includes: computer program code, when the computer program code is executed on a computer, the computer executes any satellite fault warning method as discussed above. Since the principle of solving the problem by the above-mentioned computer-readable storage medium is similar to that of the satellite fault warning method, the implementation of the above-mentioned computer-readable storage medium can refer to the implementation of the method, and the repeated parts will not be repeated.
[0133] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0134] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0135] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0136] 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 executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0137] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and 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 also intended to include these modifications and variations.
Claims
1. A satellite failure early warning method, characterized in that: Applied to a ground station, the method comprises: Obtain multiple sets of satellite attitude information at multiple times; According to the multiple groups of satellite attitude information, the attitude type of the satellite is determined; the attitude type includes three attitude types, wherein the first attitude type includes an attitude flip state, an attitude offset state and a satellite attitude failure state; the second attitude type includes an attitude maneuvering state and an attitude stable state; the third attitude type includes an attitude jitter state and an attitude fluctuation state; If the attitude type of the satellite is the first attitude type, outputting warning information for the satellite; If the attitude type of the satellite is the second attitude type, outputting information that the attitude of the satellite is normal; If the attitude type of the satellite is the third attitude type, outputting prompt information for evaluating the satellite, and receiving evaluation information for the satellite input by a user; The step of determining the attitude type of the satellite according to the multiple sets of satellite attitude information includes: Determine the deviation between each set of satellite attitude information and set standard attitude information in the multiple sets of satellite attitude information; The deviation between each set of satellite attitude information and the standard attitude information is expressed as a state characteristic value to obtain a plurality of state characteristic values; Converting the plurality of state feature values into a feature map; The feature map is input into a trained neural network to obtain the posture type output by the neural network.
2. The method according to claim 1, characterized in that Each set of satellite attitude information includes an attitude angle and an angular velocity; the standard attitude information includes a standard attitude angle and a standard angular velocity; the deviation between each set of satellite attitude information and the standard attitude information includes a deviation between the attitude angle in each set of satellite attitude information and the standard attitude angle, and a deviation between the angular velocity in each set of satellite attitude information and the standard angular velocity.
3. The method according to claim 2, characterized in that The step of expressing the deviation between each set of satellite attitude information and the standard attitude information as a state characteristic value comprises: For each set of satellite attitude information, perform the following operations respectively: The state characteristic value corresponding to the satellite attitude information is determined according to the weighted sum of the squares of each deviation component in the multiple deviation components corresponding to the satellite attitude information; the multiple deviation components include the deviation between the attitude angle in the satellite attitude information and the standard attitude angle, including the attitude angle deviation component of the X-axis, the attitude angle deviation component of the Y-axis, and the attitude angle deviation component of the Z-axis; and the deviation between the angular velocity in the satellite attitude information and the standard angular velocity includes the angular velocity deviation component of the X-axis, the angular velocity deviation component of the Y-axis, and the angular velocity deviation component of the Z-axis.
4. The method according to claim 1, characterized in that: The neural network includes multiple pooling layers, each of which is connected to a random dropout layer; the random dropout layer is used to randomly discard the output results of the pooling layer connected to the random dropout layer.
5. The method according to claim 1, characterized in that The obtaining of multiple sets of satellite attitude information of the satellite at multiple times includes: Receiving multiple groups of satellite attitude information sent by the satellite; If the multiple sets of satellite attitude information are compressed multiple sets of satellite attitude information, the compressed multiple sets of satellite attitude information are decompressed.
6. The method according to claim 1, characterized in that If the attitude type of the satellite is the first attitude type, after outputting warning information for the satellite, the method further includes: receiving a control instruction for the satellite input by a user; The control instruction is sent to the satellite; the control instruction is used to guide the satellite to adjust its attitude.
7. The method according to claim 1, characterized in that After receiving the evaluation information for the satellite input by the user, the method further includes: If the evaluation information indicates that the satellite is in a healthy state, sending an instruction to the satellite to continue to perform the mission; If the evaluation information indicates that the satellite is in an unhealthy state, an instruction to stop executing the mission is sent to the satellite.
8. The method according to claim 1, characterized in that If the attitude type of the satellite is the first attitude type, after outputting warning information for the satellite, the method further includes: Taking the Nth group of satellite attitude information from the multiple groups of satellite attitude information as a starting point, selecting the same number of satellite attitude information as the multiple groups of satellite attitude information; wherein N is a set value; The attitude type of the satellite is determined according to the satellite attitude information having the same number as the multiple groups of satellite attitude information.
9. A satellite failure warning device, characterized in that: The device comprises: An acquisition unit, used to acquire multiple sets of satellite attitude information of the satellite at multiple times; A determination unit is used to determine the attitude type of the satellite according to the multiple groups of satellite attitude information; the attitude type includes three attitude types, wherein the first attitude type includes an attitude flip state, an attitude offset state, and a satellite attitude failure state; the second attitude type includes an attitude maneuvering state and an attitude stable state; and the third attitude type includes an attitude jitter state and an attitude fluctuation state; The step of determining the attitude type of the satellite according to the multiple sets of satellite attitude information includes: Determine the deviation between each set of satellite attitude information and set standard attitude information in the multiple sets of satellite attitude information; The deviation between each set of satellite attitude information and the standard attitude information is expressed as a state characteristic value to obtain a plurality of state characteristic values; Converting the plurality of state feature values into a feature map; Inputting the feature map into a trained neural network to obtain the posture type output by the neural network; an early warning unit, configured to output early warning information for the satellite if the attitude type of the satellite is the first attitude type; If the attitude type of the satellite is the second attitude type, outputting information that the attitude of the satellite is normal; If the attitude type of the satellite is the third attitude type, prompt information for evaluating the satellite is output, and evaluation information for the satellite input by the user is received.
10. A ground station, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program running on the processor, and when the computer program is executed by the processor, the method described in any one of claims 1 to 8 is implemented.
11. A computer program product, characterized in that The computer program product comprises: a computer program code, and when the computer program code is run on a computer, the computer is enabled to execute the method as claimed in any one of claims 1 to 8.
12. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, characterized in that: When the computer program is executed by a processor, the method described in any one of claims 1 to 8 is implemented.
