A design method for spacecraft energy system

Abstract

The present invention relates to a method for designing a spacecraft energy system, comprising the following steps: S1, constructing a general spacecraft power supply and distribution system model; S2, using a priori on-orbit data, inputting a recurrent neural network RNN ​​for data generalization, and realizing the refined mining of a priori empirical data; S3, constructing an actual spacecraft power supply and distribution system model; S4, in the ground design stage, classifying the energy balance and establishing subsamples according to the spacecraft mission sequence, and generating ground test cases; S5, using a data mining method according to the actual ground test data of the spacecraft, correcting the modelable state of the actual spacecraft power supply and distribution system model, and fitting the data for the unmodelable state; S6, in the on-orbit operation stage, using a data mining method according to the on-orbit measured data, the actual spacecraft power supply and distribution system model is corrected on-orbit with high precision. The beneficial effect is that the spacecraft energy subsystem is designed according to the operating environment, the mission sequence, the working state of each single machine and the previous on-orbit data.

Description

[Technical field]

[0001] The present invention relates to the field of aerospace technology, and in particular to a design method for a spacecraft energy system. [Background technology]

[0002] As space missions become increasingly diverse and complex, spacecraft need to conduct detailed verification of each subsystem during the initial design and ground verification process. The energy subsystem is an important support and guarantee for the effective operation of spacecraft in orbit. The reasonable supply and distribution of the energy subsystem is directly related to the effectiveness of the spacecraft in completing the mission. In the past, the design of spacecraft energy subsystems was calculated according to the worst case and maximum envelope, without fully considering the charging efficiency of the sailboard under different working modes and the switching timing of each single machine.

[0003] Recurrent Neural Network (RNN) is a type of neural network that is very effective for data with sequential characteristics and can mine the temporal and semantic information in the data. Both LSTM and GRU are variants of RNN. Long Short-Term Memory (LSTM) is a temporal recurrent neural network that is specially designed to solve the long-term dependency problem of general RNN (recurrent neural network). All RNNs have a chain form of repeated neural network modules. GRU, Gated Recurrent Unit, is a gated recurrent unit that adds several gates to RNN. Its purpose is basically the same as LSTM, which is to enhance the memory capacity of RNN neural network.

[0004] The present invention makes technical improvements to the design method of the spacecraft energy system. [Summary of the invention]

[0005] The purpose of the present invention is to provide a method for designing a spacecraft energy subsystem based on the operating environment, task sequence, the working status of each single machine and previous on-orbit data.

[0006] To achieve the above object, the technical solution adopted by the present invention is a method for designing a spacecraft energy system, comprising the following steps:

[0007] S1. Construct a general spacecraft power supply and distribution system model;

[0008] S2. Use the prior on-orbit data and input it into the recurrent neural network (RNN) for data generalization to achieve refined mining of prior empirical data of the general spacecraft power supply and distribution system.

[0009] S3, build a model of the actual spacecraft power supply and distribution system;

[0010] S4, ground design phase, classifies and establishes subsamples of energy balance according to the spacecraft mission sequence, and generates ground test cases;

[0011] S5. Based on the actual ground test data of the spacecraft, the data mining method is used to correct the modelable state of the actual spacecraft power supply and distribution system model and perform data fitting for the state that cannot be modeled;

[0012] S6. During the on-orbit operation phase, the actual spacecraft power supply and distribution system model is calibrated on-orbit with high precision using data mining methods based on the on-orbit measured data.

[0013] Preferably, the aerospace energy system design method further comprises the following steps:

[0014] S7. Conduct intelligent and high-precision energy analysis on the spacecraft power supply and distribution system, and improve the prior experience general spacecraft power supply and distribution system model.

[0015] Preferably, the universal spacecraft power supply and distribution system includes an energy supply end and an energy consumption unit.

[0016] Preferably, the energy supply end includes solar panels, solar cell arrays, power controllers, and batteries, and the energy consumption units include system-level and component-level energy consumption units; the system-level energy consumption units include a load subsystem, a test / data transmission subsystem, a thermal control subsystem, and an attitude and orbit control subsystem, and the component-level energy consumption units include reaction wheels, electric propulsion systems, heaters, refrigerators, transmit / receive antennas, data processing, turntables, and detectors.

[0017] Preferably, in the aerospace energy system design method, step S2 uses a priori on-orbit data and inputs the gated recurrent unit GRU neural network to generalize the data. The low-frequency on-orbit data is subjected to data preprocessing, feature extraction, relationship inference and information reconstruction to obtain high-frequency on-orbit data, so as to realize the refined mining of the general spacecraft power supply and distribution system data model, which specifically includes the following steps:

[0018] S21. Detailed mining of the energy consumption of individual spacecraft;

[0019] S22. Detailed mining of energy output at the energy supply end of spacecraft;

[0020] S22. Detailed mining of energy output based on the lighting conditions of solar panels.

[0021] Preferably, in the aerospace energy system design method, step S3: according to the actual spacecraft configuration, call the energy supply end and energy consumption unit in the general spacecraft power supply and distribution system model library to quickly integrate the actual spacecraft power supply and distribution system model.

[0022] Preferably, in the aerospace energy system design method, step S4: ground test cases include ground test cases of spacecraft normal mode, mission mode and maneuvering mode.

[0023] Preferably, in the aerospace energy system design method, step S6: performing on-orbit high-precision calibration according to on-orbit measured data of different spacecraft mission sequences and different on-orbit operating environments.

[0024] Preferably, the different on-orbit operating environments of the spacecraft refer to the spacecraft being in the sunlit area or the earth's shadow area, and the spacecraft's attitude and the solar sail panel's incident angle are different.

[0025] Preferably, in the aerospace energy system design method, step S7: performing intelligent high-precision energy analysis by automatically generating ground and on-orbit test cases for the spacecraft power supply and distribution system and autonomously evaluating the energy balance of the spacecraft power supply and distribution system.

[0026] A spacecraft energy system design method of the present invention has the following beneficial effects: by combining the prior knowledge of previous spacecraft on the ground and in orbit, the energy consumption of different types of single machines under different working modes is deeply mined through intelligent learning, and an energy consumption pattern library is established, laying a solid foundation for the rapid design and precise testing of spacecraft energy subsystems; at the same time, during the on-orbit operation phase of the spacecraft, the model parameters and combination sequences in the model library are autonomously corrected through telemetry data, which greatly avoids the influence of human factors and ensures the accurate evaluation and effective operation of the on-orbit energy subsystem of the spacecraft.

Brief Description of the Drawings

[0027] Figure 1 It is a step-by-step diagram of a spacecraft energy system design method.

[0028] Figure 2 It is a generalized spacecraft power supply and distribution system architecture diagram constructed based on the spacecraft energy system design method.

[0029] Figure 3 It is a high-precision intelligent customized energy analysis schematic diagram of a spacecraft energy system design method.

[0030] Figure 4 It is a schematic diagram of the spacecraft energy consumption analysis results under different mission conditions of a spacecraft energy system design method.

[0031] Figure 5 It is a method for designing a spacecraft energy system and a schematic diagram of the results of a detailed energy balance analysis of a spacecraft. [Specific implementation method]

[0032] The present invention will be further described below in conjunction with embodiments and with reference to the accompanying drawings.

[0033] Example

[0034] This embodiment implements a method for designing a spacecraft energy system.

[0035] The present embodiment provides a method for designing a spacecraft energy system, which reasonably designs an energy subsystem according to the spacecraft operating environment, mission sequence, working status of each single machine and previous satellite on-orbit data, and verifies the design through the development process. It is crucial to ensure the accuracy of the design. At the same time, it also provides new solutions and plans for the optimal design of the energy subsystem. After the satellite is in orbit, the ground model can be corrected according to the satellite's actual on-orbit flight data to provide guarantee for the subsequent satellite on-orbit mission planning and execution.

[0036] In this embodiment, a spacecraft energy system design method combines the prior knowledge of previous spacecraft on the ground and on orbit, and uses intelligent learning to deeply mine the energy consumption of different types of single machines under different working modes, and establish an energy consumption model library, laying a solid foundation for the rapid design and precise testing of subsequent spacecraft energy subsystems. At the same time, during the on-orbit operation phase of the spacecraft, the model parameters and combination sequences in the model library can be independently corrected through telemetry data, which can greatly avoid the influence of human factors and ensure the accurate evaluation and effective operation of the spacecraft on-orbit energy subsystem. Figure 1 This is a step-by-step diagram of the design method of a spacecraft energy system, as shown in the attached figure. Figure 1 As shown, the specific steps of a spacecraft energy system design method in this embodiment are as follows:

[0037] one, Figure 2 This is a generalized spacecraft power supply and distribution system architecture diagram constructed using a spacecraft energy system design method. Figure 2 As shown, a method for designing a spacecraft energy system in this embodiment is based on a priori experience of a universal spacecraft power supply and distribution system, and the universal spacecraft power supply and distribution system includes an energy supply end and an energy consumption unit. The energy supply end includes a solar cell array, a power controller, and a battery. The energy consumption unit includes system-level and component-level energy consumption units. The system-level energy consumption unit includes a payload subsystem, a test / data transmission subsystem, a thermal control subsystem, and an attitude and orbit control subsystem. The component-level energy consumption unit includes a reaction wheel, an electric propulsion system, a heater, a refrigerator, a transmit / receive antenna, a data processing, a turntable, a detector, etc. A method for designing a spacecraft energy system in this embodiment constructs a priori experience of a universal spacecraft power supply and distribution system model based on computer software.

[0038] two, Figure 3 It is a high-precision intelligent customized energy analysis schematic diagram of a spacecraft energy system design method. Figure 3As shown, the present embodiment provides a method for designing a spacecraft energy system, including the steps of converting a priori general spacecraft power supply and distribution systems into high-precision intelligent customization to construct an actual spacecraft power supply and distribution system.

[0039] The present embodiment is a method for designing a spacecraft energy system, which is based on a recurrent neural network (RNN) and performs data mining on data with sequence characteristics, mining the temporal information and semantic information in the data. Both long short-term memory networks and gated recurrent unit (GRU) neural networks are acceptable, and gated recurrent unit (GRU) neural networks are preferred. Specifically, the present embodiment uses a priori on-orbit data and inputs it into a recurrent neural network (RNN) for data generalization. The low-frequency on-orbit data is subjected to data preprocessing, feature extraction, relationship inference, and information reconstruction to obtain high-frequency on-orbit data, thereby realizing the refined mining of the general spacecraft power supply and distribution system data model.

[0040] The specific refinement mining process of the general spacecraft power supply and distribution system data model in this embodiment includes:

[0041] Classify the energy consumption of spacecraft and conduct detailed mining of energy consumption of each unit;

[0042] Classify the design schemes of different energy subsystems of spacecraft (solar arrays, power controllers, batteries, etc.) and complete the detailed mining of energy output;

[0043] The charging and discharging conditions of the energy subsystems of the spacecraft at different orbital altitudes and orbital inclinations are collected, and the energy output is refined according to the lighting conditions of the sailboard.

[0044] According to the actual configuration of the spacecraft, the energy consumption units and energy subsystems in the model library are called to complete rapid integration to meet the customized power supply and distribution system requirements of the actual spacecraft;

[0045] In the ground design phase, the energy balance is classified and subsampled according to all mission sequences of the entire spacecraft, and ground test cases are generated, including normal mode, mission mode, maneuver mode and other ground test cases;

[0046] Based on the measured data on the ground, data mining methods are used to correct the states that can be modeled, and data fitting is performed on the states that cannot be modeled, to ensure the correctness of each ground model and module.

[0047] Figure 4 This is a schematic diagram of the analysis results of spacecraft energy consumption under different mission conditions in a spacecraft energy system design method. Figure 4As shown, a method for designing a spacecraft energy system in this embodiment takes a certain type of satellite as an example, and analyzes the energy consumption of the satellite under different mission conditions in the mission sequence. A method for designing a spacecraft energy system in this embodiment uses an intelligent data mining method based on on-orbit measured data such as different spacecraft mission sequences and different on-orbit operating environments (spacecraft location (sunlit area or earth shadow area), spacecraft attitude / solar sail panel incidence angle), and further completes on-orbit high-precision correction of a customized power supply and distribution system model that meets the actual spacecraft.

[0048] Figure 5 This is a spacecraft energy system design method. It is a schematic diagram of the detailed energy balance analysis results of the spacecraft. Figure 5 As shown, this embodiment provides a method for designing a spacecraft energy system, which ultimately realizes the automated generation of ground and on-orbit test cases for the energy subsystem and the autonomous evaluation of the balance of the energy subsystem, provides support for on-orbit missions, and provides experience for subsequent spacecraft models.

[0049] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0050] The above is only a preferred embodiment of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and supplements without departing from the principle of the present invention. These improvements and supplements should also be regarded as the scope of protection of the present invention.

Claims

1. A method for designing a spacecraft energy system, characterized in that The following steps are involved: S1. Construct a general spacecraft power supply and distribution system model; S2. Use the prior on-orbit data and input it into the recurrent neural network (RNN) for data generalization to achieve refined mining of prior empirical data of the general spacecraft power supply and distribution system. S3, build a model of the actual spacecraft power supply and distribution system; S4, ground design phase, classifies and establishes subsamples of energy balance according to the spacecraft mission sequence, and generates ground test cases; S5. Based on the actual ground test data of the spacecraft, the data mining method is used to correct the modelable state of the actual spacecraft power supply and distribution system model and perform data fitting for the state that cannot be modeled; S6. During the on-orbit operation phase, the actual spacecraft power supply and distribution system model is calibrated on-orbit with high precision using data mining methods based on the on-orbit measured data.

2. A method for designing a spacecraft energy system according to claim 1, characterized in that The following steps are also included: S7. Conduct intelligent and high-precision energy analysis on the spacecraft power supply and distribution system, and improve the prior experience general spacecraft power supply and distribution system model.

3. A method for designing a spacecraft energy system according to claim 2, characterized in that: The universal spacecraft power supply and distribution system includes an energy supply end and an energy consumption unit.

4. A method for designing a spacecraft energy system according to claim 3, characterized in that: The energy supply end includes solar panels, solar cell arrays, power controllers, and batteries; the energy consumption units include system-level and component-level energy consumption units; the system-level energy consumption units include load subsystems, test / data transmission subsystems, thermal control subsystems, and attitude and orbit control subsystems; the component-level energy consumption units include reaction wheels, electric propulsion systems, heaters, refrigerators, transmit / receive antennas, data processing, turntables, and detectors.

5. A method for designing a spacecraft energy system according to claim 4, characterized in that Step S2: Use the prior on-orbit data and input the gated recurrent unit GRU neural network to generalize the data. The low-frequency on-orbit data is processed through data preprocessing, feature extraction, relationship inference and information reconstruction to obtain high-frequency on-orbit data, so as to realize the refined mining of the data model of the general spacecraft power supply and distribution system. Specifically, the following steps are included: S21. Detailed mining of the energy consumption of individual spacecraft; S22. Detailed mining of energy output at the energy supply end of spacecraft; S22. Detailed mining of energy output based on the lighting conditions of solar panels.

6. A method for designing a spacecraft energy system according to claim 5, characterized in that Step S3: According to the actual spacecraft configuration, call the energy supply end and energy consumption unit in the general spacecraft power supply and distribution system model library to quickly integrate the actual spacecraft power supply and distribution system model.

7. A method for designing a spacecraft energy system according to claim 6, characterized in that Step S4: Ground test cases, including ground test cases of spacecraft normal mode, mission mode, and maneuvering mode.

8. A method for designing a spacecraft energy system according to claim 7, characterized in that Step S6: Perform on-orbit high-precision correction based on on-orbit measured data of different spacecraft mission sequences and different on-orbit operating environments.

9. A method for designing a spacecraft energy system according to claim 8, characterized in that: Different on-orbit operating environments for spacecraft refer to the spacecraft being in the sunlit area or the earth's shadow area, and the spacecraft's attitude and the solar sail's incident angle are different.

10. A method for designing a spacecraft energy system according to claim 9, characterized in that Step S7: Perform intelligent and high-precision energy analysis by automatically generating ground and on-orbit test cases for the spacecraft power supply and distribution system and autonomously evaluating the energy balance of the spacecraft power supply and distribution system.

Images