Satellite remote measurement, operation and control system

Abstract

The invention discloses a satellite remote measurement, operation and control system, which relates to the technical field of measurement, operation and control, and comprises the following steps: adopting a measurement and control link combination of a low-orbit satellite, a geosynchronous orbit satellite and a high-altitude platform, dynamically selecting an optimal measurement and control path through reinforcement learning, and optimizing measurement and control scheduling based on a link state, a task priority and a resource competition condition. The measurement and control coverage rate and the resource utilization rate are improved; in the task execution process, the integrated AI model monitors the satellite task state in real time, obtains the task state, the equipment load and the communication link delay data, and predicts the abnormal condition in combination with the Bayesian probability model and the LSTM neural network, thereby avoiding the potential risk in advance and improving the stability of the measurement and control task. The measurement and control mode can be automatically adjusted after anomaly detection, track parameters are dynamically corrected through a predictive control model, or ground instruction adjustment is executed after a measurement and control link is recovered, and the track maintaining capacity is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of measurement, operation and control, and in particular to a satellite remote measurement, operation and control system. Background Art

[0002] The satellite remote measurement, operation and control system is responsible for tracking and measuring spacecraft, collecting telemetry data, issuing remote control commands and task scheduling. For a long time, this system mainly relies on ground measurement and control stations to establish measurement and control links for data interaction and command transmission when the satellite passes by.

[0003] However, the distribution of ground measurement and control stations is restricted by geographical factors. The satellite can only establish a measurement and control link within a short visible window. For high-orbit satellites, this window can reach dozens of minutes, while for low-orbit satellites, there are often only a few minutes of measurement and control time. Task scheduling must be adjusted around these limited windows, resulting in limited flexibility and real-time performance of task execution. In case of sudden failures, if the satellite fails to enter the measurement and control range in time, the anomaly may further deteriorate and even affect normal operation. In deep space exploration missions, due to communication delays, the execution efficiency of measurement and control commands is more affected. For example, the data round-trip time of a Mars probe can reach dozens of minutes, seriously reducing the response ability to emergencies. Secondly, the insufficient measurement and control coverage is also a key factor affecting task stability. The layout of traditional ground-based measurement and control stations is mainly concentrated in the mid-latitudes and low-latitudes, resulting in polar-orbiting satellites being in the measurement and control blind area for a long time in the north and south polar regions. In addition, global missions such as ocean observation and remote exploration require all-weather data transmission, but the limited coverage of ground measurement and control stations cannot support continuous measurement and control.

[0004] Some solutions adopt space-based measurement and control relay satellites, such as TDRSS and Tianlian, to enhance coverage. However, due to their fixed orbital altitude, it is difficult to provide full-time measurement and control support for all orbit satellites. For low-orbit satellite constellations, there is great pressure on link resource competition. In addition, high-security missions such as deep space exploration and military spaceflight have extremely high requirements for the anti-interference ability of the measurement and control system. However, traditional measurement and control systems are prone to signal loss in a strong interference environment, affecting the continuity of tasks. With the increase in the number of low-orbit satellites and the improvement of task complexity, how to build a more flexible, efficient and autonomous measurement, operation and control system has become an urgent problem to be solved. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] The present invention provides a satellite remote measurement, operation and control system to solve the problems that ground measurement and control are restricted by coverage and time windows, and relay satellites are restricted by orbits, making it difficult to meet the continuous measurement and control requirements of low-orbit and deep space missions.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] An embodiment of the present invention provides a satellite remote measurement, operation and control system, which includes:

[0009] An orbit measurement and control module, configured to collect initial orbit state data and plan a measurement and control link;

[0010] An orbit monitoring and prediction module, which performs orbit monitoring and correction based on orbit state data and predicts task execution parameters;

[0011] A measurement and control link optimization module, which optimizes the measurement and control link according to task execution parameters and performs orbit adjustment;

[0012] The optimization of the measurement and control link includes dynamically selecting a link based on reinforcement learning;

[0013] The measurement and control link includes a ground measurement and control station, a relay satellite, a high-altitude platform and an inter-satellite link;

[0014] The orbit adjustment includes correcting the attitude, maintaining the orbit and calculating an anomaly warning;

[0015] A task monitoring module, configured to monitor the satellite state in real time during task execution, obtain task execution data, and predict abnormal measurement and control situations;

[0016] The task execution data includes task status, equipment workload and communication link delay;

[0017] The abnormal measurement and control prediction includes signal loss, link interruption and task failure probability;

[0018] A measurement and control adjustment module, which adjusts the measurement and control mode based on the abnormal prediction result of the task monitoring module, optimizes the data return path, and performs orbit adjustment; the adjustment of the measurement and control mode includes autonomously executing a measurement and control strategy.

[0019] As a preferred solution of the satellite remote measurement, operation and control system according to the present invention, wherein: the orbit measurement and control module adopts a combination of measurement and control links of low-earth orbit satellites (LEO), geostationary orbit satellites (GEO) and high-altitude platforms (HAPS) to provide full-time domain measurement and control coverage.

[0020] As a preferred solution of the satellite remote measurement, operation and control system according to the present invention, wherein: the measurement and control link optimization module adopts a reinforcement learning algorithm to dynamically adjust the measurement and control path based on the real-time measurement and control link state.

[0021] As a preferred solution of the satellite remote measurement, operation and control system according to the present invention, wherein: the task monitoring module integrates an AI model for abnormal state prediction and fault mode identification to avoid risks that may affect task execution in advance.

[0022] As a preferred solution of the satellite remote measurement, operation and control system according to the present invention, wherein: the measurement, operation and control method of this system is:

[0023] Step S1, collect initial orbital state data and plan the TT&C link;

[0024] The orbital state data includes satellite orbital parameters, attitude angles, and TT&C signal quality;

[0025] The TT&C link planning includes the TT&C availability assessment of ground TT&C stations, relay satellites, and inter-satellite links;

[0026] Step S2, perform orbital monitoring and correction based on the orbital state data, and predict mission execution parameters;

[0027] The orbital monitoring includes collecting telemetry data and signal integrity analysis;

[0028] The mission execution parameters include orbital drift trend, attitude change range, and TT&C signal attenuation degree;

[0029] Step S3, optimize the TT&C link in Step S1 according to the mission execution parameters, adjust the TT&C link, and perform orbital adjustment;

[0030] The TT&C link optimization includes dynamically selecting a link based on reinforcement learning;

[0031] The TT&C link includes ground TT&C stations, relay satellites, high-altitude platforms, and inter-satellite links;

[0032] The orbital adjustment includes correcting the attitude, maintaining the orbit, and calculating anomaly warnings;

[0033] Step S4, monitor the satellite state in real time during the mission execution, obtain mission execution data, and predict abnormal TT&C situations;

[0034] The mission execution data includes mission status, equipment workload, and communication link delay;

[0035] The abnormal TT&C prediction includes signal loss, link interruption, and mission failure probability;

[0036] Step S5, adjust the TT&C mode based on the abnormal prediction result in S4, optimize the data return path, and perform orbital adjustment;

[0037] The TT&C mode adjustment includes autonomously executing TT&C strategies;

[0038] The orbital adjustment includes autonomously correcting the orbit or executing ground command adjustment after the TT&C link is restored.

[0039] As a preferred solution of the satellite remote TT&C system described in the present invention, wherein: in Step S3, the step of dynamically selecting a link based on reinforcement learning is

[0040] Model the state space and define the state vector of the TT&C link to describe the current state of the link, expressed as:

[0041] s t =(b t , d t , q t , r t ),

[0042] where s t represents the state of the TT&C link at time t, b t represents the bandwidth of the TT&C link at time t, d t represents the time delay of the TT&C link at time t, q t represents the quality of the TT&C signal at time t, and r t represents the occupancy rate of TT&C resources at time t;

[0043] Define the action space, expressed as:

[0044] a t ∈{L1, L2,..., L N},

[0045] where a t represents the TT&C link selected at time t, and L i represents the i-th TT&C link, where i ranges from i ∈ {1, 2,..., N}, and N is the total number of available TT&C links, including ground TT&C stations, relay satellites, high-altitude platforms, and inter-satellite links;

[0046] Set the reward function R t to calculate the quality of the TT&C path, with the formula:

[0047] R t =α1f1(b t ) + α2f2(d t ) + α3f3(q t ) + α4f4(r t ),

[0048] where R t is the reward value, and f1(b t ) represents the bandwidth gain function, expressed as a normalized linear function:

[0049] f1(b t ) = b t / b max , where b max is the maximum bandwidth of the TT&C link,

[0050] f2(d t ) represents the time delay loss function, expressed as a negative exponential decay function:

[0051] where λ d is the time-delay weight factor,

[0052] f3(q t ) represents the signal quality function, which is represented by logarithmic mapping:

[0053] f3(q t ) = log(1 + q t ),

[0054] f4(r t ) represents the resource occupancy penalty function, which is represented by a linear penalty model:

[0055] f4(r t ) = -r t / r max , α1, α2, α3, α4 are the weight factors of the reward function, satisfying the normalization condition: α1 + α2 + α3 + α4 = 1,

[0056] where α1 weighs the importance of bandwidth, α2 weighs the impact of time delay on the link, α3 weighs the impact of signal quality on the TT&C link, and α4 weighs the impact of resource competition.

[0057] As an optimal solution of the satellite remote TT&C system described in the present invention, wherein: the step of dynamically selecting a link based on reinforcement learning further includes:

[0058] Define the task priority function, and the function formula is:

[0059] P t = β1T d + β2W c + β3C o ,

[0060] where P t is the task priority at time t, T d is the task deadline, W c is the task computing load, C o is the occupancy rate of the TT&C link, β1, β2, β3 are the priority weight factors, satisfying the normalization condition:

[0061] β1 + β2 + β3 = 1,

[0062] where β1 weighs the importance of the task deadline, β2 weighs the impact of the computing load on the priority, and β3 weighs the impact of the TT&C link occupancy rate;

[0063] In the case of limited TT&C link resources, adjust the allocation strategy, and the adjustment formula is:

[0064] U t = γ1Pt +γ2R t ,

[0065] Among them, U t is the comprehensive score, γ1 and γ2 are adjustment factors, satisfying: γ1 + γ2 = 1, P t is the task priority, and R t is the reward value of the measurement and control link;

[0066] Deep Q-learning is used for link optimization to update the Q value. The update formula is:

[0067] Among them,

[0068] represents the Q value of selecting the best action in state s t+1 ,

[0069] Q(s t , a t ) represents the Q value of selecting action a t in state s t . η is the learning rate, which determines the update step of the Q value, and R t is the current reward value, and λ is the discount factor.

[0070] As a preferred solution of the satellite remote measurement, transportation and control system described in the present invention, among them: in step S4, during the task execution process, the satellite state is monitored in real time, the task execution data is obtained, and the method for predicting abnormal measurement and control situations is:

[0071] The satellite task state is monitored in real time, and the state vector is defined as S' t :

[0072] S′ t =(X t , Y t , Z t , Θ t , P′ t , L′ t ),

[0073] Among them, S' ′t is the satellite state monitored at time t, X t , Y t , Z t are the three-dimensional positions of the satellite in the orbital coordinate system,

[0074] Θ t is the satellite attitude angle, P' t is the task state, and L' t is the link communication delay;

[0075] Anomaly detection is carried out using the Bayesian method to calculate the probability of an anomaly occurring. The calculation formula is as follows:

[0076]

[0077] Among them, P(A|S' t ) is the anomaly probability under the given state S' t , P(S' t |A) is the observed state probability under the anomaly condition, P(A) is the prior probability of the anomaly, and P(S' t ) is the total probability of the current state.

[0078] As a preferred solution of a satellite remote measurement, operation and control system described in the present invention, wherein: during the process of predicting the abnormal measurement and control situation in step S4, a misjudgment control mechanism is introduced.

[0079] The chi-square detection method is used for anomaly discrimination. The formula is as follows:

[0080]

[0081] Among them, X 2 is the statistic for anomaly detection, O i is the observed value, E i is the expected value.

[0082] Set a threshold If then it is determined as an anomaly;

[0083] The LSTM neural network is used to predict the abnormal state, and the state is updated as follows:

[0084] h' t = f(W' h h' t-1 + W' x S' t + b'),

[0085] Among them, h' t is the hidden state predicted by the LSTM, h' t-1 represents the hidden state of the previous time step, W' h is the hidden state weight matrix, W' x is the input weight matrix, b' is the bias term, and f(·) is the activation function.

[0086] As a preferred solution of a satellite remote measurement, operation and control system described in the present invention, wherein: in step S5, the steps of autonomously executing the measurement and control strategy are as follows.

[0087] After predicting an anomaly, autonomously adjust the measurement and control strategy:

[0088] Optimize the data feedback path to have the lowest cost and the fastest response time. The optimization formula is as follows:

[0089]

[0090] Among them, R opt is the optimal data feedback path, C(L i ) is the transmission cost of link L i , T(L i ) is the transmission time of link L i , μ is the adjustment factor,

[0091] Use a hybrid encryption method for encryption. The encryption process is as follows:

[0092]

[0093] Among them, C' is the encrypted data, E k (D') represents encrypting data D' with key k, denotes the exclusive OR operation, which is used to combine the encrypted data and the hash value, and H(S' t ) is the data integrity hash value;

[0094] Adjust the measurement and control mode according to abnormal situations. Define the adjustment function, which is expressed as:

[0095] M′ t =δ1A′ t +δ2R′ t ,

[0096] Among them, M' t is the measurement and control strategy adjustment parameter, A' t is the abnormal level, R' t is the current measurement and control link state, and δ1, δ2 are adjustment factors;

[0097] Based on the predictive control model, perform orbit adjustment. The adjustment formula is as follows:

[0098] X′ t+1 =A′X′ t +B′U′ t ,

[0099] Among them, X' t+1 is the adjusted orbit state, A' is the orbit dynamics matrix, B' is the control input matrix, and U' t is the orbit adjustment strategy.

[0100] The beneficial effects of the present invention are as follows: The present invention adopts a combined measurement and control link of low-earth orbit satellites, geostationary orbit satellites and high-altitude platforms, dynamically selects the optimal measurement and control path through reinforcement learning, optimizes the measurement and control scheduling based on link status, task priority and resource competition, and improves the measurement and control coverage rate and resource utilization rate; during the task execution process, an AI model is integrated to monitor the satellite task status in real time, obtain task status, equipment load and communication link delay data, and combine the Bayesian probability model and the LSTM neural network to predict abnormal situations, so as to avoid potential risks in advance and improve the stability of the measurement and control tasks.

[0101] In the present invention, after anomaly detection, the measurement and control mode can be adjusted autonomously. The predictive control model is used to dynamically correct the orbit parameters, or the ground instructions are executed to adjust after the measurement and control link is restored, so as to improve the orbit maintenance ability; in addition, to improve the security and reliability of the measurement and control data transmission, the present invention optimizes the data return path based on the optimal path selection algorithm, reduces the data transmission delay while reducing the communication cost, and uses a hybrid encryption method to ensure the security of the data and prevent link attacks and data tampering.

[0102] The present invention adopts a task priority scheduling mechanism, calculates the task importance according to the task deadline, computing load and link occupancy rate, and intelligently allocates measurement and control resources through the DQN algorithm, so that high-priority tasks are preferentially guaranteed under resource constraints, thereby improving the measurement and control stability of key tasks.

[0103] In summary, the present invention improves the stability of the measurement and control link, the reliability of task execution and the security of data transmission, while reducing the communication delay and the response time of orbit adjustment. Brief Description of the Drawings

[0104] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0105] Figure 1 It is a schematic framework diagram of the satellite remote measurement, operation and control system of the present invention. Detailed Embodiments

[0106] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the detailed embodiments of the present invention will be described below with reference to the drawings of the specification.

[0107] In the following description, numerous specific details are set forth to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways than those specifically described herein, and those skilled in the art can make similar generalizations without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0108] Secondly, as used herein, an "embodiment" or "embodiments" refers to specific features, structures, or characteristics that may be included in at least one implementation of the present invention. The phrase "in an embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it an individual or alternative embodiment that is mutually exclusive of other embodiments.

[0109] Embodiment 1, referring to Figure 1 , this embodiment provides a satellite remote measurement, operation, and control system, including:

[0110] An orbit measurement and control module, configured to collect initial orbit state data and plan the measurement and control link;

[0111] The orbit measurement and control module adopts a combined measurement and control link of low Earth orbit (LEO) satellites, geostationary Earth orbit (GEO) satellites, and high-altitude platforms (HAPS) to provide full-time-domain measurement and control coverage;

[0112] An orbit monitoring and prediction module, which performs orbit monitoring and correction based on the orbit state data and predicts the task execution parameters;

[0113] A measurement and control link optimization module, which optimizes the measurement and control link according to the task execution parameters and performs orbit adjustment;

[0114] The optimization of the measurement and control link includes dynamically selecting the link based on reinforcement learning;

[0115] The measurement and control link includes a ground measurement and control station, a relay satellite, a high-altitude platform, and an inter-satellite link;

[0116] The orbit adjustment includes attitude correction, orbit maintenance, and calculation of anomaly warnings;

[0117] The measurement and control link optimization module adopts a reinforcement learning algorithm to dynamically adjust the measurement and control path based on the real-time measurement and control link state;

[0118] A task monitoring module, configured to monitor the satellite state in real time during the task execution process, obtain the task execution data, and predict abnormal measurement and control situations;

[0119] The task execution data includes task status, device workload, and communication link delay;

[0120] The abnormal measurement and control prediction includes signal loss, link interruption, and task failure probability;

[0121] The task monitoring module integrates an AI model for abnormal state prediction and fault mode identification to avoid risks that may affect task execution in advance;

[0122] The measurement and control adjustment module adjusts the measurement and control mode based on the abnormal prediction results of the task monitoring module, optimizes the data transmission path, and performs orbit adjustment; the adjustment of the measurement and control mode includes autonomously executing measurement and control strategies.

[0123] This embodiment also provides a measurement, operation, and control method for the above satellite remote measurement, operation, and control system, including:

[0124] Step S1, collect initial orbit state data and plan the measurement and control link;

[0125] The orbit state data includes satellite orbit parameters, attitude angles, and measurement and control signal quality;

[0126] The measurement and control link planning includes the measurement and control availability assessment of ground measurement and control stations, relay satellites, and inter-satellite links;

[0127] Step S2, perform orbit monitoring and correction based on the orbit state data and predict task execution parameters;

[0128] Orbit monitoring includes collecting telemetry data and signal integrity analysis;

[0129] The task execution parameters include orbit drift trend, attitude change range, and measurement and control signal attenuation degree;

[0130] Step S3, optimize the measurement and control link in Step S1 according to the task execution parameters, adjust the measurement and control link, and perform orbit adjustment;

[0131] The measurement and control link optimization includes dynamically selecting a link based on reinforcement learning;

[0132] The measurement and control link includes ground measurement and control stations, relay satellites, high-altitude platforms, and inter-satellite links;

[0133] Orbit adjustment includes correcting the attitude, maintaining the orbit, and calculating abnormal warnings;

[0134] In Step S3, the steps of dynamically selecting a link based on reinforcement learning are as follows:

[0135] Model the state space and define the state vector of the measurement and control link to describe the current state of the link, expressed as:

[0136] s t =(b t ,d t ,q t ,r t ),

[0137] where s t represents the state of the measurement and control link at time t, bt Denote the bandwidth of the TT&C link at time t as d t Denote the latency of the TT&C link at time t as q t Denote the quality of the TT&C signal at time t as r t Denote the occupancy rate of TT&C resources at time t;

[0138] Define the action space, denoted as:

[0139] a t ∈ {L1, L2, …, L N},

[0140] where a t represents the TT&C link selected at time t, and L i represents the i-th TT&C link, where i ∈ {1, 2, …, N}, and N is the total number of optional TT&C links, including ground TT&C stations, relay satellites, high-altitude platforms, and inter-satellite links;

[0141] Set the reward function R t Calculate the quality of the TT&C path, with the formula:

[0142] R t = α1f1(b t ) + α2f2(d t ) + α3f3(q t ) + α4f4(r t ),

[0143] where, R t is the reward value, f1(b t ) represents the bandwidth gain function, expressed as a normalized linear function:

[0144] f1(b t ) = b t / b max , where b max is the maximum bandwidth of the TT&C link,

[0145] f2(d t ) represents the latency loss function, expressed as a negative exponential decay function:

[0146] where λ d is the latency weight factor,

[0147] f3(q t ) represents the signal quality function, expressed as a logarithmic mapping:

[0148] f3(q t ) = log(1 + q t ),

[0149] f4(r t ) represents the resource occupancy penalty function, which is represented by a linear penalty model:

[0150] f4(r t ) = -r t / r max , where α1, α2, α3, α4 are the weight factors of the reward function and satisfy the normalization condition: α1 + α2 + α3 + α4 = 1.

[0151] Among them, α1 weighs the importance of bandwidth, α2 weighs the impact of delay on the link, α3 weighs the impact of signal quality on the TT&C link, and α4 weighs the impact of resource competition.

[0152] The steps of dynamically selecting a link based on reinforcement learning also include:

[0153] Define the task priority function, and the function formula is:

[0154] P t = β1T d + β2W c + β3C o ,

[0155] where P t is the task priority at time t, T d is the task deadline, W c is the task computing load, C o is the occupancy rate of the TT&C link, and β1, β2, β3 are the priority weight factors, which satisfy the normalization condition:

[0156] β1 + β2 + β3 = 1.

[0157] Among them, β1 weighs the importance of the task deadline, β2 weighs the impact of the computing load on the priority, and β3 weighs the impact of the occupancy rate of the TT&C link.

[0158] In the case of limited TT&C link resources, adjust the allocation strategy, and the adjustment formula is:

[0159] U t = γ1P t + γ2R t ,

[0160] where U t is the comprehensive score, γ1, γ2 are the adjustment factors, and satisfy: γ1 + γ2 = 1, P t is the task priority, and R t is the TT&C link reward value.

[0161] Adopt deep Q-learning for link optimization, update the Q value, and the update formula is:

[0162] in,

[0163] Indicates that in state s t+1 Select the Q value of the best action,

[0164] Q(s t , a t ) means in state s t Select action a t Q value, η is the learning rate, which determines the Q value update step size, R t is the current reward value, λ is the discount factor;

[0165] Specifically, reinforcement learning is used here to optimize the measurement and control link selection, and the deep Q learning method is adopted to dynamically adjust the decision according to the measurement and control link status. Through state space modeling, the bandwidth, delay, signal quality and resource occupancy of the measurement and control link are included in the optimization range, and a reward function based on bandwidth gain, delay loss, signal quality and resource competition is designed to ensure the optimality of link selection; in addition, a task priority scheduling mechanism is introduced to calculate the importance of tasks according to task deadlines, computing load and link occupancy, and high-priority tasks obtain priority measurement and control resources. In terms of resource competition management, measurement and control resources are reasonably allocated through a comprehensive scoring function to ensure the stability of the measurement and control link for key tasks; the DQN algorithm is used to train the Q value to make the dynamic adjustment of the measurement and control link more intelligent, effectively reduce communication delays, and improve the reliability of measurement and control data transmission;

[0166] Step S4, real-time monitoring of satellite status during mission execution, acquisition of mission execution data, and prediction of abnormal measurement and control conditions;

[0167] Task execution data includes task status, device workload, and communication link latency;

[0168] Abnormal measurement and control prediction includes signal loss, link interruption and mission failure probability;

[0169] In step S4, during the mission execution, the satellite status is monitored in real time, mission execution data is acquired, and abnormal measurement and control conditions are predicted in the following manner:

[0170] Monitor the satellite mission status in real time and define the state vector as S' t :

[0171] S′ t =(X t , Y t , Z T , Θ t , P′ t , L′ t ),

[0172] Among them, S't The satellite state monitored at time t, X t , Y t , Z t are the three-dimensional positions of the satellite in the orbital coordinate system,

[0173] Θ t is the satellite attitude angle, P' t is the mission status, L' t is the link communication delay;

[0174] The Bayesian method is used for anomaly detection to calculate the probability of anomaly occurrence. The calculation formula is:

[0175]

[0176] where P(A|S' t ) is the anomaly probability under the given state S' t , P(S' t |A) is the probability of the observed state under the anomaly condition, P(A) is the prior probability of anomaly, and P(S' t ) is the total probability of the current state;

[0177] During the process of predicting the abnormal TT&C situation in step S4, a misjudgment control mechanism is introduced,

[0178] The chi-square detection method is used for anomaly discrimination. The formula is:

[0179]

[0180] where X 2 is the statistic for anomaly detection, O i is the observed value, E i is the expected value,

[0181] Set a threshold If then it is determined as an anomaly;

[0182] The LSTM neural network is used to predict the abnormal state. The state update is as follows:

[0183] h' t = f(W' h h' t-1 + W' x S' t + b'),

[0184] where h' t is the hidden state predicted by the LSTM, h' t-1 represents the hidden state of the previous time step, W' h is the hidden state weight matrix, W' xW is the input weight matrix, b' is the bias term, and f(·) is the activation function;

[0185] Specifically, this paper focuses on satellite status monitoring and anomaly prediction during mission execution, and adopts a multi-level method to ensure the accuracy and real-time of TT&C data. Specifically:

[0186] Real-time monitor the satellite orbit coordinates, attitude angles, mission status, and link delay to construct the state space, calculate the probability of anomalies using the Bayesian probability model, and adopt the chi-square detection method to reduce the risk of misjudgment and ensure the accuracy of anomaly detection. Introduce the LSTM long short-term memory neural network, train the model based on historical monitoring data, and predict possible anomalies such as link interruptions and signal losses in the future, which can detect anomalies in the satellite TT&C link in advance and improve the reliability of the TT&C mission;

[0187] Step S5: Based on the anomaly prediction results in S4, adjust the TT&C mode, optimize the data transmission path, and perform orbit adjustment;

[0188] The adjustment of the TT&C mode includes autonomously executing the TT&C strategy;

[0189] The orbit adjustment includes autonomously correcting the orbit or executing ground command adjustment after the TT&C link is restored;

[0190] In step S5, the steps of autonomously executing the TT&C strategy are as follows:

[0191] After predicting an anomaly, autonomously adjust the TT&C strategy:

[0192] Optimize the data transmission path to make it have the lowest cost and the fastest response time. The optimization formula is:

[0193] Where R opt is the optimal data transmission path, C(L i ) is the transmission cost of link L i , T(L i ) is the transmission time of link L i , and μ is the adjustment factor.

[0194] Adopt a hybrid encryption method for encryption. The encryption process is as follows:

[0195]

[0196] Where C' is the encrypted data, and E k (D') represents encrypting data D' with the key k, represents the exclusive OR operation, which is used to combine the encrypted data and the hash value. H(S′ t ) is the data integrity hash value;

[0197] Adjust the measurement and control mode according to the abnormal situation, and define the adjustment function, which is expressed as:

[0198] M′ t = δ1A′ t + δ2R′ t ,

[0199] where M' t is the measurement and control strategy adjustment parameter, A' t is the abnormal level, R' t is the current measurement and control link state, and δ1 and δ2 are adjustment factors;

[0200] Based on the predictive control model, perform orbit adjustment. The adjustment formula is:

[0201] X′ t+1 = A′X′ t + B′U′ t ,

[0202] where X' t+1 is the orbit state after adjustment, A' is the orbit dynamics matrix, B' is the control input matrix, and U' t is the orbit adjustment strategy;

[0203] Specifically, here, through the optimal path selection algorithm, considering the transmission cost and delay comprehensively, dynamically optimize the feedback path of the measurement and control data to ensure the high efficiency of data transmission. Secondly, adopt a hybrid encryption method, including symmetric encryption and hash integrity verification, to improve the security of the measurement and control data, prevent link attacks and data tampering. In terms of the adjustment of the measurement and control mode, according to the abnormal level and the measurement and control link state, autonomously adjust the measurement and control strategy to enable the satellite to have autonomous measurement and control capabilities and reduce the dependence on the ground station;

[0204] In addition, here use the predictive control model to achieve orbit adjustment, including orbit correction, orbit maintenance and abnormal early warning, to improve the orbit stability of the satellite.

[0205] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A satellite remote measurement, operation and control system, characterized in that: include, Track measurement and control module, used to collect initial track status data and plan the measurement and control link; The orbit monitoring and prediction module performs orbit monitoring and correction based on orbit status data, and predicts mission execution parameters; The measurement and control link optimization module optimizes the measurement and control link according to the mission execution parameters and performs orbit adjustment; The optimization of the measurement and control link includes dynamically selecting the link based on reinforcement learning; The measurement and control link includes ground measurement and control stations, relay satellites, high-altitude platforms and intersatellite links; The orbit adjustment includes correcting attitude, maintaining orbit and calculating abnormal warning; The mission monitoring module is used to monitor the satellite status in real time during the mission execution, obtain mission execution data, and predict abnormal measurement and control situations; The task execution data includes task status, device workload and communication link delay; The abnormal measurement and control prediction includes signal loss, link interruption and mission failure probability; The measurement and control adjustment module adjusts the measurement and control mode based on the abnormal prediction results of the task monitoring module, optimizes the data return path, and performs orbit adjustment; the adjustment of the measurement and control mode includes autonomous execution of the measurement and control strategy.

2. A satellite remote measurement, operation and control system as claimed in claim 1, characterized in that: The orbit measurement and control module adopts the measurement and control link combination of low-orbit satellite LEO, geosynchronous orbit satellite GEO and high-altitude platform HAPS to provide full-time domain measurement and control coverage.

3. A satellite remote measurement, operation and control system as claimed in claim 2, characterized in that: The measurement and control link optimization module adopts a reinforcement learning algorithm to dynamically adjust the measurement and control path based on the real-time measurement and control link status.

4. A satellite remote measurement, operation and control system as claimed in claim 3, characterized in that: The task monitoring module integrates an AI model for abnormal state prediction and fault mode identification to avoid risks that may affect task execution in advance.

5. A satellite remote measurement, operation and control system as claimed in claim 4, characterized in that: The measurement, operation and control methods of this system are as follows: Step S1, collecting initial track state data and planning the measurement and control link; The orbital state data includes satellite orbital parameters, attitude angle and measurement and control signal quality; The TT&C link planning includes TT&C availability assessment of ground TT&C stations, relay satellites and intersatellite links; Step S2, performing orbit monitoring and correction based on orbit status data, and predicting mission execution parameters; The track monitoring includes collecting telemetry data and signal integrity analysis; The mission execution parameters include orbit drift trend, attitude change range and measurement and control signal attenuation degree; Step S3, optimizing the measurement and control link of step S1 according to the task execution parameters, adjusting the measurement and control link, and performing orbit adjustment; The measurement and control link optimization includes dynamically selecting links based on reinforcement learning; The measurement and control link includes ground measurement and control stations, relay satellites, high-altitude platforms and intersatellite links; The orbit adjustment includes correcting attitude, maintaining orbit and calculating abnormal warning; Step S4, real-time monitoring of satellite status during mission execution, acquisition of mission execution data, and prediction of abnormal measurement and control conditions; The task execution data includes task status, device workload and communication link delay; The abnormal measurement and control prediction includes signal loss, link interruption and mission failure probability; Step S5, adjusting the measurement and control mode based on the abnormal prediction result of S4, optimizing the data return path, and performing orbit adjustment; The measurement and control mode adjustment includes autonomously executing the measurement and control strategy; The orbit adjustment includes autonomous orbit correction or ground command adjustment after the measurement and control link is restored.

6. A satellite remote measurement, operation and control system as claimed in claim 5, characterized in that: In step S3, the step of dynamically selecting a link based on reinforcement learning is: Modeling state space, defining the state vector of the measurement and control link to describe the current state of the link, expressed as: s t =(b t ,d t ,q t ,r t ), Among them, s t represents the state of the measurement and control link at time t, b t represents the bandwidth of the measurement and control link at time t, d t represents the delay of the measurement and control link at time t, q t represents the quality of the measurement and control signal at time t, r t represents the measurement and control resource occupancy rate at time t; Define the action space, expressed as: to t ∈{L1, L2,…, L N }, where a t represents the measurement and control link selected at time t, L i represents the i-th TT&C link, i ranges from i∈{1, 2, …, N}, and N is the total number of optional TT&C links, including ground TT&C stations, relay satellites, high-altitude platforms, and intersatellite links; Set the reward function R t The formula for calculating the quality of the measurement and control path is: R t =α1f1(b t )+α2f2(d t )+α3f3(q t )+α4f4(r t ), Among them, R t is the reward value, f1(b t ) represents the bandwidth gain function, which is expressed as a normalized linear function: f1(b t )=b t / b max , where b max is the maximum bandwidth of the measurement and control link, f2(d t ) represents the delay loss function, which is expressed by a negative exponential decay function: where λ d is the delay weight factor, f3(q t ) represents the signal quality function, which is represented by logarithmic mapping: f3(q t )=log(1+q t ), f4(r t ) represents the resource occupancy rate penalty function, which is expressed using a linear penalty model: f4(r t )=-r t / r max , α1, α2, α3, α4 are the weight factors of the reward function, satisfying the normalization condition: α1+α2+α3+α4=1, Among them, α1 weighs the importance of bandwidth, α2 weighs the impact of latency on the link, α3 weighs the impact of signal quality on the measurement and control link, and α4 weighs the impact of resource contention.

7. A satellite remote measurement, operation and control system as claimed in claim 6, characterized in that: The step of dynamically selecting a link based on reinforcement learning also includes: Define the task priority function, the function formula is: P t =β1T d +β2W c +β3C o , Among them, P t is the task priority at time t, T d is the task deadline, W c Calculate the load for the task, C o is the occupancy rate of the measurement and control link, β1, β2, β3 are priority weight factors, and the normalization condition is satisfied: β1+β2+β3＝1， Among them, β1 weighs the importance of task deadline, β2 weighs the impact of computational load on priority, and β3 weighs the impact of measurement and control link occupancy; When the measurement and control link resources are limited, the allocation strategy is adjusted, and the adjustment formula is: U t =γ1P t +γ2R t , Among them, U t is the comprehensive score, γ1 and γ2 are adjustment factors, satisfying: γ1+γ2=1, P t is the task priority, R t is the reward value of the measurement and control link; Deep Q learning is used for link optimization and Q value is updated. The update formula is: in, Indicates that in state s t+1 Select the Q value of the best action, Q(s t , a t ) means in state s t Select action a t Q value, η is the learning rate, which determines the Q value update step size, R t is the current reward value, and λ is the discount factor.

8. A satellite remote measurement, operation and control system as claimed in claim 7, characterized in that: In step S4, during the mission execution, the satellite status is monitored in real time, mission execution data is acquired, and abnormal measurement and control conditions are predicted in the following manner: Monitor the satellite mission status in real time and define the state vector as S' t : S′ t (X t ,Y t ,z T ,Θ t ,P′ t ,L′ t ), Among them, S' t is the satellite status monitored at time t, X t , Y t , Z t is the three-dimensional position of the satellite in the orbital coordinate system, Θ t is the satellite attitude angle, P' t is the task status, L' t is the link communication delay; The Bayesian method is used for anomaly detection to calculate the probability of anomaly occurrence. The calculation formula is: Among them, P(A|S' t ) is a given state S' t The abnormal probability under t |A) is the observed state probability under abnormal conditions, P(A) is the prior probability of abnormality, P(S' t ) is the total probability of the current state.

9. A satellite remote measurement, operation and control system as claimed in claim 8, characterized in that: In step S4, a misjudgment control mechanism is introduced during the process of predicting abnormal measurement and control conditions. The chi-square test method is used for abnormal identification, and the formula is: Among them, X 2 is the statistic for anomaly detection, O i is the observed value, E i is the expected value, Setting Thresholds like It is judged as abnormal; The LSTM neural network is used to predict abnormal status, and the status is updated as follows: h' t =f(W' h h' t-1 +W' x S' t +b'), Among them, h' t is the hidden state predicted by LSTM, h' t-1 represents the hidden state of the previous time step, W' h is the hidden state weight matrix, W' x is the input weight matrix, b' is the bias term, and f(·) is the activation function.

10. A satellite remote measurement, operation and control system as claimed in claim 9, characterized in that: In step S5, the step of autonomously executing the measurement and control strategy is: After predicting an abnormality, the measurement and control strategy is adjusted autonomously: Optimize the data return path to have the lowest cost and fastest response time. The optimization formula is: Among them, R opt is the optimal data return path, C(L i ) is the link L i The transmission cost, T(L i ) is the link L i The transmission time, μ is the adjustment factor, The encryption is performed using a hybrid encryption method. The encryption process is as follows: Among them, C' is the encrypted data, E k (D') represents the encrypted data D' with key k, Represents an XOR operation, which is used to combine encrypted data and hash values, H(S′ t ) is the data integrity hash value; Adjust the measurement and control mode according to the abnormal situation and define the adjustment function, which is expressed as: M′ t =δ1A′ t +δ2R′ t , Among them, M' t Adjust the parameters for the measurement and control strategy, A' t is the abnormal level, R' t is the current measurement and control link status, δ1 and δ2 are adjustment factors; Track adjustment is performed based on the predictive control model, and the adjustment formula is: X′ t+1 =A′X′ t +B′U′ t , Among them, X' t+1 is the adjusted orbital state, A' is the orbital dynamics matrix, B' is the control input matrix, and U' t Adjust strategy for the track.

Images