Carrier rocket power supply data transmission and monitoring method and device and storage medium

Abstract

The invention relates to the field of aerospace data transmission and monitoring, and discloses a carrier rocket power supply data transmission and monitoring method and device and a storage medium, and the method comprises the following steps: obtaining historical data for analyzing a network state, and predicting network transmission abnormality based on a long short-term memory network model; dynamically adjusting a threshold value through a sliding window, and triggering an abnormal response mechanism; a reliability enhancement mechanism is adopted, data transmission is optimized through a TCP exponential backoff algorithm and a UDP redundancy coding technology, and data loss and retransmission are reduced; a Q-Learning algorithm in reinforcement learning is utilized to dynamically adjust transmission parameters according to a network state, the transmission parameters comprise a window size, a retransmission interval and a data packet size, and self-adaptive optimization is realized. The method is suitable for reliable transmission of key data and instructions in launch tasks of carrier rockets, and can be widely applied to data monitoring of spaceflight launch, high-concurrency network transmission and complex scenes.

Description

Technical Field

[0001] The present invention relates to the field of aerospace data transmission and monitoring technology, and in particular to a method, device and storage medium for data transmission and monitoring of a launch vehicle power supply. Background Art

[0002] With the increasing demand for space launch missions, especially in sea launches and complex network environments, the data transmission and monitoring of power equipment and onboard products of launch vehicles face higher real-time and reliability requirements. In traditional launch scenarios, the transmission of commands and parameters mostly adopts a communication mode based on TCP or UDP protocols. However, in dynamic network environments, such as high latency, frequent switching, or network congestion, the performance of traditional communication protocols is often difficult to meet the requirements. Although the TCP protocol has certain retransmission and reliability guarantees, its congestion control and fixed parameter transmission strategy can easily lead to reduced transmission efficiency or retransmission failures in complex scenarios. The UDP protocol lacks a confirmation mechanism and retransmission function. In the case of a high packet loss rate, it may lead to the loss of key instructions, thereby threatening the safety of the mission.

[0003] Another problem with existing technologies is that, in terms of network anomaly detection, most systems rely on fixed rules or static thresholds for monitoring. This approach lacks sufficient sensitivity to fluctuations in network status and is difficult to detect potential anomalies in a timely manner, especially in complex and dynamically changing network environments, with high rates of false positives and false negatives. In addition, traditional methods are usually unable to adaptively adjust transmission parameters according to actual network status, and transmission efficiency and bandwidth utilization are significantly limited. In scenarios with high concurrency or large data volume transmission, the fixed transmission strategy used in existing technologies is prone to waste of resources or degradation of network performance.

[0004] Therefore, in view of the problem that the existing technology lacks efficient and reliable anomaly detection capabilities, dynamic adjustment capabilities and transmission optimization mechanisms in complex network environments, there is an urgent need for a launch vehicle power data transmission and monitoring method that can take into account real-time, reliability and adaptability to meet the strict requirements for critical data transmission in space missions. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a method, device and storage medium for transmitting and monitoring power supply data of a launch vehicle, which solves the problems of insufficient real-time performance, reduced transmission reliability and lack of dynamic adjustment capability caused by network anomalies in launch vehicle data transmission under a complex network environment.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for transmitting and monitoring power supply data of a launch vehicle, comprising the following steps: Obtain and analyze historical data of network status and predict network transmission anomalies; Trigger response actions based on dynamically adjusted thresholds; Optimize data transmission through reliability enhancement mechanism to prevent data loss and retransmission; Dynamically adjust transmission parameters according to the network environment to achieve adaptive optimization and improve transmission efficiency.

[0007] Preferably, the step of acquiring and analyzing historical data of network status and predicting network transmission anomalies includes: Obtain historical data of network status, construct a network feature input matrix containing multiple time steps, and use the data of each time step as the input sequence; Use the long short-term memory network model to analyze the historical input sequence and extract the time correlation of the network state; Identify possible anomalies in network transmission based on the prediction results.

[0008] Preferably, the step of obtaining historical data of network status includes: Monitor network communication status in real time through the multi-protocol capture module, and obtain data packets of TCP, UDP and ICMP protocols; Extract key network features from data packets, including latency, packet loss rate, bandwidth utilization, and network congestion status; The network features are recorded and sorted in chronological order to form time series data, and then preprocessed.

[0009] Preferably, the step of triggering a response measure according to the dynamically adjusted threshold comprises: The historical data of the network status prediction value is obtained through the sliding window method, and the mean and standard deviation of the prediction value within the sliding window are calculated; Dynamically adjust the threshold based on the mean and standard deviation of historical forecast values, and use the adjustment factor to set the threshold at the current moment; Compare the current network status prediction value with the dynamically adjusted threshold, and trigger the response mechanism when the prediction value exceeds the threshold.

[0010] Preferably, the response mechanism includes automatically starting network reconnection, adjusting the transmission path or pausing instruction sending, and sending an abnormal notification signal to the monitoring system.

[0011] Preferably, the step of optimizing data transmission through a reliability enhancement mechanism and preventing data loss and retransmission includes: For TCP protocol transmission, an exponential backoff algorithm is used to dynamically adjust the retransmission interval. When the instruction transmission does not receive a confirmation message, the retransmission interval is gradually increased, and the maximum number of retransmissions is set to limit invalid retransmissions. For UDP protocol transmission, the flag bit of the instruction packet is set to indicate the instruction sending status and receiving status through the confirmation mechanism; Redundant coding technology is introduced in the UDP transmission process to restore lost data through redundant information in the event of data packet loss; When the network status changes, the packet loss rate and delay are monitored in real time, and the transmission window size is dynamically adjusted to optimize data throughput; Based on the transmission feedback information, it is determined whether the command is successfully transmitted. Unsuccessful commands are resent through the selected reliable transmission mechanism to ensure that the data reaches the target device accurately.

[0012] Preferably, the step of dynamically adjusting transmission parameters according to the network environment to achieve adaptive optimization and improve transmission efficiency includes: Monitor current network status, including latency, packet loss rate, and bandwidth utilization, and obtain real-time network performance indicators; A transmission parameter optimization model is established using a reinforcement learning algorithm to predict the current optimal transmission parameter configuration based on the network status; Dynamically adjust transmission parameters, including transmission window size, retransmission interval, and packet size, to adapt to changes in network status; The reward mechanism of reinforcement learning is updated in real time according to network status feedback, and the transmission strategy is iteratively optimized.

[0013] Preferably, the reinforcement learning algorithm is a Q-Learning algorithm, and the state and action space in the Q-Learning algorithm includes: State space: represents network characteristics, including latency, packet loss rate, and current bandwidth utilization; Action space: includes retransmission interval adjustment, window size change and path optimization scheme; The reward function is dynamically adjusted based on the comprehensive weight of transmission success rate and transmission delay.

[0014] The present invention also provides a launch vehicle power data transmission and monitoring device, comprising: The data acquisition module is used to obtain historical data of network status, construct network feature input containing multiple time steps, and use it to predict network transmission anomalies; The anomaly detection module is used to analyze historical network status data, extract the time correlation of network status through the long short-term memory network model, and predict possible network anomalies; The threshold dynamic adjustment module is used to calculate the mean and standard deviation of the predicted value based on the sliding window method, dynamically adjust the network anomaly detection threshold, and trigger response measures according to the threshold; Reliability enhancement module, which is used to optimize data transmission through reliability enhancement mechanisms, including UDP flag confirmation, TCP exponential backoff retransmission strategy and redundant coding technology to prevent data loss and retransmission; The parameter optimization module is used to adjust the transmission parameters according to the dynamic changes of the network environment, optimize the transmission window size, retransmission interval time and data packet size, and realize the improvement of transmission efficiency and optimization of network adaptability.

[0015] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the above method is implemented.

[0016] The present invention also provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0017] The present invention provides a method, device and storage medium for transmitting and monitoring power supply data of a launch vehicle. It has the following beneficial effects: 1. The present invention significantly reduces the data packet loss rate in a complex network environment by adopting the TCP exponential backoff algorithm and UDP redundant coding technology, ensuring that key instructions and data can be accurately transmitted to the target device. The flag confirmation mechanism further enhances the reliability of the UDP protocol and effectively solves the problems of data loss and low retransmission efficiency in traditional transmission schemes.

[0018] 2. The network status prediction based on the long short-term memory network (LSTM) model of the present invention can accurately capture abnormal conditions such as delays and packet loss in network transmission. The abnormality detection threshold is dynamically adjusted in combination with the sliding window, which can adapt to the changing network environment, reduce false positives and false negatives, and ensure that the system responds quickly and takes corresponding measures when an abnormality occurs.

[0019] 3. The present invention uses the Q-Learning algorithm in reinforcement learning to adjust transmission parameters such as window size, retransmission interval and packet size in real time according to the network status, and dynamically optimizes transmission efficiency. In the case of high latency or network congestion, the stability and efficiency of data transmission are guaranteed through real-time learning and adjustment strategies.

[0020] 4. The present invention can dynamically optimize the transmission path and strategy under different network conditions through real-time monitoring and analysis of network characteristics, reducing invalid retransmissions and network resource waste. The adaptive transmission window and data packet size adjustment mechanism further improves bandwidth utilization, especially in high-concurrency or large data volume transmission scenarios.

[0021] 5. The modular design of the present invention supports multi-protocol expansion and multi-scenario adaptation. It is not only applicable to TCP and UDP protocols, but also can be extended to other transmission protocols such as ICMP. According to different mission requirements, such as sea launch, high concurrency environment or cross-regional transmission, the transmission strategy can be flexibly adjusted to provide comprehensive support for complex mission scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the method flow of the present invention; Figure 2 It is a schematic diagram of the structure of the device of the present invention; Figure 3 It is a schematic diagram of the computer device structure of the present invention.

[0023] Among them, 100, data acquisition module; 200, anomaly detection module; 300, threshold dynamic adjustment module; 400, reliability enhancement module; 500, parameter optimization module; 40, computer equipment; 41, processor; 42, memory; 43, storage medium. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Please see attached Figure 1 The present invention provides a method for transmitting and monitoring power supply data of a launch vehicle, which combines deep learning, reliability optimization mechanism and reinforcement learning technology to effectively improve the reliability and efficiency of data transmission in a complex network environment. The following is a detailed description of each step of the method of the present invention.

[0026] S1. Obtain and analyze historical data of network status and predict network transmission anomalies In this embodiment, the historical data of network status is obtained and analyzed, and network transmission anomalies are predicted, which involves real-time monitoring of network status during the data transmission of the launch vehicle and the capture and analysis of historical data. Through the application of deep learning models, the network status can be dynamically predicted, providing a reliable foundation for subsequent steps.

[0027] Generally, the acquisition of historical network status data is achieved through the network driver interface, which includes the capture and analysis of TCP, UDP and ICMP protocol packets. The network driver interface supports multi-protocol extension and can accurately extract the following key network features in a complex network environment: delay, packet loss rate, bandwidth utilization and network congestion status.

[0028] Specifically, latency refers to the time required for a data packet to reach the receiving end from the sender; packet loss rate refers to the ratio of the number of data packets that are not successfully received within a certain time window to the total number of packets sent; bandwidth utilization reflects the maximum percentage of bandwidth occupied by the current transmission data traffic; and network congestion status is determined by analyzing queue length or transmission rate fluctuations.

[0029] In some embodiments, these captured network features are organized in time series and constructed into an input matrix. The input matrix is ​​in the form of:

[0030] in, Represents the time step No. network features, such as The delay or packet loss rate at a certain moment.

[0031] As an option, the input matrix can be preprocessed, including data denoising, normalization, and standardization, to eliminate the interference of noise on the prediction and ensure the uniformity of data features before inputting into the model. For example, the unit of delay can be normalized to a relative value in the millisecond range, and the packet loss rate can be normalized to a ratio value from 0 to 1.

[0032] As an option, the input matrix can be preprocessed, including data denoising, normalization, and standardization, to eliminate the interference of noise on the prediction and ensure the uniformity of data features before inputting into the model. For example, the unit of delay can be normalized to a relative value in the millisecond range, and the packet loss rate can be normalized to a ratio value from 0 to 1.

[0033] In the Gate of Oblivion, the current moment Input and the hidden state at the previous moment It is used to calculate the forgetting ratio, the formula is:

[0034] in: : Output of the forget gate (a value between 0 and 1).

[0035] : Sigmoid activation function.

[0036] and : The weight matrix and bias term of the forget gate.

[0037] : The hidden state at the previous moment.

[0038] : Input data at the current moment.

[0039] In the input gate, the new input is processed as a candidate memory. The input gate is used to control how much of the new input is written into the memory cell. The formula is:

[0040] in: : The output of the input gate, which controls the importance of the current input.

[0041] The updated candidate memories are:

[0042] in: : Candidate memory vector, representing the new memory generated by the current input.

[0043] Subsequently, the memory cell state is updated through the forget gate and the input gate, and the formula is:

[0044] in: : The memory cell state at the current time step.

[0045] Finally, the hidden state at the current moment is generated through the output gate , the formula is:

[0046]

[0047] in: : Output vector of the output gate.

[0048] : The hidden state at the current time step.

[0049] The input sequence given the network state , where each Contains multiple network features such as latency, packet loss rate and other multi-source data. After multiple layers of LSTM calculations, in the last layer, we usually use the Sigmoid activation function to finally output a predicted probability :

[0050] if , the network is judged to be abnormal and automatic reconnection is triggered.

[0051] In one possible implementation, the output of LSTM is the probability of network anomaly occurring at the current time step. Combined with the feature input of the historical time step, this probability value is used to determine whether there is a potential anomaly in the network.

[0052] As an extension, the LSTM model can be stacked in multiple layers to further extract higher-order time-related features, thereby improving the accuracy of prediction. For example, the number of hidden layers can be increased or the dimension of hidden units can be adjusted in the model to meet the prediction needs of network states of different complexities.

[0053] In this embodiment, when the output probability value of the LSTM model exceeds the threshold, it can be determined that the network is abnormal. This prediction result will be used to trigger the dynamic response mechanism in the subsequent steps. The above process can effectively obtain the dynamic characteristics of the network state and provide reliable support for abnormal prediction in complex network environments.

[0054] S2. Trigger response measures based on dynamically adjusted thresholds In this embodiment, triggering response measures based on dynamically adjusted thresholds involves dynamically calculating predicted thresholds using a sliding window method, and triggering corresponding response mechanisms based on real-time changes in network status, thereby improving the efficiency and accuracy of network anomaly processing.

[0055] Generally speaking, a sliding window is a method for processing real-time data by maintaining a fixed-size time window containing a certain number of historical data points for dynamically calculating statistical characteristics. In this embodiment, the historical data points in the sliding window are network anomaly prediction values, which are generated by the LSTM model and can reflect the changing trend of the network status.

[0056] Specifically, for the predicted value sequence in the sliding window , first calculate its mean and standard deviation , respectively represent the central trend and volatility of the data in the sliding window. The formula is as follows:

[0057]

[0058] In one possible implementation, the above mean and standard deviation are used to dynamically calculate the anomaly detection threshold for the current time step:

[0059] in, is a tuning factor used to control the sensitivity of anomaly detection. The value can reduce the false positive rate, while a smaller The value can increase the sensitivity of anomaly detection.

[0060] As an option, the sliding window size can be adjusted dynamically based on changes in network status. For example, in the case of large network fluctuations, reducing to improve the real-time response; when the network status is stable, it can be appropriately increased. To reduce calculation fluctuations.

[0061] When LSTM predicts the probability of network anomaly Exceeding dynamically calculated threshold When the system triggers the corresponding response mechanism, the response mechanism includes the following: Automatically reestablish network connection: The system automatically attempts to reestablish a stable network connection to ensure continuity of data transmission.

[0062] Switching transmission paths: In a network that supports multi-path transmission, the system can automatically select an alternative path to bypass the currently congested or interfered transmission path.

[0063] Suspend command sending: In extreme cases, if the network status cannot meet the minimum transmission requirements, the system will temporarily stop sending commands and notify the monitoring center for manual intervention.

[0064] In some embodiments, the response mechanism can be expanded according to specific scenarios. For example, in a sea launch mission, the switching strategy of the satellite network can be combined to give priority to low-latency, high-bandwidth communication links to meet the data transmission requirements during the rocket launch.

[0065] In this embodiment, the combination of dynamic threshold adjustment and response mechanism can significantly improve the accuracy and timeliness of network exception processing. At the same time, by dynamically adjusting the sliding window size and adjustment factor , enabling the system to flexibly respond to changing network environments and providing a strong guarantee for the reliable transmission of data in subsequent steps.

[0066] S3. Optimize data transmission through reliability enhancement mechanism to prevent data loss and retransmission In this embodiment, data transmission is optimized through a reliability enhancement mechanism to prevent and control data loss and retransmission, and corresponding optimization strategies are designed based on the characteristics of TCP and UDP protocols, thereby improving the reliability and efficiency of data transmission in a complex network environment.

[0067] Generally, the TCP protocol itself provides certain reliability guarantees, such as retransmission mechanism and congestion control, but in a high-latency or unstable network environment, the traditional fixed parameter retransmission strategy may lead to low efficiency and transmission failure. Therefore, this embodiment uses an exponential backoff algorithm to optimize the TCP retransmission mechanism.

[0068] Specifically, TCP's retransmission interval is dynamically adjusted according to the number of retransmissions. The initial retransmission interval is set to After each retransmission failure, the retransmission interval increases exponentially. The specific calculation formula is:

[0069] in, is the current number of retransmissions. When the number of retransmissions reaches the preset maximum value When the delay is too long, retransmission stops and the exception handling mechanism is triggered. As a possible implementation method, the delay can be dynamically adjusted according to the actual network delay. , to meet the needs of different network scenarios.

[0070] In terms of UDP protocol optimization, since UDP itself does not provide reliability guarantee, data packets may fail to reach the target device due to network loss or delay. Therefore, this embodiment introduces a flag confirmation mechanism and redundant coding technology to enhance the reliability of UDP transmission.

[0071] Specifically, the flag confirmation mechanism adds a specific status flag to each instruction packet to indicate the sending and receiving status of the data packet. After successfully receiving the data packet, the receiving end returns confirmation information to the sending end; if the sending end does not receive the confirmation within the set time, it resends the corresponding data packet.

[0072] Redundant coding technology adds redundant information to data packets, so that even if part of the data packet is lost, it can be restored by the remaining data. For example, Reed-Solomon coding can be used to The data packets are divided into data blocks, any The original data can be restored from each data block. Redundant encoding parameters and It can be dynamically adjusted according to the network packet loss rate to achieve a balance between reliability and transmission efficiency.

[0073] As an option, the network status can be monitored in real time, and the transmission parameters of TCP and UDP can be dynamically adjusted according to the delay, packet loss rate and bandwidth utilization. For example, when the network delay is high, the packet size can be reduced to reduce the probability of packet loss; when the bandwidth utilization is close to saturation, the window size can be limited to reduce network congestion.

[0074] In one possible implementation, TCP and UDP transmission optimization can be combined with reinforcement learning methods to automatically select the optimal transmission strategy by continuously learning the relationship between network status and transmission parameters. For example, the retransmission interval and data redundancy can be adjusted according to different network conditions to achieve adaptive performance optimization.

[0075] This embodiment designs a reliability enhancement mechanism for the TCP and UDP protocols, which can ensure the reliability of data transmission in a complex network environment with high latency and packet loss rate, while significantly reducing the overhead of invalid retransmissions, providing stable support for the transmission of commands and parameters of the launch vehicle.

[0076] S4. Dynamically adjust transmission parameters according to the network environment to achieve adaptive optimization and improve transmission efficiency In this embodiment, the transmission parameters are dynamically adjusted according to the network environment to achieve adaptive optimization and improved transmission efficiency, which involves a technical solution that uses real-time network status monitoring combined with a reinforcement learning algorithm to dynamically optimize the transmission parameters, thereby maintaining high efficiency and stability of data transmission under different network conditions.

[0077] Generally, the network environment may become complex and changeable due to bandwidth fluctuations, increased delays or increased packet loss rates. To address this problem, this embodiment establishes a dynamic transmission parameter adjustment mechanism by monitoring the network status (including delay, packet loss rate and bandwidth utilization) in real time as input.

[0078] Specifically, this embodiment uses the Q-Learning algorithm in reinforcement learning to dynamically adjust the transmission parameters according to changes in network status. Q-Learning is a model-free reinforcement learning algorithm, the core of which is to use state-action pairs to Reward value Update the Q value table and gradually approach the optimal strategy. The Q value update formula is:

[0079] in, is the learning rate, which controls the speed of updating; is the discount factor, which measures the importance of future rewards; For immediate rewards; The next state reached after executing the action.

[0080] In a possible implementation, the state space of this embodiment is Including network characteristics such as delay, packet loss rate and bandwidth utilization; action space This includes adjusting the retransmission interval, modifying the window size, changing the packet size, and switching the transmission path. For example, when the delay is high and the packet loss rate increases, you can choose to increase data redundancy and reduce the window size to reduce data loss and network congestion.

[0081] As an option, instant reward value It can be defined based on the real-time performance indicators of the transmission. For example, a successful transmission of a data packet can be given a positive reward, and a packet that is lost or delayed beyond the limit can be given a negative reward, thereby guiding the Q-Learning algorithm to optimize the parameter adjustment strategy.

[0082] To further improve efficiency, this embodiment incorporates a rapid response mechanism. When the network environment changes dramatically (such as a sudden drop in bandwidth or a sudden increase in latency), low-latency paths are prioritized or non-essential data transmission is suspended to ensure the transmission stability of key instructions. For example, in a sea launch mission, the system can switch to a high-priority satellite communication channel and temporarily reduce the transmission rate of low-priority data.

[0083] To further improve efficiency, this embodiment incorporates a rapid response mechanism. When the network environment changes dramatically (such as a sudden drop in bandwidth or a sudden increase in latency), low-latency paths are prioritized or non-essential data transmission is suspended to ensure the transmission stability of key instructions. For example, in a sea launch mission, the system can switch to a high-priority satellite communication channel and temporarily reduce the transmission rate of low-priority data.

[0084] In another possible implementation, in order to improve the convergence speed and effect of Q-Learning, deep reinforcement learning methods (such as DQN) can be combined to approximate the Q-value function with a neural network. The input of the neural network is the network feature state, and the output is the Q-value score of the action adjusted with different transmission parameters, from which the action with the highest score is selected for execution. This method is suitable for complex scenarios with large state and action spaces.

[0085] This embodiment dynamically adjusts key parameters such as transmission window size, retransmission interval, data redundancy, etc., and continuously optimizes the transmission strategy in combination with network feedback, thereby achieving adaptive adjustment to the dynamic network environment. In some embodiments, the trend analysis of historical data can be further combined to predict the future network status, adjust the transmission parameters in advance, and prevent performance fluctuations.

[0086] Through the above technical solution, this embodiment can maintain the stability of transmission efficiency in a complex network environment and significantly reduce data loss or delay caused by network fluctuations, thereby providing a guarantee for high-reliability transmission of launch vehicle power data.

[0087] In general, the present invention realizes network status anomaly prediction, trigger response mechanism based on dynamic adjustment threshold, reliability-enhanced transmission optimization, and dynamic adjustment of transmission parameters by reinforcement learning, so as to ensure the reliability and real-time performance of data transmission in a complex network environment. Specific technologies include using the sliding window method to calculate the anomaly detection threshold, using TCP exponential backoff and UDP redundant coding to improve data reliability, and optimizing transmission parameters through the Q-Learning algorithm to achieve adaptive adjustment of network status.

[0088] The launch vehicle power data transmission and monitoring device described below and the launch vehicle power data transmission and monitoring method described above can be referenced to each other.

[0089] Please see attached Figure 2 The present invention also provides a launch vehicle power data transmission and monitoring device, comprising: The data acquisition module 100 is used to acquire historical data of network status, construct network feature inputs containing multiple time steps, and use them to predict network transmission anomalies; Anomaly detection module 200, used to analyze historical network status data, extract the time correlation of network status through long short-term memory network model, and predict possible network anomalies; A threshold dynamic adjustment module 300, for calculating the mean and standard deviation of the predicted value based on a sliding window method, dynamically adjusting the network anomaly detection threshold, and triggering response measures according to the threshold; Reliability enhancement module 400, used to optimize data transmission through reliability enhancement mechanism, including adopting UDP flag confirmation, TCP exponential backoff retransmission strategy and redundant coding technology to prevent data loss and retransmission; The parameter optimization module 500 is used to adjust the transmission parameters according to the dynamic changes of the network environment, optimize the transmission window size, retransmission interval time and data packet size, and realize the improvement of transmission efficiency and the optimization of network adaptability.

[0090] The device of this embodiment can be used to execute the above method embodiment, and its principles and technical effects are similar, which will not be repeated here.

[0091] Please see attached Figure 3 The present invention further provides a computer device 40, comprising: a processor 41 and a memory 42, wherein the memory 42 stores a computer program executable by the processor, and when the computer program is executed by the processor, the above method is executed.

[0092] The present invention further provides a storage medium 43 on which a computer program is stored. When the computer program is run by the processor 41, the above method is executed.

[0093] Among them, the storage medium 43 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0094] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for transmitting and monitoring power supply data of a launch vehicle, characterized in that: The following steps are involved: Obtain and analyze historical data of network status and predict network transmission anomalies; Trigger response actions based on dynamically adjusted thresholds; Optimize data transmission through reliability enhancement mechanism to prevent data loss and retransmission; Dynamically adjust transmission parameters according to the network environment to achieve adaptive optimization and improve transmission efficiency.

2. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 1, characterized in that: The steps of obtaining and analyzing historical data of network status and predicting network transmission anomalies include: Obtain historical data of network status, construct a network feature input matrix containing multiple time steps, and use the data of each time step as the input sequence; Use the long short-term memory network model to analyze the historical input sequence and extract the time correlation of the network state; Identify possible anomalies in network transmission based on the prediction results.

3. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 2, characterized in that: The step of obtaining historical data of network status includes: Monitor network communication status in real time through the multi-protocol capture module, and obtain data packets of TCP, UDP and ICMP protocols; Extract key network features from data packets, including latency, packet loss rate, bandwidth utilization, and network congestion status; The network features are recorded and sorted in chronological order to form time series data, and then preprocessed.

4. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 1, characterized in that: The step of triggering a response measure according to the dynamically adjusted threshold value comprises: The historical data of the network status prediction value is obtained through the sliding window method, and the mean and standard deviation of the prediction value within the sliding window are calculated; Dynamically adjust the threshold based on the mean and standard deviation of historical forecast values, and use the adjustment factor to set the threshold at the current moment; Compare the current network status prediction value with the dynamically adjusted threshold, and trigger the response mechanism when the prediction value exceeds the threshold.

5. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 4, characterized in that: The response mechanism includes automatically starting network reconnection, adjusting the transmission path or pausing instruction sending, and sending an abnormal notification signal to the monitoring system.

6. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 1, characterized in that: The steps of optimizing data transmission and preventing data loss and retransmission by means of a reliability enhancement mechanism include: For TCP protocol transmission, an exponential backoff algorithm is used to dynamically adjust the retransmission interval. When the instruction transmission does not receive a confirmation message, the retransmission interval is gradually increased, and the maximum number of retransmissions is set to limit invalid retransmissions. For UDP protocol transmission, the flag bit of the instruction packet is set to indicate the instruction sending status and receiving status through the confirmation mechanism; Redundant coding technology is introduced in the UDP transmission process to restore lost data through redundant information in the event of data packet loss; When the network status changes, the packet loss rate and delay are monitored in real time, and the transmission window size is dynamically adjusted to optimize data throughput; Based on the transmission feedback information, it is determined whether the command is successfully transmitted. Unsuccessful commands are resent through the selected reliable transmission mechanism to ensure that the data reaches the target device accurately.

7. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 1, characterized in that: The step of dynamically adjusting the transmission parameters according to the network environment to achieve adaptive optimization and improve transmission efficiency includes: Monitor current network status, including latency, packet loss rate, and bandwidth utilization, and obtain real-time network performance indicators; A transmission parameter optimization model is established using a reinforcement learning algorithm to predict the current optimal transmission parameter configuration based on the network status; Dynamically adjust transmission parameters, including transmission window size, retransmission interval, and packet size, to adapt to changes in network status; The reward mechanism of reinforcement learning is updated in real time according to network status feedback, and the transmission strategy is iteratively optimized.

8. A method for transmitting and monitoring power supply data of a launch vehicle according to claim 7, characterized in that: The reinforcement learning algorithm is a Q-Learning algorithm, and the state and action space in the Q-Learning algorithm includes: State space: represents network characteristics, including latency, packet loss rate, and current bandwidth utilization; Action space: includes retransmission interval adjustment, window size change and path optimization scheme; The reward function is dynamically adjusted based on the comprehensive weight of transmission success rate and transmission delay.

9. A launch vehicle power data transmission and monitoring device, applied to the method according to any one of claims 1 to 8, characterized in that: include: The data acquisition module is used to obtain historical data of network status, construct network feature input containing multiple time steps, and use it to predict network transmission anomalies; The anomaly detection module is used to analyze historical network status data, extract the time correlation of network status through the long short-term memory network model, and predict possible network anomalies; The threshold dynamic adjustment module is used to calculate the mean and standard deviation of the predicted value based on the sliding window method, dynamically adjust the network anomaly detection threshold, and trigger response measures according to the threshold; Reliability enhancement module, which is used to optimize data transmission through reliability enhancement mechanisms, including UDP flag confirmation, TCP exponential backoff retransmission strategy and redundant coding technology to prevent data loss and retransmission; The parameter optimization module is used to adjust the transmission parameters according to the dynamic changes of the network environment, optimize the transmission window size, retransmission interval time and data packet size, and realize the improvement of transmission efficiency and optimization of network adaptability.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images