A Master-Slave Transparent Transmission Method Based on Intelligent Error Correction Coding and Adaptive Retransmission

Abstract

This invention discloses a transparent transmission method for master-slave stations based on intelligent error correction coding and adaptive retransmission, comprising the following steps: Deploying a transparent gateway at the plant end, using a Denoising Autoencoder to intelligently correct and encode relay protection information, compressing the original data into a low-dimensional latent representation, and removing noise during the encoding process to generate coded data suitable for transmission; during data transmission, monitoring the network status in real time based on an LSTM model, and dynamically adjusting the retransmission strategy according to the predicted network status, generating transmission control parameters adapted to the current network conditions; after receiving the coded data at the master station, restoring it to high-dimensional data using a decoder, and triggering a retransmission mechanism when data errors are detected, performing adaptive retransmission in conjunction with the aforementioned transmission control parameters, and finally outputting complete and error-free relay protection information. This invention has advantages such as accurate and efficient transmission.

Description

Technical Field

[0001] This invention mainly relates to the field of smart grid technology, specifically to a transparent transmission method for master and slave stations based on intelligent error correction coding and adaptive retransmission. Background Technology

[0002] In the construction of smart grids, the accurate transmission of relay protection information is crucial for the safe operation of the power grid. In recent years, information protection systems based on transparent transmission (transparent transmission) schemes have gradually become a new technological trend. These systems achieve full transparent forwarding of information from station-side protection devices through transparent gateway equipment, enabling the master station to directly access and control the protection equipment. However, despite the significant advantages of transparent transmission schemes in reducing on-site configuration workload and improving data access quality, they still face many challenges in practical applications.

[0003] While transparent transmission simplifies information transmission paths, the complexity of the power system communication environment still results in a high data transmission error rate. Especially in scenarios involving large-scale data exchange and parallel communication of multiple protection devices, erroneous data during transmission not only reduces system reliability but can also lead to incomplete information received from protection devices, affecting user judgment and decision-making, and ultimately impacting the stable operation of the power grid. Furthermore, existing retransmission mechanisms mostly rely on fixed rules and cannot adaptively adjust to dynamic changes in the real-time communication environment, leading to low retransmission efficiency, wasted network resources, and increased data transmission latency. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides a transparent transmission method for master and slave stations based on intelligent error correction coding and adaptive retransmission to ensure accurate data transmission.

[0005] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows: A transparent transmission method for master and slave stations based on intelligent error correction coding and adaptive retransmission includes the following steps: Deploy transparent gateways at the plant and substation end, and use Denoising Autoencoder to perform intelligent error correction encoding on relay protection information. Compress the original data into a low-dimensional potential representation, remove noise during the encoding process, and generate encoded data suitable for transmission. During data transmission, the network status is monitored in real time based on the LSTM model, and the retransmission strategy is dynamically adjusted according to the prediction results of the network status to generate transmission control parameters that are adapted to the current network conditions. After receiving the encoded data, the master station restores it to high-dimensional data through the decoder. When a data error is detected, a retransmission mechanism is triggered. Adaptive retransmission is performed in combination with the transmission control parameters, and finally complete and error-free relay protection information is output.

[0006] Preferably, the process of intelligent error correction coding via Denoising Autoencoder includes the following sub-steps: The original relay protection information is preprocessed to remove noise and standardize the data; A multi-layer neural network is constructed as an encoder to progressively compress the input data into a low-dimensional latent representation; Design a bottleneck layer that retains key features and further filters out redundant information; Construct a decoder that is symmetric to the encoder to restore the low-dimensional representation to high-dimensional data; The model is trained using mean squared error or L1 loss function, and the model's robustness is enhanced by adding noise to the input data. Regularization strategies and data augmentation techniques are introduced to improve the robustness and generalization ability of the model.

[0007] Preferably, the encoder training process includes: Add Gaussian noise or other transmission noise to the input data to simulate the actual communication environment; Use a standardized dataset as training data and target noise-free data as the output. A multi-layer fully connected neural network is used, and the number of neurons is reduced layer by layer to finally output the bottleneck layer representation. Backpropagation training is performed using the mean squared error loss function and the Adam optimizer. An early stopping mechanism is introduced to prevent model overfitting.

[0008] Preferably, the training process of the decoder includes: Construct a decoder structure that mirrors the encoder, and restore the data dimensions layer by layer; The decoder is trained to reconstruct the original data from the low-dimensional representation. By comparing the reconstructed data with the original data, the number of network layers, neurons, and activation functions are optimized.

[0009] Preferably, the LSTM-based adaptive retransmission mechanism includes the following sub-steps: Embed the LSTM model in the transparent gateway to monitor network bandwidth, latency, and packet loss rate in real time; LSTM models perform time series analysis on network states to predict future network trends. The retransmission frequency, redundancy coding strength, or transmission rate are dynamically adjusted based on the prediction results. Design a feedback control mechanism to optimize the retransmission strategy in real time based on the actual transmission results.

[0010] Preferably, the input features of the LSTM model include real-time bandwidth, latency, packet loss rate, and transmission success rate; the LSTM model includes an input layer, multiple LSTM layers, and an output layer, used to capture long-term dependencies in time series.

[0011] Preferably, the rules for dynamically adjusting the retransmission strategy include: When the network condition is predicted to be good, reduce the retransmission frequency to reduce network load; When network conditions are predicted to deteriorate, the retransmission frequency is increased, and robust transmission methods that increase data redundancy or reduce transmission rate are adopted.

[0012] Preferably, the feedback control mechanism includes: Collect packet loss rate, retransmission count, and transmission delay feedback data after each transmission; Analyze the feedback data to determine whether the current strategy has achieved the expected results; If the expected results are not achieved, adjust the retransmission strategy parameters; Through continuous learning and optimization, an adaptive strategy adjustment loop is formed.

[0013] Preferably, the data decoding and retransmission triggering process at the master station includes: After receiving the encoded data, the main station uses a decoder to restore it to high-dimensional data; Error detection is performed on the restored data, and if an error is found, the retransmission mechanism is immediately initiated; Adjust the retransmission strategy and retransmit the affected data packets according to the transmission control parameters; Finally, complete and error-free relay protection information is output, and transmission logs are recorded for subsequent model optimization.

[0014] The present invention also discloses a master-slave transparent transmission system based on intelligent error correction coding and adaptive retransmission, including a memory and a processor connected to each other. The memory stores a computer program, which executes the steps of the method described above when the processor runs the computer program.

[0015] Compared with the prior art, the advantages of the present invention are as follows: This invention employs Denoising Autoencoder technology, leveraging the noise reduction and error correction capabilities of a deep learning model to effectively improve the transmission accuracy of relay protection information in complex power system environments. During data transmission, Denoising Autoencoder automatically identifies and filters out noise, thereby correcting data deviations caused by transmission errors while restoring the original data. This technology significantly reduces the data transmission error rate, enhances the anti-interference capability of the power grid system, and ensures the reliable transmission of relay protection information.

[0016] This invention improves the flexibility and efficiency of data transmission in power systems by introducing an adaptive retransmission mechanism to dynamically adjust transmission strategies. Compared to traditional fixed retransmission mechanisms, the adaptive retransmission mechanism adjusts based on real-time network conditions, ensuring that the retransmission frequency is reduced to decrease load when network conditions are good, and increased to enhance transmission robustness when network conditions deteriorate. This dynamic adjustment mechanism effectively reduces the waste of network resources and significantly reduces data transmission latency, ensuring the timeliness and accuracy of relay protection information.

[0017] This invention combines intelligent error correction coding with an adaptive retransmission mechanism to provide an efficient and stable power system data transmission solution. By integrating the Denoising Autoencoder with the adaptive retransmission mechanism, the system can maintain high data transmission performance in the face of complex and ever-changing power communication environments. This solution not only significantly improves the overall robustness and noise resistance of the system, but also ensures long-term stable operation of the system through the analysis of historical data and model optimization. Attached Figure Description

[0018] Figure 1 This is a flowchart of the transparent transmission method between master and slave stations according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of an intelligent error correction codec model based on Denoising Autoencoder according to an embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram of a network status monitoring model based on LSTM according to an embodiment of the present invention. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0022] like Figure 1 As shown, the transparent transmission method for master and slave stations based on intelligent error correction coding and adaptive retransmission provided in this embodiment of the invention includes the following steps: A. Design intelligent error correction coding using a Denoising Autoencoder to compress data into a low-dimensional representation and remove noise, thereby improving the accuracy of data transmission.

[0023] like Figure 2 As shown, step A specifically includes the following steps: A1. Handling Data Noise and Standardization: First, the relay protection information data is preprocessed to remove obvious noise and standardize the data. This process ensures data consistency and quality, laying the foundation for subsequent coding steps.

[0024] A2. The encoder section is designed using a multi-layer neural network to compress the input data into a low-dimensional latent representation (bottleneck layer). The encoder progressively reduces dimensionality, extracting key features from the data and effectively filtering out redundant information and noise. Through this design, the Denoising Autoencoder can extract clean feature representations even when processing noisy data.

[0025] A3. Design the bottleneck layer of the model. The bottleneck layer is the core of the autoencoder. It retains the most important feature information and effectively reduces noise in the data through noise reduction function, thereby improving the system's anti-interference ability.

[0026] A4. Design a decoder using a mirrored neural network structure. The decoder restores the low-dimensional representation to a high-dimensional representation close to the original data. The decoder not only reconstructs the data but also corrects errors generated during transmission.

[0027] A5. Design the loss function and optimize the parameters, using mean squared error (MSE) or L1 loss function to measure the difference between the reconstructed data and the original data. During model training, noise is intentionally added to the input data, and the denoising autoencoder learns how to reconstruct the data in a noisy environment.

[0028] A6. Perform model regularization and enhancement. Use regularization strategies to prevent model overfitting and combine data augmentation techniques to improve the model's robustness and generalization ability in the face of complex noise environments.

[0029] B. An LSTM-based adaptive retransmission mechanism monitors network status in real time and dynamically adjusts retransmission strategies to optimize transmission reliability and efficiency.

[0030] like Figure 3 As shown, step B specifically includes the following steps: B1. Perform real-time network status monitoring by embedding a deep learning model based on Long Short-Term Memory (LSTM) networks into the transparent gateway to monitor key network transmission metrics such as bandwidth, latency, and packet loss rate in real time. Analyze these metrics to predict network status trends and adjust transmission strategies accordingly.

[0031] B2. Dynamic adjustment of retransmission strategy: Based on the prediction results of the LSTM model, the retransmission strategy is dynamically adjusted. When network conditions are good, the retransmission frequency is reduced to decrease network load; when network conditions deteriorate, the retransmission frequency is increased, and a more robust transmission method is selected, such as adding redundant coding or reducing the data transmission rate, to ensure the success rate of data transmission.

[0032] B3. Feedback control mechanism design: During data transmission, the retransmission strategy is updated in real time using feedback information. By intelligently analyzing feedback information, retransmission decisions are optimized, enabling the retransmission mechanism to flexibly respond to dynamically changing network environments and ensuring the stability and efficiency of the communication link.

[0033] C. Deploy transparent gateways at both the plant and master stations, and utilize intelligent error correction coding and adaptive retransmission mechanisms to ensure the integrity and accuracy of data transmission.

[0034] Step C specifically includes the following steps: C1. Deploy a transparent gateway at the plant / station level, responsible for initial data processing and transmission control. It reduces data transmission errors through intelligent error correction coding based on DenoisingAutoencoder and optimizes retransmission strategies using an adaptive retransmission mechanism, thereby improving the reliability and efficiency of data transmission.

[0035] C2, the master station is responsible for receiving, decoding, and processing transmitted data. When a data transmission error is detected, a retransmission mechanism is immediately initiated to ensure data integrity and accuracy. The master station also continuously optimizes the deep learning model through analysis of historical data, improving the system's intelligence and overall performance.

[0036] This invention employs Denoising Autoencoder technology, leveraging the noise reduction and error correction capabilities of a deep learning model to effectively improve the transmission accuracy of relay protection information in complex power system environments. During data transmission, Denoising Autoencoder automatically identifies and filters out noise, thereby correcting data deviations caused by transmission errors while restoring the original data. This technology significantly reduces the data transmission error rate, enhances the anti-interference capability of the power grid system, and ensures the reliable transmission of relay protection information.

[0037] This invention improves the flexibility and efficiency of data transmission in power systems by introducing an adaptive retransmission mechanism to dynamically adjust transmission strategies. Compared to traditional fixed retransmission mechanisms, the adaptive retransmission mechanism adjusts based on real-time network conditions, ensuring that the retransmission frequency is reduced to decrease load when network conditions are good, and increased to enhance transmission robustness when network conditions deteriorate. This dynamic adjustment mechanism effectively reduces the waste of network resources and significantly reduces data transmission latency, ensuring the timeliness and accuracy of relay protection information.

[0038] This invention combines intelligent error correction coding with an adaptive retransmission mechanism to provide an efficient and stable power system data transmission solution. By integrating the Denoising Autoencoder with the adaptive retransmission mechanism, the system can maintain high data transmission performance in the face of complex and ever-changing power communication environments. This solution not only significantly improves the overall robustness and noise resistance of the system, but also ensures long-term stable operation of the system through the analysis of historical data and model optimization.

[0039] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0040] The transparent transmission method for master and slave stations based on intelligent error correction coding and adaptive retransmission provided by this invention includes the following steps: S1: Intelligent error correction coding design. The Denoising Autoencoder model is used to intelligently correct and encode relay protection information, generating a low-dimensional representation suitable for transmission. During transmission, a decoder is used to correct errors. Step S1 corresponds to the aforementioned step A. This step includes the following sub-steps: S11: In the encoder part, design a multi-layer neural network structure to compress high-dimensional data into a low-dimensional representation (i.e., the bottleneck layer). The specific training process includes the following sub-steps: S111: Input data processing. First, a certain amount of random noise is added to the input data to simulate the interference that the data may encounter during actual transmission. This noise can be Gaussian noise or other common transmission noise types. The purpose of this is to allow the model to learn how to extract key information in noisy environments.

[0041] S112: Training dataset preparation. The standardized dataset OriData is used as the training data. A portion of it is used as the noisy data input to the encoder, and the other portion is used as the noiseless target output.

[0042] S113: Model structure design. The encoder employs a multi-layer fully connected neural network architecture to progressively compress the high-dimensional input into a low-dimensional representation. The number of neurons in each layer gradually decreases, eventually reaching the minimum dimension of the bottleneck layer. Each layer of the encoder uses a non-linear activation function (such as ReLU) to enhance the model's expressive power.

[0043] S114: Loss Function and Optimization. The mean squared error (MSE) is used as the loss function to measure the difference between the low-dimensional representation of the encoder output, reconstructed by the decoder, and the noise-free data. Backpropagation and the Adam optimizer are used for parameter updates, enabling the model to effectively reduce reconstruction errors and improve the accuracy of data reconstruction.

[0044] S115: Iterative training. Through multiple rounds of iterative training, the model's weights and biases are gradually adjusted to ensure efficient encoding under various noise conditions. An early stopping mechanism can be introduced during training; training is stopped when the model's performance on the validation set no longer improves, preventing overfitting.

[0045] S12: Decoder Design and Correction The decoder design is symmetrical to the encoder, aiming to reconstruct a high-dimensional representation close to the original data from a low-dimensional representation. Decoder training includes the following sub-steps: S121: Decoding structure design. The neural network structure of the decoder is a mirror image of the encoder. By increasing the number of neurons layer by layer, the dimensionality of the data is gradually restored.

[0046] S122: Reconstruction and Correction. During training, the decoder receives the low-dimensional representation compressed by the encoder and attempts to reconstruct it from the original data. The decoder uses the learned weights and biases to correct errors in the data, ensuring the integrity and accuracy of the transmitted data.

[0047] S123: Evaluation and Optimization. The decoder's reconstruction performance is evaluated by comparing it with the original noise-free data. During optimization, the number of network layers, neurons, activation functions, and regularization strategies are adjusted to ensure the decoder can effectively correct errors under various transmission conditions.

[0048] S13: Model Optimization and Robustness Enhancement. To improve the robustness and adaptability of the Denoising Autoencoder model, the following optimization strategies are adopted, including the following sub-steps: S131: Noise enhancement. During training, different types and intensities of noise are introduced to enhance the model's adaptability. By learning data characteristics under different noise environments, the model enhances its robustness against interference.

[0049] S132: Regularization strategy, employing strategies such as L2 regularization, to prevent model overfitting and improve its generalization ability in practical applications. Regularization helps the model maintain stable performance when facing unknown data.

[0050] S133: Hyperparameter tuning, which optimizes hyperparameters such as learning rate, number of network layers, and number of neurons per layer through grid search to find the best-performing model configuration.

[0051] S2: Adaptive retransmission mechanism design. An adaptive retransmission mechanism based on LSTM (Long Short-Term Memory) is designed to monitor network transmission status in real time and dynamically adjust the retransmission strategy according to network conditions. Step S2 corresponds to step B mentioned above. Step S2 includes the following sub-steps: S21: Network Status Monitoring. This involves integrating an LSTM model into the transparent gateway to monitor network transmission status in real time and capture time-series characteristics such as bandwidth, latency, and packet loss rate. Specifically, it includes the following sub-steps: S211: Input Feature Selection. Key indicators of network status are selected as input features for the LSTM model, such as real-time bandwidth, latency, packet loss rate, and transmission success rate. These indicators reflect the current network condition and provide early warnings of potential problems during transmission.

[0052] S212: LSTM Model Architecture Design. The LSTM model architecture design includes an input layer, multiple LSTM layers, and an output layer. The input layer receives real-time network state data, the LSTM layers are used to capture long-term dependencies in the time series data, and the output layer predicts future network state change trends.

[0053] S213: Model Training and Validation. The LSTM model is trained using historical network state data. Through time-series analysis of the data, the model learns to predict future network conditions. After training, the model is validated using an independent validation set to ensure it can accurately predict network conditions and provide a reliable basis for subsequent retransmission strategy adjustments.

[0054] S214: Real-time monitoring and prediction. In actual operation, the LSTM model processes the input data in real time and generates network state predictions for a future period. These predictions serve as an important reference for adjusting retransmission strategies.

[0055] S22: Dynamic adjustment of retransmission strategy. Based on the network state prediction results of the LSTM model, the retransmission strategy is dynamically adjusted to optimize the reliability and efficiency of data transmission. The specific steps are as follows: S221: Policy rule setting. Based on the predicted network conditions, set dynamically adjusted policy rules. For example, when the network conditions are good, reduce the retransmission frequency to reduce network load; when the network conditions are bad, increase the retransmission frequency, or adopt a more robust transmission method, such as increasing data redundancy or reducing the transmission rate.

[0056] S222: Policy Implementation and Monitoring. When the LSTM model predicts that network conditions will deteriorate, the system will proactively adjust the retransmission frequency and transmission parameters to ensure the reliability of data transmission. Simultaneously, the system monitors the transmission effect after the policy adjustment in real time and evaluates its effectiveness.

[0057] S223: Strategy optimization. Based on actual transmission results, the system dynamically adjusts the strategy to optimize its performance. If the strategy adjustment is found to be ineffective in certain situations, the system will further learn and adjust the rules and parameters to ensure optimal transmission performance under different network conditions.

[0058] S23: Feedback control mechanism design. During data transmission, the system receives and analyzes the transmission results in real time through the feedback control mechanism, and further optimizes the retransmission strategy based on the feedback information.

[0059] The specific steps are as follows: S231: Feedback Data Acquisition. After each data transmission is completed, the system collects relevant feedback data, including the actual packet loss rate, retransmission count, and transmission delay. This feedback data is used to evaluate the effectiveness of the current transmission strategy.

[0060] S232: Feedback Analysis and Strategy Adjustment. By analyzing feedback data, the system determines whether the current retransmission strategy has achieved the expected results. If the feedback results show that the transmission performance has not met expectations, the system will adjust the retransmission strategy parameters, such as adjusting the retransmission frequency or changing the data redundancy.

[0061] S233: Continuous optimization and learning. The system continuously optimizes its retransmission strategy by constantly collecting and analyzing feedback data. This process is similar to a self-learning system. Over time, the system can respond more and more accurately to complex and ever-changing network environments, ensuring the stability and efficiency of the communication link.

[0062] S3: System Integration. After completing the intelligent error correction coding design and adaptive retransmission mechanism design, the method of this invention ensures efficient transmission and processing of relay protection information in the power system through the integration of a transparent gateway with the master station. Step S3 corresponds to the aforementioned step C; This step S3 includes the following sub-steps: S31: Transparent Gateway Deployment and Functional Integration. A transparent gateway is deployed at the plant / station level. The transparent gateway is responsible for initial data processing and transmission control, serving as the front line of defense for the data transmission system. Specific implementation includes the following sub-steps: S311: Intelligent error correction coding integration. A Denoising Autoencoder module is integrated into the transparent gateway to encode real-time data, compressing it into a low-dimensional representation. This process not only reduces data redundancy but also improves data transmission accuracy through noise reduction mechanisms. The transparent gateway can process data streams from multiple devices and stations in real time, ensuring data quality before transmission.

[0063] S312: Adaptive retransmission mechanism integration. This integrates the LSTM prediction model and retransmission strategy module into the transparent gateway, enabling real-time monitoring and dynamic adjustment of network conditions. When network conditions change, the transparent gateway can quickly adjust its retransmission strategy, optimizing data transmission paths and parameters. Through this integration, the gateway provides dual protection before and during data transmission.

[0064] S313: Simplified on-site operation and maintenance. The deployment of transparent gateways greatly simplifies on-site operation and maintenance. Due to the automation and intelligence of intelligent error correction coding and adaptive retransmission mechanisms, on-site personnel do not need frequent intervention, reducing human error and maintenance workload. At the same time, the modular design of transparent gateways allows for rapid replacement or expansion in case of equipment failure or upgrades, ensuring high system availability and flexibility.

[0065] S32: Master station receives and decodes data. The master station is responsible for receiving data transmitted through the transparent gateway and performing data decoding and error detection. Specifically, it includes the following sub-steps: S321: Data Decoding. The low-dimensional representation data received by the master station is restored to high-dimensional data through the decoder module. In this process, the master station uses the error correction function of the decoder to further eliminate errors that may be introduced during transmission, ensuring the integrity and accuracy of the final data.

[0066] S322: Error Detection and Retransmission Trigger. If a data error is detected during decoding, the master station will immediately initiate a retransmission mechanism. At this time, the system will adjust the transmission strategy and retransmit the affected data packets through the adaptive retransmission mechanism of the transparent gateway. Through this mechanism, the master station can ensure that the data received is ultimately complete and error-free.

[0067] This invention also discloses a master-slave transparent transmission system based on intelligent error correction coding and adaptive retransmission, comprising an interconnected memory and a processor. The memory stores a computer program, which, when run by the processor, executes the steps of the method described above. The transmission system of this invention corresponds to the transmission method described above and also possesses the advantages described above.

[0068] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be implemented by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable storage medium includes: any entity or device capable of carrying computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0069] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A transparent transmission method for master and slave stations based on intelligent error correction coding and adaptive retransmission, characterized in that, Includes the following steps: Deploy transparent gateways at the plant and substation end to perform intelligent error correction coding on relay protection information, compress the raw data into a low-dimensional potential representation, remove noise during the coding process, and generate coded data suitable for transmission. During data transmission, the network status is monitored in real time based on the LSTM model, and the retransmission strategy is dynamically adjusted according to the prediction results of the network status to generate transmission control parameters that are adapted to the current network conditions. After receiving the encoded data, the master station restores it to high-dimensional data through the decoder. When a data error is detected, a retransmission mechanism is triggered. Adaptive retransmission is performed in combination with the transmission control parameters, and finally complete and error-free relay protection information is output.

2. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to claim 1, characterized in that, The intelligent error correction coding process includes the following sub-steps: The original relay protection information is preprocessed to remove noise and standardize the data; A multi-layer neural network is constructed as an encoder to progressively compress the input data into a low-dimensional latent representation; Design a bottleneck layer that retains key features and further filters out redundant information; Construct a decoder that is symmetric to the encoder to restore the low-dimensional representation to high-dimensional data; The model is trained using mean squared error or L1 loss function, and the model's robustness is enhanced by adding noise to the input data. Regularization strategies and data augmentation techniques are introduced to improve the robustness and generalization ability of the model.

3. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to claim 2, characterized in that, The encoder training process includes: Add Gaussian noise or other transmission noise to the input data to simulate the actual communication environment; Use a standardized dataset as training data and target noise-free data as the output. A multi-layer fully connected neural network is used, and the number of neurons is reduced layer by layer to finally output the bottleneck layer representation. Backpropagation training is performed using the mean squared error loss function and the Adam optimizer. An early stopping mechanism is introduced to prevent model overfitting.

4. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to claim 2, characterized in that, The training process for the decoder includes: Construct a decoder structure that mirrors the encoder, and restore the data dimensions layer by layer; The decoder is trained to reconstruct the original data from the low-dimensional representation. By comparing the reconstructed data with the original data, the number of network layers, neurons, and activation functions are optimized.

5. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to any one of claims 1-4, characterized in that, The LSTM-based adaptive retransmission mechanism includes the following sub-steps: Embed the LSTM model in the transparent gateway to monitor network bandwidth, latency, and packet loss rate in real time; LSTM models perform time series analysis on network states to predict future network trends. The retransmission frequency, redundancy coding strength, or transmission rate are dynamically adjusted based on the prediction results. Design a feedback control mechanism to optimize the retransmission strategy in real time based on the actual transmission results.

6. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to claim 5, characterized in that, The input features of the LSTM model include real-time bandwidth, latency, packet loss rate, and transmission success rate; the LSTM model includes an input layer, multiple LSTM layers, and an output layer, used to capture long-term dependencies in time series.

7. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to claim 5, characterized in that, The rules for dynamically adjusting the retransmission strategy include: When the network condition is predicted to be good, reduce the retransmission frequency to reduce network load; When network conditions are predicted to deteriorate, the retransmission frequency is increased, and robust transmission methods that increase data redundancy or reduce transmission rate are adopted.

8. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to claim 5, characterized in that, Feedback control mechanisms include: Collect packet loss rate, retransmission count, and transmission delay feedback data after each transmission; Analyze the feedback data to determine whether the current strategy has achieved the expected results; If the expected results are not achieved, adjust the retransmission strategy parameters; Through continuous learning and optimization, an adaptive strategy adjustment loop is formed.

9. The master-slave transparent transmission method based on intelligent error correction coding and adaptive retransmission according to any one of claims 1-4, characterized in that, The data decoding and retransmission triggering process at the main station includes: After receiving the encoded data, the main station uses a decoder to restore it to high-dimensional data; Error detection is performed on the restored data, and if an error is found, the retransmission mechanism is immediately initiated; Adjust the retransmission strategy and retransmit the affected data packets according to the transmission control parameters; Finally, complete and error-free relay protection information is output, and transmission logs are recorded for subsequent model optimization.

10. A master-slave transparent transmission system based on intelligent error correction coding and adaptive retransmission, comprising an interconnected memory and a processor, wherein the memory stores a computer program, characterized in that... The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1-9.

Images