Intelligent power grid communication optimization method and system based on HPLC and HRF dual-mode communication

Abstract

The invention relates to an intelligent power grid communication optimization method based on HPLC and HRF dual-mode communication, and the method is characterized in that the method comprises the following steps: S1, obtaining power grid communication data, and carrying out the preprocessing of the data; s2, according to the preprocessed intelligent power grid communication data, a feature extractor based on a deep neural network automatically extracts two-channel time-frequency domain features, feature importance of different modes is dynamically weighted through an attention mechanism, and a unified channel state matrix CSM after fusion is obtained; s3, constructing a routing strategy model based on reinforcement learning, and obtaining an optimal path decision according to the CSM matrix, the power grid network topology and the service priority label; s4, according to the path obtained by the optimal path decision, adaptive modulation coding is adopted to obtain an optimal modulation coding scheme and a dynamic adaptive channel condition; and S5, performing communication according to the optimal modulation coding scheme and the dynamic adaptive channel condition. According to the method, the communication state sensing capability in a complex scene is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of communication optimization, and in particular to a smart grid communication optimization method and system based on HPLC and HRF dual-mode communication. Background Art

[0002] With the global energy structure transitioning to renewable energy and the advancement of the "dual carbon" goal, smart grids are upgrading from traditional one-way power supply networks to integrated energy systems with two-way interaction and intelligent control capabilities. In the new power system, a large number of diversified devices such as distributed power sources, energy storage devices, and electric vehicle charging piles are connected, giving rise to massive real-time monitoring, remote control and two-way data interaction needs. For example, the distribution network needs to support more than 1,000 smart terminals communicating simultaneously per square kilometer, and the relay protection service requires end-to-end delay less than 20ms and bit error rate controlled within 10 -9 This places extremely high demands on the reliability, real-time performance and capacity of the communication system.

[0003] Traditional smart grid communications primarily rely on single technologies: high-speed power line carrier (HPLC) or high-frequency wireless communication (HRF). HPLC utilizes power lines as a transmission medium, offering advantages such as low deployment costs and wide coverage. However, it is susceptible to harmonic interference from power equipment and signal attenuation caused by line aging, resulting in insufficient transmission stability over long distances or in complex topologies. While HRF technologies (such as 2.4GHz / 5.8GHz wireless communication) offer high bandwidth and low latency, they suffer from significant signal loss through walls and are susceptible to interference from weather (such as heavy rain and dust) and electromagnetic interference, making full coverage difficult in remote mountainous areas or with underground cables.

[0004] Existing communication systems often rely on static routing protocols (such as OSPF and RIP) and fixed modulation and coding schemes, making them unable to dynamically adapt to changes in grid operation. For example, when a surge in grid load causes increased noise in the power line channel, or when a sudden downpour causes wireless signal fading, static strategies cannot promptly switch communication paths or adjust transmission parameters, leading to packet loss and service interruptions. This makes it difficult to meet the QoS requirements of smart grids for real-time services. Summary of the Invention

[0005] In order to solve the above problems, the purpose of the present invention is to provide a smart grid communication optimization method and system based on HPLC and HRF dual-mode communication, which can effectively improve the quality and efficiency of smart grid communication.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] A smart grid communication optimization method based on HPLC and HRF dual-mode communication includes the following steps:

[0008] S1: Obtain grid communication data and preprocess it;

[0009] S2: Based on the pre-processed smart grid communication data, the feature extractor based on the deep neural network automatically extracts the dual-channel time-frequency domain features, and dynamically weights the feature importance of different modalities through the attention mechanism to obtain the fused unified channel state matrix CSM;

[0010] S3: Build a routing strategy model based on reinforcement learning and obtain the optimal path decision according to the CSM matrix, power grid network topology and service priority labels;

[0011] S4: Adaptive modulation and coding is used based on the path obtained by the optimal path decision to obtain the optimal modulation and coding scheme and dynamically adapt to the channel conditions;

[0012] S5: Communicate based on the optimal modulation and coding scheme and dynamically adapt to channel conditions.

[0013] Furthermore, the power grid communication data, including original channel parameters, real-time business data and environmental data, are as follows:

[0014] The original channel parameters include signal-to-noise ratio, attenuation coefficient, bit error rate, and carrier-to-interference ratio; the HPLC module reads the SNR, attenuation coefficient, and bit error rate calculated in real time in the PHY layer register through the physical layer monitoring interface of the power line carrier communication chip; the HRF module obtains the signal-to-noise ratio, path loss, and carrier-to-interference ratio of the wireless channel through the MAC layer or PHY layer driver interface of the wireless communication chip;

[0015] The real-time service data includes traffic type and QoS requirements. The power distribution and utilization service system connects to the power distribution automation master station, metering master station, and load control master station, and obtains service type tags and corresponding QoS requirements through OPC UA or HTTP API. The communication terminal embeds a service identification module in the MAC layer of the HPLC / HRF dual-mode terminal, and identifies the priority of the real-time service by parsing the DSCP field or application layer protocol of the IP data packet.

[0016] The environmental data includes noise spectrum and weather data. The noise spectrum is scanned by the integrated spectrum analyzer in the HPLC module at a preset frequency resolution for power line noise; the spectrum is scanned in the idle time slot of the wireless module in the HRF module to record the center frequency and power of the interference signal;

[0017] Weather data is obtained through locally deployed meteorological sensors or by accessing real-time data from the Meteorological Bureau through the REST API to obtain regional weather parameters.

[0018] Furthermore, preprocessing data cleaning, data standardization and multimodal feature fusion processing are as follows:

[0019] The data cleaning includes missing value processing and outlier detection;

[0020] The missing value handling:

[0021] For missing channel parameters, forward filling is used; for missing environmental data, the time series prediction model of historical data is used to complete the missing data.

[0022] Outlier detection combines statistical methods and clustering methods, as follows:

[0023] Based on the statistical method Z-score: set the range of SNR, and mark outliers outside the range as invalid;

[0024] Based on the clustering method DBSCAN: cluster the noise spectrum data and filter out isolated high-power outliers;

[0025] Data standardization:

[0026] Channel parameters,are standardized by Z-score to eliminate the dimensional differences,between different modalities;

[0027] For service data, the numerical parameters of latency and bandwidth in QoS requirements are normalized using Min-Max and mapped to intervals. Service type labels are converted into vectors using one-hot encoding.

[0028] Environmental data: The power values ​​of the noise spectrum are log-normalized, and weather parameters are converted into discrete features using the binning method;

[0029] The cleaned and standardized HPLC channel parameters, HRF channel parameters, business characteristics, and environmental characteristics are aligned in time and spliced ​​into a multidimensional feature vector.

[0030] Furthermore, based on the preprocessed smart grid communication data, a feature extractor based on a deep neural network automatically extracts dual-channel time-frequency domain features. The importance of features of different modalities is dynamically weighted through the attention mechanism to obtain the fused unified channel state matrix CSM, as follows:

[0031] The preprocessed multimodal data needs to be converted into a three-dimensional tensor, the structure of which is defined as: input tensor = [number of samples, time step T, feature dimension F];

[0032] The deep neural network is based on the joint feature extraction of CNN module and LSTM module, where CNN extracts frequency domain features and LSTM extracts time domain features;

[0033] The CNN module extracts the frequency domain local features of the HPLC / HRF dual-mode channel. The input layer splits the features of time step t into HPLC sub-features, HRF sub-features, and environmental business features. The noise spectra of HPLC and HRF are reorganized into a two-dimensional matrix as the input of the convolution layer.

[0034] Convolutional layer: For the HPLC spectrum matrix: use a 1×3 convolution kernel, a step size of 1, the same padding, and output 64 feature maps;

[0035] For the HRF spectrum matrix: use a 1×2 convolution kernel, a step size of 1, the same padding, and output 32 feature maps;

[0036] Pooling layer: performs maximum pooling on the convolution output to reduce the dimension and retain the main frequency domain features;

[0037] Fully connected layer: Flattens the pooling results of HPLC / HRF, concatenates them with environmental business features, and outputs 128-dimensional frequency domain fusion features;

[0038] The LSTM module captures the dynamic changes of channel parameters and business requirements over time. The input layer: the frequency domain fusion features of each time step t are input into the LSTM layer in time series. The LSTM layer adopts Bi-LSTM to output the hidden state h(t) of each time step to capture the temporal dependency between the past and the future. The output layer: through the time-distributed fully connected layer, the temporal hidden state output by the LSTM layer is converted into a time domain feature sequence F_lstm=[h(1),h(2),...,h(T)], where each h(t) corresponds to the time-frequency domain joint feature of the time step t.

[0039] Adaptively adjust the weights of HPLC / HRF dual-mode features based on real-time channel conditions and business needs;

[0040] Multiply the HPLC / HRF sub-features of each time step t by the corresponding weights, and then sum them by the weighted time step to obtain the global fusion feature F attention ; Finally, all the time steps F attention (t) splicing into fusion signature sequences;

[0041] The feature sequence after attention fusion is concatenated with the business environment features and mapped into a CSM matrix through linear transformation.

[0042] Furthermore, a routing strategy model is built based on reinforcement learning, and the optimal path decision is obtained according to the CSM matrix, power grid network topology, and service priority labels. The details are as follows:

[0043] The state space S consists of three parts: the channel state matrix CSM, the power grid network topology, and the service priority label;

[0044] The action space is defined as the set of discrete actions A = {a1, a2, a3}:

[0045] a1: Select HPLC single-mode path; a2: Select HRF single-mode path; a3: Select hybrid relay path;

[0046] The reward function R adopts a multi-objective weighted reward mechanism:

[0047] R=ω1*Rqos+ω2*Refficiency+ω3*Rstability;

[0048] Among them, Rqos is the delay reward: if the service delay is less than the threshold and the bit error rate is less than the preset value, then Rqos = 10×Priority, where Priority is the normalized priority of 0-1; otherwise, Rqos = -5×(1-Priority); Refficiency is the resource efficiency reward; Rstability is the stability reward;

[0049] Obtain the optimal path decision based on the deep Q network DQN.

[0050] Furthermore, the path obtained according to the optimal path decision adopts adaptive modulation and coding to obtain the optimal modulation and coding scheme and dynamically adapt the channel conditions, as follows:

[0051] Obtain the real-time SNR, attenuation coefficient, noise density or path loss, fading factor, and QoS requirements of each link from the routing decision results;

[0052] Based on the SNR-BER safety margin, high-order modulation is prioritized in descending order of SNR, and the lowest-order protection mode is the default;

[0053] A lightweight machine learning model is introduced to optimize MCS decisions in non-stationary channels by combining historical SNR, noise entropy, and service priority.

[0054] Inner loop: Calculates the instantaneous BER every 1ms. If the measured value exceeds twice the target, MCS degradation is triggered.

[0055] Outer loop: Count the retransmission rate every 50ms. If RR>10%, synchronously downgrade the MCS and increase the FEC redundancy (for example, the coding rate from 3 / 4 to 2 / 3).

[0056] FEC adaptation: LDPC is used for low-order modulation, while high-order modulation is switched to convolutional code and the HARQ retransmission mechanism is enabled;

[0057] Based on the QoS priority modulation strategy, high-delay-sensitive services are forced to use low-order modulation, sacrificing throughput to ensure reliability; high-throughput services allow for higher BER and prioritize high-order modulation; low-power services select the lowest-order modulation to reduce transmit power;

[0058] The weighting function is used to integrate spectrum efficiency, reliability, and power consumption, and dynamically adjust the weights;

[0059] Path parsing, state caching, and MCS decision engines are integrated into edge terminals, which interact directly with the physical layer chip through the SPI / UART / PCIe interface to pre-generate the SNR-MCS mapping table, which is updated only when the SNR changes by more than 2dB or when a service is switched.

[0060] A main distribution microgrid integrated real-time coordinated risk scheduling system includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically performs the steps of the main distribution microgrid integrated real-time coordinated risk scheduling method as described above.

[0061] The present invention has the following beneficial effects:

[0062] 1. The present invention synchronously collects original channel parameters, real-time business data and environmental data, and combines the CNN+LSTM deep neural network to construct a dual-channel time-frequency domain feature extractor. This can effectively capture the dynamic changes in the time domain (such as business flow fluctuations) and the frequency domain noise distribution characteristics (such as the environmental interference spectrum) in power grid communications. The attention mechanism is introduced to dynamically weight the dual-mode features of HPLC (high-speed power line carrier) and HRF (high-frequency radio). It can adaptively focus on key modal information (such as prioritizing the enhancement of wireless channel characteristics in severe weather) and generate a fused unified channel state matrix (CSM). This breaks through the limitations of traditional single-modal feature analysis, makes the channel state representation more comprehensive and accurate, provides a reliable data foundation for subsequent routing and modulation optimization, and significantly improves the communication state perception capability in complex scenarios.

[0063] 2. The routing strategy model constructed by this invention, based on reinforcement learning, can combine the CSM matrix, power grid network topology, and service priority tags (such as the high reliability requirements of distribution network automation and the low latency requirements of user metering) in real time. Through deep neural network mapping of the state-action space, it dynamically generates optimal path decisions for HPLC / HRF / hybrid relays. This method breaks away from the reliance of traditional static routing protocols on predefined rules and can adapt to changes in power grid topology (such as equipment switching and fault isolation) and fluctuations in service traffic, prioritizing the transmission quality of high-priority services.

[0064] 3. This invention dynamically selects the modulation and coding scheme (MCS) based on the optimal path decision results and real-time channel status (such as signal-to-noise ratio and attenuation coefficient). When channel conditions are favorable, it can use high-order modulation (such as 64QAM) to improve spectral efficiency, while switching to low-order modulation (such as QPSK) to enhance anti-interference capabilities when noise interference intensifies. This mechanism dynamically matches the modulation strategy with channel conditions through closed-loop feedback, avoiding the performance bottlenecks of fixed modulation modes in complex environments (such as the spectrum waste caused by the fixed use of BPSK in traditional power line carriers). BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 Flow chart of the method of the present invention. DETAILED DESCRIPTION

[0066] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0067] A smart grid communication optimization method based on HPLC and HRF dual-mode communication includes the following steps:

[0068] S1: Obtain grid communication data and preprocess it;

[0069] S2: Based on the pre-processed smart grid communication data, the feature extractor based on the deep neural network automatically extracts the dual-channel time-frequency domain features, and dynamically weights the feature importance of different modalities through the attention mechanism to obtain the fused unified channel state matrix CSM;

[0070] S3: Build a routing strategy model based on reinforcement learning and obtain the optimal path decision according to the CSM matrix, power grid network topology and service priority labels;

[0071] S4: Adaptive modulation and coding is used based on the path obtained by the optimal path decision to obtain the optimal modulation and coding scheme and dynamically adapt to the channel conditions;

[0072] S5: Communicate based on the optimal modulation and coding scheme and dynamically adapt to channel conditions.

[0073] In this embodiment, the power grid communication data includes original channel parameters, real-time service data, and environmental data, as follows:

[0074] The original channel parameters include signal-to-noise ratio (SNR), attenuation coefficient (path loss of power line channel / large-scale fading of wireless channel), bit error rate (BER), and carrier-to-interference ratio (CIR); the HPLC module reads the SNR, attenuation coefficient, and bit error rate calculated in real time in the PHY layer register through the physical layer monitoring interface of the power line carrier communication chip (such as ST7538, HPLC-IoT chip); the HRF module obtains the signal-to-noise ratio (through RSSI conversion), path loss (calculated based on the free space path loss model), and carrier-to-interference ratio of the wireless channel through the MAC layer or PHY layer driver interface of the wireless communication chip (such as Sub-1GHz SX1262, 2.4GHz CC2530);

[0075] The real-time service data includes traffic type (control instructions / metering data / video surveillance, etc.), QoS requirements (latency ≤ 10ms / reliability ≥ 99.9% / bandwidth ≥ 1Mbps, etc.);

[0076] The power distribution and utilization service system connects to the distribution automation master station, metering master station, and load control master station (LC), obtaining service type tags (such as "protection control," "remote meter reading," and "distributed power supply monitoring") and corresponding QoS requirements through OPC UA or HTTP API. The communication terminal embeds a service identification module in the MAC layer of the HPLC / HRF dual-mode terminal. This module identifies the priority of real-time services (for example, "emergency control" corresponds to the highest priority) by parsing the DSCP (Differentiated Services Code Point) field of the IP packet or the application layer protocol (such as IEC 104 and DL / T 645).

[0077] The environmental data includes noise spectrum and weather data. The noise spectrum is scanned by the integrated spectrum analyzer in the HPLC module at a preset frequency resolution (100kHz to 1MHz); the spectrum is scanned in the idle time slot of the wireless module in the HRF module to record the center frequency and power of the interference signal;

[0078] Weather data is obtained through locally deployed meteorological sensors (temperature and humidity sensor SHT30, wind speed sensor AMR100) or by accessing real-time data from the Meteorological Bureau through the REST API to obtain regional weather parameters (such as "thunderstorm probability 30%" and "humidity 85%").

[0079] In this embodiment, preprocessing data cleaning, data standardization and multimodal feature fusion processing are as follows:

[0080] The data cleaning includes missing value processing and outlier detection;

[0081] The missing value handling:

[0082] For missing channel parameters (such as occasional disconnection of the communication module), forward filling is used; for missing environmental data, time series prediction models (such as LSTM) based on historical data are used to complete the missing data.

[0083] Outlier detection combines statistical methods and clustering methods, as follows:

[0084] Based on the statistical method Z-score: set the SNR range (HPLC: 0 to 40dB, HRF: -120 to -50dBm), and mark outliers outside the range as invalid;

[0085] Clustering method DBSCAN: Cluster noise spectrum data to filter out isolated high-power outliers (such as power line burst noise);

[0086] Data standardization:

[0087] Channel parameters were normalized using Z-score to eliminate the dimensional differences between different modalities (HPLC / HRF) (e.g., the dB value of SNR and the linear value of attenuation coefficient;

[0088] For service data, the numerical parameters of latency and bandwidth in QoS requirements are normalized using Min-Max and mapped to intervals. Service type labels (such as "control" and "metering") are converted into vectors using one-hot encoding.

[0089] Environmental data: The power values ​​of the noise spectrum are log-normalized, and weather parameters are converted into discrete features using the binning method;

[0090] The cleaned and standardized HPLC channel parameters (such as [SNR_hplc, attenuation_hplc]), HRF channel parameters (such as [SNR_hrf, path_loss_hrf]), service characteristics (such as [QoS_delay, QoS_reliability]), and environmental characteristics (such as [noise_power, temperature]) are time aligned and spliced ​​into a multidimensional feature vector.

[0091] In this embodiment, based on the preprocessed smart grid communication data, a feature extractor based on a deep neural network automatically extracts dual-channel time-frequency domain features. The importance of features of different modalities is dynamically weighted through an attention mechanism to obtain a fused unified channel state matrix (CSM), as follows:

[0092] The preprocessed multimodal data needs to be converted into a three-dimensional tensor, the structure of which is defined as: input tensor = [number of samples, time step T, feature dimension F];

[0093] Where: Time step T: Based on the real-time requirements of power grid communication (e.g., control service latency ≤ 10ms), T = 20 (corresponding to a 200ms window and a sampling frequency of 100Hz) is used to cover short-timescale channel fluctuations. Feature dimension F: Contains the time-frequency domain parameters of the HPLC and HRF dual modes, specifically broken down as follows:

[0094] HPLC channel characteristics (F1): signal-to-noise ratio (SNR_hplc), attenuation coefficient (Att_hplc), power line noise spectrum (100kHz-50MHz, divided into 10 frequency bins) → 12 dimensions in total.

[0095] HRF channel features (F2): wireless signal-to-noise ratio (SNR_hrf), path loss (PL_hrf), wireless noise spectrum (sub-1 GHz, divided into 8 frequency bins) → a total of 10 dimensions.

[0096] Environmental and service characteristics (F3): Temperature (Temp), humidity (Hum), service priority (QoS_level, 0-1 normalized) → a total of 3 dimensions. Total feature dimensions F = F1 + F2 + F3 = 25 dimensions.

[0097] The deep neural network is based on the joint feature extraction of CNN module and LSTM module, where CNN extracts frequency domain features and LSTM extracts time domain features;

[0098] The CNN module extracts the frequency domain local features of the HPLC / HRF dual-mode channel (such as the frequency band distribution of power line noise and the center frequency of wireless interference). The input layer splits the features at time step t into HPLC sub-features (12 dimensions), HRF sub-features (10 dimensions), and environmental business features (3 dimensions). The noise spectra of HPLC and HRF (a total of 18 dimensions) are reorganized into a two-dimensional matrix as input to the convolutional layer.

[0099] Convolutional layer: For the HPLC spectrum matrix: use a 1×3 convolution kernel (covering 3 adjacent frequency bins), a stride of 1, the same padding, and output 64 feature maps (activation function ReLU);

[0100] For the HRF spectrum matrix: use a 1×2 convolution kernel (covering 2 adjacent frequency bins), a stride of 1, the same padding, and output 32 feature maps (activation function ReLU);

[0101] Pooling layer: Perform maximum pooling on the convolution output (pooling kernel 1×2, stride 2) to reduce the dimension and retain the main frequency domain features;

[0102] Fully connected layer: Flattens the pooling results of HPLC / HRF, concatenates them with the environmental business features (3D), and outputs 128-dimensional frequency domain fusion features;

[0103] The LSTM module captures the dynamic changes of channel parameters and business demands over time (such as the instantaneous jump of power line attenuation and the sudden growth of business traffic). The input layer: the frequency domain fusion features of each time step t are input into the LSTM layer in time series. The LSTM layer adopts Bi-LSTM to output the hidden state h(t) (128 dimensions) of each time step to capture the temporal dependency between the past and the future (such as predicting the SNR trend of the next time step); the output layer: through the time distributed fully connected layer, the temporal hidden state output by the LSTM layer is converted into a time domain feature sequence F_lstm = [h(1), h(2), ..., h(T)], where each h(t) corresponds to the time-frequency domain joint feature of time step t.

[0104] Adaptively adjust the importance of HPLC / HRF dual-mode features based on real-time channel conditions (e.g., increasing the weight of HRF features when the HPLC link fails) and service requirements (e.g., prioritizing low-latency features for high-priority control services);

[0105] Multiply the HPLC / HRF sub-features of each time step t by the corresponding weights, and then sum them by the weighted time step to obtain the global fusion feature F attention ; Finally, all the time steps F attention (t) splicing into fusion signature sequences;

[0106] The feature sequence after attention fusion is spliced ​​with the business environment features (3 dimensions, expanded to 10 dimensions through full connection) and mapped into a T×D CSM matrix (20×60 dimensions) through linear transformation.

[0107] In this embodiment, a routing strategy model is constructed based on reinforcement learning, and the optimal path decision is obtained according to the CSM matrix, power grid network topology, and service priority labels as follows:

[0108] The state space S consists of three parts: the channel state matrix CSM, the power grid network topology, and the service priority label: a one-hot encoded 5-dimensional vector (corresponding to 5 levels of QoS, such as protection control, distribution network automation, video surveillance, remote meter reading, and non-real-time services).

[0109] The action space is defined as the set of discrete actions A = {a1, a2, a3}:

[0110] a1: Select HPLC single-mode path (power line carrier communication); a2: Select HRF single-mode path (high-frequency wireless communication); a3: Select hybrid relay path (HPLC+HRF multi-hop forwarding, hop count ≤ 3);

[0111] The reward function R adopts a multi-objective weighted reward mechanism:

[0112] R=ω1*Rqos+ω2*Refficiency+ω3*Rstability;

[0113] Among them, Rqos is the delay reward: if the service delay is less than the threshold and the bit error rate is less than the preset value, then Rqos = 10×Priority, where Priority is the normalized priority of 0-1; otherwise, Rqos = -5×(1-Priority); Refficiency is the resource efficiency reward; Rstability is the stability reward;

[0114] Obtain the optimal path decision based on the deep Q network DQN.

[0115] In this embodiment, based on the path obtained by the optimal path decision, adaptive modulation and coding is used to obtain the optimal modulation and coding scheme and dynamically adapt the channel conditions, as follows:

[0116] Obtain the real-time SNR, attenuation coefficient, noise density or path loss, fading factor, and QoS requirements of each link from the routing decision results;

[0117] Based on the SNR-BER safety margin (the target BER is one order of magnitude lower than the theoretical value), higher-order modulation is prioritized in descending SNR order, with the lowest-order protection mode (BPSK+1 / 2 coding) being the default.

[0118] A lightweight machine learning model (decision tree / linear regression) is introduced to optimize MCS decisions in non-stationary channels by combining historical SNR, noise entropy, and service priority.

[0119] Inner loop (physical layer): Calculates the instantaneous BER every 1ms. If the measured value exceeds twice the target, it triggers MCS downgrade (e.g., 16QAM to QPSK).

[0120] Outer loop (link layer): Count the retransmission rate every 50ms. If the RR is greater than 10%, the MCS is synchronously downgraded and the FEC redundancy is increased (for example, the coding rate is changed from 3 / 4 to 2 / 3).

[0121] FEC adaptation: LDPC (high redundancy) is used for low-order modulation, and convolutional code (CC) is used for high-order modulation with HARQ retransmission mechanism enabled;

[0122] Based on the modulation strategy of QoS priority, high delay-sensitive services (such as differential protection) are forced to use low-order modulation (QPSK+1 / 2 coding), sacrificing throughput to ensure reliability; high throughput services (such as video surveillance) allow higher BER (≤5×10 -6 ), prioritize high-order modulation (64QAM). Low-power services (such as remote meter reading) select the lowest-order modulation and reduce the transmit power by 30%;

[0123] The weighting function is used to integrate spectrum efficiency, reliability, and power consumption, and dynamically adjust the weights;

[0124] The edge terminal integrates path parsing, state caching, and MCS decision engines, interacts directly with physical layer chips (such as ST7538 and SX1262) through SPI / UART / PCIe interfaces, and pre-generates SNR-MCS mapping tables (with an accuracy of 0.5dB). These tables are updated only when the SNR changes by more than 2dB or when services are switched.

[0125] A system for real-time coordinated risk scheduling of a main distribution microgrid and an integrated system includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, the system specifically performs the steps of the method for real-time coordinated risk scheduling of a main distribution microgrid and an integrated system.

[0126] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0127] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0128] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0129] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0130] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

Claims

1. A smart grid communication optimization method based on HPLC and HRF dual-mode communication, characterized in that: The following steps are involved: S1: Obtain grid communication data and preprocess it; S2: Based on the pre-processed smart grid communication data, the feature extractor based on the deep neural network automatically extracts the dual-channel time-frequency domain features, and dynamically weights the feature importance of different modalities through the attention mechanism to obtain the fused unified channel state matrix CSM; S3: Build a routing strategy model based on reinforcement learning and obtain the optimal path decision according to the CSM matrix, power grid network topology and service priority labels; S4: Adaptive modulation and coding is used based on the path obtained by the optimal path decision to obtain the optimal modulation and coding scheme and dynamically adapt to the channel conditions; S5: Communicate based on the optimal modulation and coding scheme and dynamically adapt to channel conditions.

2. The smart grid communication optimization method based on HPLC and HRF dual-mode communication according to claim 1, characterized in that: The power grid communication data includes original channel parameters, real-time service data and environmental data, as follows: The original channel parameters include signal-to-noise ratio, attenuation coefficient, bit error rate, and carrier-to-interference ratio; the HPLC module reads the SNR, attenuation coefficient, and bit error rate calculated in real time in the PHY layer register through the physical layer monitoring interface of the power line carrier communication chip; The HRF module drives the interface through the MAC layer or PHY layer of the wireless communication chip to obtain the signal-to-noise ratio, path loss, and carrier-to-interference ratio of the wireless channel; The real-time service data includes traffic type and QoS requirements. The power distribution and utilization service system connects to the power distribution automation master station, metering master station, and load control master station, and obtains service type tags and corresponding QoS requirements through OPC UA or HTTP API. The communication terminal embeds a service identification module in the MAC layer of the HPLC / HRF dual-mode terminal, and identifies the priority of the real-time service by parsing the DSCP field or application layer protocol of the IP data packet. The environmental data includes noise spectrum and weather data. The noise spectrum is scanned by the integrated spectrum analyzer in the HPLC module at a preset frequency resolution for power line noise; the spectrum is scanned in the idle time slot of the wireless module in the HRF module to record the center frequency and power of the interference signal; Weather data is obtained through locally deployed meteorological sensors or by accessing real-time data from the Meteorological Bureau through the REST API to obtain regional weather parameters.

3. The smart grid communication optimization method based on HPLC and HRF dual-mode communication according to claim 2, characterized in that: The pre-processing data cleaning, data standardization and multimodal feature fusion processing are as follows: The data cleaning includes missing value processing and outlier detection; The missing value handling: For missing channel parameters, forward filling is used; for missing environmental data, the time series prediction model of historical data is used to complete the missing data. Outlier detection combines statistical methods and clustering methods, as follows: Based on the statistical method Z-score: set the range of SNR, and mark outliers outside the range as invalid; Based on the clustering method DBSCAN: cluster the noise spectrum data and filter out isolated high-power outliers; Data standardization: Channel parameters,are standardized by Z-score to eliminate the dimensional differences,between different modalities; For service data, the numerical parameters of latency and bandwidth in QoS requirements are normalized using Min-Max and mapped to intervals. Service type labels are converted into vectors using one-hot encoding. Environmental data: The power values ​​of the noise spectrum are log-normalized, and weather parameters are converted into discrete features using the binning method; The cleaned and standardized HPLC channel parameters, HRF channel parameters, business characteristics, and environmental characteristics are aligned in time and spliced ​​into a multidimensional feature vector.

4. The smart grid communication optimization method based on HPLC and HRF dual-mode communication according to claim 1, characterized in that: According to the pre-processed smart grid communication data, the feature extractor based on the deep neural network automatically extracts the dual-channel time-frequency domain features, and dynamically weights the feature importance of different modes through the attention mechanism to obtain the fused unified channel state matrix CSM, which is as follows: The preprocessed multimodal data needs to be converted into a three-dimensional tensor, the structure of which is defined as: input tensor = [number of samples, time step T, feature dimension F]; The deep neural network is based on the joint feature extraction of CNN module and LSTM module, where CNN extracts frequency domain features and LSTM extracts time domain features; The CNN module extracts the frequency domain local features of the HPLC / HRF dual-mode channel. The input layer splits the features of time step t into HPLC sub-features, HRF sub-features, and environmental business features. The noise spectra of HPLC and HRF are reorganized into a two-dimensional matrix as the input of the convolution layer. Convolutional layer: For the HPLC spectrum matrix: use a 1×3 convolution kernel, a step size of 1, the same padding, and output 64 feature maps; For the HRF spectrum matrix: use a 1×2 convolution kernel, a step size of 1, the same padding, and output 32 feature maps; Pooling layer: performs maximum pooling on the convolution output to reduce the dimension and retain the main frequency domain features; Fully connected layer: Flattens the pooling results of HPLC / HRF, concatenates them with environmental business features, and outputs 128-dimensional frequency domain fusion features; The LSTM module captures the dynamic changes of channel parameters and business requirements over time. The input layer: the frequency domain fusion features of each time step t are input into the LSTM layer in time series. The LSTM layer adopts Bi-LSTM to output the hidden state h(t) of each time step to capture the temporal dependency between the past and the future. The output layer: through the time-distributed fully connected layer, the temporal hidden state output by the LSTM layer is converted into a time domain feature sequence F_lstm=[h(1),h(2),...,h(T)], where each h(t) corresponds to the time-frequency domain joint feature of the time step t. Adaptively adjust the weights of HPLC / HRF dual-mode features based on real-time channel conditions and business needs; Multiply the HPLC / HRF sub-features of each time step t by the corresponding weights, and then sum them by the weighted time step to obtain the global fusion feature F attention ; Finally, all the time steps F attention (t) splicing into fusion signature sequences; The feature sequence after attention fusion is concatenated with the business environment features and mapped into a CSM matrix through linear transformation.

5. The smart grid communication optimization method based on HPLC and HRF dual-mode communication according to claim 1, characterized in that: The reinforcement learning-based routing strategy model is constructed, and the optimal path decision is obtained according to the CSM matrix, power grid network topology and service priority labels as follows: The state space S consists of three parts: the channel state matrix CSM, the power grid network topology, and the service priority label; The action space is defined as the set of discrete actions A = {a1, a2, a3}: a1: Select HPLC single mode path; a2: Select HRF single mode path; a3: Select a hybrid relay path; The reward function R adopts a multi-objective weighted reward mechanism: R=ω1*Rqos+ω2*Refficiency+ω3*Rstability; Among them, Rqos is the delay reward: if the service delay is less than the threshold and the bit error rate is less than the preset value, then Rqos = 10×Priority, where Priority is the normalized priority of 0-1; otherwise, Rqos = -5×(1-Priority); Refficiency is the resource efficiency reward; Rstability is the stability reward; Obtain the optimal path decision based on the deep Q network DQN.

6. The smart grid communication optimization method based on HPLC and HRF dual-mode communication according to claim 1, characterized in that: The path obtained according to the optimal path decision adopts adaptive modulation and coding to obtain the optimal modulation and coding scheme and dynamically adapt the channel conditions, as follows: Obtain the real-time SNR, attenuation coefficient, noise density or path loss, fading factor, and QoS requirements of each link from the routing decision results; Based on the SNR-BER safety margin, high-order modulation is prioritized in descending order of SNR, and the lowest-order protection mode is the default; A lightweight machine learning model is introduced to optimize MCS decisions in non-stationary channels by combining historical SNR, noise entropy, and service priority. Inner loop: Calculates the instantaneous BER every 1ms. If the measured value exceeds twice the target, MCS degradation is triggered. Outer loop: Count the retransmission rate every 50ms. If the RR is greater than 10%, the MCS is synchronously downgraded and FEC redundancy is increased. FEC adaptation: LDPC is used for low-order modulation, while high-order modulation is switched to convolutional code and the HARQ retransmission mechanism is enabled; Based on the QoS priority modulation strategy, high-delay-sensitive services are forced to use low-order modulation, sacrificing throughput to ensure reliability; high-throughput services allow for higher BER and prioritize high-order modulation; low-power services select the lowest-order modulation to reduce transmit power; The weighting function is used to integrate spectrum efficiency, reliability, and power consumption, and dynamically adjust the weights; Path parsing, state caching, and MCS decision engines are integrated into edge terminals, which interact directly with the physical layer chip through the SPI / UART / PCIe interface to pre-generate the SNR-MCS mapping table, which is updated only when the SNR changes by more than 2dB or when a service is switched.

7. A real-time coordinated risk dispatching system for integrated main and distribution microgrids, characterized by: The method comprises a processor, a memory and a computer program stored in the memory. When the processor executes the computer program, the method specifically performs the steps of the method for real-time coordinated risk scheduling of a main-distribution-microgrid integration as described in any one of claims 1 to 6.

Images