Dynamic power distribution method for power line carrier communication heterogeneous network

Abstract

The invention relates to the technical field of intelligent power grid communication, discloses a dynamic power distribution method for a power line carrier communication heterogeneous network, and solves the problems of insufficient reliability of single medium communication and dynamic resource distribution bottleneck of heterogeneous converged communication. According to the technical scheme, the method comprises the four steps of channel quality dynamic evaluation, constraint type self-adaptive judgment, optimal medium selection and intelligent power distribution, a water injection algorithm, an iterative reverse water injection method and the like are combined, real-time characteristics of power lines and wireless channels and different power constraint scenes are adapted, differentiation strategies are designed for services with different priorities, and the optimal power distribution is realized. And an energy efficiency optimization feedback mechanism is also arranged. The method can improve the power distribution flexibility and the channel resource utilization rate of the heterogeneous network, meets the differentiated communication requirements of the smart grid, guarantees the communication reliability, and provides support for the reliable operation of the smart grid.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart grid communication, more particularly, it relates to a dynamic power allocation method for power line carrier communication heterogeneous network. BACKGROUND

[0002] The existing communication technology mainly relies on a single medium, among which the power line communication utilizes the wiring-free advantage of the existing power line, but is limited by the dynamic change of load impedance, pulse noise interference and cross transformer signal blocking problem, resulting in serious channel attenuation and significant time variation. Although wireless communication is flexible to deploy, it is easy to be affected by multipath fading, co-frequency interference and obstacle shielding, and it is difficult to guarantee the reliability in complex environment. Although the heterogeneous fusion communication can complement the medium defects and improve the system robustness, there is still a bottleneck in dynamic resource allocation. At the same time, the traditional power allocation is mostly based on the assumption of ideal channel state, which does not adapt to the real-time coupling characteristics of power line load transient and wireless channel fast fading, resulting in low resource utilization, and the single power constraint cannot take into account the differentiated requirements of power line electromagnetic compatibility specification and wireless spectrum regulation, which is limited in actual deployment.

[0003] Therefore, the present application provides a dynamic power allocation method for power line carrier communication heterogeneous network, which improves the above technical problems. SUMMARY

[0004] The present application aims at the deficiencies of the prior art, and provides a dynamic power allocation method for power line carrier communication heterogeneous network. The present application combines water injection algorithm, iterative reverse water injection method and other technologies through the whole process design of "channel quality dynamic evaluation-constraint type adaptive judgment-optimal medium selection-intelligent power allocation", which adapts to the real-time characteristics of power line and wireless channel and different power constraint scenes, reduces resource waste through dynamic optimization and accurate allocation, so as to realize the purpose of improving channel resource utilization and meeting the differentiated communication needs of smart grid.

[0005] The above technical purpose of the present application is realized by the following technical scheme: a dynamic power allocation method for power line carrier communication heterogeneous network, comprising the following steps: channel quality dynamic evaluation, constraint type adaptive judgment, optimal medium selection and intelligent power allocation four steps, which realizes the collaborative optimization of communication quality and energy efficiency by adapting to the real-time characteristics of power line and wireless channel and different power constraint scenes.

[0006] As a preferred technical scheme of the present application, the channel quality dynamic evaluation comprises: real-time acquisition of the frequency response parameters of power line and wireless channel, extraction of channel gain and noise statistics, wherein the channel gain is expressed as: , wherein, It was a power line communication channel at the time, and This is a wireless communication channel. For discrete frequency response, This represents noise power.

[0007] As a preferred technical solution of the present invention, in the dynamic assessment of channel quality, an improved weighted Middleton Class A model is used for refined modeling of power line communication impulse noise. The specific process is as follows: short-time Fourier transform and peak detection are performed on the time-domain noise waveform to distinguish between periodic and burst impulse noise. The impulse exponents of the Middleton Class A model for the two types of noise are estimated respectively and dynamic weight factors are assigned. The equivalent impulse exponent is generated by weighted fusion. At the same time, the equivalent model parameters are recursively updated using a sliding time window and maximum likelihood estimation algorithm.

[0008] As a preferred technical solution of the present invention, in the dynamic channel quality assessment, a dynamic interference identification and weighted calculation method based on spectrum sensing is adopted for wireless channel noise. The specific process is as follows: RSSI values ​​are collected in the idle time slot of the receiver and spectrum analysis is performed to distinguish between static and dynamic interference. Static interference power is calculated by long-time averaging and dynamic interference power is calculated by short-time variance. The comprehensive noise power is obtained by combining the dynamically adjusted weighting factor. The RSSI sampling period is automatically shortened when the dynamic interference intensity is abnormal.

[0009] As a preferred embodiment of the present invention, the constraint type adaptive judgment includes: extracting hardware configuration parameters and defining a hardware configuration state vector. ;in, To detect the first Amplifier module set, Not detected; Calculate the total number of power amplifiers and parameter set Dimensions ,in accordance with and The numerical relationship is used to determine the total power constraint or the total power-channel dual constraint scenario.

[0010] As a preferred embodiment of the present invention, when the scenario is determined to be a total power constraint scenario, the constraint condition for the optimization problem is as follows: , When the scenario is determined to be a dual constraint of total power and channel, the optimization problem constraints are as follows: , The objective function is the same in both scenarios: , in, Is assigned to the first The first media The power of each subcarrier, It is the first The first media Normalized signal-to-noise ratio of each subcarrier.

[0011] As a preferred embodiment of the present invention, the optimal medium selection includes: arranging the channel gain ratio set generated based on dynamic channel quality assessment in ascending order to obtain an ordered sequence; and, under the total power constraint scenario, selecting the single medium with the best channel quality for each subcarrier. If the power line medium is selected, the wireless medium is selected otherwise. In the dual-constraint scenario, the inefficient medium is shut down by traversing the ordered sequence, and a forced reselection of the medium is forced when a sudden increase in power line impulse noise power or excessive duration of wireless co-channel interference is detected.

[0012] As a preferred technical solution of the present invention, the optimal medium selection adopts a differentiated adaptation strategy for different priority services: Level 1 services use wireless as the primary mode and power line as the backup, forming a 1+1 redundant transmission architecture; Level 2 services prioritize high-quality channels and reserve dynamic backup power in a fixed proportion of their total power requirements; Level 3 services are allocated according to load balancing, prioritizing the allocation of data packets to media with lower load rates, and when system resources are tight, the transmission power of Level 3 services can be temporarily reduced to ensure high-priority services.

[0013] As a preferred embodiment of the present invention, the intelligent power allocation includes: constructing an equivalent channel gain set under a total power constraint scenario. The water injection algorithm is executed, and the power allocation formula is as follows: , in, For Lagrange multipliers, ; In the dual-constraint scenario, the iterative reverse water injection method is executed. After initializing the power and remaining power, the available power of each medium is calculated iteratively and engineering corrections are made until all power constraints are met.

[0014] As a preferred technical solution of the present invention, it also includes an energy efficiency optimization feedback mechanism: after each power allocation is completed, the global energy efficiency index is calculated, the energy efficiency history queue is maintained and the moving average is dynamically calculated as the performance baseline, when the current energy efficiency value is continuously lower than the set value of the historical baseline, the reasons for the performance decline are analyzed and 1-2 fine-tuning schemes are selected from the built-in strategy library for application, and at the same time, the experience samples of "environmental state-fine-tuning action-efficiency change" are saved to realize the autonomous learning and application of the optimal strategy.

[0015] In summary, the present invention has the following beneficial effects: Firstly, through the entire process design of "dynamic channel quality assessment - adaptive judgment of constraint type - optimal medium selection - intelligent power allocation", combined with technologies such as water-filling algorithm and iterative reverse water-filling method, it can accurately adapt to the real-time characteristics of power lines and wireless channels and different power constraint scenarios, effectively improve the problems of poor adaptability and lack of constraint compatibility of traditional power allocation, and greatly improve the flexibility and effectiveness of power allocation in heterogeneous networks.

[0016] Secondly, this method can reduce resource waste through dynamic optimization and precise allocation, significantly improve channel resource utilization, and at the same time meet the differentiated communication needs of different priority services such as fault tripping and safety control in smart grids, ensuring the continuity and reliability of high-priority service transmission.

[0017] Third, the system's built-in three-level early warning mechanism, dynamic adjustment process of sampling period, and energy efficiency optimization feedback mechanism can not only provide timely warnings and take countermeasures when power consumption approaches the constraint threshold, but also continuously optimize transmission energy efficiency, ensuring the stable and efficient operation of the smart grid power distribution and communication network, and providing strong underlying support for the reliable operation of the smart grid. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a dynamic power allocation method for heterogeneous power line carrier communication networks, as provided in an embodiment of the present invention. Detailed Implementation

[0019] The present application will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present application, but do not limit the present application in any way. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present application. These all fall within the protection scope of the present application.

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0021] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0022] Furthermore, the technical features involved in the various embodiments of this application described below can be combined with each other as long as they do not conflict with each other.

[0023] This invention aims to address the problems in smart grid power distribution communication networks, such as insufficient reliability of single-medium communication and bottlenecks in dynamic resource allocation for heterogeneous converged communication (poor adaptability and lack of constraint compatibility in traditional power allocation), which make it difficult to meet the requirements for coordinated optimization of communication quality and energy efficiency in multiple scenarios. Therefore, this invention proposes a dynamic power allocation method for heterogeneous power line carrier communication networks to improve the flexibility and effectiveness of power allocation in heterogeneous networks and ensure the reliability of smart grid communication. This method, through a full-process design of "dynamic channel quality assessment - adaptive constraint type judgment - optimal medium selection - intelligent power allocation," combines techniques such as water-filling algorithms and iterative reverse water-filling methods to adapt to the real-time characteristics of power lines and wireless channels and different power constraint scenarios. By dynamically optimizing and accurately allocating resources, it reduces resource waste, thereby improving channel resource utilization and meeting the differentiated communication needs of smart grids.

[0024] Please refer to Figure 1 , Figure 1 A flowchart illustrating the vehicle repair shop operation evaluation method according to an embodiment of this disclosure is shown. The overall process mainly includes the following four steps: S1: Dynamic assessment of channel quality.

[0025] S1.1, acquire frequency response parameters of power lines and wireless channels in real time and extract parameters; Parameter extraction includes: extracting channel gain and noise statistics. The channel gain is expressed as: , in, It was a power line communication channel at the time, and This is a wireless communication channel. For discrete frequency response, This represents noise power.

[0026] S1.2, for power line communication impulse noise, an improved weighted MiddletonClassA model is used for refined modeling to distinguish between periodic and sudden interference and to set weights.

[0027] The specific process is as follows: First, perform Short-Time Fourier Transform (STFT) and peak detection on the time-domain noise waveform to distinguish between periodic impulse noise and sudden impulse noise. Then, estimate the impulse exponent of the Middleton Class A model for both periodic and sudden impulse noise. and This reflects the pulse intensity of the corresponding noise source. Subsequently, dynamic weighting factors are assigned to the two types of noise based on their actual impact within the current observation window. and And satisfy By weighted fusion, an equivalent impulse exponent is generated. This is used to characterize the overall impulse noise characteristics of the current channel: When periodic loads dominate, the model focuses more on When the randomness of the load changes frequently, the model will automatically switch. The primary function is to overcome the time-varying nature of the channel. Simultaneously, to overcome this time-varying nature, a sliding time window and maximum likelihood estimation (MLE) algorithm are employed. During each channel coherence time interval, the equivalent model parameters are recursively updated using the latest noise sampling data, ensuring that the model always accurately matches the current real noise environment.

[0028] S1.3, for wireless channel noise, adopts a spectrum-aware dynamic interference identification and weighted calculation method to distinguish between static and dynamic interference and adjust the weights.

[0029] The specific process is as follows: During idle time slots at the receiving end, not only are RSSI values ​​collected, but further spectral analysis is performed on the collected signals. By comparing the current spectrum with historical spectra, interference is classified into static interference and dynamic interference. Then, the weighted noise power is calculated: static interference power... Dynamic interference power is obtained through long-term averaging. The overall noise power is calculated using short-time variance. Weighting factors and Dynamic adjustment based on the time-varying characteristics of interference: Through Doppler spread estimation, when the Doppler spread value exceeds a preset threshold, it is determined to be a rapidly changing environment. When the channel is in a rapidly changing environment, the [interference level] is increased. Weighting; when the Doppler spread value is less than a preset threshold, it is determined to be a relatively stable environment; in a relatively stable environment, the weighting is increased. Weighting. The preset threshold for Doppler expansion is set according to the business scenario, and the weighting factors meet the following requirements. Initially, a standard sampling period is used. When the intensity of dynamic interference is detected to exceed the threshold for N consecutive periods or the rate of change is abnormal, the RSSI sampling period is automatically shortened to capture the dynamic characteristics of the interference with higher accuracy and provide more timely decision-making basis for power allocation.

[0030] S1.4, first perform normalized SNR dynamic calculation: , in, It serves as a forgetting factor to compensate for the time-varying characteristics of the channel.

[0031] Next, time-varying characteristic processing is performed, specifically, impulse noise transient suppression is applied to power line communication, and Doppler frequency shift correction is applied to the wireless channel.

[0032] S1.5, Generate and sort the channel gain ratios, and calculate the cross-medium gain ratio: , Among them, the subcarrier is characterized The quality advantages of power line communication over wireless channels.

[0033] Constructing an ordered sequence Update cycle ; in, For power line communication channel coherence bandwidth, The equivalent average fading period of the channel is used to characterize the speed of the channel's time-varying characteristics. Its value can be estimated by observing the time it takes for the autocorrelation function of the channel gain to fall below a certain threshold.

[0034] S1.6, Construct the channel matrix and generate the evaluation matrix. : , in, for In an ordered sequence The index position in the table.

[0035] S2: Constraint type adaptive judgment.

[0036] S2.1 Extract hardware configuration parameters and define hardware configuration state vector. ; in, To detect the first Amplifier module set, Not detected.

[0037] Total number of power amplifiers ,in, It is the L0 norm. The upper-level management interface sends out the parameter set. Dimensions satisfy: , in, Due to total transmit power constraints, Independent power constraints for power line communication Independent power constraints for wireless media.

[0038] S2.2, Determine the constraint scenario, if This is a total power constraint, and the optimization problem constraints are as follows: , in, Is assigned to the first The first media The power of each subcarrier.

[0039] like This results in a dual constraint of total power and channel, with the following optimization problem constraints: , The objective function is: , in, It is the first The first media Normalized signal-to-noise ratio of each subcarrier.

[0040] If it is a double constraint condition, then the threshold ratio needs to be initialized: , in, It is the median. It is the threshold ratio that ultimately converges, and satisfies... ; It is a subcarrier The cross-media channel gain ratio, This represents the minimum channel gain ratio among all subcarriers. This represents the maximum value of the channel gain ratio among all subcarriers.

[0041] Simultaneously, the ratio of the difference between power consumption and the constraint threshold is calculated in real time. A three-tiered early warning mechanism is established: when the system log is below 10%, a minor early warning event is recorded, and the sampling period is dynamically adjusted; when the system log is below 5%, a moderate early warning event is recorded, and in addition to adjusting the sampling period, a notification is sent to the network management unit, and backup power is reserved for core control services; when the system log is below 2%, a severe early warning event is recorded, and data transmission of non-critical background services is forcibly suspended, the system enters a low-power protection mode, and communication resources for first-level services such as fault tripping and security control are prioritized.

[0042] The specific process for dynamically adjusting the sampling period is as follows: When an early warning is triggered, the sampling period is adaptively shortened using an exponential decay model. , in, This indicates the system initial sampling period, with an attenuation factor of 0.7 for a smooth transition. A lower limit for the sampling period is also set. for To avoid excessive system overhead.

[0043] Meanwhile, the system maintains a cycle based on 10 periods. If the sliding window of the value shows a consistently negative rate of change, indicating that power consumption is rapidly approaching its upper limit, then the sampling period is predictively shortened in advance. Furthermore, after each power allocation operation, the power is updated and the calculation is recalculated. If no warning is triggered within 10 consecutive sampling periods, the system is considered to have stabilized and the sampling period will be automatically restored gradually.

[0044] S2.3, Dynamic switching trigger mechanism, if the power amplifier module state changes Triggering a re-evaluation: ;like This triggers a re-determination; when a channel mutation causes... Clear historical constraint parameters and restart the process.

[0045] S3: Select the optimal medium.

[0046] S3.1, Based on the channel gain ratio set generated in step S1 ,in, For subcarriers Up Media The normalized signal-to-noise ratios are then arranged in ascending order to form an ordered sequence: ,in .

[0047] S3.2, In the scenario of total power constraint, for each subcarrier Choose the single medium with the best channel quality. , Choose the power line as the medium; otherwise Choose a wireless medium. In a dual-constraint scenario, traverse the ordered sequence. Shut down inefficient media, specifically if Turn off the wireless medium, or turn off the power line medium.

[0048] S3.3 When a sudden increase in power line impulse noise power or excessive duration of wireless co-channel interference is detected, a media reselection is forced.

[0049] The specific process is as follows: For high-priority service transmissions, the system activates a service-aware intelligent media adaptation engine. The system has a built-in service priority mapping table: fault tripping commands and safety control commands are classified as Level 1 services; power regulation and demand response commands for new energy power plants are classified as Level 2 services; and routine services such as electricity consumption information collection and equipment status monitoring are classified as Level 3 services. Each level of service corresponds to a different channel quality SNR threshold value, with Level 1 services having the highest threshold to ensure their transmission reliability.

[0050] The primary service uses wireless as the main method, with power line as a backup. The main transmission channel prioritizes subcarriers with an SNR higher than the primary threshold in the wireless medium. At the same time, the system automatically selects several subcarriers with the best SNR on the power line medium and pre-allocates a small amount of power to them as dedicated backup channels, forming a 1+1 redundant transmission architecture.

[0051] Secondary services are prioritized and reserved for backup. The system compares the SNR of the power line and the radio medium on the same subcarrier in parallel and selects the one with better quality as the primary channel. At the same time, the power allocation algorithm is required to reserve 10% of the total power requirement of the service as dynamic backup power. This part of the power does not participate in the initial allocation and is only used when the quality of the primary channel deteriorates, for switching to the backup medium or increasing the transmission power.

[0052] Level 3 services are distributed based on load balancing: A load-based distribution strategy is adopted. The system calculates the overall load rate of the current power lines and wireless media, and prioritizes allocating data packets to media with lower load rates and more idle resources, thereby improving the overall resource utilization of the system and avoiding congestion on a single media.

[0053] When system resources are scarce and high-priority services need to be transmitted simultaneously, intelligent resource scheduling is initiated. If power is insufficient, the transmission power of level 3 services can be temporarily reduced, and some resources can be used to prioritize level 1 and level 2 services. If the quality of the main channel is detected to drop below the reliability threshold, there is no need to reapply for resources. Instead, pre-configured backup channels and backup power are used for millisecond-level rapid switching to ensure the continuity and reliability of high-priority services.

[0054] S4: Intelligent power allocation.

[0055] S4.1, Constructing the equivalent channel gain set under the total power constraint scenario. The water injection algorithm is executed, and the power allocation formula is as follows: , in, For Lagrange multipliers, In a dual-constraint scenario, the reverse water injection method is executed by first initializing... and remaining power ;when Execute an iterative loop to calculate the available power for each medium: , in, , These are the upper limits for power lines and wireless media power, respectively.

[0056] S4.2, make engineering modifications, if Then place and from If the corresponding power is deducted, ( (Single media power limit), then set And update Iterate until both conditions are met simultaneously: .

[0057] S4.3, Perform energy efficiency optimization feedback. After each power allocation, the system immediately calculates the global energy efficiency index for this transmission: total transmitted data volume / total power consumed. The system maintains a fixed-length energy efficiency history queue and dynamically calculates the moving average of this queue as the performance baseline. When the system detects that the current energy efficiency value is consistently lower than the historical baseline by 15% or more, the optimization process is immediately triggered.

[0058] After optimization is triggered, the system first analyzes the potential causes of performance degradation. It quickly matches the current channel state characteristics with those of historical high-energy-efficiency periods to determine whether the degradation stems from widespread channel deterioration or a mismatch between the current power allocation strategy and the environment. If a strategy mismatch is identified, parameter fine-tuning is initiated. The system has a built-in lightweight strategy library containing several fine-tuning schemes: Scheme A (Tilted Allocation): Slightly increases the power allocation weight for high-quality subcarriers; Scheme B (Service Assurance): Reserves more spare power for high-priority services in the total power, sacrificing some efficiency for reliability in extreme conditions; Scheme C (Exploratory Adjustment): Randomly selects a small subset of subcarriers and slightly perturbs their power allocation to explore better allocation possibilities. The system will then select 1-2 of the most likely effective schemes from the strategy library based on the current service type and channel characteristics.

[0059] Furthermore, the system saves the result of this "environmental state - fine-tuning action - efficiency change" as an experience sample. By continuously collecting a large number of samples, the system can autonomously learn which fine-tuning actions will bring the greatest energy efficiency improvement in which scenarios. Ultimately, the system can achieve the function of "predicting the environment and proactively adopting the optimal strategy".

[0060] The above description is merely a preferred embodiment 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 also be considered within the scope of protection of the present invention.

Claims

1. A dynamic power allocation method for heterogeneous networks using power line carrier communication, characterized in that, The method includes four steps: dynamic channel quality assessment, adaptive constraint type judgment, optimal medium selection, and intelligent power allocation. By adapting to the real-time characteristics of power lines and wireless channels and different power constraint scenarios, it achieves coordinated optimization of communication quality and energy efficiency.

2. The dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 1, characterized in that, The dynamic channel quality assessment includes: real-time acquisition of frequency response parameters of the power line and wireless channels, extraction of channel gain and noise statistics, wherein the channel gain is expressed as: ， in, It was a power line communication channel at the time, and This is a wireless communication channel. For discrete frequency response, This represents noise power.

3. The dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 2, characterized in that, In the aforementioned dynamic assessment of channel quality, the process of refining the model for power line communication impulse noise using an improved weighted Middleton Class A model is as follows: short-time Fourier transform and peak detection are performed on the time-domain noise waveform to distinguish between periodic and sudden impulse noise; the impulse exponents of the Middleton Class A model for the two types of noise are estimated separately and dynamic weight factors are assigned; weighted fusion is performed to generate an equivalent impulse exponent; and the equivalent model parameters are recursively updated using a sliding time window and maximum likelihood estimation algorithm.

4. A dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 2, characterized in that, In the aforementioned dynamic channel quality assessment, the process of using a spectrum-aware dynamic interference identification and weighted calculation method for wireless channel noise is as follows: RSSI values ​​are collected in idle time slots at the receiving end and spectrum analysis is performed to distinguish between static and dynamic interference. Static interference power is calculated by long-time averaging and dynamic interference power is calculated by short-time variance. The comprehensive noise power is obtained by combining dynamically adjusted weighting factors, and the RSSI sampling period is automatically shortened when the dynamic interference intensity is abnormal.

5. A dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 1, characterized in that, The adaptive constraint type determination includes: extracting hardware configuration parameters and defining a hardware configuration state vector. ;in, To detect the first Amplifier module set, Not detected; Calculate the total number of power amplifiers and parameter set Dimensions ,in accordance with and The numerical relationship is used to determine the total power constraint or the total power-channel dual constraint scenario.

6. A dynamic power allocation method for heterogeneous power line carrier communication networks according to claim 5, characterized in that, When the scenario is determined to be a total power constraint scenario, the optimization problem constraints are as follows: ， When the scenario is determined to be a dual constraint of total power and channel, the optimization problem constraints are as follows: ， The objective function is the same in both scenarios: , in, Is assigned to the first The first media The power of each subcarrier, It is the first The first media Normalized signal-to-noise ratio of each subcarrier.

7. A dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 1, characterized in that, The optimal medium selection includes: arranging the set of channel gain ratios generated based on dynamic channel quality assessment into an ordered sequence in ascending order; and, under the total power constraint scenario, selecting the single medium with the best channel quality for each subcarrier. If the power line medium is selected, the wireless medium is selected otherwise. In the dual-constraint scenario, the inefficient medium is shut down by traversing the ordered sequence, and a forced reselection of the medium is forced when a sudden increase in power line impulse noise power or excessive duration of wireless co-channel interference is detected.

8. A dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 7, characterized in that, In the optimal medium selection, a differentiated adaptation strategy is adopted for different priority services: Level 1 services use wireless as the primary mode and powerline as the backup, forming a 1+1 redundant transmission architecture; Level 2 services prioritize high-quality channels and reserve dynamic backup power in a fixed proportion of their total power requirements; Level 3 services are allocated according to load balancing, prioritizing the allocation of data packets to media with lower load rates, and when system resources are tight, the transmission power of Level 3 services can be temporarily reduced to ensure high-priority services.

9. A dynamic power allocation method for heterogeneous networks using power line carrier communication according to claim 1, characterized in that, The intelligent power allocation includes: constructing an equivalent channel gain set under a total power constraint scenario. The water injection algorithm is executed, and the power allocation formula is as follows: ， in, For Lagrange multipliers, ; In the dual-constraint scenario, the iterative reverse water injection method is executed. After initializing the power and remaining power, the available power of each medium is calculated iteratively and engineering corrections are made until all power constraints are met.

10. A dynamic power allocation method for heterogeneous power line carrier communication networks according to claim 1, characterized in that, It also includes an energy efficiency optimization feedback mechanism: after each power allocation is completed, the global energy efficiency index is calculated, the energy efficiency history queue is maintained and the moving average is dynamically calculated as the performance baseline. When the current energy efficiency value is continuously lower than the set value of the historical baseline, the reasons for the performance degradation are analyzed and 1-2 fine-tuning schemes are selected from the built-in strategy library for application. At the same time, the experience samples of "environmental state-fine-tuning action-efficiency change" are saved to realize the autonomous learning and application of the optimal strategy.

Images