Power data dynamic distribution method and device based on OSU and electronic equipment

Abstract

The invention relates to the technical field of intelligent power networks, in particular to an OSU-based power data dynamic distribution method and device and electronic equipment. The method comprises the following steps: firstly, collecting electric power communication data based on an OSU, and performing preprocessing including data cleaning and normalization; and secondly, constructing a flow prediction model based on a regression model by using electric power communication data, and predicting data flow transmitted by the electric meter in future. Then, establishing a dynamic bandwidth allocation mechanism, and setting a trigger condition for the dynamic bandwidth allocation mechanism; wherein the dynamic bandwidth allocation mechanism adjusts a bandwidth allocation strategy through a priority scheduling algorithm. And finally, collecting network state data in real time, comparing the network state data with the data traffic predicted by the traffic prediction model, calculating a prediction error, and updating the traffic prediction model. Under the condition of high load of the power network, the bandwidth is dynamically allocated for different data streams, and the transmission efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart power network, and specifically to an OSU-based power data dynamic distribution method, device and electronic equipment. Background Art

[0002] OSU, or Optical Service Unit, is a fiber optic access device designed for power communication systems to achieve efficient and reliable data transmission. With the advancement of intelligent power equipment, data traffic has shown an exponential growth, and traditional communication technologies have been unable to meet the strict requirements for real-time and stability. OSU supports higher bandwidth and lower latency through optical fiber transmission, enabling the power system to exchange information efficiently in different environments. It can handle a large amount of real-time data streams, such as equipment status, fault alarms, and remote monitoring information. In addition, OSU also has flexible network adaptability, and can dynamically adjust bandwidth allocation according to the current network load to improve resource utilization efficiency. Therefore, the introduction of OSU can not only improve the overall performance of power communication, but also ensure the rapid response capability of the power system in emergency situations and ensure the safety and stability of power supply.

[0003] Existing technologies for power communication data transmission mainly include static bandwidth allocation, time-slice-based scheduling, and dynamic bandwidth management. Static bandwidth allocation processes data streams through fixed bandwidth settings, which is simple and easy to use, but lacks flexibility; time-slice-based scheduling allocates bandwidth to different data streams and rotates them in units of time, which is suitable for scenarios with small traffic fluctuations in a short period of time. However, these technologies often fail to provide satisfactory results when faced with diversified data streams of power equipment. Although dynamic bandwidth management technology can adjust bandwidth allocation according to real-time network status, it still has shortcomings. Especially under high load conditions, these technologies do not fully consider the priority differences between different data streams, so key data streams are suppressed, thereby affecting the safety and real-time performance of the power system.

[0004] To this end, an OSU-based power data dynamic allocation method, device and electronic equipment are proposed. Summary of the invention

[0005] The object of the present invention is to provide a method, device and electronic device for dynamically allocating power data based on OSU, which is used to dynamically allocate bandwidth for different data streams under high load conditions to improve transmission efficiency. First, power communication data is collected, and the power communication data includes power equipment operation data and network status data. Secondly, the power communication data is preprocessed, and the preprocessing includes data cleaning and normalization. Then, using the power communication data, a flow prediction model is constructed based on a regression model to predict the data flow used by the meter transmission in the future. Next, the thresholds of delay and bandwidth utilization are preset. If the delay and bandwidth utilization monitored in real time based on OSU exceed the preset thresholds, the dynamic bandwidth allocation mechanism is triggered. Among them, the dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through a priority scheduling algorithm based on the data flow predicted by the flow prediction model and the monitoring data of the power communication data collected in real time. Finally, the network status data is collected in real time, compared with the data flow predicted by the flow prediction model, and the prediction error is calculated; the flow prediction model is updated using error feedback.

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

[0007] The OSU-based power data dynamic allocation method includes:

[0008] Collecting power communication data, wherein the power communication data includes power equipment operation data and network status data;

[0009] Preprocessing the power communication data, wherein the preprocessing includes data cleaning and normalization;

[0010] Using the power communication data, a flow prediction model is constructed based on a regression model to predict the data flow of the electric meter transmission to be used in the future;

[0011] Establishing a dynamic bandwidth allocation mechanism and setting trigger conditions for the dynamic bandwidth allocation mechanism; the trigger conditions include delay exceeding a set threshold and bandwidth utilization exceeding a set threshold;

[0012] The dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through a priority scheduling algorithm based on the data traffic predicted by the traffic prediction model and the monitoring data based on the power communication data collected in real time by the OSU;

[0013] The network status data is collected in real time based on the OSU, compared with the data flow predicted by the flow prediction model, and the prediction error is calculated; the flow prediction model is updated using error feedback.

[0014] Furthermore, the power communication data is collected based on OSU; the power equipment operation data includes current, voltage and power; and the network status data includes delay, bandwidth utilization and data flow.

[0015] Furthermore, the data cleaning uses a Z-score method to identify and remove outliers; the normalization uses Min-Max standardization to scale the power communication data to the [0, 1] interval.

[0016] Furthermore, the input of the traffic prediction model includes power load, delay and bandwidth usage, and the power load includes active power; the output of the traffic prediction model is the predicted data traffic; the model formula of the traffic prediction model is:

[0017] ;

[0018] in, represents the predicted data traffic; represents the intercept of the traffic prediction model, that is, the data traffic predicted by the traffic prediction model when all independent variables are zero; and It represents the regression coefficient, which is the influence of each independent variable on the data flow; Indicates the power load; Indicates delay; Indicates the current bandwidth usage; Represents the error term; the regression coefficient is updated based on historical data.

[0019] Furthermore, by analyzing historical data and combining business needs, the thresholds of delay and bandwidth utilization are preset as and .

[0020] Furthermore, the priority scheduling algorithm includes three steps: traffic classification, priority calculation and bandwidth allocation strategy adjustment;

[0021] In the traffic classification step, the data streams are classified using the network status data and the predicted data traffic based on the real-time monitoring of the OSU, and a priority is assigned to each data stream;

[0022] In the step of calculating the priority, the calculation formula of the priority is:

[0023] ;

[0024] in, Indicates The priority of each data stream; Indicates the response time of the power communication data collected in real time; represents the data traffic predicted based on the traffic prediction model; Indicates the maximum available bandwidth; Indicates that the current allocation is The bandwidth of each data stream; Indicates The delay of each data stream; Indicates the maximum allowed delay; Indicates Importance score of each data stream; , , , and Represents the weight coefficient, which is used to adjust the impact of various factors on the priority.

[0025] Furthermore, in the step of adjusting the bandwidth allocation strategy, the total priority is calculated according to the priority:

[0026] ;

[0027] in, Indicates the overall priority; Indicates The priority of each data stream;

[0028] For each of the data streams , bandwidth is allocated as follows ;

[0029] ;

[0030] in, Indicates that the The bandwidth of each data stream; Indicates the total bandwidth currently available.

[0031] Furthermore, the network status data is collected in real time and compared with the predicted data flow of the flow prediction model; wherein, it is set to collect the network status data in real time every preset time.

[0032] The present invention also proposes an OSU-based power data dynamic distribution device, comprising:

[0033] A data acquisition module, which collects power communication data, wherein the power communication data includes power equipment operation data and network status data;

[0034] A data processing module, which pre-processes the power communication data, wherein the pre-processing includes data cleaning and normalization;

[0035] A flow prediction model building module, using the power communication data to build a flow prediction model based on a regression model, to predict the data flow of the electric meter transmission to be used in the future;

[0036] A dynamic bandwidth allocation module is used to establish a dynamic bandwidth allocation mechanism and set trigger conditions for the dynamic bandwidth allocation mechanism; the trigger conditions include delay exceeding a set threshold and bandwidth utilization exceeding a set threshold; the dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through a priority scheduling algorithm based on the data flow predicted by the flow prediction model and the monitoring data of the power communication data collected in real time;

[0037] The feedback update module collects the network status data in real time based on the OSU, compares it with the data flow predicted by the flow prediction model, calculates the prediction error, and uses the error feedback to update the flow prediction model.

[0038] The present invention also proposes an OSU-based power data dynamic distribution electronic device, comprising a receiver, a transmitter, a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the OSU-based power data dynamic distribution method described in any one of the embodiments of the present invention.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention constructs a traffic prediction model to analyze power load, delay and bandwidth usage, so as to accurately predict future data traffic. This prediction capability can identify potential network congestion and resource shortage problems in advance, and then effectively adjust the bandwidth allocation strategy and optimize the utilization of network resources. In addition, the regression model based on power load, delay and bandwidth usage can effectively capture the dynamic characteristics of the system and provide a scientific basis for subsequent decision-making, thereby reducing prediction errors and improving overall performance.

[0041] 2. The present invention provides a clear trigger condition for the system by presetting the threshold, making the response mechanism of bandwidth allocation more intelligent. The setting of the threshold provides an active monitoring and response mechanism, which can effectively cope with the network pressure under high load conditions, ensure the reliability and real-time performance of power communication, and help to achieve more efficient and reliable data transmission in the power communication network.

[0042] 3. The present invention uses a dynamic bandwidth allocation mechanism to flexibly adjust the bandwidth allocation strategy based on the traffic prediction model and real-time monitoring data based on OSU, which can give priority to meeting the needs of high-priority data flows and improve the utilization efficiency of network resources. Through the priority scheduling algorithm, it can effectively cope with network congestion under high load conditions, ensure the stability and timeliness of key communications, and thus improve the performance and reliability of the overall power communication system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A method flow chart of the OSU-based power communication data transmission dynamic allocation method of the present invention;

[0044] Figure 2 is a flow chart of the priority scheduling algorithm of the present invention;

[0045] Figure 3 It is a structural diagram of the OSU-based power communication data transmission dynamic allocation device of the present invention;

[0046] Figure 4 It is a schematic diagram of the OSU-based power communication data transmission dynamic allocation electronic device of the present invention.

[0047] In the figure: 1. Computer equipment; 11. Communication bus; 12. Receiver; 13. Processor; 14. Memory; 15. Computer program. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] In order to improve the transmission efficiency of the power communication data transmission system, the present invention provides a method, device and electronic device for dynamically allocating power communication data transmission based on OSU. In order to illustrate the function of the present invention, the effectiveness of the present invention will be illustrated from the following embodiments.

[0050] Embodiment 1

[0051] In the smart grid of City A, various power equipment, such as substations, distribution networks, smart meters, etc., transmit operating data and network status data in real time through wireless communication networks. These devices face the problem of insufficient bandwidth under high load conditions, resulting in data transmission delays or losses, which greatly affects the efficiency of power management. Therefore, City A adopted an OSU-based power data dynamic allocation method to solve this problem. The method flow chart of the OSU-based power data dynamic allocation method is shown below: Figure 1 shown.

[0052] Reference Figure 1Step S01 in the embodiment of the present invention is to collect power communication data, including power equipment operation data and network status data. The power communication data is collected based on the OSU; the power equipment operation data includes current, voltage and power; and the network status data includes delay, bandwidth utilization and data flow.

[0053] Collecting various types of data provides a rich source of information for the power communication data transmission system, making the decision-making process more scientific and efficient. By analyzing these data, bandwidth allocation strategies and resource management plans can be formulated more reasonably to adapt to the changing network and load conditions.

[0054] Reference Figure 1 In step S02, the power communication data is preprocessed, including data cleaning and normalization. In actual power communication data, there will be outliers introduced by sensor failure, network interference or other factors, which will have a negative impact on subsequent data analysis and model training. In addition, directly using these data for model training causes the model to focus on certain features and ignore other features, so data cleaning and normalization are required. The data cleaning uses the Z-score method to identify and remove outliers; the normalization uses Min-Max standardization to scale the power communication data to the [0, 1] interval.

[0055] Data cleaning and normalization are the basis for subsequent modeling and analysis. If there are problems with the original data, any model built based on this data will not be able to effectively reflect the actual situation.

[0056] Reference Figure 1 Step S03 in the example uses power communication data to construct a flow prediction model based on a regression model to predict the data flow that the electric meter will transmit in the future. The input of the flow prediction model includes power load, delay, and bandwidth usage, and the power load includes active power. The output of the flow prediction model is the predicted data flow. The model formula of the flow prediction model is:

[0057] ;

[0058] in, represents the predicted data traffic; represents the intercept of the traffic prediction model, that is, the data traffic predicted by the traffic prediction model when all independent variables are zero; and It represents the regression coefficient, which is the influence of each independent variable on the data flow; Indicates the power load; Indicates delay; Indicates the current bandwidth usage; Represents the error term; the regression coefficient is updated based on historical data. The introduction of the error term allows the model to better adapt to random fluctuations in data in practical applications, thereby improving the generalization ability of the model.

[0059] The traffic prediction model is fitted by using the collected historical data as a training set, with the goal of minimizing the error between the predicted value and the actual observed value. In this embodiment, the traffic prediction model is compared with the time series models ARIMA and ETS to verify the accuracy of the traffic prediction model in predicting data traffic. The autoregressive term of ARIMA is set to 2, the difference order is set to 1, and the sliding average term is set to 2; the horizontal smoothing parameter of ETS is set to 0.8, and the trend smoothing parameter is set to 0.2. The results of the comparative experiment are shown in Table 1.

[0060] Table 1 Comparative experimental results

[0061] As can be seen from Table 1, the traffic prediction model used in this embodiment has the lowest prediction error, while the prediction results of ARIMA and ETS are not ideal. Experiments have proved that the traffic prediction model can effectively capture the dynamic characteristics of the system and reduce the prediction error, thus providing a scientific basis for subsequent decision-making.

[0062] By combining the linear and nonlinear relationships of active power, delay, and bandwidth usage, the power communication data traffic can be predicted more accurately. By introducing interaction terms and polynomial terms, the model can capture the complex relationship between the input variables, thereby optimizing the prediction of data traffic and improving the dynamic allocation efficiency of the power communication system.

[0063] Reference Figure 1 Step S04 in the embodiment is to establish a dynamic bandwidth allocation mechanism and set trigger conditions for the dynamic bandwidth allocation mechanism; the trigger conditions include delay exceeding a set threshold and bandwidth utilization exceeding a set threshold. The delay and bandwidth utilization thresholds are preset by analyzing historical data and combining business needs, and are respectively preset as and :

[0064] ;

[0065] in, Indicates the latency of real-time monitoring based on OSU; Indicates the bandwidth utilization based on real-time monitoring of OSU; Indicates that the dynamic bandwidth allocation mechanism is triggered.

[0066] In this embodiment, the smart meters in the smart grid of City A transmit power data in real time through a wireless network. Under high load conditions, the delay and bandwidth utilization of data transmission will increase significantly, affecting real-time monitoring and control. Therefore, it is necessary to reasonably set the thresholds of delay and bandwidth utilization by analyzing historical data and combining business needs and real-time feedback mechanisms. By analyzing the transmission data of the past few months, it is found that the average delay during the peak hours of electricity consumption in the morning and evening is 120 milliseconds and the maximum is 300 milliseconds. Therefore, the delay threshold is set to the sum of the mean and 1.5 times the standard deviation, and finally the threshold is 150 milliseconds. When the delay exceeds the threshold, dynamic adjustment is triggered. In addition, according to past monitoring, the bandwidth utilization of the smart grid during peak hours is 80% on average. Analysis shows that when the utilization rate exceeds 85%, the reliability of data transmission decreases significantly. Therefore, the bandwidth utilization threshold is set to 85% to ensure that the system can remain stable under high load.

[0067] The thresholds for latency and bandwidth utilization can also be adjusted dynamically. For example, once the latency exceeds 150 milliseconds or the bandwidth utilization exceeds 85%, the system automatically reduces the frequency of non-critical data transmission and prioritizes the bandwidth requirements of critical applications, such as fault reporting and real-time monitoring information, to ensure that they have higher priority than general data traffic. At the same time, the backup network path is enabled to dynamically allocate traffic to other idle network nodes.

[0068] By presetting delay and bandwidth utilization thresholds, it is possible to ensure that the system can operate stably under high load conditions, avoid excessive delay or bandwidth overload, and thus trigger an appropriate dynamic adjustment mechanism to ensure the reliability and real-time performance of the power communication system.

[0069] Reference Figure 1 In step S05, the dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through a priority scheduling algorithm based on the data traffic predicted by the traffic prediction model and the monitoring data of the power communication data collected in real time by the OSU.

[0070] Specifically, the priority scheduling algorithm includes three steps: traffic classification, priority calculation and bandwidth allocation strategy adjustment. The flowchart of the priority scheduling algorithm is as follows: Figure 2 In the traffic classification step, the data flow is classified using the network status data and the predicted data flow based on the real-time monitoring of the OSU, and a priority is assigned to each data flow;

[0071] Furthermore, in the step of calculating the priority, the calculation formula of the priority is:

[0072] ;

[0073] in, Indicates The priority of each data stream; Indicates the response time of the power communication data collected in real time; represents the data traffic predicted based on the traffic prediction model; Indicates the maximum available bandwidth; Indicates that the current allocation is The bandwidth of each data stream; Indicates The delay of each data stream; Indicates the maximum allowed delay; Indicates Importance score of each data stream; , , , and Represents the weight coefficient, which is used to adjust the impact of various factors on the priority.

[0074] In this embodiment, the priority scheduling algorithm used is combined with the data based on real-time monitoring of the OSU and the data traffic prediction, so that efficient allocation of dynamic bandwidth can be achieved.

[0075] Furthermore, in the step of adjusting the bandwidth allocation strategy, the total priority is calculated according to the priority:

[0076] ;

[0077] in, Indicates the overall priority; Indicates The priority of each data stream;

[0078] For each of the data streams , bandwidth is allocated as follows :

[0079] ;

[0080] in, Indicates that the The bandwidth of each data stream; Indicates the total bandwidth currently available.

[0081] In this embodiment, there are 1,000 smart meters in the smart grid of City A, and the amount of data uploaded per meter per hour is 0.5 KB, and the available bandwidth resources are 400 Mbps. Due to the large number of meters, the data traffic increases significantly during the peak hours of electricity consumption in the morning and evening, resulting in bandwidth shortage. The current bandwidth resource utilization rate based on OSU real-time monitoring is 90%, and the real-time data traffic is 350 Mbps.

[0082] In this case, City A uses the traffic prediction model to predict the changes in data traffic in the next few hours. The result is that the data traffic is expected to be 300 Mbps at 6 o'clock; 250 Mbps at 7 o'clock; and 150 Mbps at other times. Then, priorities are assigned to different data streams based on the importance and timeliness of the data: real-time fault alarms are high-priority data streams that need to maintain a bandwidth of 30 Mbps at peak times; daily power consumption reports are medium-priority data streams that require 15 Mbps; non-critical statistics are low-priority data streams that require 5 Mbps. During the peak period from 6 to 7 o'clock, the bandwidth data of current data streams of different priorities based on real-time monitoring by OSU are: high-priority data streams use 30Mbps bandwidth; medium-priority data streams use 13 Mbps bandwidth; low-priority data streams use 4 Mbps bandwidth. Dynamic bandwidth adjustment is performed using real-time monitoring data based on OSU and predicted data traffic. After adjustment using the bandwidth allocation strategy: high priority uses 30 Mbps; medium priority uses 10 Mbps; low priority uses 2 Mbps; and a traffic margin of 8 Mbps is reserved to ensure the stability and timeliness of high-priority data stream transmission.

[0083] The priority calculation formula and bandwidth allocation strategy used in this embodiment ensure that high-priority data streams can obtain sufficient bandwidth resources, thereby optimizing the performance of the power communication system. This method not only improves the reliability of the system, but also meets the needs of real-time communication.

[0084] Reference Figure 1 In step S06, the network status data is collected in real time based on the OSU, and compared with the data flow predicted by the flow prediction model to calculate the prediction error; and the flow prediction model is updated using error feedback.

[0085] Among them, it is set to collect the network status data in real time every preset time. Under high load conditions, the preset time, i.e., the frequency of data collection, is determined based on a comprehensive analysis of network load characteristics, system performance limitations, real-time response capabilities, data processing capabilities, and other aspects. Therefore, it is necessary to flexibly adjust the collection frequency and combine it with intelligent data processing technology to ensure that under high load conditions, the system can effectively monitor the network status and respond in a timely manner to optimize data transmission performance and stability.

[0086] For example, the smart grid in City A faces extremely high network load during peak power consumption periods, such as hot summer weather. By analyzing historical data, it is found that the power load increases dramatically between 2 pm and 5 pm in the summer. During this high load period, the data collection frequency is increased from every 5 minutes to every 1 minute to monitor the line delay and bandwidth utilization in real time. In addition, artificial intelligence models can be used to analyze historical load data, predict future load changes, and dynamically adjust the data collection frequency when the load is expected to increase. Among them, artificial intelligence models can be but are not limited to machine learning models and deep learning models.

[0087] Therefore, the frequency of data collection should be considered comprehensively based on the real-time requirements of the power communication system, the characteristics of network state changes, computing resources, and model update requirements. The appropriate collection frequency can improve the response speed and accuracy of the dynamic bandwidth allocation mechanism while ensuring system stability.

[0088] Furthermore, the prediction error is calculated using the following formula:

[0089] ;

[0090] in, represents the prediction error; Indicates the actual data traffic based on real-time monitoring of OSU; Indicates the predicted data traffic given by the traffic prediction model.

[0091] Furthermore, the absolute value of the prediction error is calculated and relative error , which helps quantify the accuracy of the prediction:

[0092] ;

[0093] ;

[0094] Furthermore, after obtaining the error, it is fed back to the traffic prediction model to update the model parameters. After each new data arrives, the prediction error is used to adjust the model parameters. In addition, the model is retrained regularly based on the real-time monitored data to ensure that the model can always adapt to changes in the data flow. The update of model parameters can improve the accuracy of the model, thereby making more accurate decisions in dynamic bandwidth allocation.

[0095] The OSU-based power data dynamic allocation method used in this embodiment optimizes bandwidth allocation under high load conditions through real-time monitoring and data flow prediction based on OSU. This method effectively improves data transmission efficiency and reduces latency. Based on the use of OSU, the bandwidth allocation algorithm is further improved, and the priority is dynamically adjusted according to the real-time network status to achieve more accurate resource allocation. At the same time, the traffic prediction model is used to improve the adaptability to complex data flows, thereby enhancing the overall performance and reliability of the power communication network and improving the efficiency of data transmission.

[0096] Embodiment 2

[0097] Reference Figure 3 As shown, the present invention also provides an OSU-based power data dynamic distribution device.

[0098] The device includes a data acquisition module for collecting power communication data, wherein the power communication data includes power equipment operation data and network status data. The power communication data is collected based on the OSU; the power equipment operation data includes current, voltage and power; and the network status data includes delay, bandwidth utilization and data flow.

[0099] The device also includes a data processing module for preprocessing the power communication data, wherein the preprocessing includes data cleaning and normalization. The data cleaning uses a Z-score method to identify and remove outliers; the normalization uses Min-Max standardization to scale the power communication data to a [0, 1] interval.

[0100] The device also includes a flow prediction model construction module, which uses the power communication data to construct a flow prediction model based on a regression model to predict the data flow that the electric meter will transmit in the future. The input of the flow prediction model includes power load, delay and bandwidth usage, and the power load includes active power; the output of the flow prediction model is the predicted data flow; the model formula of the flow prediction model is:

[0101] ;

[0102] in, represents the predicted data traffic; represents the intercept of the traffic prediction model, that is, the data traffic predicted by the traffic prediction model when all independent variables are zero; and It represents the regression coefficient, which is the influence of each independent variable on the data flow; Indicates the power load; Indicates delay; Indicates the current bandwidth usage; Represents the error term; the regression coefficient is updated based on historical data.

[0103] The device also includes a dynamic bandwidth allocation module, which establishes a dynamic bandwidth allocation mechanism and sets trigger conditions for the dynamic bandwidth allocation mechanism; the trigger conditions include delay exceeding a set threshold and bandwidth utilization exceeding a set threshold; the dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through a priority scheduling algorithm based on the data flow predicted by the traffic prediction model and the monitoring data of the power communication data collected in real time by the OSU. Among them, it is set to perform real-time data collection on the network status data every preset time. In addition, by analyzing historical data and combining business needs to preset the thresholds of delay and bandwidth utilization, they are preset as and :

[0104] ;

[0105] in, Indicates the latency of real-time monitoring based on OSU; Indicates the bandwidth utilization based on real-time monitoring of OSU; Indicates that the dynamic bandwidth allocation mechanism is triggered.

[0106] The priority scheduling algorithm includes three steps: traffic classification, priority calculation and bandwidth allocation strategy adjustment;

[0107] In the traffic classification step, the data streams are classified using the network status data and the predicted data traffic based on the real-time monitoring of the OSU, and a priority is assigned to each data stream;

[0108] In the step of calculating the priority, the calculation formula of the priority is:

[0109] ;

[0110] in, Indicates The priority of each data stream; Indicates the response time of the power communication data collected in real time; represents the data traffic predicted based on the traffic prediction model; Indicates the maximum available bandwidth; Indicates that the current allocation is The bandwidth of each data stream; Indicates The delay of each data stream; Indicates the maximum allowed delay; Indicates Importance score of each data stream; , , , and Represents the weight coefficient, which is used to adjust the impact of various factors on the priority.

[0111] In the step of adjusting the bandwidth allocation strategy, the total priority is calculated according to the priority:

[0112] ;

[0113] in, Indicates the overall priority; Indicates The priority of each data stream;

[0114] For each of the data streams , bandwidth is allocated as follows :

[0115] ;

[0116] in, Indicates that the The bandwidth of each data stream; Indicates the total bandwidth currently available.

[0117] The device also has a feedback update module, which collects the network status data in real time based on the OSU, compares it with the data flow predicted by the flow prediction model, calculates the prediction error; and uses the error feedback to update the flow prediction model.

[0118] The prediction error is calculated using the following formula:

[0119] ;

[0120] in, represents the prediction error; Indicates the actual data traffic based on real-time monitoring of OSU; Indicates the predicted data traffic given by the traffic prediction model.

[0121] Furthermore, the network status data is collected in real time and compared with the predicted data flow of the flow prediction model; wherein, it is set to collect the network status data in real time every preset time.

[0122] Furthermore, the absolute value of the prediction error is calculated and relative error , which helps quantify the accuracy of the prediction:

[0123] ;

[0124] ;

[0125] Furthermore, after obtaining the error, it is fed back to the traffic prediction model to update the model parameters. After each new data arrives, the prediction error is used to adjust the model parameters. In addition, the model is retrained regularly based on the real-time monitored data to ensure that the model can always adapt to changes in the data flow. The update of model parameters can improve the accuracy of the model, thereby making more accurate decisions in dynamic bandwidth allocation.

[0126] Embodiment 3

[0127] Reference Figure 4 As shown, the present invention also provides an OSU-based power data dynamic distribution electronic device 1, including a receiver 12, a memory 14 and a processor 13. The receiver 12, the memory 14 and the processor 13 communicate with each other through a communication bus 11. The receiver 12 is used to execute commands of the processor 13, and the memory 14 is used to store computer programs. When the processor 13 executes the computer program, it implements the OSU-based power data dynamic distribution method as any one of the embodiments in Example 1.

[0128] The receiver 12 receives the collected power communication data, and the processor 13 obtains the computer program in the memory 14 to pre-process the power communication data, and reads the computer program to dynamically allocate bandwidth to different data streams. The specific process is as follows:

[0129] First, the processor 13 reads the computer program in the memory 14, including preprocessing, traffic prediction model and priority scheduling algorithm;

[0130] Secondly, the receiver 12 receives the collected power communication data and reads the computer program in the memory 14 to pre-process the collected power communication data;

[0131] Finally, the processor 13 reads the computer program in the memory 14 and performs bandwidth adjustment on the monitoring data of the real-time collected power communication data and the data flow predicted based on the flow prediction model.

[0132] The processor 13 reads the computer program in the memory 14, sets the conditions for triggering the dynamic bandwidth allocation mechanism, adjusts the bandwidth allocation strategy through the priority scheduling algorithm; calculates the prediction error; and uses the error feedback to update the traffic prediction model. The specific process is as follows:

[0133] The processor 13 reads the computer program in the memory 14 and uses the traffic prediction model to predict the data traffic using a regression model;

[0134] The processor 13 reads the computer program in the memory 14 and compares the real-time collected network status data with the predicted data flow of the flow prediction model;

[0135] The processor 13 reads the computer program in the memory 14 and presets the thresholds of delay and bandwidth utilization; if one of the delay and bandwidth utilization monitored in real time based on the OSU exceeds the preset threshold, the dynamic bandwidth allocation mechanism is triggered.

[0136] The processor 13 reads the computer program in the memory 14, and the dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through the priority scheduling algorithm, including three steps: traffic classification, priority calculation and bandwidth allocation strategy adjustment.

[0137] The communication bus 11 used by the computer device 1 includes but is not limited to a system bus, an external bus, a storage bus, a serial bus, a parallel bus, a control bus, etc. Figure 4 It is represented by only one thick line, but it does not mean that there is only one bus or one type of bus;

[0138] The communication bus 11 is used for communication between the above-mentioned computer device and other devices.

[0139] The memory 14 may be an Electrically Erasable Programmable Read-Only Memory (EEPROM or E²PROM for short), a Flash Memory, an Optical Storage, or other types of dynamic storage devices that can store information and instructions, or other media that can be used to carry or store desired program codes in the form of instructions or data structures and can be stored by a computer, but is not limited thereto.

[0140] The processor 13 may be an integrated circuit chip having the ability to process signals. The processor 13 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0141] Computer equipment 1 includes but is not limited to mobile terminals such as laptop computers and tablet computers (PADs), fixed terminals such as desktop computers, and servers in data centers. Figure 4 This is only an example and should not bring any limitation to the function and scope of use of this public embodiment.

[0142] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The method for dynamic distribution of power data based on OSU is characterized by: include: Collecting power communication data, wherein the power communication data includes power equipment operation data and network status data; Preprocessing the power communication data, wherein the preprocessing includes data cleaning and normalization; Using the power communication data, a flow prediction model is constructed based on a regression model to predict the data flow of the electric meter transmission to be used in the future; Establishing a dynamic bandwidth allocation mechanism and setting trigger conditions for the dynamic bandwidth allocation mechanism; the trigger conditions include delay exceeding a set threshold and bandwidth utilization exceeding a set threshold; The dynamic bandwidth allocation mechanism adjusts the bandwidth allocation strategy through a priority scheduling algorithm based on the data traffic predicted by the traffic prediction model and the monitoring data based on the power communication data collected in real time by the OSU; The network status data is collected in real time based on the OSU, compared with the data flow predicted by the flow prediction model, and the prediction error is calculated; the flow prediction model is updated using error feedback.

2. The method for dynamic distribution of power data based on OSU according to claim 1, characterized in that: The power communication data is collected based on the OSU; the power equipment operation data includes current, voltage and power; the network status data includes delay, bandwidth utilization and data flow.

3. The method for dynamic distribution of power data based on OSU according to claim 1, characterized in that: The data cleaning uses the Z-score method to identify and remove outliers; the normalization uses Min-Max standardization to scale the power communication data to the [0, 1] interval.

4. The method for dynamic distribution of power data based on OSU according to claim 1, characterized in that: The input of the traffic prediction model includes power load, delay and bandwidth usage, and the power load includes active power; the output of the traffic prediction model is the predicted data traffic; the model formula of the traffic prediction model is: ； in, represents the predicted data traffic; represents the intercept of the traffic prediction model, that is, the data traffic predicted by the traffic prediction model when all independent variables are zero; and It represents the regression coefficient, which is the influence of each independent variable on the data flow; Indicates the power load; Indicates delay; Indicates the current bandwidth usage; Represents the error term; the regression coefficient is updated based on historical data.

5. The method for dynamic distribution of power data based on OSU according to claim 1, characterized in that: By analyzing historical data and combining business needs, we preset the thresholds for latency and bandwidth utilization, which are respectively and .

6. The method for dynamic distribution of power data based on OSU according to claim 1, characterized in that: The priority scheduling algorithm includes three steps: traffic classification, priority calculation and bandwidth allocation strategy adjustment; In the traffic classification step, the data streams are classified using the network status data and the predicted data traffic based on the real-time monitoring of the OSU, and a priority is assigned to each data stream; In the step of calculating the priority, the calculation formula of the priority is: ； in, Indicates The priority of each data stream; Indicates the response time of the power communication data collected in real time; represents the data traffic predicted based on the traffic prediction model; Indicates the maximum available bandwidth; Indicates that the current allocation is The bandwidth of each data stream; Indicates The delay of each data stream; Indicates the maximum allowed delay; Indicates Importance score of each data stream; , , , and Represents the weight coefficient, which is used to adjust the impact of various factors on the priority.

7. The method for dynamic distribution of power data based on OSU according to claim 6, characterized in that: In the step of adjusting the bandwidth allocation strategy, the total priority is calculated according to the priority: ； in, Indicates the overall priority; Indicates The priority of each data stream; For each of the data streams , bandwidth is allocated as follows : ； in, Indicates that the The bandwidth of each data stream; Indicates the total bandwidth currently available.

8. The method for dynamic distribution of power data based on OSU according to claim 1, characterized in that: The network status data is collected in real time and compared with the predicted data flow of the flow prediction model; wherein, it is set to collect the network status data in real time every preset time.

9. The OSU-based power data dynamic distribution device is characterized by: include: A data acquisition module, which collects power communication data, wherein the power communication data includes power equipment operation data and network status data; A data processing module, which pre-processes the power communication data, wherein the pre-processing includes data cleaning and normalization; A flow prediction model building module, using the power communication data to build a flow prediction model based on a regression model, to predict the data flow of the electric meter transmission to be used in the future; A dynamic bandwidth allocation module is used to establish a dynamic bandwidth allocation mechanism and set trigger conditions for the dynamic bandwidth allocation mechanism; the trigger conditions include delay exceeding a set threshold and bandwidth utilization exceeding a set threshold; the dynamic bandwidth allocation mechanism is based on the data flow predicted by the flow prediction model and the monitoring data of the power communication data collected in real time by the OSU, and adjusts the bandwidth allocation strategy through a priority scheduling algorithm; The feedback update module collects the network status data in real time based on the OSU, compares it with the data flow predicted by the flow prediction model, calculates the prediction error, and uses the error feedback to update the flow prediction model.

10. An electronic device for dynamically distributing power data based on OSU, characterized in that: The invention comprises a receiver, a transmitter, a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method as claimed in any one of claims 1 to 8.

Images