A concentrator adaptive electric energy meter data transmission method

Abstract

The present application provides a concentrator adaptive electric energy meter data transmission method, which relates to the technical field of power grid data transmission, including: based on a regional communication twin network, an electromagnetic interference data set, and an environmental prediction data set, with the communication quality as a constraint and the device power consumption as the purpose, multiple communication time periods within the electric meter collection cycle are optimized to determine the optimal communication time period; power correction analysis is performed based on a sample data set to generate a power adjustment strategy; based on the optimal communication time period, the concentrator is controlled to collect electric meter data, and during the collection process, feedback control is performed according to the power adjustment strategy. This application can solve the existing electric energy meter data transmission method based on the concentrator, which has the problem of poor accuracy in setting the communication time period and the inability to adjust the concentrator output power in time according to the communication quality, resulting in poor data transmission quality and stability. It can achieve the effect of improving data transmission quality and efficiency and reducing device power consumption.

Description

Technical Field

[0001] The present application relates to the technical field of power grid data transmission, and in particular to a concentrator adaptive electric energy meter data transmission method. Background Art

[0002] With the continuous development and popularization of smart grid technology, the data transmission method of electric energy meters plays an increasingly important role in the power system. The traditional data transmission method of electric energy meters usually relies on manual on-site meter reading, which is not only inefficient, but also has large errors and safety hazards. Therefore, the electric energy meter data acquisition method based on the concentrator came into being.

[0003] At present, the existing concentrator-based electricity meter data transmission method does not take into account the interference of the regional environment on the communication quality, resulting in poor accuracy in setting the communication period. At the same time, the concentrator output power cannot be adjusted in time according to the communication quality during the data transmission process, resulting in poor data transmission quality and stability. At the same time, there is a technical problem of high equipment power consumption. Summary of the invention

[0004] The purpose of this application is to provide a concentrator adaptive electric energy meter data transmission method to solve the existing concentrator-based electric energy meter data transmission method. Since the interference of the regional environment on the communication quality is not taken into account, the accuracy of the communication time period setting is poor. At the same time, during the data transmission process, the concentrator output power cannot be adjusted in time according to the communication quality, resulting in poor data transmission quality and stability. At the same time, there is a technical problem of high power consumption of the equipment.

[0005] In view of the above problems, the present application provides a concentrator adaptive electric energy meter data transmission method, including: in a cloud system, according to the concentrator attribute data and the electric energy meter attribute data of the target area, constructing a regional communication twin network; based on the electric meter collection cycle, data is collected from the target area according to the preset electromagnetic interference factor to obtain an electromagnetic interference data set, and the target area is environmentally sensed according to the preset environmental interference factor to obtain an environmental prediction data set; based on the regional communication twin network, the electromagnetic interference data set and the environmental prediction data set, with the constraint of meeting the expected communication quality and with the purpose of minimizing the power consumption of the device, multiple communication time periods within the electric meter collection cycle are optimized to determine the optimal communication time period; with the concentrator attribute data, the electric energy meter attribute data and the expected communication quality as constraints, a sample communication data set is retrieved and obtained, power correction analysis is performed based on the sample communication data set, and a power adjustment strategy is generated; the optimal communication time period and the power adjustment strategy are decentralized to the edge computing unit of the concentrator, the concentrator is controlled to collect electric meter data based on the optimal communication time period, and during the collection process, the concentrator is fixed-point feedback regulated according to the power adjustment strategy.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] 1. By conducting electromagnetic interference analysis and environmental perception prediction on the target area, the communication period with the least interference is selected for data transmission, which can reduce the interference of the external environment on the communication quality; improve the accuracy of the communication period setting; then during the data transmission process, the concentrator output power is timely fed back and regulated through the pre-built power adjustment strategy, which can improve the adaptability of the output power and communication quality, thereby achieving the technical effect of significantly improving the data transmission quality and efficiency and reducing the power consumption of the equipment.

[0008] 2. Based on the cloud-edge collaborative method, the communication period optimization and power correction analysis are carried out in the cloud system, and the optimal communication period and power adjustment strategy are delegated to the edge computing unit of the concentrator. This can reduce the data processing consumption inside the concentrator and reduce operating costs, while improving the accuracy and timeliness of the concentrator power feedback control.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0011] Figure 1 This is a flow chart of a concentrator adaptive electric energy meter data transmission method of the present application.

[0012] Figure 2 This is a schematic diagram of a flow chart of obtaining an electromagnetic interference data set in a concentrator adaptive electric energy meter data transmission method of the present application. DETAILED DESCRIPTION

[0013] This application provides a concentrator adaptive power meter data transmission method, which solves the existing concentrator-based power meter data transmission method. Since the interference of the regional environment on the communication quality is not considered, the accuracy of the communication time period setting is poor. At the same time, the concentrator output power cannot be adjusted in time according to the communication quality during the data transmission process, resulting in poor data transmission quality and stability. At the same time, there is a technical problem of high device power consumption. By performing electromagnetic interference analysis and environmental perception prediction on the target area, the communication time period with the least interference is selected for data transmission, which can reduce the interference of the external environment on the communication quality; improve the accuracy of the communication time period setting; then in the data transmission process, the concentrator output power is timely fed back and regulated through the pre-built power adjustment strategy, which can improve the adaptability of the output power and the communication quality, thereby achieving the technical effect of significantly improving the data transmission quality and efficiency and reducing the power consumption of the equipment.

[0014] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0015] For examples, please see the attached Figure 1 The present application provides a concentrator adaptive electric energy meter data transmission method, which specifically includes the following steps:

[0016] Step 1: In the cloud system, build a regional communication twin network based on the concentrator attribute data and electricity meter attribute data of the target area.

[0017] Specifically, the cloud system refers to the power grid data processing center system built on the cloud platform. It is part of the smart grid and is used to centrally manage and process electricity consumption data from concentrators and other smart devices, including data collection, storage, processing, analysis and decision-making functions.

[0018] In the cloud system, the concentrator attribute data and the electric energy meter attribute data of the target area are obtained, wherein the target area is a customized power consumption area, such as a residential area, wherein the number of concentrators in the target area may be one or more. In this application, one concentrator is used as an example to facilitate the understanding of those skilled in the art, wherein each concentrator corresponds to multiple electric energy meters, which are used to collect the power consumption data of multiple electric energy meters. First, the attribute data of the concentrator and the electric energy meter are collected, including device model, communication capability, location coordinates, processing capability, operating parameters, etc. Then, based on the digital twin technology, simulation modeling is performed according to the concentrator attribute data and the electric energy meter attribute data to construct a regional communication twin network, that is, according to the collected data, the topological structure of the concentrator communication network is constructed, including the connection relationship between the concentrators and the connection relationship between the concentrator and the electric energy meter, and the communication protocol between the concentrator and the electric energy meter is simulated, including data transmission, protocol negotiation, etc.

[0019] By constructing a regional communication twin network, it provides support for subsequent regional communication simulation and improves the authenticity and accuracy of the communication simulation.

[0020] Step 2: Based on the meter collection cycle, data is collected on the target area according to the preset electromagnetic interference factor to obtain an electromagnetic interference data set, and environmental perception is performed on the target area according to the preset environmental interference factor to obtain an environmental prediction data set.

[0021] Specifically, configure the meter collection cycle, which refers to the time range for meter collection, and can be set according to actual conditions, such as every Tuesday, that is, complete the meter data collection within the time period of every Tuesday; first, obtain the preset interference factor, which refers to the external factor that affects the communication quality between the concentrator and the electric energy meter, including the preset electromagnetic interference factor and the preset environmental interference factor, wherein the preset electromagnetic interference factor includes at least wireless interference, radio frequency interference, electrical noise and mechanical interference, and the preset environmental interference factor includes at least weather type, ambient temperature, ambient humidity and lighting characteristics, and additional interference factors can also be added according to actual conditions.

[0022] Then, based on the meter collection cycle, the historical electromagnetic interference data of the target area are analyzed according to the preset electromagnetic interference factors to obtain an electromagnetic interference data set, wherein the electromagnetic interference data set includes multiple time intervals and corresponding multiple electromagnetic interference data, and the electromagnetic interference data includes interference type, interference intensity and duration; on the other hand, based on the meter collection cycle, the environment of the target area is perceived according to the preset environmental interference factors, that is, the environment is predicted for multiple time intervals of the meter collection cycle to obtain an environmental prediction data set, which includes multiple time intervals and corresponding multiple environmental prediction data.

[0023] Step three: Based on the regional communication twin network, electromagnetic interference data set and environmental prediction data set, with the expected communication quality as a constraint and the purpose of minimizing device power consumption, multiple communication time periods within the meter collection cycle are optimized to determine the optimal communication time period.

[0024] Specifically, first, the regional communication twin network is conditionally rendered according to the electromagnetic interference data set and the environmental prediction data set to obtain a rendered communication twin network; then, the time period of a single communication time period is determined based on the historical collection time analysis of the concentrator, and multiple communication time periods are generated within the meter collection period. For example, the time period interval can be set to 1 minute, that is, a communication time period is set every 1 minute; then, in the rendered communication twin network, with the expected communication quality as the constraint condition, simulated communication is performed for multiple communication time periods respectively, where the expected communication quality includes multiple communication quality indicators, such as expected signal strength, expected signal-to-noise ratio, expected bit error rate, etc., to obtain multiple simulated communication results; finally, the multiple simulated communication results are evaluated according to the communication evaluation function, and the communication time period with the largest evaluation coefficient is selected as the optimal communication time period.

[0025] By optimizing multiple communication periods within the meter collection cycle to obtain the optimal communication period, the interference of the external environment on the communication quality can be reduced, thereby improving the quality and efficiency of data transmission, while reducing the output power of the concentrator, thereby achieving the effect of reducing equipment power consumption.

[0026] Step 4: taking the concentrator attribute data, the energy meter attribute data and the expected communication quality as constraints, retrieve and obtain a sample communication data set, perform power correction analysis based on the sample communication data set, and generate a power adjustment strategy.

[0027] Specifically, the concentrator attribute data, the energy meter attribute data and the expected communication quality are used as retrieval constraints, and a sample communication data set is retrieved online, wherein the sample communication data includes a sample communication quality deviation and a sample adjustment power. Then, a first power adjustment unit is constructed based on a BP neural network, and a second power adjustment unit is constructed based on a random decision forest; and the sample communication data set is used as training data to perform supervised training and cross-validation on the first power adjustment unit and the second power adjustment unit to obtain a first converged power adjustment unit and a second converged power adjustment unit, and finally a power adjustment strategy is constructed based on the first converged power adjustment unit and the second converged power adjustment unit.

[0028] By constructing a power adjustment strategy based on the fusion of BP neural network and random decision forest, the advantages of multiple models can be combined, the prediction error of a single model can be reduced, and the accuracy and reliability of power adjustment can be improved.

[0029] Step 5: Decentralize the optimal communication period and power adjustment strategy to the edge computing unit of the concentrator, control the concentrator to collect meter data based on the optimal communication period, and during the collection process, perform fixed-point feedback control on the concentrator according to the power adjustment strategy.

[0030] Specifically, first, the optimal communication period and power adjustment strategy are decentralized and transmitted to the edge computing unit of the concentrator; then the concentrator is controlled to collect meter data within the optimal communication period to achieve the purpose of improving data transmission quality and reducing equipment power consumption; at the same time, during the data collection process, the communication quality between the concentrator and the electric energy meter is monitored at a fixed point, and the communication quality deviation is determined based on the expected communication quality, and the communication quality deviation is further input into the power adjustment strategy to obtain the adjusted power; finally, the concentrator is subjected to fixed-point feedback control according to the adjusted power, so as to improve the adaptability of the concentrator output power and the communication quality, and finally achieve the technical effect of significantly improving data transmission quality and efficiency and reducing equipment power consumption.

[0031] By providing a concentrator adaptive electric energy meter data transmission method, the existing concentrator-based electric energy meter data transmission method can solve the problem that the accuracy of communication time period setting is poor due to the failure to consider the interference of regional environment on communication quality. At the same time, the concentrator output power cannot be adjusted in time according to the communication quality during the data transmission process, resulting in poor data transmission quality and stability. At the same time, there is a technical problem of high equipment power consumption. By performing electromagnetic interference analysis and environmental perception prediction on the target area, the communication time period with the least interference is selected for data transmission, which can reduce the interference of the external environment on the communication quality; improve the accuracy of communication time period setting; then, during the data transmission process, the concentrator output power is timely fed back and regulated through the pre-built power adjustment strategy, which can improve the adaptability of output power and communication quality, thereby achieving the technical effect of significantly improving data transmission quality and efficiency and reducing equipment power consumption.

[0032] Further, the electromagnetic interference data set is obtained, as shown in the attached Figure 2 As shown, step 2 of this application includes:

[0033] The meter collection cycle is divided according to a preset step size to obtain multiple meter collection intervals, and a first meter collection interval is randomly selected. The target area environmental monitoring log is queried with the first meter collection interval as a constraint to obtain a first sample monitoring data set, wherein the first sample monitoring data is marked with a time interval; the first sample monitoring data set is selected according to a preset electromagnetic interference factor to obtain a first sample electromagnetic interference data set, wherein the preset electromagnetic interference factor includes at least wireless interference, radio frequency interference, electrical noise and mechanical interference; dynamic weights are configured based on the time interval, and the first sample electromagnetic interference data set is weightedly fused, and the first electromagnetic interference data is obtained based on the weighted fusion result analysis, wherein the first electromagnetic interference data includes a first interference type set, a first interference intensity set and a first duration set; the electromagnetic interference data set is constructed based on the first meter collection interval and the first electromagnetic interference data.

[0034] Specifically, first, configure a preset step size, where the preset step size refers to a standard time interval, which can be set according to the analysis accuracy requirement, wherein the higher the accuracy requirement, the shorter the preset step size; then divide the meter collection cycle according to the preset step size to obtain multiple meter collection intervals. Further, randomly select a first meter collection interval from the multiple meter collection intervals, and the first meter collection interval is any one of the multiple meter collection intervals; then, with the first meter collection interval as a constraint, query the target area environmental monitoring log to obtain a first sample monitoring data set, wherein the first sample monitoring data is marked with a time interval, and the time interval refers to the time interval between the collection time of the sample monitoring data and the current time node. The shorter the time interval, the higher the data credibility and adaptability.

[0035] The first sample monitoring data set is further selected according to preset electromagnetic interference factors, wherein the preset electromagnetic interference factors include at least wireless interference, radio frequency interference, electrical noise and mechanical interference, to obtain a first sample electromagnetic interference data set; on the other hand, a dynamic weight is configured based on the time interval, wherein the time interval is inversely proportional to the weight, and the shorter the time interval, the higher the credibility and adaptability of the characterization data, and the corresponding weight is greater; then the first sample electromagnetic interference data set is clustered according to the same interference type, and then based on the dynamic weight, the clustering result is weightedly calculated, and the first electromagnetic interference data is obtained according to the weighted calculation result, wherein the first electromagnetic interference data includes a first interference type set, a first interference intensity set and a first duration set, and the first interference type, the first interference intensity and the first duration have a corresponding relationship.

[0036] Finally, the same method as that for obtaining the first electromagnetic interference data is used to obtain multiple meter collection intervals and multiple electromagnetic interference data in turn, and based on the binary tree principle, the electromagnetic interference data set is constructed according to the mapping relationship between the meter collection intervals and the electromagnetic interference data.

[0037] Further, to obtain the environmental prediction data set, step 2 of this application includes:

[0038] A first sample monitoring data set is selected according to preset environmental interference factors to obtain a first sample environmental interference data set, wherein the preset environmental interference factors include at least weather type, environmental temperature, environmental humidity and lighting characteristics, and the first sample environmental interference data includes starting point environmental interference data and a mean value of environmental interference data; the starting point environmental interference data is used as input, and the mean value of environmental interference data is used as output, and supervised learning of the BP neural network is performed through the first sample environmental interference data set to obtain a first environmental perception model that meets convergence conditions; based on the weather forecast information of the target area, the first starting point environmental interference prediction data of the first electric meter collection interval is obtained, the first starting point environmental interference prediction data is analyzed through the first environmental perception model, and the first environmental interference prediction data mean is output; the environmental prediction data set is constructed based on the first electric meter collection interval and the first environmental interference prediction data mean.

[0039] Specifically, first, the first sample monitoring data set is selected according to preset environmental interference factors, wherein the preset environmental interference factors include at least weather type, ambient temperature, ambient humidity and lighting characteristics, to obtain a first sample environmental interference data set, wherein the first sample environmental interference data includes starting environmental interference data and an average of environmental interference data, the starting environmental interference data refers to the environmental data of the starting time node of the historical meter collection interval, and the average of environmental interference data refers to the average of environmental interference data of the historical meter collection interval.

[0040] Then, a first environmental perception model is constructed based on the BP neural network. The first environmental perception model is a BP neural network model that can be iteratively optimized in machine learning. It is obtained through supervised training with sample data, and includes an input layer, a hidden layer, and an output layer. The input data of the input layer is the starting environmental interference data, and the output data of the output layer is the mean of the environmental interference data. Then, the first sample environmental interference data set is used as sample training data to perform supervised training on the first environmental perception model. The training can be performed using a gradient descent algorithm to obtain a first environmental perception model that meets convergence conditions. The convergence conditions can be set according to actual needs, such as: meeting a preset number of training times or a preset accuracy index.

[0041] On the other hand, based on the weather forecast information of the target area, the first starting point environmental interference prediction data of the first meter collection interval is obtained, for example, it can be obtained through weather forecast and other means; then the first starting point environmental interference prediction data is analyzed by the first environmental perception model, and the first environmental interference prediction data mean is output; finally, multiple environmental interference prediction data means of multiple meter collection intervals are obtained in turn, and the environmental prediction data set is constructed based on the mapping relationship between the meter collection interval and the environmental interference prediction data mean.

[0042] Further, to obtain the communication period, step three of this application includes:

[0043] Retrieve the historical collection records of the concentrator to obtain a sample collection duration set; perform mean calculation on the sample collection duration set to determine the sample duration mean, and perform deviation calculation on the sample collection duration set based on the sample duration mean to obtain a sample duration deviation set; select the sample collection duration corresponding to the sample duration deviation that is less than a predetermined duration deviation threshold to construct a standard duration set, and set the mean of the standard duration set as the communication period.

[0044] Specifically, first, query the historical collection records of the concentrator, and extract the collection time from the historical collection records. The collection time refers to the time taken by the concentrator to complete the data collection of all electric energy meters. It is determined based on the processing capacity and communication quality of the concentrator to obtain a sample collection time set, wherein the sample collection time set includes multiple historical collection time sets. Then, the sample collection time set is averaged to determine the sample time mean, and the sample collection time set is further deviation-calculated based on the sample time mean to obtain a sample time deviation set. Then, the sample collection time corresponding to the sample time deviation that is less than a predetermined time deviation threshold is selected and set as the standard collection time to obtain a standard time set. By setting a predetermined time deviation threshold to screen the sample collection time, the influence of individual data can be avoided, and the accuracy and reliability of the standard time set setting can be improved. Finally, the mean of the standard time set is set as the communication period.

[0045] Further, to determine the optimal communication time period, step three of this application includes:

[0046] Obtain expected communication quality indicators, wherein the expected communication quality indicators include at least expected signal strength, expected signal-to-noise ratio, expected bit error rate, expected throughput, expected packet loss rate and expected transmission delay; enumerate the communication time periods within the meter collection cycle to determine a plurality of expected communication time periods; conditionally render the regional communication twin network according to the electromagnetic interference data set and the environmental prediction data set to generate a rendered communication twin network; subjecting the expected communication quality indicators to a constraint, perform simulated communications of the plurality of expected communication time periods through the rendered communication twin network to obtain a plurality of simulated communication results, wherein each simulated communication result includes simulated communication data of a plurality of electric energy meters; evaluate the plurality of simulated communication results based on a communication evaluation function, and select the expected communication time period corresponding to the maximum evaluation coefficient as the optimal communication time period.

[0047] Specifically, first, the expected communication quality indicators are obtained. The expected communication quality indicators can be set by technicians in this field according to actual needs, wherein the expected communication quality indicators include at least expected signal strength, expected signal-to-noise ratio, expected bit error rate, expected throughput, expected packet loss rate and expected transmission delay. Then, according to the communication time period, enumeration is performed within the meter acquisition cycle to determine multiple expected communication time periods. For example, the interval between adjacent communication time periods can be set to 1 minute for enumeration. On the other hand, the regional communication twin network is conditionally rendered according to the electromagnetic interference data set and the environmental prediction data set to generate a rendered communication twin network.

[0048] Taking the expected communication quality index as the constraint condition of the simulated communication, the simulated communication of the multiple expected communication time periods is performed through the rendering communication twin network to obtain multiple simulated communication results, wherein each simulated communication result includes the overall device power consumption, the overall data collection duration and the simulated communication data of multiple electric energy meters. Finally, the multiple simulated communication results are evaluated according to the communication evaluation function, and the expected communication time period corresponding to the maximum evaluation coefficient is selected as the optimal communication time period.

[0049] Further, the expression of the communication evaluation function is as follows:

[0050] ;in, is the evaluation coefficient of the ith simulation communication result, is the communication quality weight, is the device power consumption weight, is the communication duration weight, and Greater than Greater than , is the communication fitness of the ith simulated communication result, is the device power consumption of the ith simulated communication result, is the overall communication duration of the i-th simulated communication result; ; Where N is the number of electric energy meters, is the signal strength weight, is the signal-to-noise ratio weight, is the bit error rate weight, is the throughput weight, is the packet loss rate weight, is the transmission delay weight, is the analog signal strength of the nth electric energy meter, is the expected signal strength, is the analog signal-to-noise ratio of the nth electric energy meter, is the expected signal-to-noise ratio, is the expected bit error rate, is the simulated bit error rate of the nth electric energy meter, is the simulated throughput of the nth electric energy meter, is the expected throughput, is the expected packet loss rate, is the simulated packet loss rate of the nth electric energy meter, is the expected transmission delay, is the simulated transmission delay of the nth electric energy meter.

[0051] Specifically, in the communication evaluation function, is the evaluation coefficient of the i-th simulation communication result. The larger the evaluation coefficient is, the better the simulation communication effect is. is the communication quality weight, is the device power consumption weight, is the communication duration weight, and Greater than Greater than , is the communication fitness of the ith simulated communication result, is the device power consumption of the ith simulated communication result, is the overall communication duration of the i-th simulated communication result.

[0052] In the communication fitness calculation function, N is the number of electric energy meters, is the signal strength weight, is the signal-to-noise ratio weight, is the bit error rate weight, is the throughput weight, is the packet loss rate weight, is the transmission delay weight, is the analog signal strength of the nth electric energy meter, is the expected signal strength, is the analog signal-to-noise ratio of the nth electric energy meter, is the expected signal-to-noise ratio, is the expected bit error rate, is the simulated bit error rate of the nth electric energy meter, is the simulated throughput of the nth electric energy meter, is the expected throughput, is the expected packet loss rate, is the simulated packet loss rate of the nth electric energy meter, is the expected transmission delay, is the simulated transmission delay of the nth electric energy meter.

[0053] Further, a power adjustment strategy is generated. Step 4 of this application includes:

[0054] Taking the concentrator attribute data, the energy meter attribute data and the expected communication quality as constraints, a sample communication data set is retrieved online to obtain the sample communication data set, wherein the sample communication data includes a sample communication quality deviation and a sample adjustment power; the sample communication data set is simulated and screened through the rendering communication twin network to obtain a qualified sample communication data set; the qualified sample communication data set is divided into two equal parts, and the BP neural network and the random decision forest are supervised and cross-validated respectively to obtain a first convergence power adjustment unit and a second convergence power adjustment unit; the power adjustment strategy is generated based on the first convergence power adjustment unit and the second convergence power adjustment unit, wherein the output of the power adjustment strategy is the mean of the output results of the first convergence power adjustment unit and the second convergence power adjustment unit.

[0055] Specifically, the concentrator attribute data, the energy meter attribute data and the expected communication quality are used as retrieval constraints, and homologous information retrieval is performed based on the power big data network to obtain a sample communication data set, wherein the sample communication data includes a sample communication quality deviation and a sample adjustment power. Then, the sample communication data set is simulated and screened through the rendering communication twin network, that is, whether the sample adjustment power can meet the corresponding sample communication quality deviation, and the sample communication data that can meet the corresponding sample communication quality deviation is set as qualified sample communication data to obtain a qualified sample communication data set. By using the rendering communication twin network to screen the sample communication data set, the accuracy and reliability of the setting of the qualified sample communication data set can be improved, thereby improving the accuracy of subsequent model training.

[0056] BP neural network and random decision forest are two commonly used machine learning algorithms, which are widely used in data analysis and predictive modeling. BP neural network is a multi-layer feedforward neural network, which is trained by back propagation algorithm and minimizes prediction error by adjusting weights and biases in the network; random decision forest is an integrated learning algorithm composed of multiple decision trees, each decision tree is constructed by randomly selected features and node splitting, wherein each decision tree predicts the data set, and then the final result is obtained by voting or taking the average. Then the qualified sample communication data set is divided into two parts to obtain a first data set and a second data set, and the first data set and the second data set are used to supervise the BP neural network and the random decision forest respectively. After meeting the predetermined number of training times, the data sets are exchanged for verification training to obtain the first convergence power adjustment unit and the second convergence power adjustment unit that meet the predetermined convergence conditions.

[0057] Finally, the power adjustment strategy is generated based on the first convergence power adjustment unit and the second convergence power adjustment unit, wherein the output of the power adjustment strategy is the average of the output results of the first convergence power adjustment unit and the second convergence power adjustment unit. By constructing a power adjustment strategy based on the fusion of BP neural network and random decision forest, the advantages of multiple models can be combined to reduce the prediction error of a single model, thereby improving the accuracy and reliability of power adjustment.

[0058] Further, the concentrator is subjected to fixed-point feedback control according to the power adjustment strategy, and step five of the present application includes:

[0059] At a predetermined node, real-time communication quality data is monitored and acquired, deviation calculation is performed on the real-time communication quality data based on the expected communication quality, and the real-time communication deviation is determined; the real-time communication deviation is analyzed through the power adjustment strategy, the real-time adjustment power is output, and feedback compensation is performed on the current power of the concentrator based on the real-time adjustment power.

[0060] Specifically, first, at a predetermined node, the real-time communication quality data between the concentrator and the electric energy meter is monitored and acquired, wherein the predetermined node can be set according to the feedback control accuracy, and the higher the control accuracy requirement, the shorter the interval of the predetermined node; then, the deviation of the real-time communication quality data is calculated based on the expected communication quality to determine the real-time communication deviation. The real-time communication deviation is further input into the power adjustment strategy for analysis, and the real-time adjustment power is output; finally, the current power of the concentrator is feedback compensated according to the real-time adjustment power. The above method can be used to timely feedback and control the output power of the concentrator, thereby improving the adaptability of the output power and the communication quality.

[0061] In summary, the concentrator adaptive electric energy meter data transmission method provided by the present application has the following technical effects:

[0062] 1. By conducting electromagnetic interference analysis and environmental perception prediction on the target area, the communication period with the least interference is selected for data transmission, which can reduce the interference of the external environment on the communication quality; improve the accuracy of the communication period setting; then during the data transmission process, the concentrator output power is timely fed back and regulated through the pre-built power adjustment strategy, which can improve the adaptability of the output power and communication quality, thereby achieving the technical effect of significantly improving the data transmission quality and efficiency and reducing the power consumption of the equipment.

[0063] 2. Based on the cloud-edge collaborative method, the communication period optimization and power correction analysis are carried out in the cloud system, and the optimal communication period and power adjustment strategy are delegated to the edge computing unit of the concentrator. This can reduce the data processing consumption inside the concentrator and reduce operating costs, while improving the accuracy and timeliness of the concentrator power feedback control.

[0064] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0065] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A concentrator adaptive electric energy meter data transmission method, characterized in that: include: In the cloud system, a regional communication twin network is constructed based on the concentrator attribute data and electric energy meter attribute data of the target area; Based on the energy meter collection cycle, data is collected on the target area according to the preset electromagnetic interference factor to obtain an electromagnetic interference data set, and environmental perception is performed on the target area according to the preset environmental interference factor to obtain an environmental prediction data set; Based on the regional communication twin network, the electromagnetic interference data set and the environmental prediction data set, with the expected communication quality as a constraint and the device power consumption as a purpose, multiple preset communication time periods within the electric energy meter collection cycle are optimized to determine the optimal communication time period; Retrieving and acquiring a sample communication data set based on the concentrator attribute data, the electric energy meter attribute data and the expected communication quality, performing power correction analysis based on the sample communication data set, and generating a power adjustment strategy; The optimal communication period and power adjustment strategy are delegated to the edge computing unit of the concentrator, the concentrator is controlled to collect electric energy meter data based on the optimal communication period, and during the collection process, the concentrator is controlled by fixed-point feedback according to the power adjustment strategy; Generate power adjustment strategies, including: Taking the concentrator attribute data, the electric energy meter attribute data and the expected communication quality as constraints, acquiring a sample communication data set through online retrieval, wherein the sample communication data includes a sample communication quality deviation and a sample adjustment power; Simulating and screening the sample communication data set by rendering the communication twin network to obtain a qualified sample communication data set; The qualified sample communication data set is equally divided into two parts, and supervised learning and cross-validation are performed on the BP neural network and the random decision forest respectively to obtain a first convergence power adjustment unit and a second convergence power adjustment unit; Generating the power adjustment strategy based on the first convergence power adjustment unit and the second convergence power adjustment unit, wherein the output of the power adjustment strategy is the average of the output results of the first convergence power adjustment unit and the second convergence power adjustment unit; The concentrator is controlled by fixed-point feedback according to the power adjustment strategy, including: At a predetermined node, monitoring and acquiring real-time communication quality data, performing deviation calculation on the real-time communication quality data based on the expected communication quality, and determining a real-time communication deviation; The real-time communication deviation is analyzed through the power adjustment strategy, the real-time adjustment power is output, and feedback compensation is performed on the current power of the concentrator based on the real-time adjustment power.

2. According to claim 1, a concentrator adaptive electric energy meter data transmission method is characterized in that: Get the electromagnetic interference data set, including: Divide the electric energy meter collection period according to a preset step length to obtain a plurality of electric energy meter collection intervals, randomly select a first electric energy meter collection interval, query the target area environment monitoring log with the first electric energy meter collection interval as a constraint, and obtain a first sample monitoring data set, wherein the first sample monitoring data is marked with a time interval; The first sample monitoring data set is selected according to a preset electromagnetic interference factor to obtain a first sample electromagnetic interference data set, wherein the preset electromagnetic interference factor includes at least wireless interference, radio frequency interference, electrical noise and mechanical interference; Based on the time interval, dynamic weights are configured, weighted fusion is performed on the first sample electromagnetic interference data set, and first electromagnetic interference data is obtained based on the weighted fusion result analysis, wherein the first electromagnetic interference data includes a first interference type set, a first interference intensity set, and a first duration set; The electromagnetic interference data set is constructed based on the first electric energy meter collection interval and the first electromagnetic interference data.

3. According to claim 2, a concentrator adaptive electric energy meter data transmission method is characterized in that: Get environmental prediction datasets, including: The first sample monitoring data set is selected according to the preset environmental interference factors to obtain a first sample environmental interference data set, wherein the preset environmental interference factors at least include weather type, ambient temperature, ambient humidity and light characteristics, and the first sample environmental interference data includes starting point environmental interference data and an average value of the environmental interference data; Taking the starting environmental interference data as input and the mean of the environmental interference data as output, the BP neural network is supervised and learned through the first sample environmental interference data set to obtain a first environmental perception model that meets the convergence condition; Obtaining first starting point environmental interference prediction data of the first electric energy meter collection interval based on weather forecast information of the target area, analyzing the first starting point environmental interference prediction data through the first environmental perception model, and outputting a mean value of the first environmental interference prediction data; The environmental prediction data set is constructed based on the first electric energy meter collection interval and the first environmental interference prediction data mean.

4. According to claim 1, a concentrator adaptive electric energy meter data transmission method is characterized in that: Determine the optimal communication time, including: Obtaining expected communication quality indicators, wherein the expected communication quality indicators include at least expected signal strength, expected signal-to-noise ratio, expected bit error rate, expected throughput, expected packet loss rate, and expected transmission delay; Determine a plurality of expected communication time periods by enumerating the communication time periods within the electric energy meter acquisition cycle; Conditionally rendering the regional communication twin network according to the electromagnetic interference data set and the environmental prediction data set to generate a rendered communication twin network; Taking the expected communication quality index as a constraint, performing simulated communications of the multiple expected communication time periods through the rendering communication twin network to obtain multiple simulated communication results, wherein each simulated communication result includes simulated communication data of multiple electric energy meters; The plurality of simulated communication results are evaluated based on a communication evaluation function, and an expected communication period corresponding to a maximum evaluation coefficient is selected and set as the optimal communication period.

5. A concentrator adaptive electric energy meter data transmission method according to claim 4, characterized in that: Get the communication period, including: Retrieve the historical collection records of the concentrator and obtain the sample collection duration set; Performing mean calculation on the sample collection duration set to determine the sample duration mean, and performing deviation calculation on the sample collection duration set based on the sample duration mean to obtain a sample duration deviation set; The sample collection durations corresponding to the sample duration deviations that are smaller than a predetermined duration deviation threshold are selected to construct a standard duration set, and the mean of the standard duration set is set as the communication period.

6. A concentrator adaptive electric energy meter data transmission method according to claim 4, characterized in that: The expression of the communication evaluation function is: ； in, is the evaluation coefficient of the ith simulation communication result, is the communication quality weight, is the device power consumption weight, is the communication duration weight, and Greater than Greater than , is the communication fitness of the ith simulated communication result, is the device power consumption of the ith simulated communication result, is the overall communication duration of the i-th simulated communication result; ； Where N is the number of electric energy meters, is the signal strength weight, is the signal-to-noise ratio weight, is the bit error rate weight, is the throughput weight, is the packet loss rate weight, is the transmission delay weight, is the analog signal strength of the nth electric energy meter, is the expected signal strength, is the analog signal-to-noise ratio of the nth electric energy meter, is the expected signal-to-noise ratio, is the expected bit error rate, is the simulated bit error rate of the nth electric energy meter, is the simulated throughput of the nth electric energy meter, is the expected throughput, is the expected packet loss rate, is the simulated packet loss rate of the nth electric energy meter, is the expected transmission delay, is the simulated transmission delay of the nth electric energy meter, and P is the communication fitness calculation function.

Images