Virtual power plant data monitoring method and system based on power line carrier

Abstract

The invention relates to the technical field of power systems, and discloses a virtual power plant data monitoring method and system based on a power carrier, and the method comprises the steps: deploying monitoring and management equipment in DERs; using monitoring and management equipment to collect waveform data and operation data, and performing preprocessing; constructing a data monitoring instruction prediction model according to the waveform data and historical data of the operation data, and applying the data monitoring instruction prediction model to the preprocessed waveform data to obtain a prediction data monitoring instruction; based on the prediction data monitoring instruction and the preprocessed operation data, configuring and generating a monitoring configuration instruction; executing the monitoring configuration instruction, and collecting real-time operation state data; using the real-time operation state data to iteratively update the data monitoring instruction prediction model through a federated learning algorithm; according to the invention, local training of each edge node based on the local real-time operation state data is realized, and the beneficial effects of improving the overall performance and protecting the privacy of the user are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a virtual power plant data monitoring method and system based on power carrier. Background Art

[0002] With the development of smart grids, distributed energy resources (DERs) are increasingly used in power systems. Traditional power systems mainly rely on centralized power plants, while modern virtual power plants (VPPs) achieve more flexible and efficient energy management by integrating multiple decentralized DERs, such as solar photovoltaic panels, wind turbines, and energy storage devices. The core advantage of VPP is that it can aggregate these distributed resources and participate in the dispatch and operation of the power market as a whole, thereby improving the stability and reliability of the system. However, in order to achieve this goal, it is necessary to have strong data monitoring capabilities to grasp the status of each DER in real time and make corresponding adjustments.

[0003] In recent years, data monitoring technology based on power line communication (PLC) has developed rapidly. PLC uses existing power lines as transmission media, reducing the cost of additional wiring and maintaining stable communication performance in complex electromagnetic environments. In addition, the application of edge computing nodes, smart meters, environmental sensors, and communication gateway devices has made data collection and processing more efficient. In particular, edge computing technology allows data to be initially analyzed close to the source, alleviating the pressure on the central server and reducing latency. Despite this, existing technologies still have some shortcomings, especially in terms of data processing capabilities and model update mechanisms when faced with a large number of heterogeneous DERs. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a virtual power plant data monitoring method based on power carrier to solve the problems of imprecise data preprocessing and lack of dynamic adjustment capability of the model.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a virtual power plant data monitoring method based on power carrier, which includes deploying monitoring and management equipment in DERs; using the monitoring and management equipment to collect waveform data and operation data, and preprocessing them; constructing a data monitoring instruction prediction model based on historical data of waveform data and operation data, and applying the data monitoring instruction prediction model to the preprocessed waveform data to obtain predicted data monitoring instructions; configuring and generating monitoring configuration instructions based on the predicted data monitoring instructions and the preprocessed operation data; executing the monitoring configuration instructions and collecting real-time operation status data; and using the real-time operation status data to iteratively update the data monitoring instruction prediction model through a federated learning algorithm.

[0007] As a preferred solution of the virtual power plant data monitoring method based on power carrier described in the present invention, the monitoring and management equipment is deployed in DERs, and the specific steps are as follows: Edge computing nodes, smart meters, PLC modems, environmental sensors, security modules, BMS, inverters, and communication gateways are installed and configured according to the locations of DERs and power networks.

[0008] As a preferred solution of the virtual power plant data monitoring method based on power carrier described in the present invention, wherein: the waveform data and operation data are collected by using monitoring and management equipment, and pre-processed, the specific steps are as follows: Use PLC modems to collect waveform data, use smart meters, BMS and inverters to collect operating data, and use MQTT for encrypted transmission through security modules; Edge computing nodes are used to remove outliers and unify timestamps of waveform data and operation data. Minimum and maximum normalization is applied to waveform data for standardization, and Z-score is used to standardize operation data.

[0009] As a preferred solution of the virtual power plant data monitoring method based on power carrier described in the present invention, wherein: the data monitoring instruction prediction model is constructed according to the historical data of waveform data and operation data, and the data monitoring instruction prediction model is applied to the preprocessed waveform data to obtain the predicted data monitoring instruction. The specific steps are as follows: Extract power line noise samples and the corresponding optimal monitoring instruction combination from the historical data of waveform data and operation data, divide the power line noise samples and the corresponding optimal monitoring instruction combination into a training set and a validation set, use the training set to train the random forest regression model, use the validation set to validate the trained random forest regression model, and build a data monitoring instruction prediction model; Based on the preprocessed waveform data, a data monitoring instruction prediction model is used to obtain predicted data monitoring instructions.

[0010] As a preferred solution of the virtual power plant data monitoring method based on power carrier of the present invention, wherein: the monitoring configuration instruction is configured based on the predicted data monitoring instruction and the pre-processed operation data, and the specific steps are as follows: Extract parameter suggestions from the prediction data monitoring instructions, introduce task instruction functions based on the prediction data monitoring instructions and preprocessed operation data, and configure and generate monitoring configuration instructions; Construct monitoring configuration instructions in JSON format according to the monitoring configuration instructions.

[0011] As a preferred solution of the virtual power plant data monitoring method based on power carrier of the present invention, the specific steps of executing monitoring configuration instructions and collecting real-time operation status data are as follows: According to the monitoring configuration instructions in the constructed JSON format, the communication gateway is used to send update commands to all monitoring and management devices through MQTT, and the new monitoring configuration is applied to collect real-time operation status data.

[0012] As a preferred solution of the virtual power plant data monitoring method based on power carrier described in the present invention, wherein: the real-time operation status data is used to iteratively update the data monitoring instruction prediction model through the federated learning algorithm, and the specific steps are as follows: Send the current global monitoring parameters to all edge nodes, and instruct each node to start training based on the existing global monitoring parameters using real-time operating status data, and use edge nodes to record the loss value and accuracy during training; Based on the loss value and accuracy during training, a training quality score is obtained through a training quality score function; Define training quality thresholds based on training quality scores , for training quality scores greater than The global monitoring parameters are updated for the training results with a training quality score less than or equal to the training quality threshold, and the training rounds are increased for the training results.

[0013] In the second aspect, the present invention provides a virtual power plant data monitoring system based on power carrier, including an equipment deployment module, a data acquisition and preprocessing module, a data monitoring instruction prediction model construction and application module, a monitoring instruction configuration module, an instruction execution and data collection module and a federated learning algorithm iterative update module; the equipment deployment module is used to deploy monitoring and management equipment in DERs; the data acquisition and preprocessing module is used to use the monitoring and management equipment to collect waveform data and operation data, and perform preprocessing; the data monitoring instruction prediction model construction and application module is used to construct a data monitoring instruction prediction model according to the historical data of waveform data and operation data, and apply the data monitoring instruction prediction model to the preprocessed waveform data to obtain predicted data monitoring instructions; the monitoring instruction configuration module is used to generate monitoring configuration instructions based on the predicted data monitoring instructions and the preprocessed operation data; the instruction execution and data collection module is used to execute the monitoring configuration instructions and collect real-time operation status data; the federated learning algorithm iterative update module is used to use the real-time operation status data to iteratively optimize the global monitoring parameters through the federated learning algorithm.

[0014] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the virtual power plant data monitoring method based on power carrier as described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the virtual power plant data monitoring method based on power carrier as described in the first aspect of the present invention.

[0016] The beneficial effects of the present invention are as follows: through the construction and application of the data monitoring instruction prediction model, intelligent analysis of the preprocessed waveform data is realized, and then a data monitoring instruction prediction model that can be dynamically adjusted is trained based on the historical data of the waveform data and the operation data, and finally the beneficial effect of improving the monitoring accuracy and adaptability is achieved. Through the iterative update mechanism of the federated learning algorithm, each edge node is locally trained based on the local real-time operation status data, and a new global optimization parameter is generated through a comprehensive aggregation formula, and finally the beneficial effect of improving the overall performance and protecting the user's privacy is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 This is a flow chart of the virtual power plant data monitoring method based on power carrier in Example 1.

[0019] Figure 2 This is a schematic diagram of a virtual power plant data monitoring system based on power carrier in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a virtual power plant data monitoring method based on power carrier, comprising the following steps: S1: Deploy monitoring and management equipment in DERs; Edge computing nodes, smart meters, PLC modems, environmental sensors, security modules, BMS, inverters, and communication gateways are installed and configured according to the locations of DERs and power networks.

[0022] It is also important to note that the specific type of DERs should be considered when installing monitoring and management equipment to ensure that each device is in the optimal location for best performance. For example, edge computing nodes should be close to data sources to reduce latency, while communication gateways need to have a bandwidth of at least 100 Mbps and a packet loss rate of less than 1% to support real-time data transmission.

[0023] S2: Use monitoring and management equipment to collect waveform data and operation data, and perform preprocessing; Use PLC modems to collect waveform data, use smart meters, BMS and inverters to collect operating data, and use MQTT for encrypted transmission through security modules; Edge computing nodes are used to remove outliers and unify timestamps of waveform data and operation data. Minimum and maximum normalization is applied to waveform data for standardization, and Z-score is used to standardize operation data.

[0024] It should also be noted that: encrypted transmission ensures data security and prevents information leakage and tampering; In order to ensure the consistency and accuracy of the data, the clock synchronization problem between different devices needs to be considered during the timestamp unification process. This can be solved through NTP (Network Time Protocol) or other synchronization mechanisms. The use of minimum and maximum normalization and Z-score for standardization not only improves the quality of the data, but also provides a better foundation for subsequent data analysis and model training.

[0025] S3: constructing a data monitoring instruction prediction model based on the historical data of the waveform data and the operation data, and applying the data monitoring instruction prediction model to the preprocessed waveform data to obtain a predicted data monitoring instruction; Extract power line noise samples and the corresponding optimal monitoring instruction combination from the historical data of waveform data and operation data, divide the power line noise samples and the corresponding optimal monitoring instruction combination into a training set and a validation set, use the training set to train the random forest regression model, use the validation set to validate the trained random forest regression model, and build a data monitoring instruction prediction model; Based on the preprocessed waveform data, the data monitoring instruction prediction model is used to obtain the predicted data monitoring instruction, which is expressed as: ; in, For the Predictive data monitoring instructions at each DER, is the number of decision trees, For the A decision tree for the input feature vector The predicted output is is the index of the decision tree, The index of DER.

[0026] It should also be noted that the random forest regression model was chosen because it can perform well when processing high-dimensional data and has strong resistance to overfitting. The random forest regression model can also handle nonlinear relationships and has a certain tolerance for missing values ​​and outliers, thereby improving the reliability of the prediction results. Eigenvector is a vector of voltage and current values ​​of the preprocessed waveform data.

[0027] S4: Based on the predicted data monitoring instructions and the preprocessed operation data, configure and generate monitoring configuration instructions; Extract parameter suggestions from the prediction data monitoring instructions. Based on the prediction data monitoring instructions and preprocessed operation data, introduce the task instruction function and configure the generated monitoring configuration instructions. The expression is: ; in, For the Data monitoring instructions at each DER, For the Predictive data monitoring instructions at each DER, For the The pre-processed running data at DER, For the Configure the conflict index at each DER. The weighting coefficient for predicting data monitoring instructions is set to 0.6. is the weighting coefficient of the preprocessed running data, set to 0.4, To configure the weighting coefficient of the conflict index, set it to 0.2; Construct monitoring configuration instructions in JSON format according to the monitoring configuration instructions.

[0028] It should also be noted that the task instruction function comprehensively considers the impact of the predicted data monitoring instructions and the actual operation data, and introduces the configuration conflict index to ensure the consistency and stability between the new and old configurations. The expression is: ; in, is the quantized value of the parameter range conflict, is the quantitative value of the logical conflict, is the quantitative value of time conflict, is the weight coefficient of the quantized value of the parameter range conflict, set to 0.5, is the weight coefficient of the quantization value of the logical conflict, set to 0.3, The weight coefficient of the quantized value of time conflict is set to 0.2. For parameter range conflict, a quantitative indicator is defined to indicate the degree of conflict. If a parameter in the new configuration exceeds the allowed range, the conflict value of the parameter is set to 1; otherwise, it is 0. For logical conflict, a Boolean function is used to evaluate the logical consistency between the new and old configurations. The result is 1 for conflict and 0 for no conflict. For time conflict, check whether the time schedule of the new configuration overlaps with other tasks. If there is overlap, the conflict value is 1, otherwise it is 0. In this way, not only can the accuracy of monitoring configuration be improved, but also failures caused by configuration conflicts can be avoided. Monitoring configuration instructions in JSON format are easy to parse and execute, and are suitable for various types of monitoring and management devices.

[0029] S5: Execute monitoring configuration instructions and collect real-time operation status data; According to the monitoring configuration instructions in the constructed JSON format, the communication gateway is used to send update commands to all monitoring and management devices through MQTT, and the new monitoring configuration is applied to collect real-time operation status data.

[0030] It should also be noted that when sending update commands through the communication gateway, the timeliness and accuracy of the commands must be ensured so that all devices can respond quickly and apply the new monitoring configuration. The collected real-time operating status data should be uploaded to the coordination server in a timely manner for further analysis to ensure the real-time and effectiveness of the real-time operating status data.

[0031] S6: Use the real-time operation status data to iteratively update the data monitoring instruction prediction model through the federated learning algorithm.

[0032] Send the current global monitoring parameters to all edge nodes, and instruct each node to start training based on the existing global monitoring parameters using real-time operating status data, and use edge nodes to record the loss value and accuracy during training; Based on the loss value and accuracy during training, the training quality score is obtained through the training quality score function, which is expressed as: ; in, For the The training quality score of edge nodes, For the The average loss value of edge nodes during training, For the The average accuracy of edge nodes during training, For the The variance of parameter changes during training of edge nodes, is the weighting coefficient of the average loss value during training, set to 0.7, For the The weighting coefficient of the average accuracy of edge nodes during training is set to 0.3. For the The weighting coefficient of the variance of parameter changes during training of edge nodes is set to 0.1; Define training quality thresholds based on training quality scores , for training quality scores greater than The global monitoring parameters are updated for the training results with a training quality score less than or equal to the training quality threshold, and the training rounds are increased for the training results.

[0033] It should also be noted that: the current global monitoring parameters are sent to all edge nodes through the coordination server, and each node is instructed to start training based on the existing global monitoring parameters using the real-time operating status data; The federated learning algorithm mechanism allows the global model to be jointly optimized without sharing the original data, thereby protecting user privacy and data security; The training quality threshold The value of is set to 0.85 to ensure that only high-quality training results are included in the update, thereby ensuring the stability and high performance of the monitoring and management process. During operation, this threshold can be dynamically adjusted by monitoring the overall performance (such as prediction accuracy and response time) to find the optimal balance.

[0034] The present embodiment also provides a virtual power plant data monitoring system based on power carrier, including: an equipment deployment module, a data acquisition and preprocessing module, a data monitoring instruction prediction model construction and application module, a monitoring instruction configuration module, an instruction execution and data collection module and a federated learning algorithm iterative update module; the equipment deployment module is used to deploy monitoring and management equipment in DERs; the data acquisition and preprocessing module is used to use the monitoring and management equipment to collect waveform data and operation data, and perform preprocessing; the data monitoring instruction prediction model construction and application module is used to construct a data monitoring instruction prediction model according to the historical data of waveform data and operation data, and apply the data monitoring instruction prediction model to the preprocessed waveform data to obtain predicted data monitoring instructions; the monitoring instruction configuration module is used to generate monitoring configuration instructions based on the predicted data monitoring instructions and the preprocessed operation data; the instruction execution and data collection module is used to execute the monitoring configuration instructions and collect real-time operation status data; the federated learning algorithm iterative update module is used to use the real-time operation status data to iteratively optimize the global monitoring parameters through the federated learning algorithm.

[0035] This embodiment also provides a computer device, which is suitable for the case of a virtual power plant data monitoring method based on power carrier, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the virtual power plant data monitoring method based on power carrier proposed in the above embodiment.

[0036] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0037] In summary, the present invention achieves intelligent analysis of preprocessed waveform data through the construction and application of a data monitoring instruction prediction model, and then trains a data monitoring instruction prediction model that can be dynamically adjusted based on the historical data of waveform data and operation data, and finally achieves the beneficial effect of improving monitoring accuracy and adaptability. Through the iterative update mechanism of the federated learning algorithm, each edge node performs local training based on local real-time operation status data, and generates new global optimization parameters through a comprehensive aggregation formula, and finally achieves the beneficial effect of improving overall performance and protecting user privacy.

[0038] It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A virtual power plant data monitoring method based on power carrier, characterized in that: include, Deploy monitoring and management equipment on DERs; Use monitoring and management equipment to collect waveform data and operation data, and perform pre-processing; Building a data monitoring instruction prediction model based on the historical data of the waveform data and the operation data, and applying the data monitoring instruction prediction model to the preprocessed waveform data to obtain a predicted data monitoring instruction; Generate monitoring configuration instructions based on the predicted data monitoring instructions and the preprocessed operation data; Execute monitoring configuration instructions and collect real-time operation status data; Use real-time operating status data to iteratively update the data monitoring instruction prediction model through a federated learning algorithm.

2. The method for monitoring data of a virtual power plant based on power carrier according to claim 1, characterized in that: The specific steps for deploying monitoring and management equipment on DERs are as follows: Set up edge computing nodes, smart meters, PLC modems, environmental sensors, security modules, BMS, inverters, and communication gateways and configure parameters according to the locations of DERs and power networks.

3. The method for monitoring data of a virtual power plant based on power carrier according to claim 2, characterized in that: The monitoring and management equipment is used to collect waveform data and operation data, and pre-process them. The specific steps are as follows: Use PLC modems to collect waveform data, use smart meters, BMS and inverters to collect operating data, and use MQTT for encrypted transmission through security modules; Edge computing nodes are used to remove outliers and unify timestamps of waveform data and operation data. Minimum and maximum normalization is applied to waveform data for standardization, and Z-score is used to standardize operation data.

4. The method for monitoring data of a virtual power plant based on power carrier according to claim 3, characterized in that: The data monitoring instruction prediction model is constructed based on the historical data of the waveform data and the operation data, and the data monitoring instruction prediction model is applied to the preprocessed waveform data to obtain the predicted data monitoring instruction. The specific steps are as follows: Extract power line noise samples and the corresponding optimal monitoring instruction combination from the historical data of waveform data and operation data, divide the power line noise samples and the corresponding optimal monitoring instruction combination into a training set and a validation set, use the training set to train the random forest regression model, use the validation set to validate the trained random forest regression model, and build a data monitoring instruction prediction model; Based on the preprocessed waveform data, a data monitoring instruction prediction model is used to obtain predicted data monitoring instructions.

5. The method for monitoring data of a virtual power plant based on power carrier according to claim 4, characterized in that: The monitoring configuration instructions based on the predicted data monitoring instructions and the pre-processed operation data are configured to generate the monitoring configuration instructions. The specific steps are as follows: Extract parameter suggestions from the prediction data monitoring instructions, and obtain monitoring configuration instructions through task instruction functions based on the prediction data monitoring instructions and preprocessed operation data and parameter suggestions; Construct monitoring configuration instructions in JSON format according to the monitoring configuration instructions.

6. The method for monitoring data of a virtual power plant based on power carrier according to claim 5, characterized in that: The execution monitoring configuration instructions collect real-time operation status data. The specific steps are as follows: Use the communication gateway to send monitoring configuration instructions in JSON format to all monitoring and management devices through MQTT, and apply the monitoring configuration instructions to collect real-time operation status data.

7. The method for monitoring data of a virtual power plant based on power carrier according to claim 6, characterized in that: The specific steps of using real-time operation status data to iteratively update the data monitoring instruction prediction model through the federated learning algorithm are as follows: Send the current global monitoring parameters to all edge nodes, and instruct each node to start training based on the existing global monitoring parameters using real-time operating status data, and use edge nodes to record the loss value and accuracy during training; Based on the loss value and accuracy during training, a training quality score is obtained through a training quality score function; A training quality threshold is obtained based on the training quality score. For training results whose training quality scores are greater than the training quality threshold, global monitoring parameters are updated. For training results whose training quality scores are less than or equal to the training quality threshold, training rounds are increased.

8. A virtual power plant data monitoring system based on power carrier, based on the virtual power plant data monitoring method based on power carrier according to any one of claims 1 to 7, characterized in that: It includes equipment deployment module, data collection and preprocessing module, data monitoring instruction prediction model construction and application module, monitoring instruction configuration module, instruction execution and data collection module and federated learning algorithm iterative update module; Equipment deployment module, used to deploy monitoring and management equipment in DERs; The data acquisition and preprocessing module is used to collect waveform data and operation data using monitoring and management equipment, and perform preprocessing; A data monitoring instruction prediction model construction and application module is used to construct a data monitoring instruction prediction model based on the historical data of waveform data and operation data, and apply the data monitoring instruction prediction model to the preprocessed waveform data to obtain the predicted data monitoring instruction; A monitoring instruction configuration module, used for generating monitoring configuration instructions based on the predicted data monitoring instructions and the preprocessed operation data; Instruction execution and data collection module, used to execute data monitoring instructions and collect real-time operation status data; The federated learning algorithm iterative update module is used to iteratively optimize global monitoring parameters through the federated learning algorithm using real-time operating status data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the power carrier-based virtual power plant data monitoring method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the power carrier-based virtual power plant data monitoring method described in any one of claims 1 to 7 are implemented.

Images