An energy management and control method, system, device and medium based on a virtual power plant

Abstract

The application discloses an energy management and control method, system, device and medium based on a virtual power plant, comprising the following steps: S1, parameter acquisition: based on a standardized interface protocol, connecting with each branch load monitoring device of a power distribution network, collecting multi-dimensional operation data of each branch load, wherein the multi-dimensional operation data comprises electrical parameters, environmental parameters and equipment state parameters; S2, data preprocessing: preprocessing the collected multi-dimensional operation data; S3, identifying the load abnormal type and level: constructing a monitoring and evaluation model based on the branch load characteristics, inputting the standardized monitoring data, real-time evaluating the operation state of each branch load, identifying the load abnormal type and level; S4: for the identified load abnormality, generating a fault positioning result and a control suggestion in combination with the power distribution network topology structure. The application realizes the interconnection and intercommunication of multi-source branch load data: the application adopts an IEC61850 standardized interface protocol and combines the protocol conversion function of an interface adaptation sub-module.

Description

Technical Field

[0001] This invention relates to the field of power, and more specifically to an energy management method, system, equipment, and medium based on a virtual power plant. Background Technology

[0002] Virtual power plants (VPS) aggregate diverse energy resources such as distributed power sources, energy storage devices, and controllable loads to achieve coordinated operation with the main power grid. This is a core technological means to enhance the absorption capacity of new energy sources and ensure the safety and stability of the power system. With the acceleration of energy transition, the types of resources aggregated by VPS are becoming increasingly diverse, and the distribution of branch loads is becoming more dispersed. Inconsistent interface protocols and data formats among energy equipment from different manufacturers lead to widespread problems in energy management, including "data silos," "inefficient collaboration," and "insufficient control precision," severely restricting the operational efficiency and control capabilities of VPS.

[0003] Existing energy management technologies based on virtual power plants suffer from several critical shortcomings: First, poor interoperability of multi-source load data. Traditional management systems lack standardized interface adaptation mechanisms, making them incompatible with load monitoring devices and energy equipment from different manufacturers. This hinders the effective integration of load data and distributed power output data, making it impossible to comprehensively grasp the overall energy operation status of the virtual power plant. Second, insufficient data processing accuracy. The complex operating environment of virtual power plants makes collected data susceptible to electromagnetic interference and equipment fluctuations, resulting in outliers and noise. Traditional simple filtering methods are insufficient for accurate removal, leading to distorted subsequent management decisions. Third, anomaly identification... The problems include: low efficiency in fault handling, the existing control model not being fully adapted to the operating characteristics of different types of branch loads, low accuracy in anomaly identification, and vague fault location, making it impossible to quickly pinpoint the fault branch and root cause, thus prolonging fault handling time; fourth, the disconnect between control and dispatch, the lack of effective linkage between virtual power plant energy control data and the power grid dispatch system, making it difficult to generate accurate control suggestions, resulting in insufficient consumption of new energy and excessive operating losses in the distribution network; fifth, weak data security and traceability, the lack of encryption protection and efficient storage mechanisms for massive energy control data, the ease of data leakage or loss, and the inability to achieve full life cycle traceability, which is not conducive to fault review and operation optimization.

[0004] Therefore, developing a virtual power plant-based energy management method, system, equipment, and medium that can achieve interconnection and interoperability of multi-source branch load data, improve data processing accuracy and anomaly handling efficiency, enhance control-dispatch coordination capabilities, and ensure data security has become an urgent need to promote the large-scale development of virtual power plants and improve the level of energy management. Summary of the Invention

[0005] In view of the problems mentioned in the background art, the purpose of this invention is to provide an energy management method, system, equipment and medium based on a virtual power plant to solve the problems existing in the prior art.

[0006] The above-mentioned technical objective of the present invention is achieved through the following technical solution: an energy management and control method based on a virtual power plant, comprising the following steps: S1: constructing a standardized data access system, connecting to the monitoring devices of each branch load, distributed power source, and energy storage device in the virtual power plant through an interface adaptation module, and collecting multi-source energy data based on a preset standardized protocol, wherein the multi-source energy data includes branch load operation data, distributed power source output data, energy storage device status data, and environmental correlation data.

[0007] S2: Combine and preprocess the collected multi-source energy data, and sequentially complete the data format standardization conversion, abnormal data removal and noise reduction to obtain high-quality standardized energy data.

[0008] S3: Construct an energy management and assessment model that adapts to the characteristics of multiple types of branch loads, input standardized energy data, evaluate the energy operation status of the virtual power plant in real time, and identify the types and levels of abnormal branch loads.

[0009] S4: Based on the identified abnormal information, and in conjunction with the virtual power plant topology and grid dispatching requirements, generate fault location results and collaborative control strategies, and push the control assessment results, abnormal early warning information, fault location results and control strategies to the grid dispatching system.

[0010] S5: Receives real-time feedback results from the grid dispatching system, iteratively updates the parameters of the energy management and assessment model, and forms a closed-loop energy management and control process.

[0011] The present invention also discloses an energy management and control system based on a virtual power plant, comprising: a standardized data access module, a data combination preprocessing module, an energy status assessment module, an anomaly handling and collaborative scheduling module, and a closed-loop optimization module.

[0012] The standardized data access module is equipped with an interface adaptation submodule, which is used to connect to various branch load monitoring devices and energy equipment to collect multi-source energy data based on standardized protocols.

[0013] The data combination preprocessing module is used to perform format standardization conversion, abnormal data removal, and noise reduction on multi-source energy data.

[0014] The energy status assessment module is used to assess the energy operating status in real time and identify the type and level of anomalies by using a control and assessment model that adapts to the characteristics of multiple types of branch loads.

[0015] The anomaly handling and coordinated scheduling module is used to generate fault location results and coordinated control strategies, and to achieve data interaction with the power grid dispatching system.

[0016] The closed-loop optimization module is used to update the parameters of the control and management assessment model based on the feedback results of the control and management execution, and to optimize the energy control and management process.

[0017] Preferably, in step S1, the branch load operation data includes voltage, current, active power, reactive power and harmonic content, and the environmental data includes temperature, humidity and wind speed at the monitoring point.

[0018] Preferably, in step S2, the combined preprocessing uses an adaptive wavelet threshold denoising algorithm to first remove outliers that exceed the normal data range, and then eliminate electromagnetic interference noise through the adaptive wavelet threshold denoising algorithm. The adaptive wavelet threshold is dynamically adjusted according to the data fluctuation amplitude.

[0019] Preferably, the energy management and assessment model described in step S3 is an improved gradient boosting tree model, which adapts to the operating characteristics of different branch loads of residential, commercial and industrial users by introducing branch load type weight coefficients.

[0020] Preferably, the abnormality level in step S4 is divided into four levels: Level 1 minor abnormality, Level 2 general abnormality, Level 3 severe abnormality, and Level 4 emergency abnormality. The Level 4 emergency abnormality triggers the emergency load transfer process of the power grid dispatching system.

[0021] Ideally, it also includes a data encryption and storage module, which uses the AES encryption algorithm to encrypt multi-source energy data and control results, and uses a distributed time-series database to achieve efficient storage of massive amounts of data, supporting fast retrieval and traceability by timestamp and branch number.

[0022] Preferably, the interface adaptation submodule supports protocol conversion for non-standardized interfaces.

[0023] The present invention also discloses an energy management and control device based on a virtual power plant, including a processor, a memory, and a communication interface. The memory stores a computer program, and when the processor executes the computer program, it implements any of the energy management and control methods based on the virtual power plant described above. The communication interface is used to interact with a branch load monitoring device, energy equipment, and a power grid dispatching system.

[0024] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements an energy management method based on a virtual power plant.

[0025] In summary, the present invention has the following beneficial effects: The present invention realizes the interconnection and interoperability of multi-source branch load data: The present invention adopts the IEC61850 standardized interface protocol, combined with the protocol conversion function of the interface adapter sub-module, effectively solving the "data island" problem of branch load monitoring devices from different manufacturers in traditional technology, improving equipment compatibility by more than 80%, and realizing the comprehensive fusion and collection of multi-branch load data.

[0026] This invention improves the accuracy and reliability of monitoring data: by using a combination algorithm of "3σ criterion + adaptive wavelet threshold noise reduction", abnormal data is accurately removed and electromagnetic interference noise is eliminated, improving data processing accuracy by more than 25%, providing high-quality data support for subsequent status assessment, and avoiding false monitoring caused by data distortion.

[0027] This invention improves the efficiency of branch load anomaly identification and fault location: based on the improved XGBoost model adapted to the characteristics of different types of branch loads, the anomaly identification accuracy is increased to over 96%, the response time for level four anomalies is shortened to less than 3 minutes, and the fault location accuracy reaches the branch line level. Compared with traditional technology, the fault handling efficiency is improved by more than 70%, reducing power outage losses.

[0028] This invention enhances the monitoring and scheduling coordination capabilities: through the anomaly handling and linkage module, it achieves seamless integration with the distribution network scheduling system, generates precise control suggestions, and assists in the optimized implementation of strategies such as reactive power compensation and load transfer in the distribution network. This reduces distribution network operating losses by 12%-18%, increases the amount of renewable energy absorbed by more than 15%, and improves the economic efficiency of power grid operation.

[0029] This invention ensures data security and full lifecycle traceability: it adopts AES encryption and distributed time-series database storage to ensure the security of monitoring data transmission and storage, realizes full lifecycle traceability of data, provides reliable data support for power distribution network operation optimization, fault review and equipment maintenance, and reduces system operation and maintenance costs. Attached Figure Description

[0030] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example 1

[0032] This embodiment applies to a virtual power plant in an industrial and commercial complex. The park aggregates diverse resources such as distributed photovoltaic power, residential energy storage, industrial loads, commercial and office loads, and residential supporting loads, and connects to the distribution network branch line for operation. The energy management method, system, and supporting equipment of this invention are used for daily monitoring, anomaly identification, and coordinated scheduling. The overall process is as follows: Figure 1 .

[0033] System Overall Architecture and Hardware Deployment: This embodiment is equipped with a complete energy management and control system, comprising six functional modules: standardized data access module, data combination and preprocessing module, energy status assessment module, anomaly handling and collaborative scheduling module, closed-loop optimization module, and data encryption and storage module. Dedicated management and control equipment is deployed, consisting of a high-performance processor, large-capacity memory, and multiple types of communication interfaces, and is located in the park's energy management computer room. The management and control equipment connects to various branch load monitoring terminals, distributed photovoltaic inverters, energy storage converters, and environmental monitoring sensors within the park via communication interfaces, and simultaneously connects to the regional power grid dispatch system to achieve full-domain data interaction and command communication.

[0034] Step S1: Standardized Data Access: This embodiment uniformly adopts the IEC61850 standard protocol as the core communication protocol, and the standardized data access module has a built-in interface adaptation submodule. Addressing the non-standard interfaces and proprietary protocols of some older equipment and third-party equipment within the park, the interface adaptation submodule completes protocol conversion and interface adaptation, completely breaking down data barriers. The system comprehensively collects multi-source energy data at a sampling frequency of 1 second / time.

[0035] Branch load operation data: voltage, current, active power, reactive power, and harmonic content of each branch, covering industrial production branches, commercial office branches, and residential power branches; Distributed power source data: real-time output of distributed photovoltaic power, inverter operating status, and power generation curves; Energy storage equipment data: energy storage charging and discharging power, battery state of charge, equipment operating temperature, and fault alarm signals; Environmental correlation data: temperature, humidity, and outdoor wind speed at each monitoring point, used to analyze the impact of sunlight, ambient temperature, and humidity on photovoltaic output and load. All raw data are uniformly encapsulated in a standard message format and transmitted to the next module.

[0036] Step S2: Multi-source data combination preprocessing: The data combination preprocessing module executes a two-stage processing flow to ensure data quality: Abnormal data removal: Using the 3σ criterion, jump values, missing values, and out-of-limit abnormal data exceeding the normal operating range of the equipment are screened and removed, such as invalid data like instantaneous voltage spikes / drops and power surges; Adaptive wavelet threshold denoising: An adaptive wavelet threshold denoising algorithm is enabled to address noise interference caused by electromagnetic interference in the plant area and line signal loss. The system monitors data fluctuation amplitude in real time and dynamically adjusts the wavelet threshold. Small fluctuations use a low threshold to retain effective details, while large disturbances use a high threshold to filter out noise. After processing, all data undergoes format unification and noise reduction to form a standardized, high-quality dataset, which is then input into the energy status assessment module.

[0037] Step S3: Load Status Assessment and Anomaly Identification: The energy status assessment module is equipped with an improved gradient boosting tree (XGBoost) assessment model. Based on the load characteristics of the industrial park, independent weight coefficients are set for the three types of branch loads: industrial, commercial, and residential, matching the electricity consumption patterns and fluctuation characteristics of different loads. The model loads standardized data in real time, conducts operational status assessments for each branch, and classifies anomalies into four levels according to preset rules: Level 1 (Minor Anomaly): Parameters deviate slightly from the rated range, with no safety risk, only log entries are made; Level 2 (General Anomaly): Load fluctuations and power deviations are significant, triggering local maintenance warnings; Level 3 (Severe Anomaly): Harmonic exceedances, power exceeding limits, and abnormal equipment temperature rise immediately trigger local control measures; Level 4 (Emergency Anomaly): Major faults such as short circuits, overloads, and leakage currents activate emergency response logic. In this embodiment, the model's anomaly identification accuracy can reach over 96%, accurately distinguishing between normal load fluctuations and actual faults.

[0038] Step S4: Fault Location and Coordinated Control: The anomaly handling and coordinated dispatch module retrieves the topology diagram of the park's distribution network and, combined with the abnormal branch number and parameter characteristics, accurately locates the fault to a single branch line or a single piece of electrical equipment. Based on the anomaly level and the real-time dispatch requirements of the power grid, differentiated coordinated control strategies are generated: Level 1 and 2 anomalies: Only early warning information and optimized operation suggestions are pushed, such as adjusting the charging and discharging time of energy storage and optimizing load staggering operation; Level 3 anomalies: Local control instructions are issued to limit the power of the faulty branch and isolate the abnormal equipment; Level 4 emergency anomalies: The fault location report and anomaly details are pushed to the power grid dispatch system immediately, and the emergency load transfer process on the power grid side is triggered, linking with surrounding energy storage and adjustable loads to complete power compensation and prevent the fault from escalating. Simultaneously, the module uploads the evaluation results, early warning information, and control strategies to the upper-level power grid dispatch system, achieving data linkage between the control and dispatch ends.

[0039] Step S5: Closed-Loop Iterative Optimization: The closed-loop optimization module continuously receives control execution results and on-site fault handling feedback information returned by the power grid dispatching system. Based on the feedback data, the system dynamically corrects the weight parameters and judgment thresholds of the improved gradient boosting tree model, continuously optimizes the anomaly identification logic and control strategy, and forms a fully closed-loop control process of data acquisition—processing—evaluation—control—feedback—model update.

[0040] Data security and storage: The data encryption and storage module uses the AES encryption algorithm throughout the entire process to encrypt and protect all transmitted and stored energy data, management logs, and fault records. The backend uses a distributed time-series database to store massive amounts of data, supporting fast retrieval by timestamp, branch number, and equipment type, enabling historical data backtracking, fault review, and operational trend analysis.

[0041] The operational results of this embodiment are as follows: the interoperability and compatibility of park equipment is improved by more than 80%, and the data processing accuracy is improved by 25%; the overall response time for Level 4 emergency anomalies is controlled within 3 minutes, and the fault handling efficiency is improved by 70% compared with the traditional solution; the overall operating loss of the distribution network is reduced by 12%~15%, and the amount of distributed new energy consumption is increased by about 15%. Example 2

[0042] refer to Figure 1 An energy management and control method based on a virtual power plant includes the following steps: S1: Construct a standardized data access system, and connect to the monitoring devices of each branch load, distributed power source, and energy storage device in the virtual power plant through an interface adaptation module. Collect multi-source energy data based on a preset standardized protocol. The multi-source energy data includes branch load operation data, distributed power source output data, energy storage device status data, and environmental correlation data.

[0043] S2: Combine and preprocess the collected multi-source energy data, and sequentially complete the data format standardization conversion, abnormal data removal and noise reduction to obtain high-quality standardized energy data.

[0044] S3: Construct an energy management and assessment model that adapts to the characteristics of multiple types of branch loads, input standardized energy data, evaluate the energy operation status of the virtual power plant in real time, and identify the types and levels of abnormal branch loads.

[0045] S4: Based on the identified abnormal information, and in conjunction with the virtual power plant topology and grid dispatching requirements, generate fault location results and collaborative control strategies, and push the control assessment results, abnormal early warning information, fault location results and control strategies to the grid dispatching system.

[0046] S5: Receives real-time feedback results from the grid dispatching system, iteratively updates the parameters of the energy management and assessment model, and forms a closed-loop energy management and control process.

[0047] An energy management and control system based on a virtual power plant includes: a standardized data access module, a data combination and preprocessing module, an energy status assessment module, an anomaly handling and collaborative scheduling module, and a closed-loop optimization module.

[0048] The standardized data access module has an interface adaptation sub-module for connecting to various branch load monitoring devices and energy equipment, and collecting multi-source energy data based on standardized protocols.

[0049] The data combination preprocessing module is used to perform format standardization conversion, outlier data removal, and noise reduction on multi-source energy data.

[0050] The energy status assessment module is used to assess the energy operating status in real time and identify the type and level of anomalies by using a control and assessment model that adapts to the characteristics of various types of branch loads.

[0051] The anomaly handling and coordinated scheduling module is used to generate fault location results and coordinated control strategies, and to achieve data interaction with the power grid dispatching system.

[0052] The closed-loop optimization module is used to update the parameters of the control and management assessment model based on the feedback results of the control and management execution, thereby optimizing the energy control and management process.

[0053] In step S1, the branch load operation data includes voltage, current, active power, reactive power and harmonic content, and the environmental data includes temperature, humidity and wind speed at the monitoring point.

[0054] In step S2, the combined preprocessing uses an adaptive wavelet threshold denoising algorithm to first remove outliers that exceed the normal data range, and then eliminate electromagnetic interference noise through the adaptive wavelet threshold denoising algorithm. The adaptive wavelet threshold is dynamically adjusted according to the data fluctuation amplitude.

[0055] In step S3, the energy management and assessment model is an improved gradient boosting tree model, which adapts to the operating characteristics of different branch loads of residential, commercial and industrial users by introducing branch load type weight coefficients.

[0056] In step S4, the abnormality level is divided into four levels: Level 1 minor abnormality, Level 2 general abnormality, Level 3 severe abnormality, and Level 4 emergency abnormality. Level 4 emergency abnormality triggers the emergency load transfer process of the power grid dispatching system.

[0057] It also includes a data encryption and storage module, which uses the AES encryption algorithm to encrypt multi-source energy data and management results, and uses a distributed time-series database to achieve efficient storage of massive amounts of data, supporting fast retrieval and traceability by timestamp and branch number.

[0058] The interface adaptation submodule supports protocol conversion for non-standard interfaces.

[0059] An energy management and control device based on a virtual power plant includes a processor, a memory, and a communication interface. The memory stores a computer program, and when the processor executes the computer program, it implements any one of the energy management and control methods based on the virtual power plant. The communication interface is used to interact with branch load monitoring devices, energy equipment, and power grid dispatching systems.

[0060] A computer-readable storage medium storing a computer program that, when executed by a processor, implements an energy management method based on a virtual power plant.

[0061] Among them, the present invention realizes the interconnection and interoperability of multi-source branch load data: the present invention adopts the IEC61850 standardized interface protocol, combined with the protocol conversion function of the interface adapter sub-module, effectively solves the "data island" problem of branch load monitoring devices from different manufacturers in traditional technology, improves equipment compatibility by more than 80%, and realizes the comprehensive fusion and collection of multi-branch load data.

[0062] This invention improves the accuracy and reliability of monitoring data: by using a combination algorithm of "3σ criterion + adaptive wavelet threshold noise reduction", abnormal data is accurately removed and electromagnetic interference noise is eliminated, improving data processing accuracy by more than 25%, providing high-quality data support for subsequent status assessment, and avoiding false monitoring caused by data distortion.

[0063] This invention improves the efficiency of branch load anomaly identification and fault location: based on the improved XGBoost model adapted to the characteristics of different types of branch loads, the anomaly identification accuracy is increased to over 96%, the response time for level four anomalies is shortened to less than 3 minutes, and the fault location accuracy reaches the branch line level. Compared with traditional technology, the fault handling efficiency is improved by more than 70%, reducing power outage losses.

[0064] This invention enhances the monitoring and scheduling coordination capabilities: through the anomaly handling and linkage module, it achieves seamless integration with the distribution network scheduling system, generates precise control suggestions, and assists in the optimized implementation of strategies such as reactive power compensation and load transfer in the distribution network. This reduces distribution network operating losses by 12%-18%, increases the amount of renewable energy absorbed by more than 15%, and improves the economic efficiency of power grid operation.

[0065] This invention ensures data security and full lifecycle traceability: it adopts AES encryption and distributed time-series database storage to ensure the security of monitoring data transmission and storage, realizes full lifecycle traceability of data, provides reliable data support for power distribution network operation optimization, fault review and equipment maintenance, and reduces system operation and maintenance costs.

[0066] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An energy management and control method based on a virtual power plant, characterized in that, Includes the following steps: S1: Construct a standardized data access system, and connect to the monitoring devices of each branch load, distributed power source, and energy storage device in the virtual power plant through the interface adaptation module. Collect multi-source energy data based on the preset standardized protocol. The multi-source energy data includes branch load operation data, distributed power source output data, energy storage device status data, and environmental correlation data. S2: Combine and preprocess the collected multi-source energy data, and sequentially complete the data format standardization conversion, abnormal data removal and noise reduction to obtain high-quality standardized energy data. S3: Construct an energy management and assessment model that adapts to the characteristics of multiple types of branch loads, input standardized energy data, evaluate the energy operation status of the virtual power plant in real time, and identify the abnormal types and levels of branch loads; S4: Based on the identified abnormal information, and in combination with the virtual power plant topology and grid dispatching requirements, generate fault location results and collaborative control strategies, and push the control assessment results, abnormal early warning information, fault location results and control strategies to the grid dispatching system. S5: Receives real-time feedback results from the power grid dispatching system, iteratively updates the parameters of the energy management and assessment model, and forms a closed-loop energy management and control process.

2. An energy management and control system based on a virtual power plant, characterized in that, include: Standardized data access module, data combination and preprocessing module, energy status assessment module, anomaly handling and collaborative scheduling module, and closed-loop optimization module; The standardized data access module is equipped with an interface adaptation submodule, which is used to connect to various branch load monitoring devices and energy equipment, and collect multi-source energy data based on standardized protocols. The data combination preprocessing module is used to perform format standardization conversion, abnormal data removal, and noise reduction on multi-source energy data. The energy status assessment module is used to assess the energy operating status in real time and identify the type and level of anomalies by using a control and assessment model that adapts to the characteristics of multiple types of branch loads. The anomaly handling and coordinated scheduling module is used to generate fault location results and coordinated control strategies, and to achieve data interaction with the power grid dispatching system. The closed-loop optimization module is used to update the parameters of the control and management assessment model based on the feedback results of the control and management execution, and to optimize the energy control and management process.

3. The energy management method based on a virtual power plant according to claim 1, characterized in that, In S1, the branch load operation data includes voltage, current, active power, reactive power and harmonic content, and the environmental correlation data includes temperature, humidity and wind speed at the monitoring point.

4. The energy management method based on a virtual power plant according to claim 1, characterized in that, In step S2, the combined preprocessing uses an adaptive wavelet threshold denoising algorithm to first remove outliers that exceed the normal data range, and then eliminate electromagnetic interference noise through the adaptive wavelet threshold denoising algorithm. The adaptive wavelet threshold is dynamically adjusted according to the data fluctuation amplitude.

5. The energy management method based on a virtual power plant according to claim 1, characterized in that, The energy management and assessment model described in S3 is an improved gradient boosting tree model, which adapts to the operating characteristics of different branch loads such as residential, commercial, and industrial loads by introducing weight coefficients for branch load types.

6. The energy management method based on a virtual power plant according to claim 1, characterized in that, The abnormality level in S4 is divided into four levels: Level 1 minor abnormality, Level 2 general abnormality, Level 3 severe abnormality, and Level 4 emergency abnormality. Level 4 emergency abnormality triggers the emergency load transfer process of the power grid dispatching system.

7. An energy management and control system based on a virtual power plant according to claim 2, characterized in that, It also includes a data encryption and storage module, which uses the AES encryption algorithm to encrypt multi-source energy data and management results, and uses a distributed time-series database to achieve efficient storage of massive data, supporting fast retrieval and traceability by timestamp and branch number.

8. An energy management and control system based on a virtual power plant according to claim 2, characterized in that, The interface adaptation submodule supports protocol conversion for non-standard interfaces.

9. An energy management and control device based on a virtual power plant, characterized in that, The device includes a processor, a memory, and a communication interface. The memory stores a computer program. When the processor executes the computer program, it implements the energy management method based on a virtual power plant as described in any one of claims 1 to 3-6. The communication interface is used to interact with a branch load monitoring device, energy equipment, and a power grid dispatching system.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements an energy management method based on a virtual power plant as described in any one of claims 1, 3-6.

Images