An adjustable resource aggregation model for virtual power plant

Abstract

The application relates to the technical field of adjustable resource aggregation, and provides an adjustable resource aggregation model for a virtual power plant, which comprises a distributed energy resource parameter acquisition module, a communication module, a distributed energy resource parameter processing module and a scheduling decision optimization module; the distributed energy resource parameter acquisition module acquires various data of the distributed energy resource in real time through sensors and monitoring equipment and transmits the data to the communication module; the communication module is responsible for transmitting the data to the distributed energy resource parameter processing module and transmitting prediction values and scheduling decision support information generated by the distributed energy resource parameter processing module to the scheduling decision optimization module; the distributed energy resource parameter processing module is used for storing and analyzing the data, generating the prediction values and the scheduling decision support information; and the scheduling decision optimization module constructs a scheduling decision optimization model by using a particle swarm optimization algorithm according to the prediction results and the support information, so as to generate a scheduling strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of adjustable resource aggregation, and more particularly to an adjustable resource aggregation model for a virtual power plant. BACKGROUND

[0002] With the continuous adjustment of global energy structure and the increase of renewable energy proportion, virtual power plant as a new energy management mode has attracted widespread attention. Virtual power plant integrates various distributed energy resources such as distributed power generation equipment, energy storage devices and adjustable load resources to form a unified scheduling and control system to realize the optimal utilization of distributed energy resources. Adjustable resources are an important part of virtual power plants, including load adjustment, power generation adjustment and energy storage adjustment. These resources can effectively balance the supply and demand relationship in the power system through flexible scheduling strategies, improve energy utilization efficiency and system stability.

[0003] However, although some existing technologies introduce optimization algorithms for scheduling decisions, these algorithms are mostly based on simple linear programming or heuristic methods, which cannot fully consider the dynamic characteristics and complex constraints of distributed power systems, especially in the case of large load fluctuations, frequent state changes of distributed power generation and energy storage devices. Existing optimization algorithms are difficult to quickly and effectively generate high-quality scheduling decisions; secondly, the current scheduling decision support information is mostly based on static rules, lacking intelligent and dynamic adjustment capabilities. In the face of complex and variable power system environment, traditional scheduling support information is difficult to provide accurate scheduling suggestions, thereby affecting the scheduling efficiency and reliability of the overall system.

[0004] In order to solve the above problems, a technical scheme is provided. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art, the present application provides an adjustable resource aggregation model for a virtual power plant, comprising a distributed energy resource parameter acquisition module, a communication module, a distributed energy resource parameter processing module and a scheduling decision optimization module. By introducing a long short-term memory network to analyze and predict the distributed energy parameters, a particle swarm optimization algorithm is used to construct a scheduling decision optimization model, realizing real-time scheduling and control of distributed energy resources, and solving the problem of insufficient scheduling decision optimization in the prior art.

[0006] To achieve the above object, the present application provides the following technical scheme:

[0007] An adjustable resource aggregation model for a virtual power plant includes a distributed energy resource parameter acquisition module, a communication module, a distributed energy resource parameter processing module, and a scheduling decision optimization module. The distributed energy resource parameter acquisition module is connected to the communication module, the communication module is bidirectionally connected to the distributed energy resource parameter processing module, and the communication module is connected to the scheduling decision optimization module.

[0008] The distributed energy resource parameter processing module includes a distributed energy resource parameter storage unit and a distributed energy resource parameter analysis unit, which are used to store and analyze the collected distributed energy resource parameters, wherein the distributed energy resource parameter storage unit is connected to the distributed energy resource parameter analysis unit; the distributed energy resource parameter analysis unit uses a long short-term memory network to output the load demand of the distributed power system, the power generation of the distributed power generation equipment, and the charge and discharge power of the energy storage device. The prediction results are used to calculate and generate scheduling decision support information, wherein the scheduling decision support information includes a load demand balancing suggestion for the distributed power system, a power generation adjustment suggestion value, and a charge and discharge power suggestion value for the energy storage device. The load demand balancing suggestion includes a load reduction suggestion value and a load transfer suggestion value; wherein the formula for calculating the load reduction suggestion value is:

[0009]

[0010] Where ΔL cut (t) is the recommended load reduction value at time t, ΔL max is the maximum load reduction, L peak is the peak load threshold, L(t) is the load demand forecast value of the distributed power system;

[0011] The formula for calculating the recommended load transfer value is:

[0012]

[0013] Where, L shift (t) is the load transfer recommendation value at time t, L(t) is the load demand forecast value of the distributed power system, and L peak is the peak load threshold, T shift is the total time of the load transfer period, and t is the current time in the load transfer period;

[0014] The formula for calculating the recommended power generation adjustment value is:

[0015]

[0016] Where, P gen , i (t) is the recommended value for power generation adjustment of the i-th power generation equipment at time t, P predict，iis the predicted value of power generation of the ith distributed generation equipment, L(t) is the predicted value of load demand of the distributed power system, and L total is the total load of all power generation equipment at time t;

[0017] The formula for calculating the recommended charging power value of the energy storage device is:

[0018]

[0019] Where, P charge (t) is the recommended charging power value of the energy storage device at time t, P predict，charge (t) is the predicted value of charging power of energy storage device, L(t) is the predicted value of load demand of distributed power system, L low is the low load threshold;

[0020] The formula for calculating the recommended discharge power value of the energy storage device is:

[0021]

[0022] Where, P discharge (t) is the recommended discharge power value of the energy storage device at time t, P predict，discharge is the discharge power forecast value of the energy storage device, L(t) is the load demand forecast value of the distributed power system, and L high is the high load threshold;

[0023] The scheduling decision optimization module is used to construct a scheduling decision optimization model based on the prediction results and scheduling decision support information generated by the distributed energy resource parameter processing module, using the particle swarm optimization algorithm, output and execute the scheduling strategy, and perform real-time scheduling and control of distributed energy resources.

[0024] As a further solution of the present invention, the maximum load reduction is the difference between the 95th percentile and the 75th percentile of the total load of the distributed power system; the high load threshold is the 80th percentile of the total load of the distributed power system; the low load threshold is the 20th percentile of the total load of the distributed power system; and the peak load threshold is the 90th percentile of the total load of the distributed power system.

[0025] As a further solution of the present invention, the distributed energy resource parameter acquisition module obtains the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, as well as the total load of the distributed power system and the charging and discharging power of the energy storage device in real time through sensors and monitoring equipment installed on the distributed energy resources, and transmits the above data to the communication module through a standardized interface.

[0026] As a further solution of the present invention, the distributed energy resource parameter analysis unit receives the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, the total load of the distributed power system and the charge and discharge power of the energy storage device collected by the distributed energy resource parameter acquisition module through the communication module, and inputs the constructed long and short-term memory network to obtain the load demand forecast value of the distributed power system, the power generation forecast value of the distributed power generation equipment and the charge and discharge power forecast value of the energy storage device.

[0027] As a further solution of the present invention, the steps of constructing the long short-term memory network are:

[0028] Step S1, collecting historical data on power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of distributed power generation equipment, load of distributed power system, and charge and discharge power of energy storage devices;

[0029] Step S2, preprocessing the above data and performing feature extraction;

[0030] Step S3, designing the model structure of the long short-term memory network, determining that the input layer is used to receive the extracted feature data; the long short-term memory network layer is used to process time series data and capture long-term and short-term dependencies; the fully connected layer is used to convert the output of the long short-term memory network layer into a predicted value; and the output layer is used to generate a load demand forecast value for the distributed power system, a power generation forecast value for the distributed power generation equipment, and a charge and discharge power forecast value for the energy storage device;

[0031] Step S4: divide the historical data into a training set and a test set, use the training set to train the designed long short-term memory network model, and use the test set to test the trained long short-term memory network.

[0032] As a further solution of the present invention, the scheduling decision optimization module uses a particle swarm optimization algorithm to construct a scheduling decision optimization model. The model generates a scheduling decision optimization plan by inputting the load demand of the distributed power system, the power generation of the distributed power generation equipment, and the predicted value of the charge and discharge power of the energy storage device, as well as the real-time data of the scheduling decision support information. The scheduling decision optimization plan includes the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, the total load of the distributed power system, and the charge and discharge power setting value of the energy storage device. The steps of constructing the scheduling decision optimization model using the particle swarm optimization algorithm are as follows:

[0033] Step A1: Determine that the model inputs are the real-time predicted values ​​of the load demand of the distributed power system, the real-time predicted values ​​of the power generation of the distributed power generation equipment, the real-time predicted values ​​of the charge and discharge power of the energy storage device, and the real-time data of the scheduling decision support information; determine that the model objectives are to minimize the load fluctuation range, maximize the equipment operating time, and maximize the power generation utilization efficiency; and determine that the model output is the scheduling decision optimization plan;

[0034] Step A2: The dispatching decision optimization module collects historical forecast values ​​of the load demand of the distributed power system, historical forecast values ​​of the power generation of the distributed power generation equipment, historical forecast values ​​of the charge and discharge power of the energy storage device, and historical data of the dispatching decision support information;

[0035] Step A3: preprocessing the above data and performing feature extraction, including parameter normalization and principal component analysis;

[0036] In step A4, the historical data is divided into a training set and a test set. The designed model is trained using the training set. The state of the individual particle swarms is continuously optimized through the iterative process of the particle swarm optimization algorithm until the preset number of iterations is reached. The trained model is tested using the test set.

[0037] As a further solution of the present invention, the distributed energy resource parameter storage unit manages and stores the basic attributes, prediction information and parameter format of distributed energy resources through a unified parameter model. The basic attributes include the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of distributed power generation equipment, as well as the total load of the distributed power system and the charge and discharge power of the energy storage device; the prediction information includes the load demand prediction value of the distributed power system, the power generation prediction value of the distributed power generation equipment, the charge and discharge power prediction value of the energy storage device and scheduling decision support information; the parameter format includes parameter type, parameter unit and parameter accuracy.

[0038] As a further solution of the present invention, the communication module is used to transmit the parameters collected by the distributed energy resource parameter collection module to the distributed energy resource parameter processing module, and transmit the load demand forecast value of the distributed power system, the power generation forecast value of the distributed power generation equipment, the charging and discharging power forecast value of the energy storage device and the scheduling decision support information generated by the distributed energy resource parameter processing module to the scheduling decision optimization module.

[0039] Compared with the prior art, the adjustable resource aggregation model for virtual power plants of the present invention has the following advantages:

[0040] The present invention realizes parameter collection and transmission through the efficient collaboration of the distributed energy resource parameter acquisition module and the communication module, thereby improving the timeliness and reliability of the data; adopts the long-short-term memory network to process and analyze the distributed energy resource parameters, which can fully explore the complex patterns and long-term and short-term dependencies in the data, and improve the accuracy of load demand, power generation and energy storage charging and discharging power prediction; uses the particle swarm optimization algorithm to construct a scheduling decision optimization model, taking into account the dynamic characteristics and complex constraints of the power system, and realizes the scheduling and intelligent control of distributed energy resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a structural schematic diagram of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0043] Example 1

[0044] An adjustable resource aggregation model for a virtual power plant includes a distributed energy resource parameter acquisition module, a communication module, a distributed energy resource parameter processing module, and a scheduling decision optimization module.

[0045] The distributed energy resource parameter acquisition module in the embodiment of the present invention is connected to the communication module, the communication module is bidirectionally connected to the distributed energy resource parameter processing module, and the communication module is connected to the scheduling decision optimization module.

[0046] The distributed energy resource parameter acquisition module obtains the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, as well as the total load of the distributed power system and the charging and discharging power of the energy storage device in real time through sensors and monitoring equipment installed on the distributed energy resources, and transmits the above data to the communication module through a standardized interface.

[0047] The communication module is used to transmit the parameters collected by the distributed energy resource parameter acquisition module to the distributed energy resource parameter processing module, and transmit the load demand forecast value of the distributed power system, the power generation forecast value of the distributed power generation equipment, the charging and discharging power forecast value of the energy storage device and the scheduling decision support information generated by the distributed energy resource parameter processing module to the scheduling decision optimization module.

[0048] The distributed energy resource parameter processing module includes a distributed energy resource parameter storage unit and a distributed energy resource parameter analysis unit, which are used to store and analyze the collected distributed energy resource parameters, wherein the distributed energy resource parameter storage unit is connected to the distributed energy resource parameter analysis unit; the distributed energy resource parameter analysis unit uses a long short-term memory network to output the load demand of the distributed power system, the power generation of the distributed power generation equipment and the charging and discharging power of the energy storage device. The predicted values ​​are calculated and generated based on the prediction results. The scheduling decision support information includes load demand balancing suggestions for the distributed power system, power generation adjustment suggestions and charging and discharging power suggestions for the energy storage device. The load demand balancing suggestions include load reduction suggestions and load transfer suggestions.

[0049] The distributed energy resource parameter storage unit in the embodiment of the present invention manages and stores the basic attributes, prediction information and parameter format of distributed energy resources through a unified parameter model. The basic attributes include the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of distributed power generation equipment, as well as the total load of the distributed power system and the charge and discharge power of the energy storage device; the prediction information includes the load demand prediction value of the distributed power system, the power generation prediction value of the distributed power generation equipment, the charge and discharge power prediction value of the energy storage device and scheduling decision support information; the parameter format includes parameter type, parameter unit and parameter accuracy.

[0050] The distributed energy resource parameter analysis unit in the embodiment of the present invention receives the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, the total load of the distributed power system and the charge and discharge power of the energy storage device collected by the distributed energy resource parameter collection module through the communication module, and inputs the constructed long and short-term memory network to obtain the load demand prediction value of the distributed power system, the power generation prediction value of the distributed power generation equipment and the charge and discharge power prediction value of the energy storage device.

[0051] The steps of constructing the long short-term memory network in the embodiment of the present invention are:

[0052] Step S1, collecting historical data on power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of distributed power generation equipment, load of distributed power system, and charge and discharge power of energy storage devices;

[0053] Step S2, preprocessing the above data and performing feature extraction;

[0054] Step S3, designing the model structure of the long short-term memory network, determining that the input layer is used to receive the extracted feature data; the long short-term memory network layer is used to process time series data and capture long-term and short-term dependencies; the fully connected layer is used to convert the output of the long short-term memory network layer into a predicted value; and the output layer is used to generate a load demand forecast value for the distributed power system, a power generation forecast value for the distributed power generation equipment, and a charge and discharge power forecast value for the energy storage device;

[0055] Step S4: divide the historical data into a training set and a test set, use the training set to train the designed long short-term memory network model, and use the test set to test the trained long short-term memory network.

[0056] The following is a Python code example that uses a long short-term memory network model to obtain a forecast of the load demand of a distributed power system, the power generation of distributed generation equipment, and the charge and discharge power of an energy storage device. Please note that this example is only a starting point and may need to be adjusted in practice based on actual conditions and device interfaces.

[0057]

[0058]

[0059]

[0060] This code is only an example and needs to be modified and adjusted appropriately according to specific circumstances in actual applications.

[0061] The formula for calculating the load reduction recommendation value in the embodiment of the present invention is:

[0062]

[0063] Where ΔL cut (t) is the recommended load reduction value at time t, ΔL max is the maximum load reduction, L peak is the peak load threshold, and L(t) is the load demand forecast value of the distributed power system.

[0064] The formula for calculating the load transfer recommendation value in the embodiment of the present invention is:

[0065]

[0066] Where, L shift (t) is the load transfer recommendation value at time t, L(t) is the load demand forecast value of the distributed power system, and L peak is the peak load threshold, T shift is the total time of the load transfer period, and t is the current time in the load transfer period.

[0067] The formula for calculating the recommended power generation adjustment value in the embodiment of the present invention is:

[0068]

[0069] Where, P gen , i (t) is the recommended value for power generation adjustment of the i-th power generation equipment at time t, P predict，i is the predicted value of power generation of the ith distributed generation equipment, L(t) is the predicted value of load demand of the distributed power system, and L total is the total load of all power generation equipment at time t.

[0070] The formula for calculating the recommended charging power value of the energy storage device in the embodiment of the present invention is:

[0071]

[0072] Where, P charge (t) is the recommended charging power value of the energy storage device at time t, P predict，charge (t) is the predicted value of charging power of energy storage device, L(t) is the predicted value of load demand of distributed power system, L low is the low load threshold.

[0073] The formula for calculating the recommended discharge power value of the energy storage device in the embodiment of the present invention is:

[0074]

[0075] Where, P discharge (t) is the recommended discharge power value of the energy storage device at time t, P predict，discharge is the discharge power forecast value of the energy storage device, L(t) is the load demand forecast value of the distributed power system, and L high is the high load threshold.

[0076] The maximum load reduction in the embodiment of the present invention is the difference between the 95th percentile and the 75th percentile of the total load of the distributed power system; the high load threshold is the 80th percentile of the total load of the distributed power system; the low load threshold is the 20th percentile of the total load of the distributed power system; and the peak load threshold is the 90th percentile of the total load of the distributed power system.

[0077] The scheduling decision optimization module uses the particle swarm optimization algorithm to construct a scheduling decision optimization model. The model generates a scheduling decision optimization plan by inputting the load demand of the distributed power system, the power generation of distributed power generation equipment and the predicted value of the charge and discharge power of the energy storage device, as well as the real-time data of the scheduling decision support information. The scheduling decision optimization plan includes the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, as well as the total load of the distributed power system and the charge and discharge power set value of the energy storage device.

[0078] The steps of constructing a scheduling decision optimization model using the particle swarm optimization algorithm in the embodiment of the present invention are as follows:

[0079] Step A1: Determine that the model inputs are the real-time predicted values ​​of the load demand of the distributed power system, the real-time predicted values ​​of the power generation of the distributed power generation equipment, the real-time predicted values ​​of the charge and discharge power of the energy storage device, and the real-time data of the scheduling decision support information; determine that the model objectives are to minimize the load fluctuation range, maximize the equipment operating time, and maximize the power generation utilization efficiency; and determine that the model output is the scheduling decision optimization plan;

[0080] Step A2: The dispatching decision optimization module collects historical forecast values ​​of the load demand of the distributed power system, historical forecast values ​​of the power generation of the distributed power generation equipment, historical forecast values ​​of the charge and discharge power of the energy storage device, and historical data of the dispatching decision support information;

[0081] Step A3: preprocessing the above data and performing feature extraction, including parameter normalization and principal component analysis;

[0082] In step A4, the historical data is divided into a training set and a test set. The designed model is trained using the training set. The state of the individual particle swarms is continuously optimized through the iterative process of the particle swarm optimization algorithm until the preset number of iterations is reached. The trained model is tested using the test set.

[0083] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0084] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An adjustable resource aggregation model for a virtual power plant, comprising a distributed energy resource parameter acquisition module, a communication module, a distributed energy resource parameter processing module, and a scheduling decision optimization module, characterized in that: The distributed energy resource parameter acquisition module is connected to the communication module, the communication module is bidirectionally connected to the distributed energy resource parameter processing module, and the communication module is connected to the scheduling decision optimization module; The distributed energy resource parameter processing module includes a distributed energy resource parameter storage unit and a distributed energy resource parameter analysis unit, which are used to store and analyze the collected distributed energy resource parameters; wherein the distributed energy resource parameter storage unit is connected to the distributed energy resource parameter analysis unit; the distributed energy resource parameter analysis unit uses a long short-term memory network to output the load demand of the distributed power system, the power generation of the distributed power generation equipment, and the charge and discharge power of the energy storage device. The prediction results are used to calculate and generate scheduling decision support information, wherein the scheduling decision support information includes a load demand balancing suggestion for the distributed power system, a power generation adjustment suggestion value, and a charge and discharge power suggestion value for the energy storage device. The load demand balancing suggestion includes a load reduction suggestion value and a load transfer suggestion value; wherein the formula for calculating the load reduction suggestion value is: Where ΔL cut (t) is the recommended load reduction value at time t, ΔL max is the maximum load reduction, L peak is the peak load threshold, L(t) is the load demand forecast value of the distributed power system; The formula for calculating the recommended load transfer value is: Where, L shift (t) is the load transfer recommendation value at time t, L(t) is the load demand forecast value of the distributed power system, and L peak is the peak load threshold, T shift is the total time of the load transfer period, and t is the current time in the load transfer period; The formula for calculating the recommended power generation adjustment value is: Where, P gen , i (t) is the recommended value for power generation adjustment of the i-th power generation equipment at time t, P predict , i is the predicted value of power generation of the ith distributed generation equipment, L(t) is the predicted value of load demand of the distributed power system, and L total is the total load of all power generation equipment at time t; The formula for calculating the recommended charging power value of the energy storage device is: Where, P charge (t) is the recommended charging power value of the energy storage device at time t, P predict , charge (t) is the predicted value of charging power of energy storage device, L(t) is the predicted value of load demand of distributed power system, L low is the low load threshold; The formula for calculating the recommended discharge power value of the energy storage device is: Where, P discharge (t) is the recommended discharge power value of the energy storage device at time t, P predict , discharge is the discharge power forecast value of the energy storage device, L(t) is the load demand forecast value of the distributed power system, and L high is the high load threshold; The scheduling decision optimization module is used to construct a scheduling decision optimization model based on the prediction results and scheduling decision support information generated by the distributed energy resource parameter processing module, using the particle swarm optimization algorithm, output and execute the scheduling strategy, and perform real-time scheduling and control of distributed energy resources.

2. The adjustable resource aggregation model for a virtual power plant according to claim 1, characterized in that: The maximum load reduction is the difference between the 95th percentile and the 75th percentile of the total load of the distributed power system; the high load threshold is the 80th percentile of the total load of the distributed power system; the low load threshold is the 20th percentile of the total load of the distributed power system; and the peak load threshold is the 90th percentile of the total load of the distributed power system.

3. The adjustable resource aggregation model for a virtual power plant according to claim 1, characterized in that: The distributed energy resource parameter acquisition module obtains the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, as well as the total load of the distributed power system and the charging and discharging power of the energy storage device in real time through sensors and monitoring equipment installed on the distributed energy resources, and transmits the above data to the communication module through a standardized interface.

4. The adjustable resource aggregation model for a virtual power plant according to claim 1, characterized in that: The distributed energy resource parameter analysis unit receives the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, the total load of the distributed power system and the charge and discharge power of the energy storage device collected by the distributed energy resource parameter acquisition module through the communication module, and inputs the constructed long and short-term memory network to obtain the load demand forecast value of the distributed power system, the power generation forecast value of the distributed power generation equipment and the charge and discharge power forecast value of the energy storage device.

5. The adjustable resource aggregation model for a virtual power plant according to claim 1 or 4, characterized in that: The steps of constructing the long short-term memory network are: Step S1, collecting historical data on power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of distributed power generation equipment, load of distributed power system, and charge and discharge power of energy storage devices; Step S2, preprocessing the above data and performing feature extraction; Step S3, designing the model structure of the long short-term memory network, determining that the input layer is used to receive the extracted feature data; the long short-term memory network layer is used to process time series data and capture long-term and short-term dependencies; the fully connected layer is used to convert the output of the long short-term memory network layer into a predicted value; and the output layer is used to generate a load demand forecast value for the distributed power system, a power generation forecast value for the distributed power generation equipment, and a charge and discharge power forecast value for the energy storage device; Step S4: divide the historical data into a training set and a test set, use the training set to train the designed long short-term memory network model, and use the test set to test the trained long short-term memory network.

6. The adjustable resource aggregation model for a virtual power plant according to claim 1, characterized in that: The scheduling decision optimization module uses a particle swarm optimization algorithm to construct a scheduling decision optimization model. The model generates a scheduling decision optimization plan by inputting the load demand of the distributed power system, the power generation of distributed power generation equipment, and the predicted value of the charge and discharge power of the energy storage device, as well as the real-time data of the scheduling decision support information. The scheduling decision optimization plan includes the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of the distributed power generation equipment, the total load of the distributed power system, and the charge and discharge power setting value of the energy storage device. The steps of constructing the scheduling decision optimization model using the particle swarm optimization algorithm are as follows: Step A1: Determine that the model inputs are the real-time predicted values ​​of the load demand of the distributed power system, the real-time predicted values ​​of the power generation of the distributed power generation equipment, the real-time predicted values ​​of the charge and discharge power of the energy storage device, and the real-time data of the scheduling decision support information; determine that the model objectives are to minimize the load fluctuation range, maximize the equipment operating time, and maximize the power generation utilization efficiency; and determine that the model output is the scheduling decision optimization plan; Step A2: The dispatching decision optimization module collects historical forecast values ​​of the load demand of the distributed power system, historical forecast values ​​of the power generation of the distributed power generation equipment, historical forecast values ​​of the charge and discharge power of the energy storage device, and historical data of the dispatching decision support information; Step A3: preprocessing the above data and performing feature extraction, including parameter normalization and principal component analysis; In step A4, the historical data is divided into a training set and a test set. The designed model is trained using the training set. The state of the individual particle swarms is continuously optimized through the iterative process of the particle swarm optimization algorithm until the preset number of iterations is reached. The trained model is tested using the test set.

7. The adjustable resource aggregation model for a virtual power plant according to claim 1, characterized in that: The distributed energy resource parameter storage unit manages and stores the basic attributes, prediction information and parameter format of distributed energy resources through a unified parameter model. The basic attributes include the power generation, temperature, voltage, current, power frequency, vibration frequency and amplitude of distributed power generation equipment, as well as the total load of the distributed power system and the charge and discharge power of the energy storage device; the prediction information includes the load demand prediction value of the distributed power system, the power generation prediction value of the distributed power generation equipment, the charge and discharge power prediction value of the energy storage device and scheduling decision support information; the parameter format includes parameter type, parameter unit and parameter accuracy.

8. The adjustable resource aggregation model for a virtual power plant according to claim 1, characterized in that: The communication module is used to transmit the parameters collected by the distributed energy resource parameter acquisition module to the distributed energy resource parameter processing module, and transmit the load demand forecast value of the distributed power system, the power generation forecast value of the distributed power generation equipment, the charging and discharging power forecast value of the energy storage device and the scheduling decision support information generated by the distributed energy resource parameter processing module to the scheduling decision optimization module.

Images