A virtual power plant joint scheduling method, device, equipment and medium

Abstract

The application discloses a virtual power plant joint scheduling method, device, equipment and medium, relates to the technical field of virtual power plant joint scheduling, and comprises the following steps: constructing a joint scheduling model of a virtual power plant, the joint scheduling model being used for scheduling a cooperative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air conditioner loads and a power grid under the constraint of constraint conditions, with the minimum scheduling cost as an optimization target, the constraint conditions comprising power balance constraints, energy storage equipment constraints, electric vehicle charging station constraints, electric vehicle battery swap station constraints, air conditioner load constraints and power grid interactive power constraints; and then solving the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant. By constructing and solving the joint scheduling model, the application can reduce the operation risk of the virtual power plant and improve the efficiency, stability, flexibility and reliability of the virtual power plant operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of virtual power plant joint scheduling, in particular to a virtual power plant joint scheduling method, device, equipment and medium. BACKGROUND

[0002] With the rapid development of renewable energy, virtual power plant (VPP) has gradually become an important scheduling mode in the power system. Virtual power plant integrates various distributed energy resources, such as photovoltaic power generation equipment, energy storage equipment, electric vehicle charging and swapping stations (including electric vehicle charging stations and electric vehicle swapping stations), air conditioning loads, etc., to provide flexible power supply solutions. By optimizing the scheduling of these distributed energy resources, virtual power plant can balance power supply and demand, optimize grid load, and provide stability for the power market. However, virtual power plant faces many technical challenges in practical application, especially the lack of effective joint scheduling scheme.

[0003] As a key regulating means in virtual power plant, energy storage equipment can effectively balance the volatility of photovoltaic power generation. However, traditional virtual power plant scheduling methods mostly focus on the coordination between energy storage equipment and the grid, ignoring the joint scheduling of energy storage equipment, renewable energy and electric vehicle charging and swapping stations. This single scheduling method cannot fully utilize the synergy of various resources, easily leading to low system operation efficiency and failing to achieve optimal resource allocation.

[0004] As an important controllable load in virtual power plant, air conditioning load has significant flexibility and adjustability, providing potential for scheduling optimization. By adjusting the operating power, start-stop state and set temperature, air conditioning load can quickly participate in demand response. At the same time, air conditioning load needs to balance between user comfort and scheduling economy, and excessive scheduling may cause user dissatisfaction. Despite this, air conditioning load can participate in demand response and collaborative optimization through flexibility, forming joint scheduling with photovoltaic power generation equipment, electric vehicle charging and swapping stations, and energy storage equipment, improving system operation efficiency and resource utilization, and achieving stability enhancement.

[0005] Most current virtual power plant scheduling methods still lack effective joint scheduling schemes, failing to collaboratively optimize multiple resources such as photovoltaic power generation equipment, electric vehicle charging and swapping stations, energy storage equipment and air conditioning load. Related virtual power plant scheduling methods usually treat various resources as independent scheduling objects and fail to perform global optimization within a unified framework. This lack of mutual coordination between resources increases the risk of system operation, and in a complex power market environment, single resource scheduling strategies are difficult to cope with rapidly changing power demand and supply conditions, resulting in a significant reduction in system flexibility and reliability. SUMMARY

[0006] The purpose of the present application is to provide a virtual power plant joint scheduling method, device, equipment and medium, which can reduce the operation risk of the virtual power plant and improve the efficiency, stability, flexibility and reliability of the virtual power plant operation.

[0007] To achieve the above-mentioned purpose, the present application provides the following solutions.

[0008] In a first aspect, the present application provides a virtual power plant joint scheduling method, which comprises:

[0009] constructing a joint scheduling model of the virtual power plant; the joint scheduling model is used to schedule the cooperative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air conditioning loads and a power grid under the constraint of constraint conditions, with the minimum scheduling cost as the optimization target, the constraint conditions comprising power balance constraints, energy storage equipment constraints, electric vehicle charging station constraints, electric vehicle battery swap station constraints, air conditioning load constraints and power grid interactive power constraints;

[0010] solving the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant; the joint scheduling scheme comprises the charging power and discharging power of the energy storage equipment, the charging power and discharging power of the electric vehicle, the charging power and discharging power of each battery in the electric vehicle battery swap station, the reducible power of the air conditioning load and the interactive power between the virtual power plant and the power grid, the interactive power being the power purchase power of the virtual power plant from the power grid or the power selling power of the virtual power plant to the power grid.

[0011] In a second aspect, the present application provides a virtual power plant joint scheduling device, which comprises:

[0012] a model construction module configured to construct a joint scheduling model of the virtual power plant; the joint scheduling model is used to schedule the cooperative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air conditioning loads and a power grid under the constraint of constraint conditions, with the minimum scheduling cost as the optimization target, the constraint conditions comprising power balance constraints, energy storage equipment constraints, electric vehicle charging station constraints, electric vehicle battery swap station constraints, air conditioning load constraints and power grid interactive power constraints;

[0013] a model solving module configured to solve the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant; the joint scheduling scheme comprises the charging power and discharging power of the energy storage equipment, the charging power and discharging power of the electric vehicle, the charging power and discharging power of each battery in the electric vehicle battery swap station, the reducible power of the air conditioning load and the interactive power between the virtual power plant and the power grid, the interactive power being the power purchase power of the virtual power plant from the power grid or the power selling power of the virtual power plant to the power grid.

[0014] In a third aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the computer program to implement the virtual power plant joint scheduling method.

[0015] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the virtual power plant joint scheduling method.

[0016] According to the specific embodiments provided by the present application, the present application has the following technical effects:

[0017] The present application provides a virtual power plant joint scheduling method, device, equipment and medium, a joint scheduling model of a virtual power plant is constructed, the joint scheduling model is used to schedule the collaborative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air conditioning loads and power grids under the constraint of the constraint condition, the constraint condition includes power balance constraint, energy storage equipment constraint, electric vehicle charging station constraint, electric vehicle battery swap station constraint, air conditioning load constraint and power grid interaction power constraint, and then the joint scheduling model is solved to obtain a joint scheduling scheme of the virtual power plant. The present application can globally optimize various resources in the virtual power plant in a unified framework by constructing and solving the joint scheduling model, can reduce the operation risk of the virtual power plant, improve the resource utilization rate, and thus improve the efficiency, stability, flexibility and reliability of the virtual power plant operation. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0019] Figure 1 An application environment diagram of a virtual power plant joint scheduling method provided by Embodiment 1 of the present application.

[0020] Figure 2 A flowchart of a virtual power plant joint scheduling method provided by Embodiment 1 of the present application.

[0021] Figure 3 An optimization strategy framework diagram provided by Embodiment 1 of the present application.

[0022] Figure 4 A rolling horizon optimization algorithm diagram provided by Embodiment 1 of the present application.

[0023] Figure 5 A virtual power plant multi-element joint scheduling strategy flowchart based on a rolling horizon optimization algorithm and a hybrid density neural network is provided for Embodiment 1 of the present application.

[0024] Figure 6 A functional module diagram of a virtual power plant joint scheduling device is provided for Embodiment 2 of the present application.

[0025] Figure 7 A structural diagram of a computer device is provided for Embodiment 3 of the present application. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0027] Embodiment 1.

[0028] The virtual power plant joint scheduling method provided by the embodiments of the present application can be applied in an application environment as shown in Figure 1 The terminal communicates with the server through the network. The data storage system can store the data required to be processed by the server. The data storage system can be separately arranged, or integrated on the server, or placed on the cloud or other servers. The terminal can send a joint scheduling request to be processed to the server. After receiving the joint scheduling request to be processed, the server constructs a joint scheduling model of the virtual power plant for the joint scheduling request to be processed; solves the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant. The server can feed back the joint scheduling scheme obtained for the joint scheduling request to the terminal as the joint scheduling result.

[0029] In addition, in some embodiments, the virtual power plant joint scheduling method can also be realized by the server or the terminal alone, such as being directly processed by the terminal for the joint scheduling request to be processed, or being processed by the server for the joint scheduling request to be processed obtained from the data storage system.

[0030] The terminal can be, but is not limited to, various desktop computers, notebook computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things device can be a smart speaker, a smart television, a smart air conditioner, a smart vehicle-mounted device, and the like. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, and the like. The server can be implemented by an independent server or a server cluster composed of multiple servers, and can also be a cloud server.

[0031] In an exemplary embodiment, as shown in Figure 2 , a virtual power plant joint scheduling method is provided, which is executed by a computer device, specifically, can be executed by a terminal or a server, or both. In the embodiments of the present application, the method is applied to the server in Figure 1 , and includes the following steps.

[0032] Step S1, a joint scheduling model of the virtual power plant is constructed; the joint scheduling model is used to schedule the collaborative operation process between the photovoltaic power generation device, the energy storage device, the electric vehicle charging station, the electric vehicle battery swap station, the air conditioning load, and the power grid under the constraint of a constraint condition, with the minimum scheduling cost as the optimization target, the constraint condition including power balance constraint, energy storage device constraint, electric vehicle charging station constraint, electric vehicle battery swap station constraint, air conditioning load constraint, and power grid interactive power constraint.

[0033] Step S2, the joint scheduling model is solved to obtain a joint scheduling scheme of the virtual power plant; the joint scheduling scheme includes the charging power and discharging power of the energy storage device, the charging power and discharging power of the electric vehicle, the charging power and discharging power of each battery in the electric vehicle battery swap station, the reducible power of the air conditioning load, and the interactive power between the virtual power plant and the power grid, the interactive power being the power purchase power of the virtual power plant from the power grid or the power selling power of the virtual power plant to the power grid.

[0034] By implementing the above steps S1 to S2, the present embodiment provides a virtual power plant joint scheduling method based on photovoltaic- energy storage- charging- battery swap- air conditioning load collaboration. By constructing a joint scheduling model of the virtual power plant and solving the joint scheduling model, a joint scheduling scheme of the virtual power plant is obtained. Since various resources in the virtual power plant are considered when the joint scheduling model is constructed, the global optimization of the various resources in the virtual power plant can be performed in a unified framework by solving the joint scheduling model. Subsequently, the virtual power plant is operated according to the joint scheduling scheme, thereby reducing the operation risk of the virtual power plant and improving the efficiency, stability, flexibility, and reliability of the virtual power plant operation.

[0035] As shown in Figure 3As shown, in this embodiment of the virtual power plant, photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swapping stations, air conditioning loads, and the power grid are all connected to the virtual power plant's DC bus. This embodiment builds a joint scheduling model that takes into account components such as electric vehicles. Based on the uncertainty of photovoltaic power generation, the charging and discharging constraints of energy storage equipment, the curtailment characteristics of air conditioning loads, and the charging and swapping behavior of electric vehicles, this embodiment specifically constructs a comprehensive joint scheduling model. This joint scheduling model considers the coordination and optimization of multiple resources (photovoltaic power generation equipment, energy storage equipment, electric vehicle charging and swapping stations, and air conditioning loads) with the goal of minimizing the scheduling cost of the virtual power plant. In this joint scheduling model, the charging and discharging strategies of energy storage equipment are optimized based on the volatility of grid load and photovoltaic power generation. While ensuring power balance and system stability, the joint scheduling model also considers the operating costs and resource constraints of various resources, ensuring efficient and economical scheduling of all resources.

[0036] The joint scheduling model constructed in this embodiment that takes into account the uncertainty of multiple scenarios includes an objective function and constraints. The objective function is to minimize the scheduling cost. The joint scheduling model is used to schedule the collaborative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air-conditioning loads and power grids with the minimum scheduling cost as the optimization goal under the constraints.

[0037] For the objective function, since it involves scene sampling and probability issues, the expected minimum scheduling cost can be used as the optimization goal. The expression of the objective function is as follows:

[0038] ;

[0039] in, is the scheduling period; is the number of scenes; for The probability of the scenario, for The centroid of the cluster corresponding to the scene; for The benefits of charging electric vehicles in this scenario; for The benefits gained from electric vehicle battery replacement in the scenario; for The cost of trading with the grid in the electricity market under the scenario; for Compensation expenses for electric vehicles in the scenario; for Depreciation costs of energy storage equipment operation in this scenario.

[0040] ;

[0041] in, for The number of electric vehicles participating in charging in the scenario; for Scene The battery state of charge of an electric vehicle when charging is completed; for Scene The battery state of charge of an electric vehicle when it starts charging; is the scheduling step length, which can be 1h in this embodiment; for Scene Energy capacity of electric vehicles; The charging unit price for electric vehicles can be charged according to time-of-use electricity prices, thereby improving the ability of car owners to respond to demand; The number of batteries in the electric vehicle battery swap station; for Electric vehicle swap station in the scenario The battery state of charge of the new battery after the battery of an electric vehicle is replaced; for Electric vehicle swap station in the scenario The battery state of charge of the old battery (i.e. the replaced depleted battery) after the battery of an electric vehicle is replaced; For the Energy capacity of a battery; The unit price of battery replacement for electric vehicles; for Virtual power plant and power grid in the scenario Interaction power of the time period; for Time-of-use electricity price of the power grid; for The on-grid electricity price of the power grid during the time period; for Incentive expenditure costs under the scenario; A coefficient for compensating electric vehicles; For energy storage equipment Discharge power during the time period; For energy storage equipment Charging power during the time period; is the unit depreciation cost of charging and discharging of energy storage equipment.

[0042] The embodiment designs an incentive mechanism for electric vehicles. In order to encourage electric vehicle charging stations to participate in virtual power plant dispatching, the embodiment designs a comprehensive incentive mechanism. For electric vehicles, the incentive mechanism sets dynamic electricity prices in combination with charging time periods, battery health conditions, charging demands of electric vehicles, and the like, and gives electric vehicle charging stations certain electricity price incentives to guide electric vehicles to charge according to grid demands. This mechanism can not only reduce the impact of charging on grid load, but also delay battery degradation and improve the service life of electric vehicles.

[0043] Specifically, the electric vehicle incentive scheme designed in the embodiment is as follows: In order to encourage electric vehicle owners to participate in VPP dispatching, an innovative compensation plan is proposed to adjust incentive measures and maximize the use of available storage capacity in the VPP and reduce the pressure brought by dispatching uncertainty. The owner can obtain a benefit as long as he agrees to participate in VPP dispatching and makes the VPP operator discharge the battery, without considering the parking time of the electric vehicle. The participation degree of the electric vehicle can be dynamically adjusted through the incentive mechanism.

[0044] A battery aging model is established to calculate the battery wear compensation of the electric vehicle owner participating in the VPP discharging process, and the expression is as follows:

[0045] ;

[0046] wherein, is the battery wear compensation; is the attenuation rate of the battery for each charging and discharging cycle; is the battery discharging energy of the battery in the period; is the maximum energy of the battery when full; is the depth of discharge of the battery in the period; is the total cost of the battery.

[0047] Based on this, the incentive scheme is designed to compensate for the wear of the electric vehicle battery and encourage it to participate in grid services. The incentive scheme can be set as:

[0048] ;

[0049] wherein, is the incentive expenditure in the scenario; is the battery wear compensation coefficient; is the battery participation in grid service compensation coefficient; is the incentive coefficient for participating in grid service discharging (such as frequency support, load balancing, etc.) in the period; is the In the scenario, VPP Time-scheduled energy dispatching of electric vehicles; is the total capacity of the electric vehicle battery.

[0050] As for the constraints, they include power balance constraints, energy storage equipment constraints, electric vehicle charging station constraints, electric vehicle battery swap station constraints, air conditioning load constraints, and grid interaction power constraints.

[0051] The expression of the power balance constraint within the virtual power plant is as follows:

[0052] ;

[0053] in, for Photovoltaic power generation equipment in the scene PV power generation during the period; For energy storage equipment Discharge power during the time period; for The number of electric vehicles participating in charging in the scenario; For the Electric vehicles in Discharge power during the time period; for Virtual power plant and power grid in the scenario The interactive power of the time period, including the power purchased from the grid (positive value) or the power sold to the grid (negative value); The number of batteries in the electric vehicle battery swap station; For the Batteries in Discharge power during the time period; for Virtual power plant in the scenario Base load demand during the period, which may include normal load and air-conditioning load; For energy storage equipment Charging power during the time period; For the Electric vehicles in Charging power during the time period; For air conditioning load The curtailable power during the time period is the amount of air conditioning load reduction that the virtual power plant operator can implement; For the Batteries in Charging power during the time period.

[0054] Energy storage device constraints include charge and discharge power limits, SOC (State of Charge) updates, and SOC constraints. The expressions are as follows:

[0055] ;

[0056] in, is the maximum charging power of the energy storage device; is the first 0-1 state variable, representing the energy storage device The charging status of the time period is 1 for charging and 0 for discharging, ensuring that the energy storage device cannot be charged and discharged at the same time; is the maximum discharge power of the energy storage device; For energy storage equipment Battery state of charge during the time period; For energy storage equipment Battery state of charge during the time period; Charging efficiency of energy storage devices; is the discharge efficiency of the energy storage device; is the minimum battery state of charge of the energy storage device; is the maximum battery state of charge of the energy storage device; The battery state of charge of the energy storage device in one period; For energy storage equipment Battery state of charge during the time period, The scheduling period.

[0057] The constraints of electric vehicle charging stations include charging power limit, discharging power limit, SOC update, SOC constraint, energy demand satisfaction, charging window, etc. The expressions are as follows:

[0058] ;

[0059] in, For the Maximum charging power of an electric vehicle; is the second 0-1 state variable, representing the Electric vehicles in The charging status of the time period is 1 for charging and 0 for discharging, ensuring that the electric vehicle cannot be charged and discharged at the same time; For the The maximum discharge power of an electric vehicle. If the owner of the electric vehicle does not accept the VPP unified dispatch, then is 0; is the third 0-1 state variable, representing In the scenario, Is the time period in The charging window of an electric vehicle, that is, whether it is parked at an electric vehicle charging station, 1 represents yes, 0 represents no; For the The starting time of the charging window of an electric vehicle, that is, the starting time of the parking; For the The end of the charging window for each electric vehicle, i.e. the end of the docking period; for Scene Electric vehicles in Battery state of charge during the time period; for Scene Electric vehicles in Battery state of charge during the time period; For the Charging efficiency of electric vehicles; For the The discharge efficiency of an electric vehicle; For the Minimum battery state of charge for an electric vehicle; For the Maximum battery state of charge for an electric vehicle; for Scene Electric vehicles in Battery state of charge during the time period; for Scene The target battery state of charge for an electric vehicle.

[0060] The constraints of electric vehicle battery swap stations include battery swap requirements and energy storage battery constraints, which are expressed as follows:

[0061] ;

[0062] in, For the Maximum charging power of each battery; is the fourth 0-1 state variable, representing the Batteries in The charging status of the time period, 1 represents charging, and 0 represents discharging; is the fifth 0-1 state variable, representing the Batteries in Whether a charger is assigned to the time period. 1 indicates that a charger is assigned, and 0 indicates that a charger is not assigned. For the The maximum discharge power of a battery is usually taken in engineering practice. The maximum charging power and maximum discharging power of each battery are equal, and the specific value depends on the battery model; The number of batteries in the electric vehicle battery swap station; is the number of chargers; is the sixth 0-1 state variable, representing whether the battery can provide battery swap service in the time period, 1 represents that the battery can provide battery swap service, and 0 represents that the battery cannot provide battery swap service, and the battery swap service can be provided only when the state of charge of the battery is greater than 95%; is the seventh 0-1 state variable, representing whether the battery provides battery swap service in the time period, 1 represents that the battery provides battery swap service, and 0 represents that the battery does not provide battery swap service; is the eighth 0-1 state variable, representing whether the battery provides battery swap service for the electric vehicle in the time period, 1 represents that the battery provides battery swap service, and 0 represents that the battery does not provide battery swap service; is the ninth 0-1 state variable, representing whether the battery provides battery swap service for the electric vehicle in the time period, 1 represents that the battery provides battery swap service, and 0 represents that the battery does not provide battery swap service; is the tenth 0-1 state variable, representing the number of electric vehicles arriving at the battery swap station in the time period in the scenario; is the eleventh 0-1 state variable, representing the state of charge of the battery in the time period in the scenario; is the twelfth 0-1 state variable, representing the initial state of charge of the electric vehicle coming for battery swap in the time period in the scenario; is the thirteenth 0-1 state variable, representing the state of charge of the battery in the time period in the scenario; is the charging efficiency of the battery; is the discharging efficiency of the battery; is the minimum state of charge of the battery; is the maximum state of charge of the

[0063] For the air conditioning load constraint, since the air conditioning load can be reduced by a small amount in a short time without affecting the user's use, it is modeled as an air conditioning type of reducible load, including air conditioning reduction amount limit and total air conditioning reduction amount limit in the scheduling period, and the expression is as follows:

[0064] ;

[0065] ;

[0066] wherein, is the maximum value of the curtailed power of the air conditioning load in the time period; is the maximum value of the cumulative curtailed power of the air conditioning load.

[0067] The grid interaction power constraint includes a grid interaction power limit, expressed as follows:

[0068] ;

[0069] wherein, is the absolute value of the maximum value of the interaction power between the virtual power plant and the grid, i.e., the maximum value of the grid connection line power between the grid and the virtual power plant.

[0070] Before solving the joint scheduling model, the embodiment first needs to collect the basic data related to the virtual power plant, including photovoltaic power generation data, meteorological data, electric vehicle charging and battery swapping station data, energy storage device operation data, and air conditioning load data. The photovoltaic power generation data includes historical photovoltaic power generation, meteorological conditions (such as solar radiation intensity), etc. The electric vehicle charging and battery swapping station data includes the charging and battery swapping demand of the charging and battery swapping station, the charging and battery swapping time period, etc. All data needs to be preprocessed to ensure the integrity and consistency of the data. In addition, all data also needs to be normalized to eliminate the scale difference of different data sources and ensure the comparability and effectiveness of various data in the model.

[0071] The joint scheduling model is solved to obtain the joint scheduling scheme of the virtual power plant, which includes the charging power and the discharging power of the energy storage device, the charging power and the discharging power of the electric vehicle, the charging power and the discharging power of each battery in the electric vehicle battery swapping station, the curtailed power of the air conditioning load, and the interaction power between the virtual power plant and the grid, which is the power purchase power of the virtual power plant from the grid or the power sale power of the virtual power plant to the grid.

[0072] ​The virtual power plant still faces the technical challenge of source-load uncertainty in practical applications. The volatility and uncertainty of renewable energy sources such as photovoltaic power generation is one of the biggest problems in virtual power plant scheduling. Photovoltaic power generation is affected by multiple factors such as weather, time, and season, and its output has significant uncertainty. Currently, many virtual power plant scheduling models still use prediction methods based on historical data, assuming that photovoltaic power generation can be accurately predicted. However, in reality, the volatility and sudden changes in photovoltaic power generation make this assumption often invalid. Lack of sufficient consideration of uncertainty can lead to poor robustness of virtual power plant scheduling strategies, making them overly dependent on energy storage devices or the power grid, thereby affecting overall revenue and system stability. The charging and discharging behavior of electric vehicles also brings source-load uncertainty. The charging and discharging demand of electric vehicles is not only related to the real-time usage status of the vehicle and the load of the charging facility, but also affected by individual factors such as the charging and discharging habits of the vehicle owner, the selection of charging and discharging time, and the initial state of charge of the electric vehicle when it arrives at the charging station. Currently, most scheduling models do not effectively consider the dynamic demand characteristics of electric vehicles, and lack optimization management of electric vehicle charging time and charging power, making it difficult for virtual power plants to achieve optimal revenue when scheduling electric vehicles, and even potentially causing adverse effects on electric vehicle batteries.

[0073] To solve the above problems, the embodiment builds a photovoltaic power generation uncertainty prediction model based on a mixture density neural network (MDN). To accurately capture the volatility and uncertainty of photovoltaic power generation, the embodiment uses a mixture density neural network to predict photovoltaic power generation. Unlike traditional single prediction methods, MDN can output the probability distribution of photovoltaic power generation, rather than simple point estimates, which is particularly important for subsequent scheduling optimization. MDN learns the complex relationship between photovoltaic power generation and meteorological data and their time series relationship through the structure of the neural network, generating parameters (mean, variance, and weight) of multiple Gaussian distributions to describe the possible range of photovoltaic power generation. During model training, input data includes historical photovoltaic power generation data, meteorological data, and other related factors. MDN is trained through the backpropagation algorithm to ensure high accuracy and robustness of the prediction results. Similarly, an electric vehicle charging uncertainty prediction model and an electric vehicle battery replacement uncertainty prediction model are established to use MDN to predict electric vehicle load data (i.e. electric vehicle charging data and electric vehicle battery replacement data), so that the volatility of photovoltaic power generation and electric vehicle load data is modeled through MDN prediction results.

[0074] At this time, the joint scheduling model is solved in the embodiment to obtain a joint scheduling scheme of the virtual power plant, which includes the following steps.

[0075] (1) Taking the meteorological data of the scheduling period as input, the photovoltaic power generation uncertainty prediction model is used to predict the photovoltaic power generation data distribution of the scheduling period.

[0076] (2) Taking the historical scheduling period electric vehicle charging data as input, the electric vehicle charging uncertainty prediction model is used to predict the electric vehicle charging data distribution of the scheduling period.

[0077] (3) Taking the historical scheduling period electric vehicle battery swap data as input, the electric vehicle battery swap uncertainty prediction model is used to predict the electric vehicle battery swap data distribution of the scheduling period.

[0078] (4) Randomly sampling the photovoltaic power generation data distribution, the electric vehicle charging data distribution and the electric vehicle battery swap data distribution respectively, to obtain the photovoltaic power generation, the electric vehicle charging data and the electric vehicle battery swap data of the scheduling period under multiple scenarios, the number of scenarios is the same as the number of random sampling times, the electric vehicle charging data includes the initial battery state of charge of each electric vehicle participating in charging and the charging window, the charging window includes the start time period and the end time period , the electric vehicle battery swap data includes the initial battery state of charge of each electric vehicle participating in battery swap and the battery swap window, the battery swap window includes the start time period and the end time period, and the decision .

[0079] (5) Taking the photovoltaic power generation, the electric vehicle charging data and the electric vehicle battery swap data of the scheduling period under multiple scenarios as input, the joint scheduling model is optimized and solved to obtain the joint scheduling scheme of the virtual power plant.

[0080] In this embodiment, the photovoltaic power generation uncertainty prediction model, the electric vehicle charging uncertainty prediction model and the electric vehicle battery swap uncertainty prediction model all adopt the hybrid density neural network, and the hybrid density neural network adopts the negative log-likelihood loss function and the Adam (Adaptive Moment Estimation) optimizer during training.

[0081] Specifically, the embodiment first performs data acquisition and cleaning. Since it involves photovoltaic power generation and electric vehicle load data prediction tasks, before deployment, the photovoltaic power generation data, electric vehicle load data and corresponding meteorological data of the virtual power plant to be scheduled should be collected, and missing value and abnormal value processing and other cleaning steps should be performed. Optionally, all data can also be normalized to eliminate the scale difference of different data sources.

[0082] The embodiment further designs a hybrid density neural network, which specifically includes the following steps.

[0083] (1) Design the neural network structure. First, the MDN should contain an input layer, which inputs the relevant features of photovoltaic power generation and the relevant features of electric vehicle load data. Second, the MDN should contain a hidden layer, which may need to be designed in detail according to the characteristics of the data. In this embodiment, a three-layer hidden layer is adopted, and the number of neurons in each layer decreases, with 128 neurons in the first layer, 64 neurons in the second layer, and 32 neurons in the third layer. ReLU (Rectified Linear Unit) is used as the activation function of the hidden layer, which has the advantages of high computational efficiency and the ability to avoid gradient disappearance problems, and the expression is as follows:

[0084] ;

[0085] wherein x is the input value.

[0086] The connection between layers uses a standard fully connected layer (Dense Layer), that is, each neuron in the hidden layer is connected to all neurons in the previous layer. Finally, the MDN should contain an output layer, and the goal of the output layer of the MDN is to predict the mean, standard deviation, and weight of each Gaussian distribution component in the Gaussian Mixture Model (GMM). The output layer needs to output these parameters, and needs to ensure that the standard deviation is positive, the weight is non-negative, and the sum of the weights is 1. Therefore, the Softplus and Softmax activation functions are used, and the mean is directly output. The mathematical expressions are as follows:

[0087] ;

[0088] wherein K is the number of Gaussian distribution components, and the number of nodes in the output layer is K+2K+1. , respectively output the mean , standard deviation , and weight of each Gaussian distribution component. is the original output value of the i-th node; is the original output value of the i-th node; is the original output value of the i-th node; is the original output value of the i-th node.

[0089] ​​​​​​Therefore, the mixture density neural network includes an input layer, a hidden layer, and an output layer connected in sequence. If you want to capture the time series relationship between the data, you can add a long short-term memory neural network layer or a bidirectional long short-term memory neural network layer for temporal memory. Specifically, you can place it between the input layer and the hidden layer.

[0090] (2) Design the loss function. This embodiment uses the negative log-likelihood loss function. For each training sample, its log-likelihood under the current mixed Gaussian model needs to be calculated. The specific expression is as follows:

[0091] ;

[0092] in, for loss; is the number of training samples; For the The original output value of each node; For the The true label value of each sample (the true value of photovoltaic power generation or the true value of electric vehicle load data); For the The probability density formula of the Gaussian distribution components is as follows:

[0093] ;

[0094] This loss function minimizes the difference between the true label value and the mixture Gaussian distribution predicted by the model, thereby optimizing the network parameters.

[0095] (3) Design a hyperparameter tuning algorithm. This embodiment uses the currently commonly used Adam optimizer for optimization.

[0096] Finally, this embodiment uses the designed MDN model to predict the photovoltaic power generation data distribution and electric vehicle load data distribution (including electric vehicle charging data distribution and electric vehicle battery replacement data distribution) for each period in the scheduling cycle.

[0097] In this embodiment, random sampling is performed on the photovoltaic power generation data distribution, the electric vehicle charging data distribution, and the electric vehicle battery replacement data distribution, respectively. Specifically, the Monte Carlo sampling method is used to randomly sample the photovoltaic power generation data distribution, the electric vehicle charging data distribution, and the electric vehicle battery replacement data distribution, respectively.

[0098] Several random scenarios are obtained through Monte Carlo sampling, which specifically includes the following steps.

[0099] (1) Extracting components from Gaussian distribution components. Selecting one component from these components based on the weight of each Gaussian distribution component can be done by random number method.

[0100] Let the cumulative distribution function of the weights be:

[0101] ;

[0102] wherein, is the cumulative value of the th Gaussian distribution component; is the weight of the th Gaussian distribution component.

[0103] Generate a random number uniformly distributed , select the th Gaussian distribution component, so that:

[0104] ;

[0105] wherein, is the cumulative value of the th Gaussian distribution component.

[0106] (2) Sampling from the normal distribution of the selected component, once the th Gaussian distribution component is selected, a sample is extracted from the Gaussian distribution corresponding to the component.

[0107] A standard normal random variable can be generated, and a linear transformation is used to obtain, that is:

[0108] ;

[0109] The obtained is the prediction value of the current period.

[0110] (3) Repeat the above process for each period in the scheduling period to obtain a sufficient number of samples to form a scene matrix.

[0111] Assuming that the number of sampled scenes is , the obtained scene matrix is:

[0112] ;

[0113] wherein, is the scheduling period, which is also the window length in the random backoff view optimization, and in this embodiment, = 1h x 24, the scheduling step is 1h, and the scheduling period is 24 scheduling steps, that is, the scheduling period is 1 day, and the obtained scene matrix Each row is a sampled scene, and each column is the sampling result for the period.

[0114] Through the above sampling process, the distribution of photovoltaic power generation and electric vehicle load data can be converted into specific values of photovoltaic power generation and electric vehicle load data under different scenarios.

[0115] In this embodiment, the photovoltaic power generation, electric vehicle charging data and electric vehicle battery replacement data of the scheduling period under multiple scenarios are taken as inputs to solve the joint scheduling model to obtain the joint scheduling scheme of the virtual power plant, specifically including: clustering the photovoltaic power generation, electric vehicle charging data and electric vehicle battery replacement data of the scheduling period under multiple scenarios by using a clustering algorithm to obtain the photovoltaic power generation, electric vehicle charging data and electric vehicle battery replacement data of the scheduling period under multiple clusters; taking the photovoltaic power generation, electric vehicle charging data and electric vehicle battery replacement data of the scheduling period under multiple clusters as inputs to solve the joint scheduling model to obtain the joint scheduling scheme of the virtual power plant.

[0116] In this embodiment, the sampled scenarios are clustered. If the uncertainty of the virtual power plant is too high, the sampling function distribution (i.e. ) is relatively scattered, and the number of sampling scenarios needs to be increased. Then the clustering algorithm is used for scenario reduction. The clustering algorithms that can be selected include K-means algorithm and density-based spatial clustering of applications with noise (DBSCAN for short). In this embodiment, the K-means algorithm is taken as an example, and the clustering process includes the following steps.

[0117] (1) First, initialization is performed, and initial cluster centers (centroids) are selected. These initial centroids can be obtained by randomly selecting data points in the scenario matrix or other heuristic methods (such as K-means++).

[0118] (2) Each data point is assigned to the nearest cluster center, i.e., the distance between the data point and the cluster center is minimized (usually the Euclidean distance is used).

[0119] The expression of the distance is as follows:

[0120] ;

[0121] Wherein, is the distance between the th row vector in the scenario and the centroid of the th cluster; is the total number of features, i.e., the column number of ; is the th feature of the scenario ; is the th cluster center. centroid of the cluster the first feature.

[0122] Then the data points are assigned to the nearest cluster, i.e.

[0123] ;

[0124] wherein, is the index of the cluster to which the scene belongs; is the centroid of the first cluster. (3) Update the centroid.

[0125] For each cluster, calculate the new centroid of the cluster, which is the mean of all scenes in the cluster, i.e.

[0126]

[0127] ; wherein,

[0128] is the number of data points in the first cluster; represents all data points in the first cluster. (4) Repeat (2) and (3) until the position of the centroid no longer changes or changes very little, considering that the algorithm converges, obtaining

[0129] clusters, corresponding to scenes.

[0130] (5) Assign a probability to each scene.

[0131] The probability of each scene (centroid of the cluster) is proportional to the size of the cluster, and its expression is as follows:

[0132] ;

[0133] wherein, is the probability of the first cluster.

[0134] The embodiment can also predict the air conditioner load, taking the air conditioner load of the historical scheduling period and the air temperature of the current scheduling period as input, using MDN to predict the air conditioner load distribution of the current scheduling period, determining the value of the air conditioner load through sampling and clustering method. Of course, the air conditioner load can also be a known value.

[0135] ​In this embodiment, the joint scheduling model is solved to obtain the joint scheduling plan of the virtual power plant, specifically including: optimizing and solving the joint scheduling model using the RHC (Rolling Horizon Control) algorithm to obtain the joint scheduling plan of the virtual power plant.

[0136] Specifically, this embodiment uses an RHC-based scheduling algorithm for solution. In order to solve the virtual power plant scheduling problem, this embodiment uses the RHC algorithm for solution. The RHC algorithm introduces a fallback mechanism in multiple time steps and combines different forecast scenarios of photovoltaic power generation to perform dynamic optimization. In each time period, the RHC algorithm selects the optimal scheduling decision based on the current forecast information and real-time data, and continuously updates and adjusts the scheduling strategy in the future rolling optimization process. The RHC algorithm is able to handle multiple uncertain factors (such as uncertainty in photovoltaic power generation, electric vehicle charging and swapping, etc.), and perform robust optimization by considering the probability distribution of future scenarios. Through iterative solution, the RHC algorithm can provide a more stable and efficient scheduling solution under uncertain conditions to ensure the maximization of the overall benefits of the virtual power plant.

[0137] This embodiment uses the RHC algorithm to solve the joint scheduling model, such as Figure 4 and Figure 5 As shown, to effectively address the uncertainties of photovoltaic power generation, electric vehicle charging demand, and load reduction, the RHC algorithm dynamically updates the scheduling plan within each rolling time step to ensure the maximum benefit of the virtual power plant. Specifically, the following steps are used to solve the joint scheduling model: First, the scheduling period and scheduling step are set. This embodiment adopts a scheduling period of 24 hours and a scheduling step of 1 hour. Second, the established MDN model is called to predict the distribution of photovoltaic power generation and electric vehicle load data within the time horizon. Then, the proposed sampling method is used to reduce scenarios to obtain typical scenarios and their distribution probabilities. Finally, the solver is called to solve the established mixed integer linear programming model. This embodiment specifically uses the Cplex solver for this solution. After obtaining the optimal joint scheduling plan for dispatchable components such as energy storage and electric vehicles, the joint scheduling plan corresponding to the scheduling step is executed, and the above process is repeated.

[0138] In view of the obvious deficiencies of the related art in source load uncertainty modeling and joint scheduling scheme design, in order to improve the overall performance and reliability of virtual power plant scheduling, it is urgent to develop an optimization scheduling method that can fully consider the collaborative effect of various resources and effectively cope with source load uncertainty, which not only improves the economic benefit of virtual power plant, but also enhances the adaptability of the system to external disturbances. Based on this, the embodiment proposes a virtual power plant optimization scheduling method based on rolling horizon optimization and hybrid density neural network prediction. The core technology adopts RHC optimization to make scheduling decisions for virtual power plants containing photovoltaic power generation equipment, energy storage equipment, air conditioning load, and electric vehicle charging and swapping stations. MDN is used to predict photovoltaic power generation uncertainty, and an incentive mechanism for electric vehicle charging and a swapping station model are designed to realize the joint scheduling of multiple energy resources and minimize the scheduling cost of virtual power plants. This optimization method based on RHC and MDN prediction can effectively handle the volatility and uncertainty of photovoltaic power generation, fully coordinate photovoltaic power generation equipment, energy storage equipment, electric vehicle charging and swapping stations, and air conditioning load, and improve the operation efficiency and economy of virtual power plants.

[0139] Compared with the prior art, the embodiment has significant advantages in the construction and solution of the virtual power plant scheduling model, especially in the innovation of uncertainty handling, multi-component joint scheduling, and incentive mechanism design, which greatly improves the scheduling efficiency and resource optimization capability of virtual power plants.

[0140] In terms of multi-component joint scheduling, the embodiment jointly schedules multiple energy resources such as photovoltaic power generation equipment, energy storage equipment, electric vehicle charging and swapping stations, and air conditioning load. Traditional virtual power plant scheduling methods usually focus on the optimization of a single component and ignore the collaborative effect between resources, resulting in unsatisfactory scheduling results. However, the embodiment realizes dynamic collaboration between components by introducing an incentive mechanism for electric vehicles. Electric vehicles participate in load scheduling through the charging incentive mechanism, energy storage equipment performs charging and discharging operations according to the uncertainty of photovoltaic power generation, and air conditioning load reduces electricity consumption through demand response mechanisms. Through this multi-component collaborative scheduling method, not only can the power supply and demand be balanced, but also the overall economic benefit of the virtual power plant can be improved.

[0141] In terms of uncertainty processing, the embodiment effectively solves the uncertainty problem of photovoltaic power generation by using MDN to predict the probability distribution of photovoltaic power generation. Traditional photovoltaic power generation prediction can usually only give a single point estimate, while MDN can generate multiple Gaussian distribution parameters to reflect the possible fluctuation range of photovoltaic power generation, which provides more accurate input for subsequent scheduling decisions. In addition, the embodiment combines the RHC algorithm, which not only relies on the current prediction information in the solving process, but also fully considers the probability distribution of different scenarios in the future, making the scheduling scheme more robust. The RHC algorithm dynamically optimizes the scheduling decision based on the consideration of photovoltaic power generation volatility, energy storage device charging and discharging constraints, electric vehicle charging demand and air conditioning load, which can effectively cope with various uncertain factors and ensure the efficient operation of the virtual power plant.

[0142] In terms of incentive mechanism design, the embodiment proposes a flexible incentive mechanism to encourage flexible load resources such as electric vehicles to participate in the scheduling optimization of the virtual power plant. For electric vehicles, the embodiment designs a dynamic electricity price incentive scheme based on battery health status and charging demand, encouraging electric vehicles to reasonably arrange charging time and charging amount according to grid load conditions. This mechanism not only avoids the pressure on the grid caused by concentrated electric vehicle charging, but also prolongs the service life of the battery and reduces the risk of battery degradation. Specifically, when participating in scheduling, electric vehicles will be provided with corresponding incentive electricity prices according to their charging demand, battery status and charging period, ensuring that the charging process meets the load demand of the grid and takes into account the economic benefits of electric vehicle users.

[0143] In summary, the embodiment effectively solves the problems of uncertainty processing and resource optimization scheduling by using MDN to predict photovoltaic power generation uncertainty, RHC algorithm for scheduling optimization, and joint scheduling of multiple energy components. In addition, by designing a reasonable incentive mechanism, the embodiment can promote the participation of flexible load resources such as electric vehicles and air conditioning loads, further improving the scheduling efficiency and economy of the virtual power plant. These innovations make the application of the embodiment in the context of virtual power plants have high practical value, which can effectively improve the robustness, flexibility and overall benefit of virtual power plant scheduling.

[0144] The application also provides an application scenario of the virtual power plant joint scheduling method. Specifically, the virtual power plant joint scheduling method provided by the embodiment can be applied in a virtual power plant joint scheduling scenario. The virtual power plant joint scheduling scenario includes a scheme generation link and a running link, the scheme generation link is used to establish and solve a joint scheduling model of the virtual power plant to obtain a joint scheduling scheme, and the running link is used to control the virtual power plant to run according to the joint scheduling scheme. The virtual power plant joint scheduling method provided by the embodiment belongs to the scheme generation link.

[0145] Embodiment 2.

[0146] Based on the same inventive concept, the embodiments of the present application further provide a virtual power plant joint scheduling device for implementing the virtual power plant joint scheduling method described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more virtual power plant joint scheduling device embodiments provided below can refer to the limitations of the virtual power plant joint scheduling method described above, which will not be repeated here.

[0147] In an exemplary embodiment, as shown in Figure 6 A virtual power plant joint scheduling device is provided, which includes the following modules.

[0148] A model construction module M1 is configured to construct a joint scheduling model of a virtual power plant; the joint scheduling model is configured to schedule a collaborative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air conditioning loads and a power grid under the constraint of an optimization objective of minimizing scheduling cost and constraint conditions, the constraint conditions including power balance constraints, energy storage equipment constraints, electric vehicle charging station constraints, electric vehicle battery swap station constraints, air conditioning load constraints and power grid interactive power constraints.

[0149] A model solution module M2 is configured to solve the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant; the joint scheduling scheme includes charging power and discharging power of the energy storage equipment, charging power and discharging power of the electric vehicle, charging power and discharging power of each battery in the electric vehicle battery swap station, reducible power of the air conditioning load, and interactive power between the virtual power plant and the power grid, the interactive power being power purchase power of the virtual power plant from the power grid or power selling power of the virtual power plant to the power grid.

[0150] Embodiment 3.

[0151] In an exemplary embodiment, a computer device is provided, which can be a server or a terminal, and an internal structure diagram of the computer device can be as shown in Figure 7As shown in the figure. The computer device includes a processor, a memory, an Input / Output (I / O) interface and a communication interface. Among them, the processor, the memory and the input / output interface are connected through the system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with the terminal outside through the network connection. The computer program is executed by the processor to realize a virtual power plant joint scheduling method.

[0152] Those skilled in the art can understand that, Figure 7 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0153] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to realize the virtual power plant joint scheduling method in embodiment 1.

[0154] Embodiment 4.

[0155] In one exemplary embodiment, a computer readable storage medium is provided, storing a computer program, which is executed by a processor to realize the virtual power plant joint scheduling method in embodiment 1.

[0156] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0157] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of these technical features does not exist contradictory, it should be considered as the scope of the present application.

[0158] The principles and implementation manners of the present application are described herein by using specific examples, and the above examples are only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will have changes. In conclusion, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A virtual power plant joint scheduling method, characterized in that, The virtual power plant joint scheduling method comprises the following steps: a joint scheduling model of the virtual power plant is constructed; the joint scheduling model is used to schedule a cooperative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swap stations, air conditioning loads and a power grid under the constraint of a constraint condition, with a minimum scheduling cost as an optimization target, the constraint condition comprising a power balance constraint, an energy storage equipment constraint, an electric vehicle charging station constraint, an electric vehicle battery swap station constraint, an air conditioning load constraint and a power grid interactive power constraint; the joint scheduling model is solved to obtain a joint scheduling scheme of the virtual power plant; the joint scheduling scheme comprises charging power and discharging power of the energy storage equipment, charging power and discharging power of the electric vehicle, charging power and discharging power of each battery in the electric vehicle battery swap station, a reducible power of the air conditioning load and interactive power of the virtual power plant and the power grid, the interactive power being power purchase power of the virtual power plant from the power grid or power selling power of the virtual power plant to the power grid; the joint scheduling model is solved to obtain a joint scheduling scheme of the virtual power plant, specifically comprising the following steps: weather data of a scheduling period is taken as input to predict photovoltaic power generation data distribution of the scheduling period by using a photovoltaic power generation uncertainty prediction model; electric vehicle charging data of a historical scheduling period is taken as input to predict electric vehicle charging data distribution of the scheduling period by using an electric vehicle charging uncertainty prediction model; electric vehicle battery swap data of a historical scheduling period is taken as input to predict electric vehicle battery swap data distribution of the scheduling period by using an electric vehicle battery swap uncertainty prediction model; the photovoltaic power generation data distribution, the electric vehicle charging data distribution and the electric vehicle battery swap data distribution are randomly sampled respectively to obtain photovoltaic power generation, electric vehicle charging data and electric vehicle battery swap data of the scheduling period under multiple scenarios; the number of scenarios is the same as the number of random sampling times, the electric vehicle charging data comprises initial battery state of charge and a charging window of each electric vehicle participating in charging, and the electric vehicle battery swap data comprises initial battery state of charge and a battery swap window of each electric vehicle participating in battery swap; the photovoltaic power generation, the electric vehicle charging data and the electric vehicle battery swap data of the scheduling period under multiple scenarios are taken as input to solve the joint scheduling model to obtain the joint scheduling scheme of the virtual power plant.

2. The virtual power plant joint scheduling method according to claim 1, characterized in that, The power balance constraint is: ； in, for Photovoltaic power generation equipment in the scene PV power generation during the period; For energy storage equipment Discharge power during the time period; for The number of electric vehicles participating in charging in the scenario; For the Electric vehicles in Discharge power during the time period; for Virtual power plant and power grid in the scenario Interaction power of the time period; The number of batteries in the electric vehicle battery swap station; For the Batteries in Discharge power during the time period; for Virtual power plant in the scenario Basic load demand during the period, which includes normal load and air-conditioning load; For energy storage equipment Charging power during the time period; For the Electric vehicles in Charging power during the time period; For air conditioning load The power that can be reduced during the period; For the Batteries in Charging power during the time period.

3. The virtual power plant joint scheduling method according to claim 1, characterized in that, The photovoltaic power generation uncertainty prediction model, the electric vehicle charging uncertainty prediction model and the electric vehicle battery swap uncertainty prediction model all adopt a hybrid density neural network, and the hybrid density neural network adopts a negative log-likelihood loss function and an Adam optimizer during training.

4. The virtual power plant joint scheduling method according to claim 1, characterized in that, The photovoltaic power generation data distribution, the electric vehicle charging data distribution and the electric vehicle battery swap data distribution are randomly sampled respectively, specifically comprising the following steps: the photovoltaic power generation data distribution, the electric vehicle charging data distribution and the electric vehicle battery swap data distribution are randomly sampled respectively by using a Monte Carlo sampling method.

5. The virtual power plant joint scheduling method according to claim 1, characterized in that, The photovoltaic power generation amount, the electric vehicle charging data and the electric vehicle battery swapping data in the scheduling period under multiple scenarios are taken as inputs to solve the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant, and specifically includes: The photovoltaic power generation amount, the electric vehicle charging data and the electric vehicle battery swapping data in the scheduling period under multiple scenarios are clustered by using a clustering algorithm to obtain the photovoltaic power generation amount, the electric vehicle charging data and the electric vehicle battery swapping data in the scheduling period under multiple clusters. The photovoltaic power generation amount, the electric vehicle charging data and the electric vehicle battery swapping data in the scheduling period under multiple clusters are taken as inputs to solve the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant.

6. The virtual power plant joint scheduling method according to claim 5, characterized in that, The joint scheduling model is solved to obtain a joint scheduling scheme of the virtual power plant, specifically including: the joint scheduling model is optimized and solved by using an RHC algorithm to obtain a joint scheduling scheme of the virtual power plant.

7. A virtual power plant joint dispatching device, characterized in that, The virtual power plant joint scheduling device includes: A model construction module is configured to construct a joint scheduling model of a virtual power plant. The joint scheduling model is configured to schedule a collaborative operation process between photovoltaic power generation equipment, energy storage equipment, electric vehicle charging stations, electric vehicle battery swapping stations, air conditioning loads and a power grid under a constraint condition with a minimum scheduling cost as an optimization target. The constraint condition includes power balance constraints, energy storage equipment constraints, electric vehicle charging station constraints, electric vehicle battery swapping station constraints, air conditioning load constraints and power grid interactive power constraints. A model solving module is configured to solve the joint scheduling model to obtain a joint scheduling scheme of the virtual power plant. The joint scheduling scheme includes charging power and discharging power of the energy storage equipment, charging power and discharging power of the electric vehicle, charging power and discharging power of each battery in the electric vehicle battery swapping station, reducible power of the air conditioning load and interactive power between the virtual power plant and the power grid. The interactive power is power purchase power of the virtual power plant from the power grid or power selling power of the virtual power plant to the power grid. The joint scheduling model is solved to obtain a joint scheduling scheme of the virtual power plant, specifically including: Meteorological data in a scheduling period is taken as input to predict photovoltaic power generation amount data distribution in the scheduling period by using a photovoltaic power generation uncertainty prediction model. Historical electric vehicle charging data in a scheduling period is taken as input to predict electric vehicle charging data distribution in the scheduling period by using an electric vehicle charging uncertainty prediction model. Historical electric vehicle battery swapping data in a scheduling period is taken as input to predict electric vehicle battery swapping data distribution in the scheduling period by using an electric vehicle battery swapping uncertainty prediction model. randomly sample the photovoltaic power generation data distribution, the electric vehicle charging data distribution and the electric vehicle battery swap data distribution respectively to obtain photovoltaic power generation data, electric vehicle charging data and electric vehicle battery swap data of a scheduling period under multiple scenarios; the number of the scenarios is the same as the number of times of random sampling, the electric vehicle charging data includes initial battery state of charge and charging window of each electric vehicle participating in charging, and the electric vehicle battery swap data includes initial battery state of charge and battery swap window of each electric vehicle participating in battery swap; solve the joint scheduling model by taking the photovoltaic power generation data, the electric vehicle charging data and the electric vehicle battery swap data of the scheduling period under the multiple scenarios as input to obtain a joint scheduling scheme of the virtual power plant.

8. A computer device comprising: A memory, a processor and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the virtual power plant joint scheduling method of any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the virtual power plant joint scheduling method of any one of claims 1-6.

Images