Comprehensive transaction method and device of virtual power plant and electronic equipment

Abstract

The invention provides a comprehensive transaction method and device of a virtual power plant and electronic equipment, and belongs to the technical field of virtual power plants. The method comprises the steps of performing clustering processing on a plurality of power generation prediction values based on a clustering algorithm to obtain a plurality of typical power generation prediction values; calculating an adjustable space value of the virtual power plant based on the actual output value and the predicted output value of the semi-controlled resource of the virtual power plant and the actual output value and the predicted output value of the controllable resource; constructing an objective function based on the plurality of typical power generation predicted values and the adjustable spatial value; solving the objective function to obtain day-ahead transaction income, and correcting the day-ahead transaction income by using a real-time condition to obtain real-time transaction income; and calculating the total revenue of the virtual power plant based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation cost. The method is used for solving the defect that the prediction result of the prediction data of the wind power generation and photovoltaic power generation scenes is uncertain when the virtual power plant participates in the power market transaction in the existing method.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual power plants, and in particular to a comprehensive trading method for virtual power plants, a comprehensive trading device for virtual power plants, an electronic device, a machine-readable storage medium and a computer program product. Background Art

[0002] As the intersection of energy supply and consumption, Virtual Power Plant (VPP) is free from the limitations of geographic space and grid topology. Relying on advanced communication technology, it can aggregate and centrally manage widely distributed resources, and optimize the scheduling of distributed resources through power market transactions. It has become a research hotspot. In order to effectively utilize demand-side resources, it is necessary to overcome challenges such as small single capacity, poor resource compliance, and strong uncertainty. Virtual power plants aggregate demand-side resources such as distributed power sources, energy storage equipment, and loads, and regulate their flexible consumption, which helps to enhance the grid's response capabilities in peak load regulation and safety emergencies.

[0003] In the existing methods of virtual power plants participating in power market transactions, the prediction data generation methods for wind power generation and photovoltaic power generation scenarios are mostly based on direct predictions based on historical data. Due to the large training data set, the training learning efficiency is low, and it is easy to cause problems such as dimensionality disaster and overfitting, which further increases the uncertainty of the prediction results. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a comprehensive trading method, device and electronic equipment for a virtual power plant, so as to solve the defect that the prediction results of the prediction data for wind power generation and photovoltaic power generation scenarios in the existing method of virtual power plants participating in electricity market transactions are uncertain.

[0005] In order to achieve the above object, an embodiment of the present invention provides a comprehensive transaction method of a virtual power plant, including:

[0006] Obtain multiple power generation forecast values ​​of renewable energy sources of the virtual power plant;

[0007] Performing clustering processing on the multiple power generation prediction values ​​based on a clustering algorithm to obtain multiple typical power generation prediction values;

[0008] Calculate the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controllable resources of the virtual power plant, as well as the actual output value and predicted output value of the controllable resources;

[0009] Constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue;

[0010] Solving the objective function to obtain the day-ahead transaction revenue, and modifying the day-ahead transaction revenue using real-time conditions to obtain the real-time transaction revenue;

[0011] The total revenue of the virtual power plant is calculated based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee; the set deviation fee represents the deviation cost between balancing the semi-controlled resources and the controllable resources.

[0012] Optionally, clustering the multiple power generation prediction values ​​based on a clustering algorithm to obtain multiple typical power generation prediction values ​​includes:

[0013] Repeat the following steps until the set end condition is reached:

[0014] Randomly select a power generation prediction value from the multiple power generation prediction values ​​as a center point;

[0015] Calculating the shortest Euclidean distance between each power generation prediction value among the plurality of power generation prediction values ​​excluding the center point and the center point;

[0016] The probability of each of the plurality of power generation prediction values ​​excluding the center point becoming a cluster center point is calculated based on the shortest Euclidean distance, and the power generation prediction value with the highest probability is determined as the next center point and the typical power generation prediction value.

[0017] Optionally, the adjustable space value of the virtual power plant is calculated based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources, and is calculated by the following formula:

[0018]

[0019] Among them, ΔP(t) represents the adjustable space value of the virtual power plant at time t, P i (t) represents the actual output value of the controllable resource at time t, P j (t) represents the actual output value of the semi-controlled resource at time t, P i e (t) represents the predicted output value of controllable resource i at time t, represents the predicted output value of semi-controlled resource j at time t, N i is the number of controllable resources; N j is the number of semi-controlled resources.

[0020] Optionally, constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue, includes:

[0021] Determining a minimum value among the plurality of typical power generation prediction values ​​as a target power generation prediction value;

[0022] The day-ahead bidding transaction volume is calculated based on the target power generation forecast value and the adjustable space value;

[0023] Based on the day-ahead bidding transaction volume, the revenue from participating in the energy market, the revenue from participating in the reserve service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads are determined; the other grid loads represent other grid loads excluding the adjustable load cluster;

[0024] An objective function is constructed based on the revenue from participating in the energy market, the revenue from participating in standby services, the revenue from participating in ancillary services, the cost of purchasing electricity from the internal resources of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads.

[0025] Optionally, based on the revenue from participating in the energy market, the revenue from participating in the reserve service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads, an objective function is constructed and calculated by the following formula:

[0026]

[0027] in, They represent the revenue of the virtual power plant participating in the energy market, the revenue of participating in the reserve service and the revenue of participating in the ancillary service during period t respectively; represents the cost of the virtual power plant purchasing electricity from the internal resources of the virtual power plant during period t; represents the net income of the virtual power plant from selling electricity to the adjustable load cluster during period t; It represents the net income of the virtual power plant from selling electricity to other grid loads during period t, and T represents the maximum period.

[0028] Optionally, the total revenue of the virtual power plant is calculated based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee, and is calculated by the following formula:

[0029]

[0030] Where R represents the total revenue of the virtual power plant, R DA represents the day-ahead transaction revenue, R RT Indicates real-time transaction revenue, It represents the deviation cost between balancing the semi-controllable resources and the controllable resources in the time period from 1 to T.

[0031] Optionally, the multiple power generation prediction values ​​are obtained by the following steps:

[0032] Obtain multiple actual power generation values ​​and multiple prediction errors of renewable energy in a virtual power plant;

[0033] Filtering a plurality of target prediction errors from the plurality of prediction errors using Latin hypercube sampling;

[0034] A plurality of power generation prediction values ​​of renewable energy are calculated based on the plurality of actual power generation values ​​and the plurality of target prediction errors.

[0035] On the other hand, an embodiment of the present invention further provides a comprehensive trading device for a virtual power plant, including:

[0036] An acquisition module, used to acquire multiple power generation prediction values ​​of renewable energy of a virtual power plant;

[0037] A clustering module, used for clustering the plurality of power generation prediction values ​​based on a clustering algorithm to obtain a plurality of typical power generation prediction values;

[0038] A first calculation module, used to calculate the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources;

[0039] A construction module, configured to construct an objective function based on the plurality of typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximization of day-ahead transaction revenue;

[0040] A solution module, used for solving the objective function to obtain the day-ahead transaction revenue, and modifying the day-ahead transaction revenue using real-time conditions to obtain the real-time transaction revenue;

[0041] The second calculation module is used to calculate the total revenue of the virtual power plant based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee; the set deviation fee represents the deviation cost between balancing the semi-controlled resources and the controllable resources.

[0042] On the other hand, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned comprehensive trading method of the virtual power plant when executing the program.

[0043] On the other hand, the present invention also provides a machine-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned comprehensive trading method of the virtual power plant.

[0044] On the other hand, the present invention also provides a computer program product, including a computer program, which implements the above-mentioned comprehensive trading method of the virtual power plant when executed by a processor.

[0045] Through the above technical solution, the embodiment of the present invention obtains multiple power generation forecast values ​​of renewable energy of the virtual power plant; clusters the multiple power generation forecast values ​​based on the clustering algorithm to obtain multiple typical power generation forecast values, and reduces the multiple power generation forecast values ​​of renewable energy obtained by using the clustering algorithm to obtain a representative limited scene subset of wind and light power generation, thereby converting the uncertainty problem into a deterministic problem. The present invention also calculates the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, as well as the actual output value and predicted output value of the controllable resources, which is used to calculate the total revenue of the virtual power plant. The embodiment of the present invention effectively evaluates the adjustable ability of operators who aggregate multiple different resources to participate in the peak-shaving market. The adjustable space value of the virtual power plant fully considers the uncertainty of the aggregated resources of the virtual power plant. The embodiment of the present invention converts the uncertainty problem into a deterministic problem, and solves the defect that the prediction results of the prediction data of the wind power generation and photovoltaic power generation scenes in the virtual power plant participating in the power market transaction in the existing method are uncertain.

[0046] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0048] Figure 1 It is one of the flow diagrams of the comprehensive transaction method of the virtual power plant provided by the present invention;

[0049] Figure 2 It is a schematic diagram of the composition structure of the virtual power plant provided by the present invention;

[0050] Figure 3 It is a schematic diagram of a process of clustering multiple power generation prediction values ​​based on the K-means++ clustering algorithm provided by the present invention;

[0051] Figure 4 This is the second flow chart of the comprehensive transaction method of the virtual power plant provided by the present invention;

[0052] Figure 5 It is a structural schematic diagram of the comprehensive trading device of the virtual power plant provided by the present invention;

[0053] Figure 6It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0054] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0055] Method Embodiment

[0056] Please refer to Figure 1 The embodiment of the present invention provides a comprehensive transaction method of a virtual power plant, including:

[0057] Step 100: Obtain multiple power generation forecast values ​​of renewable energy of the virtual power plant.

[0058] The composition structure of the virtual power plant (VPP) of the embodiment of the present invention is as follows: Figure 2 As shown in the figure, the internal part is composed of a variety of resources such as renewable distributed energy (such as wind power generation units, photovoltaic power generation units, etc.), controllable distributed power sources (such as gas turbine generator sets, etc.), user loads, and energy storage systems (ES). Various resources within the virtual power plant report their own power generation information and operating status to the operator of the virtual power plant. The role of the operator of the virtual power plant is to summarize its own operating baseline and adjustable space in the day-ahead stage, and combine the electricity price information of the power market to formulate the next day's operation plan for various resources.

[0059] In the embodiment of the present invention, user loads are classified into price-type demand response loads and incentive-type demand response loads according to load characteristics. Price-type demand response loads refer to loads in which users actively adjust their electricity consumption behavior patterns according to real-time electricity prices and personal living habits. Incentive-type demand response loads refer to loads in which virtual power plants incentivize users through electricity price compensation, encouraging users to participate in peak shaving and valley filling to achieve optimal scheduling of the power system.

[0060] The embodiment of the present invention extracts the operating characteristics of the internal resources of the virtual power plant, and divides the internal resources into two categories: semi-controlled resources and controllable resources according to the control capability of the virtual power plant on each resource.

[0061] Among them, semi-controlled resources refer to resources that the virtual power plant has weaker flexible control capabilities, and are usually aggregated or started and stopped according to their operating status. This type of resource mainly includes renewable distributed energy such as wind power generation and photovoltaic power generation (referred to as renewable energy), as well as price-based demand response loads. Due to the significant influence of external factors such as weather, geographical location and user living habits, this type of resource has strong intermittent and random characteristics, which makes the virtual power plant's flexibility in regulating it relatively limited.

[0062] The renewable distributed energy (such as wind power generation, photovoltaic power generation, etc.) inside the virtual power plant has strong output uncertainty. In order to ensure system security in the power grid optimization scheduling, the embodiment of the present invention adopts dynamic robust constraints to model the output set W0 of the renewable distributed energy inside the virtual power plant during the day-ahead optimization scheduling process, as follows:

[0063]

[0064] Where P j (t) is the actual output value of the semi-controlled resources in the virtual power plant at time t; is the expected output value of the semi-controlled resource j in the virtual power plant at time t; is the maximum fluctuation deviation of the predicted output of the semi-controlled resource j at time t (in the embodiment of the present invention, the sign is set to be positive when the internal resources of the virtual power plant are generating electricity and negative when consuming electricity); δ j When (t) takes the value of 1 or -1, it means that the output of semi-controlled resources reaches the upper and lower limits of the prediction interval, respectively. j When (t) takes the value of 0, it means that the output of semi-controlled resources is the expected value of the prediction area; Γ S is the fluctuation parameter of the demand response output uncertainty set on the spatial scale; Γ T is the fluctuation parameter of the demand response uncertainty set on the time scale.

[0065] Since virtual power plants participate in electricity market transactions, the output of price-based demand response loads is significantly affected by residents' living habits and market electricity prices. Therefore, the embodiment of the present invention has refined the output modeling of such loads. Under the time-of-use electricity price mechanism, the response prediction problem of price-based demand response loads can be reduced to the load prediction problem under different electricity price conditions. It can be expressed as:

[0066]

[0067] Among them, ε(t) is the load self-elastic coefficient at time t; λ Δc (t) is the rate of change of electricity price at time t; P Load (t) is the total internal load of the virtual power plant at time t; P DR (t) is the load that can be translated at time t; P L (t) is the forecast value of the internal demand response load of the virtual power plant at time t.

[0068] In summary, the semi-controlled resource output set of the virtual power plant in the embodiment of the present invention can be summarized as follows:

[0069]

[0070] On the other hand, controllable resources refer to resources that can flexibly respond to internal instructions of the virtual power plant, are less affected by external environmental factors, and are highly controllable. This type of resource mainly includes controllable distributed power sources on the power supply side (such as gas turbine generator sets), incentive-type demand response loads on the load side (such as interruptible loads, shiftable loads), and energy storage systems. Considering that interruptible loads are a special type of shiftable load, the embodiment of the present invention selects representative shiftable loads for modeling. During the operation of the virtual power plant, the key information that operators need to make decisions includes the following aspects:

[0071] 1) Expected power consumption plan P of the shiftable load DR (t), the upper limit of the dispatchable range in each period Lower limit of dispatchable range and unit dispatch costs.

[0072] 2) Conventional output P of controllable distributed power source C (t), maximum dispatchable power Minimum dispatchable power

[0073] 3) Maximum charging power of the energy storage unit Minimum discharge power Maximum charge and discharge capacity Minimum charge and discharge capacity And the initial capacity required for the scheduling period

[0074] Based on the above virtual power plant modeling, an embodiment of the present invention obtains multiple power generation prediction values ​​of renewable energy of the virtual power plant. The renewable energy may be wind power generation and / or photovoltaic power generation. That is, the embodiment of the present invention obtains multiple power generation prediction values ​​of wind power generation and / or photovoltaic power generation of the virtual power plant. In one embodiment, multiple power generation prediction values ​​of wind power generation and / or photovoltaic power generation can be predicted by various artificial intelligence models. The embodiment of the present invention obtains multiple power generation prediction values ​​of wind power generation and photovoltaic power generation of the virtual power plant.

[0075] In other embodiments, the plurality of power generation prediction values ​​are obtained by the following steps:

[0076] Step 10: Obtain multiple actual power generation values ​​and multiple prediction errors of renewable energy in the virtual power plant.

[0077] Step 20: Filter out multiple target prediction errors from the multiple prediction errors using Latin hypercube sampling.

[0078] Step 30: Calculate and obtain multiple power generation prediction values ​​of renewable energy based on the multiple actual power generation values ​​and multiple target prediction errors.

[0079] Specifically, since the forecast errors of wind power generation, photovoltaic power generation and the day-ahead market electricity price obey the normal distribution, the embodiment of the present invention expresses the actual values ​​of wind power generation, photovoltaic power generation and the market electricity price as follows:

[0080]

[0081] Where P t w,DA , P t p,DA and t DA are the actual values ​​of wind power, photovoltaic power and day-ahead market electricity prices in each period, P t ew,DA , P t ep,DA and t e,DA are the forecast values ​​of wind power, photovoltaic power and day-ahead market electricity prices in different time periods, are the forecast errors of wind power, photovoltaic power and day-ahead market electricity prices in different time periods; σ is the standard deviation of normal distribution.

[0082] In order to better reflect the distribution range of prediction errors and reduce calculation time, an embodiment of the present invention uses Latin hypercube sampling to randomly sample prediction errors, screens out multiple target prediction errors from the multiple prediction errors, and then calculates multiple power generation prediction values ​​of renewable energy based on the difference between each actual power generation value and the target prediction error, thereby generating a large amount of prediction data for wind power generation, photovoltaic power generation and market electricity price scenarios.

[0083] Step 200: cluster the multiple power generation prediction values ​​based on a clustering algorithm to obtain multiple typical power generation prediction values.

[0084] The embodiment of the present invention can cluster the multiple power generation prediction values ​​based on various clustering algorithms to obtain multiple typical power generation prediction values. For example, the clustering algorithm includes K-means clustering algorithm, DBSCAN clustering algorithm, etc. In one embodiment, the embodiment of the present invention can cluster the multiple power generation prediction values ​​based on K-means++ clustering algorithm to obtain multiple typical power generation prediction values.

[0085] Specifically, clustering is performed on the multiple power generation prediction values ​​based on the K-means++ clustering algorithm to obtain multiple typical power generation prediction values, including:

[0086] Repeat the following steps until the set end condition is reached:

[0087] Step 210: randomly selecting a power generation prediction value from the multiple power generation prediction values ​​as a center point;

[0088] Step 220, calculating the shortest Euclidean distance between each power generation prediction value among the plurality of power generation prediction values ​​excluding the center point and the center point;

[0089] Step 230: Calculate the probability of each power generation prediction value among the multiple power generation prediction values ​​excluding the center point becoming a cluster center point based on the shortest Euclidean distance, and determine the power generation prediction value with the highest probability as the next center point and typical power generation prediction value.

[0090] Compared with the K-means clustering algorithm, the K-means++ clustering algorithm can traverse the data set to update the center point, getting rid of the dependence on the initial center point. Taking the wind power generation and photovoltaic power generation scenarios as an example, the K-means++ clustering algorithm process is as follows: Figure 3 As shown, where ψ=[p1,p2,…,p n ] is a scene set of wind power generation and photovoltaic power generation (the predicted power generation value of wind power generation and photovoltaic power generation). Each scene p1 is regarded as a point in Euclidean space, and its spatial coordinates can be expressed as p i =(x1,x2,…,x n );K is the number of scene categories; d i is the Euclidean distance between the center point and the other points; P(i) is the probability of each point becoming the center point, S is a typical scenario set, including multiple typical power generation forecast values.

[0091] The embodiment of the present invention achieves the reduction of power generation prediction values ​​of wind power generation and photovoltaic power generation scenarios through the K-means++ clustering algorithm. The embodiment of the present invention can obtain a representative typical scenario set S of wind power generation and photovoltaic power generation.

[0092] Step 300: Calculate the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant and the actual output value and predicted output value of the controllable resources.

[0093] The adjustable space value of a virtual power plant is a quantitative indicator for measuring the flexible adjustment capability of the virtual power plant. The embodiment of the present invention provides a basis for the subsequent virtual power plant to participate in the optimization and scheduling of the power market by establishing an adjustable space model of the virtual power plant.

[0094] In order to construct an adjustable spatial model of a virtual power plant, we first need to base The predicted net load curve of the virtual power plant is obtained by statistically analyzing the predicted output values ​​of the semi-controlled resources and controllable resources within the virtual power plant, and the operating baseline of the virtual power plant on the next day is obtained:

[0095]

[0096] in, is the predicted output value of controllable resource i at time t; is the predicted output value of semi-controlled resource j at time t; N i is the number of controllable resources; N j is the number of semi-controlled resources.

[0097] Through the above analysis, the adjustable capabilities of the two major types of resources within the virtual power plant are characterized respectively.

[0098] The adjustable capacity model of semi-controlled resources is as follows:

[0099] The next day output difference of the semi-controlled resources of the virtual power plant ΔP j (t) can be expressed as:

[0100] P j (t)∈{P WT (t),P PV (t),P Load (t)}

[0101] Among them, P PV (t) and P WT (t) are the actual output values ​​of photovoltaic power generation and wind power generation in the virtual power plant at time t, P j (t) represents the actual output value of the semi-controlled resource at time t. is the predicted output value of semi-controlled resource j at time t, P Load (t) is the total internal load of the virtual power plant at time t.

[0102] The adjustable capacity model of controllable resources is as follows:

[0103] Controllable resources have clear parameters such as capacity and output range. Virtual power plant operators can determine their adjustable capacity based on their resource parameters, which is specifically manifested as the increase / reduction of output value ΔP after the controllable resources are dispatched. i (t).

[0104]

[0105] Among them, P i (t) is the actual output value of the controllable resources after scheduling; and are the charging and discharging power of the energy storage at time t, is the predicted output value of controllable resource i at time t, P DR (t) is the expected power consumption plan of the shiftable load, P C (t) is the normal output value of the controllable distributed power source.

[0106] Then the adjustable spatial output value ΔP(t) of the virtual power plant can be expressed as:

[0107]

[0108] That is, the adjustable space value of the virtual power plant is calculated based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, as well as the actual output value and predicted output value of the controllable resources, and is calculated by the following formula:

[0109]

[0110] Among them, ΔP(t) represents the adjustable space value of the virtual power plant at time t, P i (t) represents the actual output value of the controllable resource at time t, P j (t) represents the actual output value of the semi-controlled resource at time t, P i e (t) represents the predicted output value of controllable resource i at time t, represents the predicted output value of semi-controlled resource j at time t, N i is the number of controllable resources; N j is the number of semi-controlled resources.

[0111] In summary, the embodiment of the present invention models the adjustable space of a virtual power plant that aggregates multiple resources. First, its internal resource output needs to be predicted, the operating baseline and the semi-controlled resource output set are obtained, and then the internal controllable resources are expressed in the above form. The advantage of this model is that it takes into account the uncertainty of the aggregated resources of the virtual power plant, is not limited to the establishment of the model under a specific aggregation method, and facilitates the overall participation of the virtual power plant in the optimization scheduling bidding strategy.

[0112] Step 400: construct an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue.

[0113] In one embodiment, step 400, constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue, includes:

[0114] Step 410: Determine the minimum value among the multiple typical power generation prediction values ​​as the target power generation prediction value.

[0115] In order to ensure the margin, the embodiment of the present invention selects the minimum value among multiple typical power generation prediction values ​​as the target power generation prediction value.

[0116] Step 420: Calculate the day-ahead bidding transaction volume based on the target power generation forecast value and the adjustable space value.

[0117] The virtual power plant needs to sign a bilateral contract and submit a day-ahead bid transaction volume. The day-ahead bid transaction volume is calculated based on the sum of the target power generation forecast value of the virtual power plant and the adjustable space value. During the bidding process, the day-ahead bid volume and the real-time bid volume must meet the power balance constraint. In addition, the embodiment of the present invention regards the virtual power plant as a third-party operating organization, and its core profit model is to earn the difference in power trading.

[0118] Step 430: Determine the revenue from participating in the energy market, participating in the standby service, participating in the auxiliary service, the cost of purchasing electricity from the internal resources of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads based on the day-ahead bidding transaction volume; the other grid loads represent other grid loads excluding the adjustable load cluster.

[0119] The embodiment of the present invention determines the revenue from participating in the energy market, the revenue from participating in standby services, the revenue from participating in auxiliary services, the cost of purchasing electricity from the internal resources of the virtual power plant, the net revenue from selling electricity to adjustable load clusters, and the net revenue from selling electricity to other grid loads based on the day-ahead bidding transaction volume.

[0120] Therefore, please refer to Figure 4 ,The goal of Phase 1 is to optimize the revenue of the virtual power plant in the ,energy market, reserve services and ancillary services, and determine the ,optimal day-ahead bidding amount.

[0121] Step 440, construct an objective function based on the revenue from participating in the energy market, the revenue from participating in the backup service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resources of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster and the net revenue from selling electricity to other grid loads.

[0122] In one embodiment, the objective function is constructed based on the revenue from participating in the energy market, the revenue from participating in the reserve service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads, and is calculated by the following formula:

[0123]

[0124] sP t min ≤P t ≤P t max

[0125] in, They represent the revenue of the virtual power plant participating in the energy market, the revenue of participating in the reserve service and the revenue of participating in the auxiliary service in period t respectively. This part is the net revenue of the virtual power plant from external bidding. represents the cost of the virtual power plant purchasing electricity from the internal resources of the virtual power plant during period t; It represents the net income of the virtual power plant selling electricity to the adjustable load cluster in period t, and the regulation cost of the virtual power plant regulating the adjustable load cluster has been deducted from this income; represents the net income of the virtual power plant from selling electricity to other grid loads during period t, T represents the maximum period, and P t min , P t max are the minimum and maximum day-ahead bids of virtual power plants, respectively.

[0126] Step 500: Solve the objective function to obtain the day-ahead transaction profit, and modify the day-ahead transaction profit using real-time conditions to obtain real-time transaction profit.

[0127] The embodiment of the present invention can solve the objective function through various solution methods (such as genetic algorithms) to obtain the day-ahead transaction income, and use real-time conditions to modify the day-ahead transaction income to obtain real-time transaction income. The virtual power plant needs to sign a bilateral contract and submit a real-time bidding transaction volume in stage 2. The real-time bidding transaction volume in stage 2 is based on the real-time prediction scenario data of wind power generation and photovoltaic power generation of the virtual power plant and the real-time data of the adjustable space of the virtual power plant. At this stage, the day-ahead market has been cleared, and the day-ahead transaction volume and electricity price are known quantities. The decision variables before time period t are also known quantities, so the real-time trading model for time period t is:

[0128]

[0129] sP h min ≤P h ≤P h max ;

[0130] Where: P h is the real-time bidding transaction volume of the virtual power plant; P h min , P h max are the minimum and maximum values ​​of the real-time bidding transaction volume of virtual power plants, respectively. They represent the revenue of the virtual power plant participating in the energy market, the revenue of participating in the reserve service and the revenue of participating in the auxiliary service in the h period respectively. This part is the net revenue of the virtual power plant from external bidding; represents the cost of the virtual power plant purchasing electricity from the internal resources of the virtual power plant during period h; It represents the net income of the virtual power plant selling electricity to the adjustable load cluster in period h, and the regulation cost of the virtual power plant regulating the adjustable load cluster has been deducted from this income; It represents the net income of the virtual power plant from selling electricity to other grid loads in the h period, and T represents the maximum period. All can be obtained based on the real-time bidding transaction volume. Solve the real-time transaction model in period t to obtain the real-time transaction income. Stage 2 is the model of real-time transaction income, and stage 1 is the model of day-ahead transaction income. There will be errors between the two. It can be regarded that the model of day-ahead transaction income is the total transaction cost control target, and the model of real-time transaction income is some small corrections made according to the changes in real-time conditions.

[0131] Step 600, calculating the total revenue of the virtual power plant based on the day-ahead transaction revenue, the real-time transaction revenue and a set deviation fee; the set deviation fee represents the deviation fee for balancing the semi-controlled resources and the controllable resources.

[0132] In the stage 3 of the embodiment of the present invention, the output deviation of the virtual power plant needs to be absorbed by the balancing market and fees are charged to the virtual power plant. Therefore, the total revenue of the virtual power plant is calculated based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee, and is calculated by the following formula:

[0133]

[0134] Where R represents the total revenue of the virtual power plant, R DA represents the day-ahead transaction revenue, R RT Indicates real-time transaction revenue, It represents the deviation cost between balancing the semi-controllable resources and the controllable resources in the time period from 1 to T.

[0135] Therefore, the embodiment of the present invention first obtains the power generation forecast data of wind power generation and photovoltaic power generation through the uncertain scenario generation method of wind power generation and photovoltaic power generation and market electricity price, and then obtains the adjustable space value of the virtual power plant based on the unified modeling of the adjustable space of the virtual power plant. Based on the adjustable space value of the virtual power plant and the power generation forecast data of wind power generation and photovoltaic power generation, the process of the virtual power plant participating in the power market transaction is designed.

[0136] The embodiment of the present invention obtains multiple power generation forecast values ​​of renewable energy of a virtual power plant; clusters the multiple power generation forecast values ​​based on a clustering algorithm to obtain multiple typical power generation forecast values, and reduces the multiple power generation forecast values ​​of renewable energy obtained by using a clustering algorithm to obtain a representative limited subset of wind and photovoltaic power generation scenarios, thereby converting the uncertainty problem into a deterministic problem. The present invention also calculates the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, as well as the actual output value and predicted output value of the controllable resources, for calculating the total revenue of the virtual power plant. The embodiment of the present invention effectively evaluates the adjustable capacity of operators who aggregate multiple different resources to participate in the peak-shaving market. The adjustable space value of the virtual power plant fully considers the uncertainty of the aggregated resources of the virtual power plant. The embodiment of the present invention converts the uncertainty problem into a deterministic problem, and solves the defect that the prediction results of the prediction data of the wind power generation and photovoltaic power generation scenarios in the virtual power plant participating in the power market transaction in the existing method are uncertain.

[0137] In summary, according to the prediction error of the virtual power plant, the embodiment of the present invention generates a scenario set by using a multi-scenario optimization method to convert the prediction results of wind power generation, photovoltaic power generation and the electricity price of the day-ahead power market into a scenario set, and uses the K-means++ clustering algorithm to reduce the obtained scenario set, and obtains a representative limited scenario subset of wind power generation, photovoltaic power generation and electricity price of the day-ahead power market, which converts the uncertainty problem into a deterministic problem. In addition, the embodiment of the present invention proposes an adjustable space modeling method for a virtual power plant, which can effectively evaluate the adjustable ability of operators who aggregate a variety of different resources to participate in the peak-shaving market. The adjustable space model of the virtual power plant fully considers the uncertainty of the aggregated resources of the virtual power plant, is not limited to the establishment of the model under a specific aggregation mode, and is convenient for the virtual power plant to participate in the optimization scheduling bidding strategy as a whole. In addition, the embodiment of the present invention divides the power market transaction process into three stages, uses the day-ahead-real-time double-layer optimization model to characterize the optimization scheduling model of the virtual power plant participating in the market transaction, and finally uses the deviation between the reported transaction and the actual transaction in the balance market settlement to effectively improve the scheduling potential of the virtual power plant, while fully guaranteeing the stability and reliability of the distribution network system.

[0138] Device Embodiment

[0139] Please refer to Figure 5 On the other hand, an embodiment of the present invention further provides a comprehensive trading device for a virtual power plant, comprising:

[0140] An acquisition module 501 is used to acquire a plurality of power generation prediction values ​​of renewable energy sources of a virtual power plant;

[0141] A clustering module 502 is used to perform clustering processing on the multiple power generation prediction values ​​based on a clustering algorithm to obtain multiple typical power generation prediction values;

[0142] A first calculation module 503 is used to calculate the adjustable space value of the virtual power plant based on the actual output value and the predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and the predicted output value of the controllable resources;

[0143] A construction module 504 is used to construct an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue;

[0144] A solution module 505, used for solving the objective function to obtain the day-ahead transaction revenue, and modifying the day-ahead transaction revenue using real-time conditions to obtain the real-time transaction revenue;

[0145] The second calculation module 506 is used to calculate the total revenue of the virtual power plant based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee; the set deviation fee represents the deviation cost between balancing the semi-controlled resources and the controllable resources.

[0146] Optionally, clustering the multiple power generation prediction values ​​based on a clustering algorithm to obtain multiple typical power generation prediction values ​​includes:

[0147] Repeat the following steps until the set end condition is reached:

[0148] Randomly select a power generation prediction value from the multiple power generation prediction values ​​as a center point;

[0149] Calculating the shortest Euclidean distance between each power generation prediction value among the plurality of power generation prediction values ​​excluding the center point and the center point;

[0150] The probability of each of the plurality of power generation prediction values ​​excluding the center point becoming a cluster center point is calculated based on the shortest Euclidean distance, and the power generation prediction value with the highest probability is determined as the next center point and the typical power generation prediction value.

[0151] Optionally, the adjustable space value of the virtual power plant is calculated based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources, and is calculated by the following formula:

[0152]

[0153] Among them, ΔP(t) represents the adjustable space value of the virtual power plant at time t, P i (t) represents the actual output value of the controllable resource at time t, P j (t) represents the actual output value of the semi-controlled resource at time t, represents the predicted output value of controllable resource i at time t, represents the predicted output value of semi-controlled resource j at time t, N i is the number of controllable resources; N j is the number of semi-controlled resources.

[0154] Optionally, constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue, includes:

[0155] Determining a minimum value among the plurality of typical power generation prediction values ​​as a target power generation prediction value;

[0156] The day-ahead bidding transaction volume is calculated based on the target power generation forecast value and the adjustable space value;

[0157] Based on the day-ahead bidding transaction volume, the revenue from participating in the energy market, the revenue from participating in the reserve service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads are determined; the other grid loads represent other grid loads excluding the adjustable load cluster;

[0158] An objective function is constructed based on the revenue from participating in the energy market, the revenue from participating in standby services, the revenue from participating in ancillary services, the cost of purchasing electricity from the internal resources of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads.

[0159] Optionally, based on the revenue from participating in the energy market, the revenue from participating in the reserve service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads, an objective function is constructed and calculated by the following formula:

[0160]

[0161] in, They represent the revenue of the virtual power plant participating in the energy market, the revenue of participating in the reserve service and the revenue of participating in the ancillary service during period t respectively; represents the cost of the virtual power plant purchasing electricity from the internal resources of the virtual power plant during period t; represents the net income of the virtual power plant from selling electricity to the adjustable load cluster during period t; It represents the net income of the virtual power plant from selling electricity to other grid loads during period t, and T represents the maximum period.

[0162] Optionally, the total revenue of the virtual power plant is calculated based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee, and is calculated by the following formula:

[0163]

[0164] Where R represents the total revenue of the virtual power plant, R DA represents the day-ahead transaction revenue, R RT Indicates real-time transaction revenue, It represents the deviation cost between balancing the semi-controllable resources and the controllable resources in the time period from 1 to T.

[0165] Optionally, the multiple power generation prediction values ​​are obtained by the following steps:

[0166] Obtain multiple actual power generation values ​​and multiple prediction errors of renewable energy in a virtual power plant;

[0167] Filtering a plurality of target prediction errors from the plurality of prediction errors using Latin hypercube sampling;

[0168] A plurality of power generation prediction values ​​of renewable energy are calculated based on the plurality of actual power generation values ​​and the plurality of target prediction errors.

[0169] The comprehensive trading device of the virtual power plant includes a processor and a memory. The above-mentioned acquisition module 501, clustering module 502, first calculation 503, construction module 504, solution module 505 and second calculation module 506 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0170] The processor includes a kernel, which calls the corresponding program unit from the memory. There can be one or more kernels.

[0171] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0172] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6As shown, the electronic device may include: a processor (processor) 610, a communication interface (Communications Interface) 620, a memory (memory) 630 and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 can call the logic instructions in the memory 630 to execute the comprehensive trading method of the virtual power plant, which includes: obtaining multiple power generation forecast values ​​of renewable energy of the virtual power plant; clustering the multiple power generation forecast values ​​based on a clustering algorithm to obtain multiple typical power generation forecast values; calculating the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources; constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, and the objective function represents maximizing the day-ahead trading profit; solving the objective function to obtain the day-ahead trading profit, and using real-time conditions to correct the day-ahead trading profit to obtain the real-time trading profit; calculating the total profit of the virtual power plant based on the day-ahead trading profit, the real-time trading profit and the set deviation fee; the set deviation fee represents the deviation cost for balancing the semi-controlled resources and the controllable resources.

[0173] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0174] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a machine-readable storage medium. When the computer program is executed by a processor, the computer can execute a comprehensive trading method for a virtual power plant, which method includes: obtaining multiple power generation forecast values ​​of renewable energy of the virtual power plant; clustering the multiple power generation forecast values ​​based on a clustering algorithm to obtain multiple typical power generation forecast values; calculating the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources; constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, the objective function representing the maximization of the day-ahead trading profit; solving the objective function to obtain the day-ahead trading profit, and using real-time conditions to correct the day-ahead trading profit to obtain the real-time trading profit; calculating the total profit of the virtual power plant based on the day-ahead trading profit, the real-time trading profit and the set deviation fee; the set deviation fee represents the deviation cost for balancing the semi-controlled resources with the controllable resources.

[0175] On the other hand, the present invention also provides a machine-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute a comprehensive trading method for a virtual power plant, the method comprising: obtaining multiple power generation forecast values ​​of renewable energy of the virtual power plant; clustering the multiple power generation forecast values ​​based on a clustering algorithm to obtain multiple typical power generation forecast values; calculating the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources; constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, the objective function representing maximizing the day-ahead trading revenue; solving the objective function to obtain the day-ahead trading revenue, and correcting the day-ahead trading revenue using real-time conditions to obtain the real-time trading revenue; calculating the total revenue of the virtual power plant based on the day-ahead trading revenue, the real-time trading revenue and the set deviation fee; the set deviation fee represents the deviation fee for balancing the semi-controlled resources with the controllable resources.

[0176] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0177] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A comprehensive trading method for a virtual power plant, characterized in that: include: Obtain multiple power generation forecast values ​​of renewable energy sources of the virtual power plant; Performing clustering processing on the multiple power generation prediction values ​​based on a clustering algorithm to obtain multiple typical power generation prediction values; Calculate the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controllable resources of the virtual power plant, as well as the actual output value and predicted output value of the controllable resources; Constructing an objective function based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue; Solving the objective function to obtain the day-ahead transaction revenue, and modifying the day-ahead transaction revenue using real-time conditions to obtain the real-time transaction revenue; Calculate the total revenue of the virtual power plant based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee; The set deviation cost represents the deviation cost of balancing the semi-controllable resource and the controllable resource.

2. The comprehensive transaction method of virtual power plant according to claim 1, characterized in that: The clustering algorithm is used to cluster the multiple power generation prediction values ​​to obtain multiple typical power generation prediction values, including: Repeat the following steps until the set end condition is reached: Randomly select a power generation prediction value from the multiple power generation prediction values ​​as a center point; Calculating the shortest Euclidean distance between each power generation prediction value among the plurality of power generation prediction values ​​excluding the center point and the center point; The probability of each of the plurality of power generation prediction values ​​excluding the center point becoming a cluster center point is calculated based on the shortest Euclidean distance, and the power generation prediction value with the highest probability is determined as the next center point and the typical power generation prediction value.

3. The comprehensive transaction method of virtual power plant according to claim 1, characterized in that: The adjustable space value of the virtual power plant is calculated based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources, and is calculated by the following formula: Among them, ΔP(t) represents the adjustable space value of the virtual power plant at time t, P i (t) represents the actual output value of the controllable resource at time t, P j (t) represents the actual output value of the semi-controlled resource at time t, represents the predicted output value of controllable resource i at time t, represents the predicted output value of semi-controlled resource j at time t, N i is the number of controllable resources; N j is the number of semi-controlled resources.

4. The comprehensive transaction method of virtual power plant according to claim 1, characterized in that: The objective function is constructed based on the multiple typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximizing the day-ahead transaction revenue, and includes: Determining a minimum value among the plurality of typical power generation prediction values ​​as a target power generation prediction value; The day-ahead bidding transaction volume is calculated based on the target power generation forecast value and the adjustable space value; Based on the day-ahead bidding transaction volume, the revenue from participating in the energy market, the revenue from participating in the reserve service, the revenue from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads are determined; the other grid loads represent other grid loads excluding the adjustable load cluster; An objective function is constructed based on the revenue from participating in the energy market, the revenue from participating in standby services, the revenue from participating in ancillary services, the cost of purchasing electricity from the internal resources of the virtual power plant, the net revenue from selling electricity to the adjustable load cluster, and the net revenue from selling electricity to other grid loads.

5. The comprehensive transaction method of virtual power plant according to claim 4, characterized in that: The objective function is constructed based on the income from participating in the energy market, the income from participating in the reserve service, the income from participating in the auxiliary service, the cost of purchasing electricity from the internal resource side of the virtual power plant, the net income from selling electricity to the adjustable load cluster and the net income from selling electricity to other grid loads, and is calculated by the following formula: in, They represent the revenue of the virtual power plant participating in the energy market, the revenue of participating in the reserve service and the revenue of participating in the ancillary service during period t respectively; represents the cost of the virtual power plant purchasing electricity from the internal resources of the virtual power plant during period t; represents the net income of the virtual power plant from selling electricity to the adjustable load cluster during period t; It represents the net income of the virtual power plant from selling electricity to other grid loads during period t, and T represents the maximum period.

6. The comprehensive transaction method of virtual power plant according to claim 1, characterized in that: The total revenue of the virtual power plant is calculated based on the day-ahead transaction revenue, the real-time transaction revenue and the set deviation fee, and is calculated by the following formula: Where R represents the total revenue of the virtual power plant, R DA represents the day-ahead transaction revenue, R RT Indicates real-time transaction revenue, It represents the deviation cost between balancing the semi-controllable resources and the controllable resources in the time period from 1 to T.

7. The comprehensive transaction method of virtual power plant according to claim 1, characterized in that: The multiple power generation prediction values ​​are obtained by the following steps: Obtain multiple actual power generation values ​​and multiple prediction errors of renewable energy in a virtual power plant; Filtering a plurality of target prediction errors from the plurality of prediction errors using Latin hypercube sampling; A plurality of power generation prediction values ​​of renewable energy are calculated based on the plurality of actual power generation values ​​and the plurality of target prediction errors.

8. A comprehensive trading device for a virtual power plant, characterized in that: include: An acquisition module, used to acquire multiple power generation prediction values ​​of renewable energy of a virtual power plant; A clustering module, used for clustering the plurality of power generation prediction values ​​based on a clustering algorithm to obtain a plurality of typical power generation prediction values; A first calculation module, used to calculate the adjustable space value of the virtual power plant based on the actual output value and predicted output value of the semi-controlled resources of the virtual power plant, and the actual output value and predicted output value of the controllable resources; A construction module, configured to construct an objective function based on the plurality of typical power generation forecast values ​​and the adjustable space value, wherein the objective function represents maximization of day-ahead transaction revenue; A solution module, used for solving the objective function to obtain the day-ahead transaction revenue, and modifying the day-ahead transaction revenue using real-time conditions to obtain the real-time transaction revenue; A second calculation module is used to calculate the total income of the virtual power plant based on the day-ahead transaction income, the real-time transaction income and the set deviation fee; The set deviation cost represents the deviation cost of balancing the semi-controllable resource and the controllable resource.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the comprehensive trading method of the virtual power plant described in any one of claims 1 to 7 is implemented.

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

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the comprehensive trading method of the virtual power plant described in any one of claims 1 to 7 is implemented.

Images