Power market participation-oriented virtual power plant two-stage joint optimization scheduling method, system and related device

Abstract

The invention provides a virtual power plant two-stage joint optimization scheduling method and system for participation of a power market, and a related device, and the method comprises the following steps: 1, depicting the uncertainty characteristics of renewable energy sources, and obtaining a prediction error confidence interval; step 2, constructing a virtual power plant day-ahead scheduling model taking the maximum market income as a target; step 3, obtaining a day-ahead energy declaration curve and an energy storage operation strategy corresponding to the virtual power plant based on the obtained virtual power plant day-ahead scheduling model; 4, obtaining the actual output of the renewable energy in real time, and calculating the output deviation between the obtained actual output and the date energy declaration curve obtained in the step 3; 5, constructing a virtual power plant real-time scheduling model according to the obtained output deviation; 6, generating an interaction strategy between energy storage and a power grid according to the obtained real-time scheduling model of the virtual power plant; on the premise of guaranteeing the power supply stability of the system, the economic benefits of the virtual power plant participating in the power market transaction can be remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of power system optimization and dispatching technology, specifically relating to a two-stage joint optimization and dispatching method, system and related devices for virtual power plants oriented towards power market participation. Background Technology

[0002] Driven by the global shift towards a low-carbon energy structure and the carbon neutrality goals of various countries, the efficient utilization of renewable energy has become a widely discussed topic. This has spurred the aggregation of various distributed resources such as wind power, photovoltaics, energy storage, and flexible loads. Virtual Power Plant (VPP) technology is beneficial for integrating these distributed energy sources, improving system reliability and market operational flexibility. VPPs participate in the electricity market by aggregating small-scale, flexible resources lacking individual participation capabilities to improve economic efficiency. However, the volatility and intermittency of distributed energy sources not only lead to reliability and stability deficiencies in VPPs but also reduce their competitiveness in electricity market transactions. Therefore, how to ensure reliable and stable grid connection of VPP systems while effectively improving market returns has become a crucial challenge for optimized dispatching.

[0003] To better facilitate the operation of virtual power plant systems in the electricity market, various optimization methods and control schemes have been developed for renewable energy forecasting, resource aggregation, market trading strategies, energy dispatch management, and energy storage capacity optimization. Limited by current forecasting capabilities, the accuracy of renewable energy output prediction remains unsatisfactory, necessitating consideration of uncertainties in energy optimization. A typical approach is to reserve sufficient generation reserves and energy storage capacity to mitigate energy supply volatility; however, this method lacks clear modeling and its dispatch results are often conservative and uneconomical. Other common methods for considering renewable energy uncertainties include Monte Carlo simulation (MCS), which models the statistical distribution of input parameters to generate random samples and selects the desired output parameters through statistical analysis; and robust optimization (RO), which primarily seeks the optimal solution under worst-case scenarios. While these methods can mitigate the risks posed by renewable energy uncertainties, MCS has a significant computational burden, especially when considering multi-source uncertainties; robust optimization ignores the possibility of extreme cases, and its results are typically conservative. Although the MCS and RO methods can handle the uncertainties in VPPs, the time-varying nature of electricity prices and the uncertainty of load demand still weaken the advantages of VPP resource aggregation when virtual power plants participate in electricity market operations, and cannot effectively balance the reliability of VPP operation and market benefits.

[0004] Current literature proposes robust economic dispatch methods for multi-energy virtual power plants that consider multiple uncertainties. This method addresses the multiple uncertainties arising from flexible resource forecasting and response by proposing a two-stage economic dispatch approach for multi-energy virtual power plants, including a robust economic pre-dispatch process and an online parameter self-tuning process. The robust optimization dispatch model considers the uncertainties in renewable energy generation and load demand forecasting, employing a column constraint generation algorithm to solve the VPP dispatch problem under varying robustness coefficients and forecast biases. Simultaneously, an online parameter calibration model addresses the biases in flexible resource response dispatch instructions, using a quantum genetic algorithm for parameter identification and online self-tuning. Numerical simulations validate the model's accuracy and effectiveness. Although this method attempts to simultaneously correct and reference multiple uncertainties in both stages, its dispatch strategy still suffers from overly conservative approaches, lack of dynamic time-series modeling, and insufficient physical coupling of equipment. Furthermore, the parameter identification computation is complex, its engineering practicality is low, and it does not adequately integrate real market mechanisms for systematic comparison with other optimization methods, limiting its application value in practical virtual power plants.

[0005] In existing technologies, virtual power plants enhance the ability of small and medium-sized resource entities to participate in the electricity market by aggregating various distributed energy sources and adjustable load resources. However, due to the high volatility, low forecast accuracy, and intermittent operation of renewable energy sources such as wind and solar power, deviations between power output and market plans are easily caused, leading to reduced feasibility of virtual power plant dispatch schemes, increased operating costs, and exacerbated market performance risks. Furthermore, while some current dispatch strategies employ two-stage optimization methods for day-ahead and real-time dispatch linkage, they fail to fully integrate the temporal correlation of wind and solar power output, the conditional dependence of load fluctuations, and the dynamic trends of electricity prices. This results in insufficient accuracy in uncertainty modeling, conservative dispatch results, and limited market revenue optimization capabilities, making it difficult to meet the competitive demands of virtual power plants in complex market environments under multi-source collaborative conditions. Summary of the Invention

[0006] The purpose of this invention is to provide a two-stage joint optimization dispatching method, system and related devices for virtual power plants participating in the electricity market, which solves the problems of poor robustness, unstable returns and weak real-time dispatching response capability of existing dispatching schemes under the uncertainty of renewable energy output forecasting.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation, comprising the following steps: Step 1: Characterize the uncertainty features of renewable energy and obtain the prediction error confidence interval; Step 2: Based on the obtained prediction error confidence interval, construct a virtual power plant day-ahead scheduling model with the objective of maximizing market returns; Step 3: Based on the obtained day-ahead scheduling model of the virtual power plant, obtain the day-ahead energy application curve and energy storage operation strategy corresponding to the virtual power plant; Step 4: Obtain the actual output of renewable energy in real time, and calculate the output deviation between the actual output and the date energy reporting curve obtained in Step 3; Step 5: Construct a real-time scheduling model for the virtual power plant based on the obtained output deviation; Step 6: Generate an interaction strategy between energy storage and the power grid based on the obtained virtual power plant real-time scheduling model.

[0008] Preferably, in step 1, the uncertainty characteristics of renewable energy are characterized to obtain the prediction error confidence interval. The specific method is as follows: S11, obtain day-ahead wind and solar power output forecast data; S12, the acquired historical prediction error data is fitted with the error distribution by time period and power level to obtain the error fitting result; S13. Based on the obtained error fitting results, an error prediction model is constructed using the t-copula distribution; S14. Based on the obtained error prediction model and combined with the output prediction data obtained in S11, the prediction error confidence interval is obtained.

[0009] Preferably, in step 2, based on the obtained prediction error confidence interval, a virtual power plant day-ahead dispatch model is constructed with the objective of maximizing market returns. Specifically, the method is as follows: The virtual power plant day-ahead scheduling model includes an optimization objective function and constraints, wherein the expression for the optimization objective function is:

[0010] In the formula, The total revenue of the entire system during the current day's operation; The total time interval for the plan; The price of electricity sold by the system in the market; Cost of charging and discharging batteries; The market-based electricity purchase price; The final optimized clearing declaration power of the system; To optimize the power purchase volume in the results; The interactive power of the energy storage device; The constraints include power balance constraints, energy storage SOC and charge / discharge constraints, grid interaction power constraints, and opportunity constraints. Among them, the opportunity constraints are constructed based on the obtained prediction error confidence intervals.

[0011] Preferably, the power balance constraint is: (9a) (9b) (9c) in, Let t be the wind power output at time t; Let be the photovoltaic power at time t; Let t be the output power of wind and solar power after energy storage; and These represent the charging and discharging power of the energy storage device at time t, respectively. Let be the load power at time t; Energy storage SOC and charge / discharge constraints are: (10a) (10b) (10c) in, and These represent the maximum and minimum SOC values ​​of the energy storage device, respectively. The maximum value of energy storage power interaction at time t; Indicates the initial SOC value of energy storage; This represents the summation at all points in time. This represents the state-of-the-art (SOA) variable of an energy storage battery during time period t. Indicates the battery's maximum capacity; The power grid interaction constraint is: (11) in, This indicates the maximum power traded with the grid on the previous day; The opportunity constraint is:

[0012] in, It is a probability operator. Indicates the confidence level.

[0013] Preferably, in step 3, the deviation between the actual output of renewable energy and the day-ahead plan is obtained in real time, and the energy storage and grid interaction strategy is dynamically adjusted through reserve power and compensation mechanisms. Specifically, the method is as follows: Based on the deviation between the actual output of renewable energy and the day-ahead plan, and combined with reserve power, a real-time dispatch model is constructed. The energy storage and grid interaction strategy is obtained by using the real-time scheduling model.

[0014] Preferably, in step 5, the constructed virtual power plant real-time scheduling model includes a real-time scheduling objective function and constraints, wherein: The expression for the real-time scheduling objective function is:

[0015] In the formula, The total revenue function for the virtual power plant in the real-time market; The total time interval for the plan; For electricity prices; Cost of charging and discharging batteries; The market-based electricity purchase price; and These are the compensation cost and backup cost per unit power during the real-time phase, respectively. It is the final optimized and cleared declaration power of the system; To optimize the power purchase volume in the results; The interactive power of energy storage devices in a virtual power plant; This refers to the compensation power during the real-time operation phase. This is the backup power for the real-time operation phase; The constraints include power balance constraints, energy storage SOC and charge / discharge constraints, grid interaction power constraints, and opportunity constraints. Among them, the opportunity constraints are constructed based on the obtained prediction error confidence intervals.

[0016] Secondly, the present invention provides a two-stage joint optimization dispatch system for virtual power plants oriented towards electricity market participation, comprising: The confidence interval acquisition unit is used to characterize the uncertainty of renewable energy and obtain the confidence interval of the prediction error. The date scheduling model construction unit is used to construct a virtual day-ahead scheduling model for power plants with the goal of maximizing market returns, based on the obtained prediction error confidence interval. The energy storage operation strategy acquisition unit is used to obtain the day-ahead energy declaration curve and energy storage operation strategy of the virtual power plant based on the obtained day-ahead scheduling model of the virtual power plant. The output deviation calculation unit is used to obtain the actual output of renewable energy in real time and calculate the output deviation between the actual output and the obtained date energy declaration curve. The real-time scheduling model construction unit is used to construct a virtual power plant real-time scheduling model based on the obtained output deviation. The interaction strategy generation unit is used to generate interaction strategies between energy storage and the power grid based on the obtained virtual power plant real-time scheduling model.

[0017] Thirdly, the present invention provides an electronic device including a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the method described thereon.

[0018] Fourthly, the present invention provides a computer program product, the computer program product including computer-executable instructions, which, when executed, implement the method described.

[0019] Fifthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the method described herein.

[0020] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation. It utilizes historical wind and solar power output prediction error data to extract statistical patterns under different time periods and output levels, establishing an error distribution model with time-series correlation and conditional probability characteristics. By constructing opportunity constraints based on a standard t-distribution, the prediction error is introduced into the optimization scheduling model, thereby improving the ability to express the uncertainty of renewable energy output and enhancing scheduling robustness. In the day-ahead scheduling phase, the goal is to maximize market returns, generating energy application curves that meet the opportunity constraints. In the real-time phase, the deviation between actual power generation and the scheduling plan is compensated, comprehensively considering reserve power and deviation compensation costs to achieve optimization updates. This mechanism significantly improves the market participation capability and economic efficiency of virtual power plants in high-penetration renewable energy scenarios while ensuring power supply stability.

[0021] In summary, this invention constructs a data-driven opportunity-constraint model during the day-ahead scheduling phase, introducing the uncertainty of renewable energy output such as wind and solar power. Through statistical analysis of historical forecast error data, it captures the temporal correlation and probability distribution characteristics of output deviations, thereby formulating a more robust and reliable day-ahead energy reporting strategy. During the real-time scheduling phase, it compensates for or reserves the deviation between the actual output of renewable energy and the day-ahead plan, achieving dynamic correction and resource coordination control of system operation. Based on this two-stage collaborative optimization strategy, it can significantly improve the economic benefits of virtual power plants participating in electricity market transactions while ensuring the stability of system power supply. Its potential drawback is mainly the high complexity of parameter modeling and data processing. Attached Figure Description

[0022] Figure 1 This is a virtual power plant day-ahead operation model; Figure 2 This is a two-stage scheduling flowchart; Figure 3 Here is the flowchart for the day-to-day and real-time joint optimization process; Figure 4 For the controller hardware-in-the-loop experimental platform architecture; Figure 5 Forecasts for renewable energy and day-ahead electricity prices; Figure 6 The distribution of prediction errors under different power levels; Figure 7 The prediction results are based on a 95% confidence interval. Figure 8 For energy storage dispatch strategies; Figure 9 The power curves represent the interaction between the power grid and energy storage. Figure 10 This is the final optimization result as of now; Figure 11 A comparison of the day-ahead reporting curves at different confidence levels. Detailed Implementation

[0023] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0024] Example 1 This embodiment provides a two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation, comprising the following steps: Step 1: Characterize the uncertainty features of renewable energy and obtain the prediction error confidence interval; Step 2: Based on the obtained prediction error confidence interval, construct a virtual power plant day-ahead scheduling model with the objective of maximizing market returns; Step 3: Based on the obtained day-ahead scheduling model of the virtual power plant, obtain the day-ahead energy application curve and energy storage operation strategy corresponding to the virtual power plant; Step 4: Obtain the actual output of renewable energy in real time, and calculate the output deviation between the actual output and the date energy reporting curve obtained in Step 3; Step 5: Construct a real-time scheduling model for the virtual power plant based on the obtained output deviation; Step 6: Generate an interaction strategy between energy storage and the power grid based on the obtained virtual power plant real-time scheduling model.

[0025] Example 2 This embodiment provides a two-stage joint optimization dispatch system for virtual power plants (VPPs) oriented towards electricity market participation, aiming to address the problems of poor robustness, unstable returns, and weak real-time dispatch response capabilities of existing dispatch schemes under the uncertainty of renewable energy output forecasting. The proposed system architecture includes: a wind power generation electronic system, a photovoltaic power generation electronic system, an energy storage system, a power electronic conversion device, a control center, and an interface with the power grid. The control center performs unified coordination and control of the above subsystems based on the proposed dispatch method, realizing energy dispatch management in both the day-ahead and real-time stages.

[0026] Step 1: Renewable Energy Forecasting Processing S11, obtain day-ahead wind and solar power output forecast data, load forecast data and day-ahead electricity price; S12, Based on historical prediction error data, the error distribution is fitted by time period and power level to obtain the error fitting result; The prediction accuracy of renewable power typically decreases with increasing time scale, and the prediction error tends to increase over time. When describing the prediction error characteristics using probability distributions, it is necessary to perform error fitting in different time periods to represent this temporal nature. The uncertainty of renewable power prediction error is also related to the output level. Generally, the higher the renewable power output level, the larger the prediction error, and vice versa; that is, there is a conditional correlation between prediction error and output level. While considering time-period fitting, prediction error distributions for different power levels are also fitted. Specifically, the predicted power is divided into ten levels according to rated capacity, and the prediction error distribution is fitted every 0.1 pu.

[0027] S13, based on the error fitting results obtained in S12, uses the t-copula distribution to construct an error prediction model to capture the characteristics of the prediction error. This model can better express the heavy-tailed effect of the prediction error and is suitable for subsequent chance-constrained programming.

[0028] Specifically, this includes: time correlation modeling Let the prediction error vectors for t time periods be respectively According to copula theory, there exists a copula function such that the cumulative distribution function of the prediction error over t time periods is expressed as: (1) in, It is the copula density function; Let be the marginal distribution function of the prediction error random variable for the t-th time period, used to characterize the probability distribution of the prediction error for that time period.

[0029] The joint probability density function considering time correlation can then be expressed as: (2) in, This represents the marginal distribution of the prediction error in the i-th time period; This represents the product of the marginal density functions of the prediction error for all time periods.

[0030] Conditional correlation modeling According to copula theory, the predicted power y and its corresponding actual power x, the joint probability density function of x and y under any random sequence is: (3) in, Let represent the joint probability density function of the predicted power y and the actual power x; This represents the joint distribution function of predicted power and actual power; and These represent the marginal distribution functions of the actual power and the predicted power, respectively; and These represent the marginal probability density functions of the actual power and the predicted power, respectively; , It is the partial derivative with respect to x and y, representing the joint rate of change of the joint distribution with respect to the two variables.

[0031] make ,but The conditional probability density function of the true value under the predicted value is expressed as follows: (4) in, Indicates the predicted value Under the given conditions, the conditional probability density function of the actual power x; This represents the predicted power value; This indicates that when the predicted power is The joint probability density function at time refers to the actual power x and the predicted power x. The joint probability density function at time; The predicted power is in the range of values. The marginal probability density function at that time; Actual power at Marginal distribution function at; Predicted power in Marginal distribution function at; Actual power at The marginal probability density function at the point; e represents the prediction error, that is, the deviation between the actual power and the predicted power.

[0032] S14. Based on the obtained error prediction model and combined with the output prediction data obtained in S11, the prediction error confidence interval is obtained.

[0033] Step 2: Day-ahead optimization scheduling modeling S21, Construct a day-ahead scheduling model with the goal of maximizing market returns. It represents the total revenue of the entire virtual power plant system during day-ahead operation; Indicates the total time interval of the plan; It is the electricity price of the system's electricity sold on the market; It's the cost of charging and discharging the battery; It is the market price for electricity; It is the final optimized and cleared declaration power of the system; It refers to the power purchased in the optimization results; It is the interactive power of the energy storage device.

[0034] The day-ahead optimal economic dispatch problem of this virtual power plant model can be expressed as: (5) in: (6) (7) A value greater than 0 indicates the charging status. Indicates; a value less than 0 indicates the discharge state. express, and These represent the unit power charge / discharge cost of the battery, calculated as follows: (8) in, It is the investment cost of the battery. and These represent the charging and discharging power within the time interval t, respectively. and These represent the charging efficiency and discharging efficiency of the battery, respectively. It refers to the number of battery cycles.

[0035] S22, the constraints include power balance constraints, energy storage SOC and charge / discharge constraints, and grid interaction power constraints, among which: Power balance constraints (9a) (9b) (9c) in, This represents the wind power output at time t; Indicates the photovoltaic power at time t; This represents the output power of wind and solar power at time t after energy storage; and These represent the charging and discharging power of the energy storage device at time t, respectively. This represents the load power at time t.

[0036] Energy storage SOC and charge / discharge constraints (10a) (10b) (10c) in, and These represent the maximum and minimum SOC values ​​of the energy storage device, respectively. The maximum value of energy storage power interaction at time t; Indicates the initial SOC value of energy storage; This represents the summation at all points in time. This represents the state-of-the-art (SOA) variable of an energy storage battery during time period t, where 1 indicates charging and 0 indicates discharging. This indicates the battery's maximum capacity.

[0037] Power grid interaction constraints (11) in, This indicates the maximum power traded with the grid on a given day.

[0038] S23, using the distribution parameters of the prediction error to generate confidence intervals, and incorporating this uncertainty into the optimization model through chance constraints, thus obtaining: objective function (12) Constraints

[0039] (13) in, It is a probability operator. Indicates the confidence level.

[0040] S24, solve for the day-ahead energy reporting curve and energy storage operation strategy.

[0041] Step 3: Real-time scheduling to compensate for deviations S31, real-time acquisition of renewable energy, i.e., actual wind and solar power output, compared with day-ahead plans. S32, calculate the deviation and introduce reserve compensation power to construct a real-time scheduling model, where: Real-time scheduling objective function (14) in, It is the compensation power during the real-time operation phase; It is the backup power during the real-time operation phase; This represents the total revenue function of a virtual power plant in the real-time market. and These represent the compensation cost and reserve cost per unit power during the real-time phase, respectively.

[0042] Constraints

[0043] S33, solve for the energy storage and grid interaction strategy.

[0044] Step 4: Verification Process The effectiveness of the proposed method was verified through MATLAB program simulation and a hardware-in-the-loop experimental platform for the controller, such as... Figure 2 The diagram shown is a flowchart illustrating the specific implementation of the two-stage joint optimization energy management strategy proposed in this invention. Figure 3 The flowchart shows the two-stage optimization process for the specific program execution, from the final day-to-day optimal scheduling result to the specific rolling optimization result in real time. Figure 4 This describes the architecture of the hardware-in-the-loop simulation platform. In the program case study, the key parameters of the system are set as follows: , , , , , , , , , , , The system includes all historical market electricity prices and predicted outputs for wind and solar power generation, as shown below. Figure 5 As shown, and at the same time Figure 6 The diagram illustrates a characteristic study of historical renewable energy forecasting errors, yielding a distribution characteristic map of forecasting errors that includes both time and conditional correlations. Figure 7 As shown, the prediction results of any 96h confidence interval obtained based on the fitted distribution verify the characteristics of the prediction error distribution.

[0045] Based on the distribution of prediction error characteristics, opportunity constraints are added to perform day-ahead optimization modeling to obtain the day-ahead reporting curve and energy storage operation strategy, as follows: Figure 8-10As shown, in the day-ahead market, energy storage devices are in charging mode between 2:00 and 4:00 AM. During this time, wind and solar power generation is low, and electricity prices are low. To meet the minimum local load demand, electricity is purchased from the grid for day-ahead reporting, and the low-cost purchased electricity is stored in batteries. When the day-ahead electricity price gradually increases to its two maximum peaks between 7:00 and 9:00 AM and between 1:00 and 4:00 PM, the energy storage devices supplement wind and solar power output as the main reported electricity volume, and the system no longer selects to purchase electricity from the market. Figure 9 It shows the power balance during the daytime operation of the system, the energy change status of each component in the system, and the clearing result of the output curve reported by the virtual power plant system to the dispatching agency.

[0046] To verify the superiority of the proposed model, the economic benefits under different chance constraints are compared, such as... Figure 11 As shown in Table 1, the economic benefits of adding opportunity constraints, without considering the uncertainty of renewable energy, are significantly improved compared to not having opportunity constraints, and the best economic benefits are achieved at a certain confidence level.

[0047] Table 1:

[0048] Example 3 This embodiment provides a two-stage joint optimization dispatch system for virtual power plants oriented towards electricity market participation, including: The confidence interval acquisition unit is used to characterize the uncertainty of renewable energy and obtain the confidence interval of the prediction error. The date scheduling model construction unit is used to construct a virtual day-ahead scheduling model for power plants with the goal of maximizing market returns, based on the obtained prediction error confidence interval. The energy storage operation strategy acquisition unit is used to obtain the day-ahead energy declaration curve and energy storage operation strategy of the virtual power plant based on the obtained day-ahead scheduling model of the virtual power plant. The output deviation calculation unit is used to obtain the actual output of renewable energy in real time and calculate the output deviation between the actual output and the obtained date energy declaration curve. The real-time scheduling model construction unit is used to construct a virtual power plant real-time scheduling model based on the obtained output deviation. The interaction strategy generation unit is used to generate interaction strategies between energy storage and the power grid based on the obtained virtual power plant real-time scheduling model.

[0049] Example 4 This embodiment also provides a computing device. The computing device includes a bus, a processor, a memory, and a communication interface. The processor, memory, and communication interface communicate with each other via the bus. The computing device can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memory in the computing device.

[0050] A bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, a bus can include a path for transmitting information between various components of a computing device (e.g., memory, processor, communication interfaces).

[0051] The processor may include any one or more of the following: central processing unit (CPU), graphics processing unit (GPU), tensor processing unit (TPU), application specific integrated circuit (ASIC), field-programmable gate array (FPGA), microprocessor (MP), or digital signal processor (DSP).

[0052] The memory may include volatile memory, such as random access memory (RAM). The processor may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0053] The memory stores executable program code, which the processor executes to implement the functions of the aforementioned units, thereby achieving, for example, the method described in Embodiment 1. That is, the memory may store instructions for the methods and functions relating to the computing device in any of the above embodiments.

[0054] The communication interface uses transceiver modules such as, but not limited to, network interface cards and transceivers to enable communication between computing devices and other devices or communication networks.

[0055] Example 5 This embodiment also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, cause the processor to perform the methods and functions of the computing device involved in any of the above embodiments.

[0056] Generally, the various embodiments of this disclosure can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects can be implemented in hardware, while others can be implemented in firmware or software, which can be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of this disclosure are shown and described as block diagrams, flowcharts, or represented using some other illustration, it should be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as, as non-limiting examples, in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0057] Example 6 This embodiment provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods as described above with reference to the accompanying drawings. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0058] Computer program code used to implement the methods of this disclosure may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the computer or other programmable data processing apparatus, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be performed. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0059] In the context of this disclosure, computer program code or related data may be carried on any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and so on. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0060] Computer-readable media can be any tangible medium that contains or stores programs for or relating to an instruction execution system, apparatus, or device, or a data storage device such as a data center containing one or more available media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. More detailed examples of computer-readable storage media include electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0061] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation, characterized in that, Includes the following steps: Step 1: Characterize the uncertainty features of renewable energy and obtain the prediction error confidence interval; Step 2: Based on the obtained prediction error confidence interval, construct a virtual power plant day-ahead scheduling model with the objective of maximizing market returns; Step 3: Based on the obtained day-ahead scheduling model of the virtual power plant, obtain the day-ahead energy application curve and energy storage operation strategy corresponding to the virtual power plant; Step 4: Obtain the actual output of renewable energy in real time, and calculate the output deviation between the actual output and the date energy reporting curve obtained in Step 3; Step 5: Construct a real-time scheduling model for the virtual power plant based on the obtained output deviation; Step 6: Generate an interaction strategy between energy storage and the power grid based on the obtained virtual power plant real-time scheduling model.

2. The two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation as described in claim 1, characterized in that, In step 1, the uncertainty characteristics of renewable energy are characterized to obtain the prediction error confidence interval. The specific method is as follows: S11, obtain day-ahead wind and solar power output forecast data; S12, the acquired historical prediction error data is fitted with the error distribution by time period and power level to obtain the error fitting result; S13. Based on the obtained error fitting results, an error prediction model is constructed using the t-copula distribution; S14. Based on the obtained error prediction model and combined with the output prediction data obtained in S11, the prediction error confidence interval is obtained.

3. The two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation as described in claim 1, characterized in that, In step 2, based on the obtained prediction error confidence interval, a virtual power plant day-ahead dispatch model is constructed with the objective of maximizing market returns. The specific method is as follows: The virtual power plant day-ahead scheduling model includes an optimization objective function and constraints, wherein the expression for the optimization objective function is: In the formula, The total revenue of the entire system during the current day's operation; The total time interval for the plan; The price of electricity sold by the system in the market; Cost of charging and discharging batteries; The market-based electricity purchase price; The final optimized clearing declaration power of the system; To optimize the power purchase volume in the results; The interactive power of the energy storage device; The constraints include power balance constraints, energy storage SOC and charge / discharge constraints, grid interaction power constraints, and opportunity constraints. Among them, the opportunity constraints are constructed based on the obtained prediction error confidence intervals.

4. The two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation as described in claim 3, characterized in that, The power balance constraint is: (9a) (9b) (9c) in, Let t be the wind power output at time t; Let be the photovoltaic power at time t; Let t be the output power of wind and solar power after energy storage; and These represent the charging and discharging power of the energy storage device at time t, respectively. Let be the load power at time t; Energy storage SOC and charge / discharge constraints are: (10a) (10b) (10c) in, and These represent the maximum and minimum SOC values ​​of the energy storage device, respectively. The maximum value of energy storage power interaction at time t; Indicates the initial SOC value of energy storage; This represents the summation at all points in time. This represents the state-of-the-art (SOA) variable of an energy storage battery during time period t. Indicates the battery's maximum capacity; The power grid interaction constraint is: (11) in, This indicates the maximum power traded with the grid on the previous day; The opportunity constraint is: in, It is a probability operator. Indicates the confidence level.

5. The two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation as described in claim 1, characterized in that, In step 3, the deviation between the actual output of renewable energy and the day-ahead plan is obtained in real time. The energy storage and grid interaction strategy is dynamically adjusted through reserve power and compensation mechanisms. The specific method is as follows: Based on the deviation between the actual output of renewable energy and the day-ahead plan, and combined with reserve power, a real-time dispatch model is constructed. The energy storage and grid interaction strategy is obtained by using the real-time scheduling model.

6. The two-stage joint optimization scheduling method for virtual power plants oriented towards electricity market participation as described in claim 1, characterized in that, In step 5, the constructed virtual power plant real-time scheduling model includes a real-time scheduling objective function and constraints, wherein: The expression for the real-time scheduling objective function is: In the formula, The total revenue function for the virtual power plant in the real-time market; The total time interval for the plan; For electricity prices; Cost of charging and discharging batteries; The market-based electricity purchase price; and These are the compensation cost and backup cost per unit power during the real-time phase, respectively. It is the final optimized and cleared declaration power of the system; To optimize the power purchase volume in the results; The interactive power of energy storage devices in a virtual power plant; This refers to the compensation power during the real-time operation phase. This is the backup power for the real-time operation phase; The constraints include power balance constraints, energy storage SOC and charge / discharge constraints, grid interaction power constraints, and opportunity constraints. Among them, the opportunity constraints are constructed based on the obtained prediction error confidence intervals.

7. A two-stage joint optimization dispatch system for virtual power plants oriented towards electricity market participation, characterized in that, include: The confidence interval acquisition unit is used to characterize the uncertainty of renewable energy and obtain the confidence interval of the prediction error. The date scheduling model construction unit is used to construct a virtual day-ahead scheduling model for power plants with the goal of maximizing market returns, based on the obtained prediction error confidence interval. The energy storage operation strategy acquisition unit is used to obtain the day-ahead energy declaration curve and energy storage operation strategy of the virtual power plant based on the obtained day-ahead scheduling model of the virtual power plant. The output deviation calculation unit is used to obtain the actual output of renewable energy in real time and calculate the output deviation between the actual output and the obtained date energy declaration curve. The real-time scheduling model construction unit is used to construct a virtual power plant real-time scheduling model based on the obtained output deviation. The interaction strategy generation unit is used to generate interaction strategies between energy storage and the power grid based on the obtained virtual power plant real-time scheduling model.

8. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 6.

9. A computer program product, characterized in that, The computer program product includes computer-executable instructions that, when executed, implement the method of any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method of any one of claims 1 to 6.

Images