Virtual power plant operation method based on demand response resource uncertainty condition in electricity market

Abstract

The invention discloses a virtual power plant operation method based on demand response resource uncertainty conditions in a power market, and the method takes demand response (DR) resources as an auxiliary means, determines photovoltaic generation error composition through the DR resources, and deduces an optimal operation strategy participating in future energy and a DR market. The invention uses a model that participates in an electricity market, in particular energy and demand response, and utilizes demand response resources to compensate for uncertainty of renewable energy products. Photovoltaic power generation probability distribution obtained based on stochastic programming is used as a framework for solving an optimization problem, and the uncertainty of photovoltaic power generation is reflected. By estimating the maximum potential demand response capability in the industrial load, a demand response modeling method is developed to improve flexibility. And the optimal operation strategy participating in the day-ahead market is obtained through the output of the models.

Description

Technical Field

[0001] The present invention relates to the field of electric power systems, and in particular to a method for operating a virtual power plant under uncertain resource conditions based on demand response in an electric power market. Background Art

[0002] The transition from traditional fossil fuel power generation to environmentally friendly and pollution-free technologies has been taking place worldwide, however, the intermittency of photovoltaic power generation may cause problems for grid operation and electricity markets. As the scale of photovoltaic power generation increases, corresponding countermeasures need to be taken. The concept of Virtual Power Plant (VPP) brings together behind the meter (BTM) resources including photovoltaic power generation and distributed energy resources (DERs), which can effectively deal with the intermittency problem of renewable energy sources (RESs). Participating in the market through VPP can support better performance than participating alone and make up for the weaknesses of individual resources. However, in order to support a large number of renewable energy sources (RESs), sufficient storage capacity must be provided, and there are also some uncertainties and disadvantages in aggregating RESs into VPP, including factors such as renewable energy production, demand resources and market prices. Summary of the invention

[0003] The purpose of the present invention is to address the deficiencies of the prior art and provide a method for operating a virtual power plant in an electricity market based on demand response resources under uncertainty conditions.

[0004] According to a first aspect of an embodiment of the present application, a method for operating a virtual power plant in an electricity market based on demand response resource uncertainty is provided, comprising:

[0005] Using historical data of photovoltaic power generation to build a photovoltaic sampling model, and by combining the photovoltaic sampling model with weather forecasts, generate a predicted photovoltaic power generation curve to determine the available demand response resources that can be dispatched;

[0006] Obtain demand response market operation results and load data, and calculate demand response-related constraints in the dispatch and allocation optimization of power resources;

[0007] The output of the photovoltaic power generation curve and the demand response constraints are used to build an optimization model and solve the optimization problem to obtain the theoretical value of the load curve. The theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation to derive the optimal operation strategy for participating in the electricity market.

[0008] Furthermore, the photovoltaic sampling model is constructed using the historical data of photovoltaic power generation. By combining the photovoltaic sampling model with weather forecasts, a predicted photovoltaic power generation curve is generated to determine the available demand response resources that can be dispatched, including:

[0009] Data collection and preprocessing: Collect historical data and weather data of photovoltaic power generation, associate weather data with photovoltaic power generation data, process missing values, abnormal values ​​and duplicate values ​​in the data, and normalize or standardize data of different dimensions as needed;

[0010] Feature selection: Select features related to photovoltaic power generation, including irradiance, temperature, weather conditions, and time characteristics;

[0011] Photovoltaic sampling model construction: The sampling model is established by modeling the normal distribution of each solar energy resource at time t;

[0012] Weather forecast data integration: collect future weather forecast data, including irradiance, temperature, cloud cover, and wind speed, preprocess the weather forecast data to make it consistent with the structure of historical weather data, and use the preprocessed weather forecast data as the input features of the model;

[0013] Generate photovoltaic power generation prediction curve: input the pre-processed weather forecast data into the trained photovoltaic power generation model to generate the prediction curve of future photovoltaic power generation;

[0014] Demand response resource estimation based on power generation forecast: Evaluate the dispatchable PV power generation according to the forecast curve of future PV power generation and determine the available demand response resources that can be dispatched.

[0015] Furthermore, the constraints include a bidding success rate and a maximum flexibility potential. The bidding success rate is calculated based on the DR market operation results, and the maximum flexibility potential is estimated using load data and a DR potential model.

[0016] Furthermore, the objective function of the optimization model, i.e. the optimal operation of the virtual power plant (VPP), is as follows:

[0017] max(Rpv+Rinc+Rec-Cec)(1)

[0018] In formula (1), R pv represents the income from photovoltaic power generation, calculated by the following equation:

[0019]

[0020] In formula (2), λ t represents the unit price at time t, P k,t represents the photovoltaic power generation, the subscript k is used to distinguish different photovoltaic modules, Rinc represents revenue from PV forecast incentives;

[0021]

[0022] In formula (3) It represents the forecast incentive settlement price calculated according to the forecast error rate range (FERR) at time t. represents the state variable calculated according to the prediction error rate range at time t, and T represents the time period;

[0023] Income from compensation for errors is shown;

[0024]

[0025] In formula (4), represents the photovoltaic power generation with error compensation at time t, η t represents the profit gained from compensation for error by not paying the imbalance penalty;

[0026] Opportunity Cost ec Produced by error compensation:

[0027]

[0028] In formula (5), ρ represents the average winning rate in the day-ahead economic demand response market, and the bidding success rate is used as the average winning rate. and Respectively represent the increase or decrease of demand response resources activated for error compensation.

[0029] Furthermore, the multi-integer linear programming of the optimization problem is realized by linearizing the optimization model:

[0030] The prediction error ratio range FERR is divided into five parts, the prediction error of photovoltaic power generation is graded and combined with the demand response constraints:

[0031]

[0032] The prediction error in formula (6) is the error rate ∈ t and state variables The result of multiplication;

[0033] The error rate ∈ in formula (7) tis determined by dividing the difference between the predicted value and the actual measured value by the nameplate capacity of the PV generator; if the error rate value belongs to one of the five components of FERR, then the state variable in formula (8) is equal to 1, otherwise it is zero;

[0034] The linearization process includes:

[0035]

[0036] In formula (9), pr ferr is a parameter used to distinguish the five parts, pr 1 ,pr 2 ,pr 3 ,pr 4 and pr 5 The values ​​of are -1, -0.08, -0.06, 0.06 and 0.08, and Z represents positive infinity;

[0037] The linearization is redefined as follows:

[0038]

[0039] In formula (10) It is power generation P k,t and incentive price The product of The calculation method is to convert the state variable Multiply

[0040] Therefore, the nonlinear constraint (9) is converted into a linear inequality constraint as follows:

[0041]

[0042] In formula (11), The result depends on the state variables in (9)

[0043] Prediction error rate after reorganization is the adjusted error rate θ t and state variables The product of the adjusted error rate θ t By calculation It is concluded that the error is the amount of electricity generated by error compensation, as shown in the following formulas (12) to (14):

[0044]

[0045] The linearization is redefined as follows:

[0046]

[0047] Power generation determined by error compensation It is the sum of PV power generation, DR increase, and DR decrease;

[0048]

[0049] Similarly, the constraints formed by error compensation are as follows:

[0050]

[0051] in, is the multiplication of all terms related to the prediction error interval at time t in the prediction error ratio range considering error compensation, To motivate the error compensation considered within the prediction error ratio range;

[0052] For the DR participation cap for a load resource, each value derived from the DR potential model is applied as follows:

[0053]

[0054] in, Increase the upper limit of DR participation for load resources, Reduce the upper limit of DR participation for load resources;

[0055] The constraints of the imbalance penalty are as follows:

[0056]

[0057] Solve the constraints of formulas (11), (17), and (20) to obtain the theoretical value of the load curve.

[0058] Furthermore, the theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation to derive the optimal operation strategy for participating in the power market, which is specifically:

[0059] The theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation to obtain the error rate;

[0060] The error rate is compared with a predetermined error rate threshold. If it is lower than the error rate threshold, the condition for error compensation is met. The virtual power plant operator sorts the load resources to participate in the energy market by compensating for the error of photovoltaic power generation and utilizing demand response. Otherwise, the virtual power plant operator only participates in the demand response market.

[0061] According to a second aspect of an embodiment of the present application, a virtual power plant operation device based on demand response resource uncertainty in an electricity market is provided, comprising:

[0062] The demand response resource determination module is used to construct a photovoltaic sampling model using the historical data of photovoltaic power generation, and generate a predicted photovoltaic power generation curve by combining the photovoltaic sampling model and weather forecast, so as to determine the available demand response resources that can be dispatched;

[0063] The constraint calculation module is used to obtain the demand response market operation results and load data, and calculate the constraints related to demand response in the scheduling and allocation optimization problem of power resources;

[0064] The optimization problem solving module is used to construct an optimization model and solve the optimization problem using the output of the photovoltaic power generation curve and the demand response constraints, obtain the theoretical value of the load curve, compare the theoretical value of the load curve with the load curve obtained by Monte Carlo simulation, and derive the optimal operation strategy for participating in the power market.

[0065] According to a third aspect of an embodiment of the present application, a computer program product is provided, comprising a computer program / instruction, which implements the method described in the first aspect when executed by a processor.

[0066] According to a fourth aspect of an embodiment of the present application, there is provided an electronic device, including:

[0067] one or more processors;

[0068] A memory for storing one or more programs;

[0069] When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in the first aspect.

[0070] According to a fifth aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which computer instructions are stored, and when the instructions are executed by a processor, the steps of the method described in the first aspect are implemented.

[0071] The technical solution provided by the embodiments of the present application may have the following beneficial effects:

[0072] The present invention uses a model for participating in the electricity market, especially energy and demand response, using demand response resources to compensate for the uncertainty of renewable energy products. The probability distribution of photovoltaic power generation derived based on stochastic programming is used as a framework to solve the optimization problem, reflecting the uncertainty of photovoltaic power generation. A demand response modeling method is developed to improve flexibility by estimating the maximum potential demand response capacity in industrial loads. The outputs of these models are then used to derive the optimal operation strategy for participating in the day-ahead market.

[0073] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0075] Figure 1 It is a flow chart of a method for operating a virtual power plant in an electricity market based on demand response resource uncertainty conditions according to an exemplary embodiment.

[0076] Figure 2 It is a schematic diagram showing the error rate between a theoretical value of a load curve and a load curve obtained by Monte Carlo simulation according to an exemplary embodiment.

[0077] Figure 3 It is a block diagram of a virtual power plant operation device based on demand response resource uncertainty conditions in an electricity market according to an exemplary embodiment.

[0078] Figure 4 The diagram is a schematic diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0079] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0080] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0081] It should be understood that although the terms first, second, third, etc. may be used in the present application to describe various information, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0082] The present invention is a method for operating a virtual power plant under the condition of resource uncertainty based on demand response (DR) in the power market, comprising the following steps:

[0083] Step S1: construct a photovoltaic sampling model using the historical data of photovoltaic power generation, and generate a predicted photovoltaic power generation curve by combining the photovoltaic sampling model and weather forecast, so as to determine the available DR resources that can be dispatched;

[0084] Specifically, by analyzing the historical data of photovoltaic power generation, a photovoltaic sampling model is constructed to capture the time series characteristics and changing laws of power generation. The model combines meteorological forecast information such as radiation intensity, temperature and wind speed to generate a photovoltaic power generation forecast curve for the future period. Based on this forecast curve, the dispatchable photovoltaic power generation in a specific time period can be evaluated to determine the available demand response (DR) resources. This step may include the following process:

[0085] S11: Data collection and preprocessing: Collect historical data of photovoltaic power generation and weather data, and associate weather data with photovoltaic power generation data, check missing values, outliers and duplicate values ​​in the data, use interpolation or other data filling techniques to handle missing values, and filter out obvious outliers. Normalize or standardize data of different dimensions as needed to ensure that the numerical scales of different features are consistent;

[0086] S12: Feature selection: Select features closely related to photovoltaic power generation, such as irradiance, temperature, weather conditions, time characteristics, etc.;

[0087] S13: Photovoltaic sampling model construction: The sampling model is established by modeling the normal distribution of each solar energy resource at time t;

[0088] S14: Weather forecast data integration: Collect future weather forecast data, including irradiance, temperature, cloud cover, wind speed, etc. Preprocess these weather forecast data, normalize or standardize them to make them consistent with the structure of historical weather data, and then use the weather forecast data as input features of the model;

[0089] S15: Generate photovoltaic power generation prediction curve: input future weather forecast data into the trained photovoltaic power generation model to generate a future photovoltaic power generation prediction curve;

[0090] S16: DR resource estimation based on power generation forecast: Evaluate the dispatchable photovoltaic power generation according to the forecast curve of future photovoltaic power generation, and determine the available DR resources that can be dispatched: Evaluate the current load demand according to the grid demand and dispatchable power generation, and determine the power demand that can be reduced; calculate the amount of electricity that can be used as a demand response resource by comparing the dispatchable photovoltaic power generation and load demand; optimize the dispatch strategy of DR resources according to market rules and power dispatch requirements to ensure the stability and economy of the power system.

[0091] Step S2: Obtain the demand response (DR) market operation results and load data, and calculate the DR-related constraints in the scheduling and allocation optimization problem of power resources, namely, the bidding success rate and the maximum flexibility potential;

[0092] Specifically, the bidding success rate is calculated based on the DR market operation results, and the maximum flexibility potential is estimated through the DR potential model using load data. The DR market operation results refer to the specific performance of the electricity demand adjustment achieved through demand response activities in a specific time period, which includes the reduced electricity demand of participants in DR events, market price changes, the response capabilities of participants, and the supply and demand balance of the overall system. The operation results usually reflect the impact of demand response on the stability and economy of the electricity market, and provide a basis for subsequent policy formulation and market optimization. The bidding success rate is usually calculated by analyzing historical bidding data, market demand, price fluctuations and competition. Specifically, it can include: collecting past bidding records, calculating the proportion of successful bidding, analyzing market reactions under different conditions, and evaluating influencing factors (such as price, time window and supplier capacity), and deriving a statistical model of the success rate through these data. The maximum potential calculation is implemented according to the DR potential model, and its mechanism is summarized as follows: the frequency score (FS), consistency score (CS) and operation score (OS) of power consumption are calculated from historical data, and the potential score S is defined by the product of the above three factors.

[0093] Such a design can provide clear market signals to market participants and optimize their bidding strategies, while helping the power system to dispatch resources more accurately, reduce reliance on reserve capacity, and enhance the flexibility and stability of the power grid. In addition, tapping the maximum flexibility potential of loads can help better integrate renewable energy, reduce operating costs, and provide effective data support for policymakers. All of these can improve the overall efficiency, reliability, and participation enthusiasm of the demand response market.

[0094] Step S3: construct an optimization model using the output of the photovoltaic power generation curve and the DR constraint condition and solve the multi-integer linear programming (MILP) optimization problem to obtain a theoretical value of the load curve, compare the theoretical value of the load curve with the load curve obtained by Monte Carlo simulation (MCS), and derive the optimal operation strategy for participating in the power market;

[0095] The optimization model, i.e. the objective function of the optimal operation of the virtual power plant (VPP), is as follows:

[0096] max(Rpv+Rinc+Rec-Cec)(1)

[0097] In formula (1), R pv represents the income from photovoltaic power generation, calculated by the following equation:

[0098]

[0099] In formula (2), λ t represents the unit price at time t, P k,t represents the photovoltaic power generation at time t, the subscript k is used to distinguish different photovoltaic modules, R inc Represents revenue from forecast incentives for PV.

[0100]

[0101] In formula (3) It represents the forecast incentive settlement price calculated according to the forecast error rate range (FERR) at time t. It represents the state variable calculated according to the prediction error rate range at time t, and T represents the time period.

[0102] R ec Income from error compensation is shown.

[0103]

[0104] In formula (4), represents the photovoltaic power generation with error compensation at time t, which is affected by the maximum flexibility potential; η t Represents the profit gained from error compensation by not paying imbalance penalties.

[0105] In addition, due to the error compensation, there is also an opportunity cost C ecIn other words, when flexible load resources are used as demand response resources rather than to cover PV uncertainty, demand response settlement profits are obtained, which are converted into opportunity costs.

[0106]

[0107] In formula (5), ρ represents the average winning rate of the day-ahead economic demand response market. In this embodiment, the bidding success rate is used as the average winning rate. Even if a load resource is used as an economic demand response resource, it cannot be guaranteed to participate in 100% of the demand response market; therefore, market performance is calculated and reflected. At the same time, and They represent the increase and decrease of DR resources activated for error compensation respectively.

[0108] R inc Contains a nonlinear unit, so it must be linearized to model the multi-integer linear programming (MILP) problem. FERR is divided into five parts. If the error exceeds 8%, the FERR is 1; if the error is between 6-8%, the FERR is 2, and so on.

[0109]

[0110] The prediction error in formula (6) is the error rate ∈ t and state variables The result of multiplication.

[0111] The error rate ∈ in formula (7) t The predicted value of photovoltaic power generation is Compared with the actual measured value P k,t Divide the difference by the nameplate capacity C of the PV generator PV If the error rate value belongs to one of the five parts of FERR, the state variable in formula (8) is equal to 1, otherwise it is zero.

[0112] To perform linearization, the following process is performed according to the proposed process:

[0113]

[0114] In formula (9), pr ferr is a parameter used to distinguish the five parts, pr 1 ,pr 2 ,pr 3 ,pr 4 and pr 5The values ​​of are -1, -0.08, -0.06, 0.06 and 0.08 respectively, and Z represents positive infinity.

[0115] The linearization is redefined as follows:

[0116]

[0117] In formula (10) It is power generation P k,t and incentive price The product of The calculation method is to convert the state variable Multiply

[0118] Therefore, the nonlinear constraint (9) is converted into a linear inequality constraint as follows:

[0119]

[0120] In formula (11), The result depends on the state variables in (9)

[0121] In addition, the prediction error rate after reorganization is the adjusted error rate θ t and state variables The adjusted error rate θ t By calculation It is concluded that the error is the amount of electricity generated by error compensation, as shown in the following formulas (12) to (14):

[0122]

[0123] also, The linearization is redefined as follows:

[0124]

[0125] Power generation determined by error compensation is the photovoltaic power generation P k,t , the sum of DR increase and DR decrease.

[0126]

[0127] Similar to the constraints of the linearization process mentioned above, the constraints formed by error compensation are as follows:

[0128]

[0129] in, is the multiplication of all terms related to the forecast error interval at time t in the forecast error ratio range (FREE) considering error compensation, To motivate the error compensation considered within the prediction error ratio range (FREE).

[0130] For the DR participation cap for a load resource, each value derived from the DR potential model is applied as follows:

[0131]

[0132] in, Increase the upper limit of DR participation for load resources, Reduce the upper limit of DR participation for load resources.

[0133] The constraints of the imbalance penalty are as follows:

[0134]

[0135] Solve the constraints of formulas (11), (17), and (20) to obtain the theoretical value of the load curve.

[0136] Virtual power plant (VPP) operators use Monte Carlo simulation (MCS) to formulate day-ahead market participation strategies. In this process, the theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation to obtain the error rate. The error rate for each period is pre-calculated through the above mechanism, and the error rate distribution is as follows: Figure 2 shown.

[0137] As Figure 2 As shown, the maximum error rate in each period is before 8 am and after 7 pm None of them exceeds the threshold ε of the electricity market th , i.e. 8%. That is, the VPP operator organizes its load resources to participate in the energy market by compensating for the photovoltaic error between 8 am and 7 pm when photovoltaic power generation is active. Therefore, the VPP operator adjusts its load curve L according to the predetermined error rate. MCS Considering the error rate In the case of MCS evaluation of L MCS Compared with the theoretical value L calculated by the optimization problem opt More practical significance. If the L output of the optimization model is opt is a feasible set, namely L opt YesL MCS The error compensation condition is satisfied if L is a subset of L. That is, the VPP operator can participate in the energy market by compensating the error of PV generation and utilizing demand response (DR). opt is not feasible, and Lopt Not L MCS If the VPP operator only participates in the DR market, Figure 1 shown.

[0138] Therefore, the VPP operator decides whether to participate in the day-ahead market based on the above strategy. If the forecast is successful, the VPP operator will participate in both the energy market and the DR market by activating its DR resources to compensate for the error in PV generation.

[0139] Corresponding to the aforementioned embodiment of the method for operating a virtual power plant in the electricity market based on demand response resource uncertainty conditions, the present application also provides an embodiment of a device for operating a virtual power plant in the electricity market based on demand response resource uncertainty conditions.

[0140] Figure 3 The block diagram of a virtual power plant operation device based on demand response resource uncertainty in an electricity market is shown according to an exemplary embodiment. Figure 3 , the device may include:

[0141] The demand response resource determination module 21 is used to construct a photovoltaic sampling model using historical data of photovoltaic power generation, and generate a predicted photovoltaic power generation curve by combining the photovoltaic sampling model and weather forecast, so as to determine the available demand response resources that can be dispatched;

[0142] The constraint calculation module 22 is used to obtain the demand response market operation results and load data, and calculate the constraints related to demand response in the scheduling and allocation optimization problem of power resources;

[0143] The optimization problem solving module 23 is used to construct an optimization model and solve the optimization problem using the output of the photovoltaic power generation curve and the demand response constraint conditions, obtain the theoretical value of the load curve, compare the theoretical value of the load curve with the load curve obtained by Monte Carlo simulation, and derive the optimal operation strategy for participating in the power market.

[0144] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0145] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, 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 present application scheme. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0146] Accordingly, the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the virtual power plant operation method based on demand response resource uncertainty conditions in the electricity market as described above.

[0147] Accordingly, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned virtual power plant operation method under the condition of resource uncertainty based on demand response in the power market. Figure 4 As shown, a hardware structure diagram of a virtual power plant operation device in an electricity market based on demand response resource uncertainty conditions provided by an embodiment of the present invention is located in any device with data processing capabilities, except Figure 4 In addition to the processor, memory and network interface shown, any device with data processing capability in which the apparatus in the embodiment is located may also include other hardware, generally based on the actual functions of the device with data processing capability, which will not be described in detail.

[0148] Accordingly, the present application also provides a computer-readable storage medium on which computer instructions are stored, and when the instructions are executed by the processor, the virtual power plant operation method under the uncertainty condition of demand response resources in the power market as described above is implemented. The computer-readable storage medium can be an internal storage unit of any device with data processing capabilities described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium can also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card (Flash Card), etc. equipped on the device. Furthermore, the computer-readable storage medium can also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and can also be used to temporarily store data that has been output or is to be output.

[0149] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the contents disclosed herein. The present application is intended to cover any variations, uses or adaptations of the present application, which follow the general principles of the present application and include common knowledge or customary technical means in the art that are not disclosed in the present application.

[0150] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for operating a virtual power plant under uncertainty conditions of demand response resources in an electricity market, characterized in that: The following steps are involved: Using historical data of photovoltaic power generation to build a photovoltaic sampling model, and by combining the photovoltaic sampling model with weather forecasts, generate a predicted photovoltaic power generation curve to determine the available demand response resources that can be dispatched; Obtain demand response market operation results and load data, and calculate demand response-related constraints in the dispatch and allocation optimization of power resources; The output of the photovoltaic power generation curve and the demand response constraints are used to build an optimization model and solve the optimization problem to obtain the theoretical value of the load curve. The theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation to derive the optimal operation strategy for participating in the electricity market.

2. The method according to claim 1, characterized in that The photovoltaic sampling model is constructed using the historical data of photovoltaic power generation. By combining the photovoltaic sampling model with weather forecasts, a predicted photovoltaic power generation curve is generated to determine the available demand response resources that can be dispatched, including: Data collection and preprocessing: Collect historical data and weather data of photovoltaic power generation, associate weather data with photovoltaic power generation data, process missing values, abnormal values ​​and duplicate values ​​in the data, and normalize or standardize data of different dimensions as needed; Feature selection: Select features related to photovoltaic power generation, including irradiance, temperature, weather conditions, and time characteristics; Photovoltaic sampling model construction: The sampling model is established by modeling the normal distribution of each solar energy resource at time t; Weather forecast data integration: collect future weather forecast data, including irradiance, temperature, cloud cover, and wind speed, preprocess the weather forecast data to make it consistent with the structure of historical weather data, and use the preprocessed weather forecast data as the input features of the model; Generate photovoltaic power generation prediction curve: input the pre-processed weather forecast data into the trained photovoltaic power generation model to generate the prediction curve of future photovoltaic power generation; Demand response resource estimation based on power generation forecast: Evaluate the dispatchable PV power generation according to the forecast curve of future PV power generation and determine the available demand response resources that can be dispatched.

3. The method according to claim 1, characterized in that: The constraints include a bidding success rate and a maximum flexibility potential. The bidding success rate is calculated based on the DR market operation results, and the maximum flexibility potential is estimated using load data and a DR potential model.

4. The method according to claim 1, characterized in that The objective function of the optimization model, i.e. the optimal operation of the virtual power plant (VPP), is as follows: max(Rpv+Rinc+Rec-Cec)(1) In formula (1), R pv represents the income from photovoltaic power generation, calculated by the following equation: In formula (2), λ t represents the unit price at time t, P k,t represents the photovoltaic power generation, the subscript k is used to distinguish different photovoltaic modules, R inc represents revenue from PV forecast incentives; In formula (3) It represents the forecast incentive settlement price calculated according to the forecasting error rate range (FERR) at time t. represents the state variable calculated according to the prediction error rate range at time t, and T represents the time period; Income from compensation for errors is shown; In formula (4), represents the photovoltaic power generation with error compensation at time t, η t represents the profit gained from compensation for error by not paying the imbalance penalty; Opportunity Cost ec Produced by error compensation: In formula (5), ρ represents the average winning rate in the day-ahead economic demand response market, and the bidding success rate is used as the average winning rate. and Respectively represent the increase or decrease of demand response resources activated for error compensation.

5. The method according to claim 4, characterized in that The multi-integer linear programming of the optimization problem is realized by linearizing the optimization model: The prediction error ratio range FERR is divided into five parts, the prediction error of photovoltaic power generation is graded and combined with the demand response constraints: The prediction error in formula (6) is the error rate ∈ t and state variables The result of multiplication; The error rate ∈ in formula (7) t is determined by dividing the difference between the predicted value and the actual measured value by the nameplate capacity of the PV generator; if the error rate value belongs to one of the five components of FERR, then the state variable in formula (8) is equal to 1, otherwise it is zero; The linearization process includes: In formula (9), pr ferr is a parameter used to distinguish the five parts, pr 1 ,pr 2 ,pr 3 ,pr 4 and pr 5 The values ​​of are -1, -0.08, -0.06, 0.06 and 0.08, and Z represents positive infinity; The linearization is redefined as follows: In formula (10) It is power generation P k,t and incentive price The product of is calculated by taking the state variable Multiply Therefore, the nonlinear constraint (9) is converted into a linear inequality constraint as follows: In formula (11), The result depends on the state variables in (9) Prediction error rate after reorganization is the adjusted error rate θ t and state variables The product of the adjusted error rate θ t By calculation It is concluded that the error is the amount of electricity generated by error compensation, as shown in the following formulas (12) to (14): The linearization is redefined as follows: Power generation determined by error compensation It is the sum of PV power generation, DR increase, and DR decrease; Similarly, the constraints formed by error compensation are as follows: in, is the multiplication of all terms related to the prediction error interval at time t in the prediction error ratio range considering error compensation, To motivate the error compensation considered within the prediction error ratio range; For the DR participation cap for a load resource, each value derived from the DR potential model is applied as follows: in, Increase the upper limit of DR participation for load resources, Reduce the upper limit of DR participation for load resources; The constraints of the imbalance penalty are as follows: Solve the constraints of formulas (11), (17), and (20) to obtain the theoretical value of the load curve.

6. The method according to claim 1, characterized in that The theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation, and the optimal operation strategy for participating in the power market is derived, which is: The theoretical value of the load curve is compared with the load curve obtained by Monte Carlo simulation to obtain the error rate; The error rate is compared with a predetermined error rate threshold. If it is lower than the error rate threshold, the condition for error compensation is met. The virtual power plant operator sorts the load resources to participate in the energy market by compensating for the error of photovoltaic power generation and utilizing demand response. Otherwise, the virtual power plant operator only participates in the demand response market.

7. A virtual power plant operation device based on demand response resource uncertainty in the power market, characterized in that: include: The demand response resource determination module is used to construct a photovoltaic sampling model using the historical data of photovoltaic power generation, and generate a predicted photovoltaic power generation curve by combining the photovoltaic sampling model and weather forecast, so as to determine the available demand response resources that can be dispatched; The constraint calculation module is used to obtain the demand response market operation results and load data, and calculate the constraints related to demand response in the scheduling and allocation optimization problem of power resources; The optimization problem solving module is used to construct an optimization model and solve the optimization problem using the output of the photovoltaic power generation curve and the demand response constraints, obtain the theoretical value of the load curve, compare the theoretical value of the load curve with the load curve obtained by Monte Carlo simulation, and derive the optimal operation strategy for participating in the power market.

8. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. An electronic device, characterized in that: include: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images