Market clearing method and system based on virtual power plant gradeability constraint

Abstract

A market clearing method and system based on virtual power plant gradeability constraint can improve the feasibility of clearing results and the new energy consumption level, and the method comprises the steps: collecting the real-time operation data of a load side, a power side and an energy storage side of a virtual power plant in a current scheduling period, using the prediction model to generate load prediction and new energy output prediction of the next scheduling period; calculating the maximum rising rate and the maximum falling rate according to the various adjustable resource operation parameters, determining the up-regulation capability and the down-regulation capability, and aggregating to form a virtual power plant foundation climbing capability curve; in combination with the output plan of the current scheduling period, effective upper climbing ability and effective lower climbing ability between adjacent scheduling periods are obtained, and a virtual power plant climbing constraint is constructed; and establishing a market clearing optimization model including electric power balance, output upper and lower limits, power grid safety and climbing constraints, and solving by taking the minimum system operation cost and / or the maximum market overall benefit as a target to obtain clearing electric quantity of each market subject and a virtual power plant output plan.

Description

Technical Field

[0001] This application relates to the field of virtual power plant market clearing technology, and in particular to a market clearing method and system based on virtual power plant ramp-up capability constraints. Background Technology

[0002] With the increasing proportion of new energy sources such as wind power and photovoltaics being integrated into the grid, the power output volatility and uncertainty of the power system have increased significantly. To improve the efficiency of the aggregation and utilization of resources such as distributed power sources and energy storage, virtual power plants, by aggregating various adjustable resources such as distributed power sources, energy storage devices, and adjustable loads, participate in the power market clearing and dispatch execution as a unified entity, becoming an important means to improve the capacity for new energy absorption and enhance the grid regulation capacity.

[0003] Existing market clearing methods typically construct optimization models based on load forecasting results and electricity price and volume bids submitted by various market participants. These models consider power balance constraints, upper and lower limits of unit output, and grid security constraints, aiming to minimize system operating costs or maximize market benefits during the clearing calculation. However, in the context of virtual power plants, existing methods often treat them as single generating units or only introduce simple ramp-up rate constraints at the unit level. They lack a mechanism for unified modeling and dynamic aggregation of the upward and downward adjustment capabilities of various adjustable resources within the virtual power plant.

[0004] Meanwhile, existing technologies, when conducting multi-period market clearing, typically do not fully integrate the output plan of the current dispatch cycle with the available ramp-up capacity of the next dispatch cycle. They lack a refined description of ramp-up constraints for output changes between adjacent dispatch cycles, making it difficult to promptly reflect the effective ramp-up and ramp-down capabilities of virtual power plants between consecutive dispatch cycles. This can easily lead to problems in actual implementation of market clearing results, such as power ramp-up exceeding physical capabilities, the need for frequent re-dispatch, or reduction of renewable energy output. This is detrimental to ensuring the feasibility of dispatch schemes and improving the virtual power plant's ability to absorb renewable energy fluctuations. Summary of the Invention

[0005] To overcome the problems in the prior art that it failed to accurately characterize and constrain the overall ramp-up capability of virtual power plants during the market clearing process, resulting in insufficient feasibility of dispatching results and low efficiency of renewable energy consumption, this invention provides a market clearing method based on the ramp-up capability constraint of virtual power plants.

[0006] The technical solution of the present invention is as follows: A market clearing method based on virtual power plant ramp-up capacity constraints includes: Real-time operating data of the load side, power supply side and energy storage side in the virtual power plant are collected in the current scheduling cycle, and load forecast results and new energy output forecast results for the next scheduling cycle are generated based on the prediction model. Obtain the operating parameters of various adjustable resources, calculate the maximum rise rate and maximum fall rate of various adjustable resources in the next scheduling cycle based on the operating parameters of various adjustable resources, determine the upward and downward adjustment capabilities of various adjustable resources based on the maximum rise rate and maximum fall rate, aggregate the upward and downward adjustment capabilities of various adjustable resources, and form the basic ramp-up capability curve of the virtual power plant in the next scheduling cycle. Obtain the output plan of various adjustable resources in the virtual power plant in the current scheduling cycle. Based on the output plan of various adjustable resources in the virtual power plant in the current scheduling cycle and the basic ramping capability curve of the next scheduling cycle, obtain the effective ramping capability and effective ramping capability of the virtual power plant between adjacent scheduling cycles. Based on the effective ramping capability and effective ramping capability, construct the ramping constraint of the virtual power plant between adjacent scheduling cycles. Based on the load forecast results, the new energy output forecast results, and the ramping constraints of virtual power plants between adjacent dispatch cycles, a market clearing optimization model is constructed with the electricity price and electricity quotation submitted by each market participant as input variables. The constraints of the market clearing optimization model include at least power balance constraints, upper and lower limits of unit output constraints, grid security constraints, and ramping constraints of virtual power plants between adjacent dispatch cycles. With the goal of minimizing system operating costs and / or maximizing overall market benefits, the market clearing optimization model is solved to obtain the cleared electricity volume of each market participant in the next scheduling cycle and the output plan of the virtual power plant.

[0007] Furthermore, after obtaining the cleared electricity volume of each market entity in the next dispatch cycle and the output plan of the virtual power plant, the upward and downward adjustment capabilities of various adjustable resources in the virtual power plant are decomposed and allocated and controlled according to the output plan of the virtual power plant in the next dispatch cycle. The actual output of the virtual power plant in the next scheduling cycle is obtained, and the actual output, together with the real-time operating data of the load side, power supply side and energy storage side, is used as a new input for the prediction model. The load prediction results, new energy output prediction results and basic ramp-up capability of the virtual power plant are updated on a rolling basis for subsequent scheduling cycles.

[0008] Furthermore, determine the upward and downward adjustment capabilities of various adjustable resources in the next scheduling cycle, including: Multiplying the maximum rate of increase of various adjustable resources by the scheduling cycle length yields the theoretical increase in adjusted output; multiplying the maximum rate of decrease of various adjustable resources by the scheduling cycle length yields the theoretical decrease in adjusted output. Obtain the upper and lower limits of output for various adjustable resources in the current scheduling cycle; The difference between the current output and the upper limit of the adjustable resource is used as the adjustment margin, and the smaller value between the theoretical increase in output and the adjustment margin is determined as the adjustment capacity of the adjustable resource in the next scheduling cycle. The difference between the current output and the lower limit of the output of various adjustable resources is used as the adjustment margin. The smaller value between the theoretical reduction in output and the adjustment margin is determined as the adjustment capacity of the adjustable resources in the next scheduling cycle.

[0009] Furthermore, a basic ramp-up capability curve for the virtual power plant in the next dispatch cycle is formed, including: According to the preset time period, the upward and downward adjustment capabilities of various adjustable resources in the next scheduling cycle are superimposed in time sequence to obtain the total upward and downward adjustment capabilities of the virtual power plant in each time period. The basic ramp-up capability curve is constructed by combining the total upward and downward adjustment capabilities of the virtual power plant in each time period.

[0010] Furthermore, based on the output plans of various adjustable resources within the virtual power plant in the current scheduling cycle and the basic ramp-up capability curve for the next scheduling cycle, the effective ramp-up capability and effective ramp-down capability of the virtual power plant between adjacent scheduling cycles are obtained, including: Based on the output plans of various adjustable resources in the current scheduling cycle, the baseline output of the virtual power plant at the end of the current scheduling cycle and the planned output at the beginning of the next scheduling cycle are determined, and the output change between the baseline output and the planned output is obtained. Under the constraint of the basic climbing capacity curve, the output change is allocated between adjacent scheduling cycles according to preset time segments to obtain the upward and downward power used to meet the output plan in each time segment; At each time segment, the total upward adjustment capacity in the basic ramp-up capacity curve is reduced by the upward adjustment power occupied by the power output plan, and the total downward adjustment capacity in the basic ramp-up capacity curve is reduced by the downward adjustment power occupied by the power output plan to obtain the remaining upward reserve capacity and the remaining downward reserve capacity. The remaining upward reserve capacity and the remaining downward reserve capacity are reduced according to a preset safety margin coefficient. The reduced upward reserve capacity is taken as the effective uphill capacity, and the reduced downward reserve capacity is taken as the effective downhill capacity.

[0011] Furthermore, based on the effective uphill and downhill capabilities, a ramp constraint is constructed for the virtual power plant between adjacent scheduling cycles, including: Determine the uplink and downlink constraints of the virtual power plant between adjacent scheduling cycles; In the aforementioned uplink constraint, the output increment between the planned output of the virtual power plant in the next scheduling cycle and the planned output of the virtual power plant in the previous scheduling cycle is limited to the effective ramp-up capability of the corresponding time period. In the downlink constraint, the output reduction between the planned output of the virtual power plant in the previous scheduling cycle and the planned output of the virtual power plant in the next scheduling cycle is limited to not exceeding the effective downhill ramping capacity of the corresponding time period. The uplink and downlink constraints are incorporated into the market clearing optimization model as ramping constraints for the virtual power plant between adjacent scheduling cycles.

[0012] Furthermore, the prediction module construction process is as follows: Historical operating data of the virtual power plant's load side, power supply side, and energy storage side are obtained, along with corresponding meteorological data, electricity price data, and holiday information. The historical operational data is aligned and cleaned with meteorological data, electricity price data, and holiday information to construct a training sample set containing multi-dimensional features; The training sample set is divided into a training set and a validation set. The training set is iteratively trained based on a preset deep learning model to obtain the model parameters of the load prediction sub-model and the new energy output prediction sub-model. The prediction accuracy of the load prediction sub-model and the new energy output prediction sub-model is evaluated on the validation set. The model structure or training hyperparameters are adjusted according to the evaluation results until the prediction accuracy meets the preset threshold, which is set according to the accuracy requirements of the model. The load prediction sub-model and the new energy output prediction sub-model that meet the requirements are merged into the prediction model.

[0013] Furthermore, with the goal of minimizing system operating costs and / or maximizing overall market benefits, the market clearing optimization model is solved, including: Construct an objective function that includes a system operating cost objective and a market overall benefit objective. The system operating cost objective is used to characterize the comprehensive operating cost of generating units, energy storage devices, adjustable loads, and renewable energy output. The market overall benefit objective is used to characterize the comprehensive benefits of the degree of matching between electricity supply and demand, electricity price level, and renewable energy consumption level. Based on preset weighting coefficients, the system operating cost objective item and the overall market benefit objective item are weighted and synthesized to obtain a comprehensive objective function; Under the conditions of satisfying power balance constraints, unit output upper and lower limit constraints, grid security constraints, and virtual power plant ramp-up constraints between adjacent scheduling cycles, the comprehensive objective function is solved to obtain the cleared power of each market participant in the next scheduling cycle and the output plan of the virtual power plant in each time period.

[0014] Furthermore, the adjustable resources include one or more of conventional generator sets, adjustable loads, energy storage devices, and distributed power sources; The operating parameters of adjustable resources include at least: rated output, maximum current output, start / stop status, maximum rate of rise, maximum rate of fall, and the capacity and state of charge of the energy storage device.

[0015] A market clearing system based on virtual power plant ramp-up capacity constraints, characterized in that it includes: Processor and memory; The processor and memory are connected via a communication bus: The processor is used to call and execute the program stored in the memory; The memory is used to store a program, which is at least used to execute a market clearing method based on virtual power plant ramping capacity constraints as described in any one of claims 1-9.

[0016] The beneficial effects of this invention are: By explicitly introducing virtual power plant ramp-up capacity modeling and constraints at the market clearing level, the feasibility of the clearing results and the level of renewable energy absorption are improved. First, real-time operational data from the load side, power generation side, and energy storage side are collected during the current dispatch cycle. Based on the predictive model, load forecasts and renewable energy output forecasts for the next dispatch cycle are generated, enabling market clearing to make decisions based on future load and renewable energy fluctuations, thus improving adaptability to uncertainty. Second, the maximum ramp-up rate and maximum ramp-down rate are calculated based on the operating parameters of various adjustable resources and converted into upward and downward adjustment capabilities. These are then aggregated within a preset time period to form a basic ramp-up capacity curve for the virtual power plant, achieving a characterization of ramp-up capacity from resource units to the entire virtual power plant. Compared to setting simple ramp-up constraints only at the single-unit level, this more accurately reflects the adjustment space of the virtual power plant. Based on this, the power output plan for the current dispatch cycle is combined with the basic ramp-up capacity curve to obtain the effective ramp-up and ramp-down capacities between adjacent dispatch cycles. Based on this, ramp-up constraints for virtual power plants between adjacent dispatch cycles are constructed, limiting power output changes in adjacent dispatch cycles to the ramp-up capacity of virtual power plants and reducing the risk of sudden power fluctuations. Finally, the ramp-up constraints of virtual power plants, along with power balance constraints, unit output upper and lower limit constraints, and grid security constraints, are incorporated into the market clearing optimization model. The solution aims to minimize system operating costs and / or maximize overall market benefits, achieving a balance between economic efficiency and feasibility in cleared power volume and virtual power plant output plans. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0018] Figure 1 This is a flowchart illustrating a market clearing method based on virtual power plant ramp-up capability constraints provided in an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0020] A market clearing method based on virtual power plant ramping capacity constraints, referring to Figure 1 ,include: The virtual power plant master station collects real-time operating data of active power output, status variables, and SOC from the load-side smart meters, the power-side unit monitoring system, and the energy storage converter controller through the communication interface in a 15-minute scheduling cycle. Based on the prediction model, it generates load forecast results and new energy output forecast results for the next scheduling cycle. Subsequently, based on the rated output, maximum current output, start / stop status, maximum ramp rate, maximum ramp rate, and capacity and state-of-charge operating parameters of various adjustable resources reported by the main station, the up-and-down capabilities of each resource in the next scheduling cycle are calculated and aggregated over time to form the basic ramp-up capability curve of the virtual power plant in the next scheduling cycle. The adjustable resources include one or more of conventional generator sets, adjustable loads, energy storage devices, and distributed power sources. The main station obtains the effective uphill and downhill capabilities of the virtual power plant between adjacent scheduling cycles based on the output plans of various adjustable resources in the virtual power plant in the current scheduling cycle and the basic ramp-up capability curve for the next scheduling cycle (the basic ramp-up capability is reduced). Based on the effective uphill and downhill capabilities, the main station constructs ramp-up constraints for the virtual power plant between adjacent scheduling cycles. Based on the above load forecast results, new energy output forecast results, and the ramping constraints of virtual power plants between adjacent dispatch cycles, the main station uses the electricity price and power quotation submitted by each market participant as input variables to construct a market clearing optimization model that includes power balance, upper and lower limits of unit output, grid security, and ramping capacity constraints of virtual power plants (ramping constraints between adjacent dispatch cycles). With the goal of minimizing system operating costs and / or maximizing overall market benefits, the market clearing optimization model is solved, and the clearing power of each market participant in the next scheduling cycle and the time-of-use output plan of the virtual power plant are output.

[0021] This invention improves the feasibility of market clearing results and the level of renewable energy absorption by explicitly introducing virtual power plant ramp-up capacity modeling and constraints at the market clearing level. First, real-time operational data from the load side, power source side, and energy storage side are collected during the current scheduling cycle. Based on a predictive model, load forecasts and renewable energy output forecasts for the next scheduling cycle are generated, enabling market clearing to make decisions based on future load and renewable energy fluctuations, thus improving adaptability to uncertainty. Second, the maximum rate of increase and the maximum rate of decrease are calculated based on the operating parameters of various adjustable resources and converted into upward and downward adjustment capabilities. These are then aggregated within a preset time period to form a basic ramp-up capacity curve for the virtual power plant, achieving a characterization of ramp-up capacity from resource units to the entire virtual power plant. Compared to setting simple ramp-up constraints only at the single-unit level, this more accurately reflects the adjustment space of the virtual power plant. Based on this, the power output plan for the current dispatch cycle is combined with the basic ramp-up capacity curve to obtain the effective ramp-up and ramp-down capacities between adjacent dispatch cycles. Based on this, ramp-up constraints for virtual power plants between adjacent dispatch cycles are constructed, limiting power output changes in adjacent dispatch cycles to the ramp-up capacity of virtual power plants and reducing the risk of sudden power fluctuations. Finally, the ramp-up constraints of virtual power plants, along with power balance constraints, unit output upper and lower limit constraints, and grid security constraints, are incorporated into the market clearing optimization model. The solution aims to minimize system operating costs and / or maximize overall market benefits, achieving a balance between economic efficiency and feasibility in cleared power volume and virtual power plant output plans.

[0022] It should be noted that this market clearing method based on virtual power plant ramp-up capacity constraints also includes: In the next scheduling cycle, the upward and downward adjustment capabilities of various adjustable resources within the virtual power plant are decomposed and allocated according to the output plan of the virtual power plant. The actual output of the virtual power plant in the next scheduling cycle is obtained, and the actual output, along with the real-time operating data of the load side, power supply side, and energy storage side, is used as new inputs to continuously update the load forecast results, new energy output forecast results, and basic ramp-up capability curve of the virtual power plant for subsequent scheduling cycles.

[0023] In one implementation, after obtaining the time-of-use output plan of the virtual power plant for the next scheduling cycle, the master station decomposes the planned output at the virtual power plant level to conventional units, energy storage devices, and adjustable loads according to the upward and downward adjustment capabilities of various adjustable resources. This forms resource-level execution instructions, which are then sent to the field control equipment via the scheduling communication channel. Each resource tracks and executes the corresponding active power output instructions based on its own unit constraints or equipment constraints. At the end of the scheduling cycle, the master station again collects the actual output of the virtual power plant during that cycle and uses it, along with real-time operating data from the load side, power supply side, and energy storage side, as new input data to update the input sequence of the prediction model and the basic ramp-up capability curve parameters of the virtual power plant, thereby achieving rolling prediction and rolling optimization in subsequent scheduling cycles.

[0024] It should be noted that determining the upward and downward adjustment capabilities of various adjustable resources in the next scheduling cycle includes: Multiplying the maximum rate of increase of various adjustable resources by the scheduling cycle length yields the theoretical increase in adjusted output; multiplying the maximum rate of decrease of various adjustable resources by the scheduling cycle length yields the theoretical decrease in adjusted output. Obtain the upper and lower limits of output for various adjustable resources in the current scheduling cycle; The difference between the current output and the upper limit of the adjustable resource is used as the adjustment margin, and the smaller value between the theoretical increase in output and the adjustment margin is determined as the adjustment capacity of the adjustable resource in the next scheduling cycle. The difference between the current output and the lower limit of output of various adjustable resources is used as the adjustment margin. The smaller value between the theoretical reduction in output and the adjustment margin is determined as the adjustment capacity of the adjustable resources in the next scheduling cycle.

[0025] The master station configures the maximum rate of increase (MW / min) and the maximum rate of decrease (MW / min) for each type of adjustable resource, and calculates the theoretical increase in output and the theoretical decrease in output based on the scheduling cycle duration T (min). At the same time, the master station obtains the upper and lower limits of output for each resource in the current scheduling cycle from the resource monitoring system. For example, for conventional units, the upper and lower limits are determined based on rated output, minimum technical output, and safety reserve. For energy storage devices, the charging and discharging limits are determined based on rated power and SOC.

[0026] For each resource, the master station calculates the difference between the current output and the upper limit of output as the upward adjustment margin, and determines the smaller value between the theoretical increase in output and the upward adjustment margin as the upward adjustment capacity of the resource in the next scheduling cycle; it also calculates the difference between the current output and the lower limit of output as the downward adjustment margin, and determines the smaller value between the theoretical decrease in output and the downward adjustment margin as the downward adjustment capacity of the resource in the next scheduling cycle.

[0027] It should be noted that the basic ramp-up capability curve for the virtual power plant in the next dispatch cycle includes: The upward and downward adjustment capabilities of various adjustable resources in the next scheduling cycle are superimposed according to the preset time period to obtain the total upward and downward adjustment capabilities of the virtual power plant in each time period. The total upward and downward adjustment capabilities of the virtual power plant in each time period constitute the basic ramp-up capability curve.

[0028] In one embodiment, the master station divides the next scheduling cycle into multiple time periods according to a preset time resolution, for example, a 15-minute cycle is divided into three 5-minute time periods. For each time period, the master station sums the upward adjustment capabilities of all resources within the same time period based on the upward and downward adjustment capabilities of various adjustable resources, obtaining the total upward adjustment capability of the virtual power plant for that time period; and sums the downward adjustment capabilities of all resources, obtaining the total downward adjustment capability of the virtual power plant for that time period. The master station arranges the total upward and downward adjustment capabilities of each time period in chronological order to form a time series, which constitutes the basic ramp-up capability curve of the virtual power plant in the next scheduling cycle, used to reflect the theoretical upward and downward adjustment space of the virtual power plant in each subdivided time period.

[0029] It should be noted that, based on the output plans of various adjustable resources within the virtual power plant in the current scheduling cycle and the basic ramp-up capability curve for the next scheduling cycle, the effective ramp-up capability and effective ramp-down capability of the virtual power plant between adjacent scheduling cycles are obtained, including: Based on the output plans of various adjustable resources in the current scheduling cycle, determine the baseline output of the virtual power plant at the end of the current scheduling cycle and the planned output at the beginning of the next scheduling cycle, and obtain the output change between the baseline output and the planned output. Under the constraint of the basic ramping capacity curve, the output change is allocated between adjacent scheduling cycles according to preset time segments to obtain the upward and downward power used to meet the output plan in each time segment; At each time segment, the total upward adjustment capacity in the basic ramp-up capacity curve is reduced by the upward adjustment power occupied by the power output plan, and the total downward adjustment capacity in the basic ramp-up capacity curve is reduced by the downward adjustment power occupied by the power output plan to obtain the remaining upward reserve capacity and the remaining downward reserve capacity. The remaining upward reserve capacity and the remaining downward reserve capacity are reduced according to the preset safety margin coefficient. The reduced upward reserve capacity is taken as the effective uphill capacity, and the reduced downward reserve capacity is taken as the effective downhill capacity.

[0030] Understandably, the master station first determines the baseline output of the virtual power plant at the end of the current dispatch cycle based on the output plan of the current dispatch cycle, and the planned output at the beginning of the next dispatch cycle based on the market clearing results, and calculates the output change between the two. Then, under the constraint of the basic ramp-up capability curve, the master station allocates the output change in each time segment between adjacent dispatch cycles to obtain the upward or downward power required to achieve the output change in each time segment.

[0031] For each time segment, the main station subtracts the occupied upward adjustment power from the total upward adjustment capacity in the basic ramp-up capacity curve, and subtracts the occupied downward adjustment power from the total downward adjustment capacity, thereby obtaining the remaining upward adjustment reserve capacity and the remaining downward adjustment reserve capacity for that time segment.

[0032] Then, based on the preset safety margin factor (e.g., reduced by 80%), the remaining upward reserve capacity and the remaining downward reserve capacity are reduced. The resulting reduced upward reserve capacity and reduced downward reserve capacity are respectively used as the effective uphill and downhill capacity of the virtual power plant in that time segment.

[0033] It should be noted that the ramping constraints of the virtual power plant between adjacent dispatch cycles are constructed based on the effective ramping and ramping capabilities, including: Determine the uplink and downlink constraints of the virtual power plant between adjacent scheduling cycles; In the uplink constraint, the output increment between the planned output of the virtual power plant in the next scheduling cycle and the planned output of the virtual power plant in the previous scheduling cycle is limited to the effective ramp-up capability of the corresponding time period. In the downlink constraint, the output reduction between the planned output of the virtual power plant in the previous scheduling cycle and the planned output of the virtual power plant in the next scheduling cycle is limited to not exceeding the effective downhill ramping capacity of the corresponding time period. The uplink and downlink constraints are incorporated into the market clearing optimization model as ramping constraints for virtual power plants between adjacent scheduling cycles.

[0034] The master station sets up upward and downward constraints on the output changes of virtual power plants between adjacent scheduling cycles: In the upward constraints, for each time segment, the output increment between the planned output of the virtual power plant in the corresponding time segment of the next scheduling cycle and the planned output of the virtual power plant in the corresponding time segment of the previous scheduling cycle is limited to not exceeding the effective ramp-up capability of that time segment; in the downward constraints, the output decrease between the planned output of the virtual power plant in the previous scheduling cycle and the planned output of the virtual power plant in the next scheduling cycle is limited to not exceeding the effective ramp-down capability of that time segment. The master station incorporates these upward and downward constraints into the multi-period market clearing optimization model in the form of linear inequalities, enabling the model to automatically satisfy the ramp-up constraints of virtual power plants between adjacent scheduling cycles during the solution process.

[0035] It should be noted that the prediction module is constructed as follows: Acquire historical operating data from the load side, power supply side, and energy storage side of the virtual power plant, and acquire meteorological data, electricity price data, and holiday information corresponding to the historical operating data; Historical operating data is aligned and cleaned with meteorological data, electricity price data and holiday information to construct a training sample set containing multi-dimensional features, and the actual load output and actual output of new energy are used as sample labels. The training sample set is divided into a training set and a validation set. The training set is iteratively trained based on a pre-set deep learning model to obtain the model parameters of the load prediction sub-model and the new energy output prediction sub-model. The prediction accuracy of the load forecasting sub-model and the new energy output forecasting sub-model is evaluated on the validation set. The model structure or training hyperparameters are adjusted according to the evaluation results until the prediction accuracy meets the preset threshold, which is set according to the accuracy requirements of the model. The load forecasting sub-model and the new energy output forecasting sub-model that meet the requirements are merged into a prediction model.

[0036] This embodiment describes the training process of the prediction model. Specifically, the virtual power plant platform extracts operational data from the load side, power source side, and energy storage side from the historical operation database, and obtains meteorological data, electricity price data, and holiday markers for the corresponding time moments from external meteorological services and electricity price centers. After aligning the above multi-source data by timestamp, outlier removal, missing value imputation, and normalization are performed to construct a training sample set containing multi-dimensional features. The platform divides the training sample set into a training set and a validation set using historical actual load output and actual renewable energy output as labels, and performs iterative training using a deep learning model (such as a two-layer LSTM network or a convolutional-recurrent hybrid network) to obtain the parameters of the load prediction sub-model and the renewable energy output prediction sub-model. The platform evaluates the prediction error on the validation set. When the error is lower than a preset threshold, the outputs of the load prediction sub-model and the renewable energy output prediction sub-model are time-aligned and combined to form a prediction model for runtime, generating the load prediction result and renewable energy output prediction result for the next scheduling cycle in real-time operation.

[0037] It should be noted that the market clearing optimization model is solved with the goal of minimizing system operating costs and / or maximizing overall market benefits, including: Construct an objective function that includes a system operating cost objective and a market overall benefit objective. The system operating cost objective is used to characterize the comprehensive operating cost of generating units, energy storage devices, adjustable loads, and renewable energy output. The market overall benefit objective is used to characterize the comprehensive benefits of the degree of matching between electricity supply and demand, electricity price level, and renewable energy consumption level. Based on preset weighting coefficients, the system operating cost objective item and the overall market benefit objective item are weighted and synthesized to obtain a comprehensive objective function; Under the conditions of satisfying power balance constraints, unit output upper and lower limit constraints, grid security constraints, and virtual power plant ramp-up constraints between adjacent dispatch cycles, the comprehensive objective function is solved to obtain the cleared power of each market participant in the next dispatch cycle and the output plan of the virtual power plant in each time period.

[0038] In one embodiment of the present invention, when constructing the market clearing optimization model, the master station linearly or piecewise models the fuel cost function of each conventional unit, start-up and shutdown cost, charging and discharging cost of energy storage devices, compensation cost of adjustable load, and penalty cost of renewable energy curtailment as system operating cost objective items; and quantifies the degree of electricity supply and demand matching (such as load shortage penalty), electricity price stability index, and renewable energy consumption ratio as overall market benefit objective items. The master station assigns weight coefficients to the two types of objective items according to the operation strategy, forming a weighted comprehensive objective function of the system operating cost objective item and the overall market benefit objective item. In this embodiment, the weight coefficient of the system operating cost objective item is 0.4, and the weight coefficient of the overall market benefit objective item is 0.6.

[0039] Under the premise of satisfying power balance constraints, unit output upper and lower limit constraints, grid security constraints, and virtual power plant ramping constraints, the main station uses a mixed integer linear programming algorithm or other optimization algorithms to solve the comprehensive objective function, thereby obtaining the cleared power of each market participant and the output plan of the virtual power plant for each time period.

[0040] A market clearing system based on virtual power plant ramp-up capacity constraints includes: Processor and memory; The processor and memory are connected via a communication bus: The processor is used to call and execute programs stored in memory. A memory is used to store a program, which is at least used to execute a market clearing method based on virtual power plant ramp-up capacity constraints as described in any of the above embodiments.

[0041] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0042] It should be noted that in the description of this application, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means at least two.

[0043] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as will be understood by those skilled in the art to which embodiments of this application pertain.

[0044] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0045] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0046] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0047] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0048] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0049] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A market clearing method based on virtual power plant ramp-up capacity constraints, characterized in that, include: Real-time operating data of the load side, power supply side and energy storage side in the virtual power plant are collected in the current scheduling cycle, and load forecast results and new energy output forecast results for the next scheduling cycle are generated based on the prediction model. Obtain the operating parameters of various adjustable resources, calculate the maximum rise rate and maximum fall rate of various adjustable resources in the next scheduling cycle based on the operating parameters of various adjustable resources, determine the upward and downward adjustment capabilities of various adjustable resources based on the maximum rise rate and maximum fall rate, aggregate the upward and downward adjustment capabilities of various adjustable resources, and form the basic ramp-up capability curve of the virtual power plant in the next scheduling cycle. Obtain the output plan of various adjustable resources in the virtual power plant in the current scheduling cycle. Based on the output plan of various adjustable resources in the virtual power plant in the current scheduling cycle and the basic ramping capability curve of the next scheduling cycle, obtain the effective ramping capability and effective ramping capability of the virtual power plant between adjacent scheduling cycles. Based on the effective ramping capability and effective ramping capability, construct the ramping constraint of the virtual power plant between adjacent scheduling cycles. Based on the load forecast results, the new energy output forecast results, and the ramping constraints of virtual power plants between adjacent dispatch cycles, a market clearing optimization model is constructed with the electricity price and electricity quotation submitted by each market participant as input variables. The constraints of the market clearing optimization model include at least power balance constraints, upper and lower limits of unit output constraints, grid security constraints, and ramping constraints of virtual power plants between adjacent dispatch cycles. With the goal of minimizing system operating costs and / or maximizing overall market benefits, the market clearing optimization model is solved to obtain the cleared electricity volume of each market participant in the next scheduling cycle and the output plan of the virtual power plant.

2. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, After obtaining the cleared electricity volume of each market entity in the next dispatch cycle and the output plan of the virtual power plant, the upward and downward adjustment capabilities of various adjustable resources in the virtual power plant are decomposed and allocated and controlled according to the output plan of the virtual power plant in the next dispatch cycle. The actual output of the virtual power plant in the next scheduling cycle is obtained, and the actual output, together with the real-time operating data of the load side, power supply side and energy storage side, is used as a new input for the prediction model. The load prediction results, new energy output prediction results and basic ramp-up capability of the virtual power plant are updated on a rolling basis for subsequent scheduling cycles.

3. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, Determine the upward and downward adjustment capabilities of various adjustable resources in the next scheduling cycle, including: Multiplying the maximum rate of increase of various adjustable resources by the scheduling cycle length yields the theoretical increase in adjusted output; multiplying the maximum rate of decrease of various adjustable resources by the scheduling cycle length yields the theoretical decrease in adjusted output. Obtain the upper and lower limits of output for various adjustable resources in the current scheduling cycle; The difference between the current output and the upper limit of the adjustable resource is used as the adjustment margin, and the smaller value between the theoretical increase in output and the adjustment margin is determined as the adjustment capacity of the adjustable resource in the next scheduling cycle. The difference between the current output and the lower limit of the output of various adjustable resources is used as the adjustment margin. The smaller value between the theoretical reduction in output and the adjustment margin is determined as the adjustment capacity of the adjustable resources in the next scheduling cycle.

4. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, The basic ramp-up capability curve of the virtual power plant in the next dispatch cycle is formed, including: According to the preset time period, the upward and downward adjustment capabilities of various adjustable resources in the next scheduling cycle are superimposed in time sequence to obtain the total upward and downward adjustment capabilities of the virtual power plant in each time period. The basic ramp-up capability curve is constructed by combining the total upward and downward adjustment capabilities of the virtual power plant in each time period.

5. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, Based on the output plans of various adjustable resources within the virtual power plant in the current scheduling cycle and the basic ramp-up capability curve for the next scheduling cycle, the effective ramp-up capability and effective ramp-down capability of the virtual power plant between adjacent scheduling cycles are obtained, including: Based on the output plans of various adjustable resources in the current scheduling cycle, the baseline output of the virtual power plant at the end of the current scheduling cycle and the planned output at the beginning of the next scheduling cycle are determined, and the output change between the baseline output and the planned output is obtained. Under the constraint of the basic climbing capacity curve, the output change is allocated between adjacent scheduling cycles according to preset time segments to obtain the upward and downward power used to meet the output plan in each time segment; At each time segment, the total upward adjustment capacity in the basic ramp-up capacity curve is reduced by the upward adjustment power occupied by the power output plan, and the total downward adjustment capacity in the basic ramp-up capacity curve is reduced by the downward adjustment power occupied by the power output plan to obtain the remaining upward reserve capacity and the remaining downward reserve capacity. The remaining upward reserve capacity and the remaining downward reserve capacity are reduced according to a preset safety margin coefficient. The reduced upward reserve capacity is taken as the effective uphill capacity, and the reduced downward reserve capacity is taken as the effective downhill capacity.

6. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, Based on the effective uphill and downhill capabilities, a ramp constraint is constructed for the virtual power plant between adjacent scheduling cycles, including: Determine the uplink and downlink constraints of the virtual power plant between adjacent scheduling cycles; In the aforementioned uplink constraint, the output increment between the planned output of the virtual power plant in the next scheduling cycle and the planned output of the virtual power plant in the previous scheduling cycle is limited to the effective ramp-up capability of the corresponding time period. In the downlink constraint, the output reduction between the planned output of the virtual power plant in the previous scheduling cycle and the planned output of the virtual power plant in the next scheduling cycle is limited to not exceeding the effective downhill ramping capacity of the corresponding time period. The uplink and downlink constraints are incorporated into the market clearing optimization model as ramping constraints for the virtual power plant between adjacent scheduling cycles.

7. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, The prediction module is constructed as follows: Historical operating data of the virtual power plant's load side, power supply side, and energy storage side are obtained, along with corresponding meteorological data, electricity price data, and holiday information. The historical operational data is aligned and cleaned with meteorological data, electricity price data, and holiday information to construct a training sample set containing multi-dimensional features; The training sample set is divided into a training set and a validation set. The training set is iteratively trained based on a preset deep learning model to obtain the model parameters of the load prediction sub-model and the new energy output prediction sub-model. The prediction accuracy of the load prediction sub-model and the new energy output prediction sub-model is evaluated on the validation set. The model structure or training hyperparameters are adjusted according to the evaluation results until the prediction accuracy meets the preset threshold, which is set according to the accuracy requirements of the model. The load prediction sub-model and the new energy output prediction sub-model that meet the requirements are merged into the prediction model.

8. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, The market clearing optimization model is solved with the goal of minimizing system operating costs and / or maximizing overall market benefits, including: Construct an objective function that includes a system operating cost objective and a market overall benefit objective. The system operating cost objective is used to characterize the comprehensive operating cost of generating units, energy storage devices, adjustable loads, and renewable energy output. The market overall benefit objective is used to characterize the comprehensive benefits of the degree of matching between electricity supply and demand, electricity price level, and renewable energy consumption level. Based on preset weighting coefficients, the system operating cost objective item and the overall market benefit objective item are weighted and synthesized to obtain a comprehensive objective function; Under the conditions of satisfying power balance constraints, unit output upper and lower limit constraints, grid security constraints, and virtual power plant ramp-up constraints between adjacent scheduling cycles, the comprehensive objective function is solved to obtain the cleared power of each market participant in the next scheduling cycle and the output plan of the virtual power plant in each time period.

9. The market clearing method based on virtual power plant ramp-up capability constraints according to claim 1, characterized in that, The adjustable resources include one or more of conventional generator sets, adjustable loads, energy storage devices, and distributed power sources; The operating parameters of adjustable resources include at least: rated output, maximum current output, start / stop status, maximum rate of rise, maximum rate of fall, and the capacity and state of charge of the energy storage device.

10. A market clearing system based on virtual power plant ramp-up capability constraints, characterized in that, include: Processor and memory; The processor and memory are connected via a communication bus: The processor is used to call and execute the program stored in the memory; The memory is used to store a program, which is at least used to execute a market clearing method based on virtual power plant ramping capacity constraints as described in any one of claims 1-9.

Images