Methods, apparatus, computer equipment, and storage media for determining the application strategy of distributed photovoltaic aggregation entities

Abstract

This application provides a method for determining the application strategy of distributed photovoltaic aggregation entities. The method includes: obtaining the predicted time-of-use (TOU) unpriced output for each target time period based on various daily statistical data corresponding to the centrally dispatched load and market-based unit load for the predicted date, and the unpriced output prediction function for each target time period; obtaining the effective bidding capacity for each target time period based on the time-of-use market-based unit load and the predicted TOU unpriced output for the predicted date; finding the target cumulative bidding capacity distribution from the cumulative bidding capacity distribution corresponding to each historical trading day based on various bidding influencing factors for the predicted date; finding the corresponding bid as the predicted transaction price in the target cumulative bidding capacity distribution based on the effective bidding capacity for each target time period; and determining the application power for each target time period based on the predicted transaction price and the grid purchase price. This method makes the application process more scientific and efficient.

Description

Technical Field

[0001] This application relates to the field of distributed photovoltaic technology, and in particular to a method, apparatus, computer equipment, and storage medium for determining the application strategy of distributed photovoltaic aggregation entities. Background Technology

[0002] Photovoltaic (PV) power generation systems directly convert solar energy into electrical energy using solar cells. Their main components are solar cells, batteries, controllers, and inverters. They are characterized by high reliability, long service life, no environmental pollution, and the ability to generate electricity independently or connect to the grid, showing broad development prospects. Distributed solar photovoltaic (PV) power generation, also known as decentralized power generation or distributed energy supply, refers to configuring small-scale PV power supply systems at or near user sites to meet the needs of specific users, support the economic operation of existing power distribution networks, or simultaneously meet both requirements. In its operating mode, under solar radiation conditions, the PV system's solar cell array converts solar energy into electrical energy, which is then centrally fed into a DC distribution cabinet via a DC combiner box. The grid-connected inverter converts this into AC power to supply the building's own load, and any excess or insufficient power is regulated by connecting to the public grid. To better leverage the role of distributed PV, an aggregator operation model has emerged, where aggregators lead the trading of distributed PV-related electricity. However, currently, under this model, the amount of electricity to be declared for grid connection often relies on experience to predict the clearing price, leading to low prediction accuracy and unscientific declaration strategies. Summary of the Invention

[0003] The purpose of this application is to at least address one of the aforementioned technical deficiencies, and in particular to provide an efficient, accurate, and scientific scheme for determining the application strategy for distributed photovoltaic aggregation entities.

[0004] Firstly, this application provides a method for determining the application strategy for distributed photovoltaic aggregation entities, including:

[0005] Based on the central dispatch load corresponding to the forecast date, the various daily statistical data corresponding to the market-based unit load, and the non-price output forecast function corresponding to each target time, the predicted time-of-use non-price output corresponding to each target time is obtained;

[0006] Based on the time-of-use market-based unit load and the predicted time-of-use non-pricing output for each target time period on the day to be predicted, the effective bidding capacity for each target time period on the day to be predicted is obtained.

[0007] Based on the various price influencing factors corresponding to the date to be predicted, the most matching target cumulative price capacity distribution is found from the cumulative price capacity distribution corresponding to each historical trading day.

[0008] Based on the effective bidding capacity corresponding to each target time-of-use, the corresponding bid is found in the target cumulative bid capacity distribution as the predicted transaction price;

[0009] The declared power for each target time period is determined based on the predicted transaction price and grid purchase price for each target time period.

[0010] In one embodiment, the process of constructing the non-pricing output prediction function includes:

[0011] Obtain the dataset corresponding to each target time-of-use; the dataset includes a subset of datasets corresponding to all historical trading days, and the subset of datasets includes various daily statistical data on the centrally dispatched load and market-based unit load in the corresponding historical trading days, as well as the time-of-use non-pricing output corresponding to the target time-of-use;

[0012] For any target time-of-use (TOU), multiple daily statistical data of centrally dispatched load and market-based unit load are used as independent variables, and TOU unpriced output is used as the dependent variable. Linear regression is performed using the corresponding dataset to obtain the TOU prediction function corresponding to that target TOU.

[0013] In one embodiment, the daily statistics include the maximum value, minimum value, and average value.

[0014] In one embodiment, the effective bidding capacity corresponding to each target time period on the day to be predicted is obtained based on the time-of-use market-based unit load and the predicted time-of-use non-pricing output for each target time period on the day to be predicted, including:

[0015] The effective bidding capacity is obtained by subtracting the time-of-use market-based unit load from the predicted time-of-use unpriced output.

[0016] In one embodiment, based on multiple price influencing factors corresponding to the date to be predicted, the most matching target cumulative price capacity distribution is found from the cumulative price capacity distribution corresponding to each historical trading day, including:

[0017] Based on the priority order of the current coal price on the forecast date, the average load of market-based generating units, the average load of central dispatch, and the type of working day, the most similar target cumulative bid capacity distribution is found from the cumulative bid capacity distribution corresponding to each historical trading day.

[0018] In one embodiment, the most similar target cumulative bid-price capacity distribution is found from the cumulative bid-price capacity distribution corresponding to each historical trading day, in the order of priority: the current coal price on the forecast date, the average load of market-based generating units, the average load of centrally dispatched units, and the type of working day. This includes:

[0019] Calculate the similarity between the forecast date and each historical trading day in terms of current coal price, average load of market-based generating units, average load of central dispatch, and working day type, and set the weight of similarity according to priority order;

[0020] The total similarity for each historical trading day is obtained by weighted summation based on the similarity and corresponding weights.

[0021] The cumulative quote volume distribution corresponding to the historical trading day with the highest total similarity is selected as the target cumulative quote volume distribution.

[0022] In one embodiment, the declared power for each target time-of-use (TOU) is determined based on the predicted transaction price and grid purchase price for each TOU, including:

[0023] If the sum of the predicted electricity trading price and the environmental value benefits is greater than the grid purchase price, then the declared power will be determined as the sum of the predicted power of distributed photovoltaic power.

[0024] Otherwise, the declared power will be determined as the sum of the predicted power of distributed photovoltaic power minus the predicted electricity consumption of users.

[0025] Secondly, this application provides a device for determining the declaration strategy of distributed photovoltaic aggregation entities, comprising:

[0026] The time-of-use non-price output forecasting module is used to obtain the predicted time-of-use non-price output corresponding to each target time based on the central dispatch load corresponding to the forecast day, the various daily statistical data corresponding to the market-based unit load, and the non-price output forecasting function corresponding to each target time.

[0027] The effective bidding capacity prediction module is used to obtain the effective bidding capacity corresponding to each target time period on the day to be predicted, based on the time-sharing market-based unit load and the predicted time-sharing unpriced output corresponding to each target time period on the day to be predicted.

[0028] The optimal quote determination module is used to find the most matching target cumulative quote volume distribution from the cumulative quote volume distribution corresponding to each historical trading day, based on the various quote influencing factors corresponding to the date to be predicted.

[0029] The electricity price prediction module is used to find the corresponding bid in the target cumulative bid capacity distribution based on the effective bidding capacity corresponding to each target time-of-use, and use it as the predicted electricity price.

[0030] The power declaration determination module is used to determine the power declaration for each target time period based on the predicted transaction price and grid purchase price for each target time period.

[0031] Thirdly, this application provides a computer device including one or more processors and a memory storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, they perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method in any of the above embodiments.

[0032] Fourthly, this application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method in any of the above embodiments.

[0033] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0034] The method for determining the bidding strategy of distributed photovoltaic aggregation entities based on this application analyzes the central dispatch load, market-based unit load, and unpriced output prediction function for the forecast date to obtain the predicted time-of-use unpriced output for each target time period. Then, combining the predicted unpriced output and market-based unit load, the effective bidding capacity for each target time period is calculated. Next, based on multiple bidding influencing factors, the most matching cumulative bidding capacity distribution is found from historical transaction data. Using this distribution and the previously calculated effective bidding capacity, the transaction price for each target time period is predicted. Finally, by comparing the predicted transaction price with the grid purchase price, the bidding power for each target time period is determined, providing an objective reference for distributed aggregation entities to determine their bidding strategies. This scheme improves the decision-making accuracy of electricity market participants, enabling them to better respond to market changes and optimize their trading strategies. By accurately predicting unpriced output and market prices, the supply and demand relationship of the power system can be balanced more effectively, improving system stability and reliability. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart illustrating a method for determining the application strategy of distributed photovoltaic aggregation entities in one embodiment of this application;

[0037] Figure 2 This is a schematic diagram of the quotation capacity distribution in one embodiment of this application;

[0038] Figure 3 This is a schematic diagram of the cumulative bid capacity distribution in one embodiment of this application;

[0039] Figure 4 This is a flowchart illustrating the process of constructing an indeterminate output prediction function in one embodiment of this application;

[0040] Figure 5This is a flowchart illustrating the process of determining the target cumulative bid capacity distribution in one embodiment of this application;

[0041] Figure 6 This is a schematic diagram of the process for determining the declared power in one embodiment of this application;

[0042] Figure 7 This is an internal structural diagram of a computer device provided in one embodiment of this application. Detailed Implementation

[0043] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] This application provides a method for determining the application strategy for distributed photovoltaic aggregation entities. Please refer to [link / reference]. Figure 1 This includes steps S102 to S110.

[0045] S102, based on the central dispatch load corresponding to the day to be predicted, the various daily statistical data corresponding to the market-based unit load, and the non-price output prediction function corresponding to each target time, the predicted time-of-use non-price output corresponding to each target time is obtained.

[0046] It is understandable that the forecast date refers to the date on which the clearing price forecast needs to be made, generally the day before, to forecast the clearing price for the following day. Centralized dispatch load refers to the load uniformly dispatched by the power grid within a certain area, including the generating load of various dispatched power plants and the load of inter-regional power grid interconnections at the same time. Market-based unit load refers to the generating load of generating units qualified for electricity market trading, including coal-fired, gas-fired, nuclear power, and new energy units. Among them, new energy units include distributed photovoltaic. These units can declare trading volume and price in the electricity trading market (spot market) to participate in market bidding. The managers of the electricity trading market will build, operate, and manage the electricity market technical support system (trading center), in which the forecasts for centralized dispatch load and market-based unit load for the following day are uniformly released. Daily statistical data is the statistical result obtained by statistically analyzing the forecast results of centralized dispatch load and market-based unit load within a day. Examples include average, maximum, and minimum values. The maximum and minimum values ​​of centralized dispatch load represent the peak and valley values ​​of electricity load, while the average value represents the average level of electricity load. The maximum, minimum, and average load values ​​of market-based generating units represent the maximum, minimum, and average levels of electricity that market-based generating units need to provide.

[0047] In the energy trading market, the forecast day is divided into multiple consecutive time periods, and transactions are settled separately in each time period (spot settlement of all electricity). In this embodiment, these time periods are referred to as target time periods, for example, each hour of the 24-hour period is defined as a target time period. Time-of-use (TOU) non-priceable output refers to the minimum technical output that some market-based generating units, such as coal-fired and gas-fired units, must maintain after startup; this output does not participate in pricing in the spot market. The non-priceable output prediction function is used to predict non-priceable output. Each target time period has a corresponding non-priceable output prediction function. The non-priceable output prediction function is established based on the correlation between historical non-priceable TOU output and various daily statistical data of central dispatch load and market-based generating unit load. It reflects the trend of TOU pricing output changing with these daily statistical data. Therefore, by inputting these daily statistical data corresponding to the forecast day into the non-priceable output prediction function, the TOU non-priceable output of each target time period on the forecast day can be predicted.

[0048] S104. Based on the time-of-use market-based unit load and the predicted time-of-use non-pricing output corresponding to each target time period on the day to be predicted, the effective bidding capacity corresponding to each target time period on the day to be predicted is obtained.

[0049] It is understood that effective bidding capacity refers to the amount of electricity that needs to be traded through competitive bidding among market-based generating units in the electricity trading market. The total load of the entire power system is divided into a market-based portion and a non-market-based portion. This application mainly focuses on the market-based portion. The total market-based load is satisfied by both non-priced output and the output of market-based generating units. The trading center will publish the forecast results of the time-of-use market-based generating unit load for each target time of the forecast date, which is the total market-based load of the entire power system. By removing the load that does not participate in the bidding from the total market-based load, the effective bidding capacity can be obtained. Therefore, for each target time of use, the effective bidding capacity for that target time of use can be obtained by subtracting the forecasted time-of-use non-priced output from the time-of-use market-based generating unit load.

[0050] S106. Based on the various price influencing factors corresponding to the date to be predicted, find the most matching target cumulative price capacity distribution from the cumulative price capacity distribution corresponding to each historical trading day.

[0051] It's understandable that pricing influencing factors refer to factors that affect the pricing of market participants, potentially related to factors such as power generation costs, electricity consumption patterns, and supply and demand. Cumulative pricing capacity reflects the change in the accumulated bidding capacity as prices increase. That is, by dividing the price range into multiple price segments according to preset intervals and summing the bidding capacity of each price segment, the pricing capacity distribution can be obtained, such as... Figure 2As shown. The bidding capacity is collected for each price segment in the bidding capacity distribution. Then, multiple price points are divided at preset intervals. The bidding capacity corresponding to all price segments below each price point is summed to obtain the cumulative bidding capacity corresponding to that price point, thus yielding the cumulative bidding capacity distribution, as shown. Figure 3 As shown, the trading center publishes daily price volume distribution. Based on the price volume distribution of each historical trading day, the cumulative price volume distribution for each historical trading day can be obtained using the method described above.

[0052] Since the daily cumulative price volume distribution is difficult to predict, this embodiment adopts a method of selecting multiple key price influencing factors and analyzing historical data to find the cumulative price volume distribution of historical trading days that are most similar to the price influencing factors on the day to be predicted. Because the price influencing factors are most similar, the behavior of market participants when participating in pricing on the day to be predicted will also be most similar. Therefore, the most matching cumulative price volume distribution can be used as the target cumulative price volume distribution, which is regarded as the pricing pattern on the day to be predicted.

[0053] S108, based on the effective bidding capacity corresponding to each target time-of-use, find the corresponding bid in the target cumulative bid capacity distribution as the predicted transaction price.

[0054] It can be understood that the predicted electricity trading price is the predicted value of the electricity clearing price for the corresponding target time period. The target cumulative bid capacity distribution predicts the cumulative bid capacity corresponding to each bid point. After obtaining the effective bidding capacity for each target time period through step S104, the bid closest to the effective bidding capacity can be found from the cumulative bid capacity corresponding to each bid point of the target cumulative bid capacity, and the corresponding bid is used as the predicted electricity trading price.

[0055] S110, based on the predicted transaction price and grid purchase price for each target time period, determine the declared power for each target time period.

[0056] It's understandable that distributed photovoltaic (PV) owners can choose to self-consume their generated electricity or apply for grid connection. The distributed PV aggregator needs to make decisions for each owner to ensure profitability. The grid purchase price refers to the price at which owners directly purchase electricity from the grid, equivalent to the unit price of electricity for distributed PV users. The predicted trading price is the projected unit revenue an owner can obtain by applying for grid connection for their generated electricity. Based on the relationship between revenue and cost, the distributed PV aggregator chooses how much of its managed distributed PV systems should apply for grid connection and how much should be self-consumed. It's also important to note that distributed PV accounts for a small proportion of the overall power system; the aggregator's bids do not affect the electricity market, acting only as a clearing price acceptor.

[0057] Based on the distributed photovoltaic aggregation entity declaration strategy determination method in this embodiment, the predicted time-of-use (TOU) non-price output is obtained by analyzing the central dispatch load, market-based unit load, and non-price output prediction function for the predicted date. Then, combining the predicted non-price output and market-based unit load, the trading capacity for each target TOU is calculated. Next, based on multiple pricing influencing factors, the most matching cumulative bid capacity distribution is found from historical trading data. Using this distribution and the previously calculated trading capacity, the trading price for each target TOU is predicted. Finally, by comparing the predicted trading price with the grid purchase price, the declared power for each target TOU is determined, providing an objective reference for distributed aggregation entities to determine their declaration strategies. This scheme improves the decision-making accuracy of electricity market participants, enabling them to better respond to market changes and optimize their trading strategies. By accurately predicting non-price output and market prices, the supply and demand relationship of the power system can be balanced more effectively, improving system stability and reliability.

[0058] In one embodiment, please refer to Figure 4 The process of constructing the power prediction function without pricing includes steps S402 and S404.

[0059] S402, Obtain the dataset corresponding to each target time-of-use. The dataset includes a subset corresponding to all historical trading days. The subset includes various daily statistical data on the centrally dispatched load and market-based unit load for the corresponding historical trading day, as well as the time-of-use non-pricing output corresponding to the target time-of-use.

[0060] It is understandable that each target time-of-day period has a corresponding non-price output prediction function, therefore the construction process also relies on a corresponding dataset. The dataset for each target time-of-day period includes subsets of data corresponding to all historical trading days since the establishment of the trading center. Each subset includes various daily statistical data for the centrally dispatched load and market-based unit load, such as maximum, minimum, and average values. It also includes the non-price output of that target time-of-day period on the corresponding historical trading days. For example, if the historical trading days are from January 1st to February 1st, a total of 31 days, then the dataset corresponding to 12:00 would include 31 subsets. Each subset includes the maximum, minimum, and average values ​​of the centrally dispatched load and market-based unit load for the corresponding date, as well as the non-price output of that date at 12:00.

[0061] On each historical trading day, the trading center clears data at a specific granularity, such as every 15 minutes. During clearing, it releases the corresponding centrally dispatched load, market-based generating unit load, and clearing price. Generally, the granularity of clearing can be unified with the granularity of settlement. That is, the arithmetic average of the centrally dispatched load, market-based generating unit load, and clearing price corresponding to multiple clearing cycles within each settlement cycle is calculated to obtain the centrally dispatched load, market-based generating unit load, and clearing price for each target time-of-use (TOU). Regarding the acquisition of various daily statistical data from the subset, the daily statistical data, such as maximum, minimum, and average values, can be determined based on these target TOU centrally dispatched loads and market-based generating unit loads. Regarding the acquisition of the time-of-use non-pricing output corresponding to the target TOU in the subset, the cumulative bid capacity for that target TOU needs to be subtracted from the market-based generating unit load for that target TOU. The market-based unit load for each target time period has been obtained. The cumulative bid capacity for each target time period can be obtained by using the method described in step S106 to obtain the cumulative bid capacity distribution for that historical trading day. Then, based on the clearing price for each target time period, the corresponding cumulative bid capacity can be found from the cumulative bid capacity distribution as the cumulative bid capacity for that target time period.

[0062] S404. For any target time-of-use, multiple daily statistical data of centrally dispatched load and market-based unit load are used as independent variables, and time-of-use unpriced output is used as the dependent variable. Linear regression is performed using the corresponding dataset to obtain the unpriced output prediction function corresponding to the target time-of-use.

[0063] It's understandable that linear regression is used to establish a time-of-use (TOU) power output prediction function. Linear regression is a fundamental yet powerful statistical method that assumes a linear relationship between the dependent variable (in this case, TOU power output) and the independent variables (various daily statistical data on centrally dispatched load and market-based unit load). Using a large amount of historical data corresponding to each target TOU, a corresponding TOU power output prediction function can be fitted. When it's necessary to predict the TOU power output for each target TOU on the forecast date, the corresponding TOU power output prediction function is selected, and the various daily statistical data on centrally dispatched load and market-based unit load for the forecast date are input to obtain the predicted TOU power output for each target TOU. The linear regression process involves adjusting the parameters of the constant term and each independent variable. The value of R² can be used to determine whether further parameter tuning is needed; for example, if the R² value is greater than 0.7, the function is selected; otherwise, parameter tuning is performed.

[0064] In one embodiment, based on the various price influencing factors corresponding to the date to be predicted, the most matching target cumulative price capacity distribution is found from the cumulative price capacity distribution corresponding to each historical trading day. This includes finding the most similar target cumulative price capacity distribution from the cumulative price capacity distribution corresponding to each historical trading day in order of priority: the current coal price of the date to be predicted, the average load of market-based generating units, the average load of central dispatch, and the type of working day.

[0065] It is understandable that the current price of coal is a significant cost factor, directly impacting the power generation cost of thermal power units and their pricing strategies in the market. The pricing of thermal power units has a substantial influence on the clearing price, thus driving changes in the clearing price. Guiding prices for coal are published on regional coal trading websites, and these guiding prices can be obtained through data scraping and other methods. Coal can also be classified according to calorific value, source, quality, etc. Guiding prices for the corresponding types can be selected based on the mainstream usage of thermal power units, such as the guiding price for 5500 kcal coal. The average load of market-based generating units reflects the overall output level of generating units participating in market transactions, which is directly related to market supply. The average load of centrally dispatched units reflects the demand level of the entire power system. Both of these average loads can be obtained by referring to the explanation in step S402. The type of working day is closely related to electricity demand patterns, with typical classifications including: weekdays, Saturdays, Sundays, statutory holidays, and adjusted holidays. The selection of these factors reflects a comprehensive consideration of both the supply and demand sides of the electricity market. When matching, the influence of each price influencing factor is different. In this embodiment, the factors will be sorted in order of priority: the current coal price on the day to be predicted, the average load of market-based units, the average load of central dispatch, and the type of working day. That is, the higher the priority, the greater its influence in the matching process.

[0066] In one embodiment, the most similar target cumulative bid-price capacity distribution is found from the cumulative bid-price capacity distribution corresponding to each historical trading day, in the order of priority: the current coal price on the forecast date, the average load of market-based generating units, the average load of centrally dispatched units, and the type of working day. (See [link to relevant documentation]). Figure 5 This includes steps S502 to S506.

[0067] S502 calculates the similarity between the forecast date and each historical trading day in terms of current coal price, average load of market-based generating units, average load of central dispatch, and working day type, and sets the weight of similarity according to priority order.

[0068] Understandably, this step requires calculating the similarity between the day to be predicted and each historical trading day across these factors. Different methods may be used for similarity calculation. For example, for continuous variables (such as current coal price, average load of market-based generating units, and average load of centrally dispatched units), the difference can be directly calculated and then normalized; for discrete variables (such as workday type), a simple matching degree (1 for identical, 0 for different) can be used. The weight values ​​corresponding to each price-influencing factor will be different, and the higher the priority, the higher the corresponding weight. In this embodiment, the current coal price has the highest weight, and the workday type has the lowest weight.

[0069] S504: The total similarity for each historical trading day is obtained by weighted summation based on the similarity and corresponding weights.

[0070] S506: Select the cumulative quote volume distribution corresponding to the historical trading day with the highest total similarity as the target cumulative quote volume distribution.

[0071] In one embodiment, the declared power for each target time-of-use (TOU) is determined based on the predicted trading price and grid purchase price for each TOU. Please refer to [link to relevant documentation]. Figure 6 This includes steps S602 and S604.

[0072] S602 If the sum of the predicted trading price and the environmental value benefits is greater than the grid purchase price, then the declared power will be determined as the sum of the predicted power of distributed photovoltaic power.

[0073] This refers to the monetization of the environmental benefits of distributed photovoltaic (PV) power generation. It may include carbon reduction revenue, renewable energy subsidies, etc. The approach adopted in this embodiment is to maximize revenue. When the predicted electricity trading price plus the environmental value revenue exceeds the grid purchase price, it means that submitting the electricity to the grid is more profitable than self-consumption. In this case, the aggregator should choose to submit all predicted power generation to the grid and participate in market transactions.

[0074] S604, otherwise, the declared power will be determined as the sum of the predicted power of distributed photovoltaic power minus the predicted electricity consumption of users.

[0075] It's understandable that when the predicted electricity price plus environmental value is less than the grid purchase price, it means the return on grid connection is negative, and the return on self-consumption is higher than the return on grid connection. In this case, the aggregator should prioritize meeting the electricity needs of all owners and declare the remaining generation to the grid for market trading.

[0076] This application provides a distributed photovoltaic aggregation entity application strategy determination device, including a time-of-use non-pricing output prediction module, an effective bidding capacity prediction module, an optimal bid determination module, a transaction electricity price prediction module, and an application power determination module.

[0077] The time-of-use (TOU) unpricing output prediction module is used to obtain the predicted TOU unpricing output for each target time based on various daily statistical data corresponding to the central dispatch load and market-based unit load for the day to be predicted, as well as the unpricing output prediction function for each target time.

[0078] The effective bidding capacity prediction module is used to obtain the effective bidding capacity corresponding to each target time period on the day to be predicted, based on the time-sharing market-based unit load and the predicted time-sharing unpriced output corresponding to each target time period on the day to be predicted.

[0079] The optimal quote determination module is used to find the most matching target cumulative quote volume distribution from the cumulative quote volume distribution corresponding to each historical trading day, based on various quote influencing factors corresponding to the date to be predicted.

[0080] The electricity price prediction module is used to find the corresponding bid in the target cumulative bid capacity distribution based on the effective bidding capacity corresponding to each target time-of-use, and use it as the predicted electricity price.

[0081] The power declaration determination module is used to determine the power declaration for each target time period based on the predicted transaction price and grid purchase price for each target time period.

[0082] Specific limitations regarding the device for determining the application strategy of distributed photovoltaic (PV) aggregation entities can be found in the above-described limitations of the method for determining the application strategy of distributed PV aggregation entities, and will not be repeated here. The above modules can be embedded in hardware or independent of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the operations corresponding to each module. It should be noted that the module division in this embodiment is illustrative and only represents a logical functional division; other division methods may be used in actual implementation.

[0083] This application provides a computer device including one or more processors and a memory storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, they perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method in any of the above embodiments.

[0084] Indicatively, such as Figure 7 As shown, Figure 7 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. (Refer to...) Figure 7The computer device 700 includes a processing component 702, which further includes one or more processors, and memory resources represented by memory 701 for storing instructions, such as application programs, that can be executed by the processing component 702. The application programs stored in memory 701 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 702 is configured to execute instructions to perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method in any of the above embodiments.

[0085] The computer device 700 may also include a power supply component 703 configured to perform power management of the computer device 700, a wired or wireless network interface 704 configured to connect the computer device 700 to a network, and an input / output (I / O) interface 705.

[0086] Those skilled in the art will understand that Figure 7 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0087] This application provides a storage medium storing computer-readable instructions. When executed by one or more processors, the computer-readable instructions cause the one or more processors to perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method in any of the above embodiments.

[0088] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0089] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for determining the application strategy for distributed photovoltaic aggregation entities, characterized in that, include: Based on the central dispatch load corresponding to the forecast day, the various daily statistical data corresponding to the market-based unit load, and the non-price output forecast function corresponding to each target time, the predicted time-of-use non-price output corresponding to each target time is obtained; Based on the time-of-use market-based unit load corresponding to each of the target time periods on the day to be predicted and the predicted time-of-use non-price output, the effective bidding capacity corresponding to each of the target time periods on the day to be predicted is obtained. Based on the various price influencing factors corresponding to the date to be predicted, the most matching target cumulative price capacity distribution is found from the cumulative price capacity distribution corresponding to each historical trading day. Based on the effective bidding capacity corresponding to each target time-of-use, the corresponding bid is found in the target cumulative bid capacity distribution as the predicted transaction price; The declared power for each target time period is determined based on the predicted transaction electricity price and grid purchase price for each target time period.

2. The method for determining the application strategy for distributed photovoltaic aggregation entities according to claim 1, characterized in that, The construction process of the non-price output prediction function includes: Obtain the dataset corresponding to each target time-of-use; the dataset includes a subset of datasets corresponding to all the historical trading days, and the subset of datasets includes various daily statistical data of the centrally dispatched load and the market-based unit load in the corresponding historical trading day, as well as the time-of-use non-pricing output corresponding to the target time-of-use; For any of the target time-of-use loads, the daily statistical data of the centrally dispatched load and the market-based unit load are used as independent variables, and the time-of-use unpriced output is used as the dependent variable. Linear regression is performed using the corresponding dataset to obtain the unpriced output prediction function corresponding to the target time-of-use load.

3. The method for determining the application strategy for distributed photovoltaic aggregation entities according to claim 2, characterized in that, The daily statistics include the maximum, minimum, and average values.

4. The method for determining the application strategy for distributed photovoltaic aggregation entities according to claim 1, characterized in that, The step of obtaining the effective bidding capacity corresponding to each of the target time periods on the day to be predicted, based on the time-of-use market-based unit load corresponding to each of the target time periods on the day to be predicted and the predicted time-of-use non-pricing output, includes: The effective bidding capacity is obtained by subtracting the time-of-use market-based unit load from the predicted time-of-use non-price output.

5. The method for determining the application strategy for distributed photovoltaic aggregation entities according to claim 1, characterized in that, The step of finding the most matching target cumulative price capacity distribution from the cumulative price capacity distribution corresponding to each historical trading day, based on the various price influencing factors corresponding to the date to be predicted, includes: Based on the priority order of the current coal price, average load of market-based generating units, average load of central dispatch, and working day type for the predicted date, the most similar target cumulative bid capacity distribution is found from the cumulative bid capacity distribution corresponding to each of the historical trading days.

6. The method for determining the application strategy for distributed photovoltaic aggregation entities according to claim 5, characterized in that, The process involves sequentially prioritizing the current coal price on the forecast date, the average load of market-based generating units, the average load of centrally dispatched units, and the type of working day, to find the most similar target cumulative bid-price capacity distribution from the cumulative bid-price capacity distribution corresponding to each of the historical trading days. This includes: Calculate the similarity between the predicted date and each of the historical trading days in terms of the current coal price, the average load of the market-based generating units, the average load of the central dispatch, and the type of working day, and set the weight of the similarity according to the priority order. The total similarity for each of the historical trading days is obtained by weighted summation based on the similarity and the corresponding weight. The cumulative quote volume distribution corresponding to the historical trading day with the highest total similarity is selected as the target cumulative quote volume distribution.

7. The method for determining the application strategy for distributed photovoltaic aggregation entities according to claim 1, characterized in that, The step of determining the declared power for each target time slot based on the predicted transaction price and grid purchase price for each target time slot includes: If the sum of the predicted electricity price and the environmental value benefits is greater than the grid purchase price, then the declared power will be determined as the sum of the predicted power of distributed photovoltaic power. Otherwise, the declared power will be determined as the sum of the predicted distributed photovoltaic power minus the predicted user electricity consumption.

8. A device for determining the application strategy of a distributed photovoltaic aggregation entity, characterized in that, include: The time-of-use non-price output prediction module is used to obtain the predicted time-of-use non-price output corresponding to each target time based on the central dispatch load corresponding to the forecast day, the various daily statistical data corresponding to the market-based unit load, and the non-price output prediction function corresponding to each target time. The effective bidding capacity prediction module is used to obtain the effective bidding capacity corresponding to each of the target time periods on the day to be predicted based on the time-sharing market-based unit load corresponding to each of the target time periods on the day to be predicted and the predicted time-sharing unpriced output. The optimal quote determination module is used to find the most matching target cumulative quote volume distribution from the cumulative quote volume distribution corresponding to each historical trading day, based on the various quote influencing factors corresponding to the date to be predicted. The electricity price prediction module is used to find the corresponding bid in the target cumulative bid capacity distribution based on the effective bidding capacity corresponding to each target time-of-use segment and use it as the predicted electricity price. The power declaration determination module is used to determine the power declaration for each target time period based on the predicted transaction electricity price and the grid purchase price for each target time period.

9. A computer device, characterized in that, It includes one or more processors and a memory storing computer-readable instructions, which, when executed by the one or more processors, perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method as described in any one of claims 1-7.

10. A storage medium, characterized in that, The storage medium stores computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the distributed photovoltaic aggregation subject declaration strategy determination method as described in any one of claims 1-7.

Images