Method, device and equipment for generating day-ahead declaration optimization strategy of electric power spot market

Abstract

The invention relates to the technical field of electric power spot transaction, in particular to an electric power spot market day-ahead declaration optimization strategy generation method, device and equipment, and the method comprises the steps: carrying out the preprocessing of the obtained operation boundary data and historical transaction data of an electric power spot market of a target region, and obtaining the processed data; based on the processed data and historical electricity price prediction data, optimizing a pre-trained initial electricity price prediction model to obtain an electricity price prediction model; generating at least one power day-ahead declaration optimization strategy according to the obtained power demand information of the user and day-ahead electricity price prediction data obtained based on a power price prediction model; and selecting a target power day-ahead declaration optimization strategy from the at least one power day-ahead declaration optimization strategy based on an analysis result obtained by analyzing each power day-ahead declaration optimization strategy in the at least one power day-ahead declaration optimization strategy. The optimal declaration scheme is provided for the user on the premise that market rules and constraint conditions are met.

Description

Technical Field

[0001] The present application relates to the technical field of electricity spot trading, and in particular to a method, device and equipment for generating a day-ahead declaration optimization strategy for the electricity spot market. Background Art

[0002] The commodity traded in the electricity spot market is electricity. The difference between the supply and demand relationship of electricity and other commodities is that the supply and demand relationship of the electricity market is not expressed by the relationship between electricity production (i.e., power generation) and sales (i.e., electricity sales). The reason is that electric energy cannot be stored, and electricity production has the characteristics of "generated and used immediately, and balanced production and sales". The supply and demand relationship of the electricity market should be expressed by the relationship between the power load demand of the entire network and the total capacity of the generator sets that can be operated in the power grid. The load of the entire network reflects the size of the power demand, and the total capacity of the operating generator sets represents the power production capacity. In fact, in one day, the marginal electricity price of the valley section with low load demand of the entire network is indeed the lowest in one day, and the marginal electricity price of the peak section with high load demand is the highest in one day. This change in the marginal electricity price in one day is basically consistent with the supply and demand relationship. Because the total capacity of the units that bear base load and mid-load power generation in one day is unchanged, only the load demand at different times changes.

[0003] The marginal price refers to the quotation of the last power supplier who trades electricity one by one in the order of low to high quotations in spot power transactions, so that the traded electricity meets the load demand. It is called the "marginal price" of the system. The power of the generator set with a quotation higher than the marginal price cannot be traded and the bidding fails; the generator set with a quotation lower than the marginal price does not settle the spot electricity with the power market according to the quotation, but settles according to the marginal price of the system. After all units have finished quoting, under the premise of meeting the safety requirements of the power grid, based on the principle of minimum power generation cost / maximum social welfare, the quotation will be traded in order from high to low according to the price difference between the power generation and the use side, until the cumulative power generation output is exactly equal to the load demand. The quotation of the last traded unit that meets the balance of power generation and use is the marginal price, and all winning units are settled according to the marginal price. However, power plants usually quote according to their own estimated costs, expected benefits and experience, and are not sensitive to the supply and demand trends in the power spot market. Usually, users cannot accurately estimate their production scale and day-ahead electricity prices, and in order to ensure production and respond to emergencies, they propose date declaration plans, which may cause waste of resources and increase in costs. Summary of the invention

[0004] The main purpose of this application is to provide a method, device and equipment for generating an optimization strategy for day-ahead declaration in the electricity spot market; used to optimize the day-ahead declaration strategy of electricity market participants.

[0005] To achieve the above-mentioned object, the present invention provides a method for generating an optimization strategy for day-ahead declaration in an electricity spot market, comprising: preprocessing the acquired operation boundary data and historical transaction data of the electricity spot market in a target area to obtain processed data; Based on the processed data and the historical electricity price forecast data, optimizing the pre-trained initial electricity price forecast model to obtain an electricity price forecast model; Generate at least one day-ahead electricity declaration optimization strategy based on the acquired user power demand information and the day-ahead electricity price forecast data obtained based on the power price forecast model; Based on the analysis result obtained by analyzing each of the at least one optimization strategy for electricity day-ahead declaration, a target electricity day-ahead declaration optimization strategy is selected from the at least one optimization strategy for electricity day-ahead declaration.

[0006] Optionally, before generating at least one power day-ahead declaration optimization strategy based on the acquired power demand information of the user and the day-ahead power price forecast data obtained based on the power price forecast model, the method further includes: Acquiring electricity market data of the target area; Preprocessing the electricity market data to obtain processed electricity market data; Based on the processed power market data, a multi-layer neural network model is constructed to obtain a power price prediction model; The processed data is input into the electricity price prediction model to obtain the day-ahead electricity price prediction data.

[0007] Optionally, the preprocessing the electricity market data to obtain processed electricity market data includes: Performing correlation analysis on the data affecting the real-time electricity price in the electricity market data, eliminating irrelevant data, and obtaining a processed data set; The processed data set is preprocessed to convert the processed data set into data in a format suitable for input into the electricity price prediction model, and the data in the format is used as the processed electricity market data.

[0008] Optionally, the electricity market data includes: power generation side data, grid side data market transaction data, power sales side data, market transaction data and policy and regulation data, wherein the power generation side data includes power generation data, power generation cost data and unit operation data; the grid side data includes transmission and distribution data, grid operation data and grid construction data; the power sales side data includes power sales data, power sales data and user data; the market transaction data includes spot market transaction data, medium- and long-term transaction data and ancillary service market transaction data; the policy and regulation data includes power market policy documents and regulatory data.

[0009] Optionally, the initial electricity price prediction model is obtained through the following training steps: Acquire a training sample set, wherein each training sample in the training sample set includes: sample historical power data and corresponding day-ahead power price forecast data; The sample historical power data of the training sample set in the training sample set is taken as input, and the day-ahead power price prediction data corresponding to the input sample historical power data is taken as expected output, so as to train and obtain the initial power price prediction model.

[0010] Optionally, the sample historical power data is obtained by the following steps: Collecting raw power data, wherein the raw power data includes historical power transaction price data, power production and consumption data, long-term weather forecast data, historical load data and new energy wind power output data, the weather forecast data includes temperature data, wind speed data and light intensity data, the power production and consumption data includes thermal power generation data, wind power generation data, photovoltaic power generation data, power transportation loss data and power exchange data of various regions; The original power data is averaged to obtain the processed original power data, and the processed original power data is used as the sample historical power data, wherein the weather forecast data at least includes average sunshine, average temperature and average wind speed.

[0011] Optionally, the day-ahead electricity price forecast data is obtained through the following steps: Converting the preprocessed raw power data into an adjacency matrix topology graph, and performing a graph convolution operation on the adjacency matrix to obtain embedding vectors of different dimensions; Splicing the embedding vectors of different dimensions with the preprocessed original power data to obtain spliced ​​data, and obtaining a weighted vector according to the spliced ​​data; The power transaction price data to be predicted is obtained according to the weighted vector, and the power transaction price data to be predicted is used as the day-ahead electricity price prediction data.

[0012] In addition, to achieve the above-mentioned purpose, the present invention also provides a device for generating an optimization strategy for day-ahead declaration in a power spot market, comprising: A preprocessing module is used to preprocess the acquired operation boundary data and historical transaction data of the power spot market in the target area to obtain processed data; A model optimization module, used to optimize the pre-trained initial electricity price prediction model based on the processed data and the historical electricity price prediction data to obtain an electricity price prediction model; A strategy generation module, used to generate at least one power day-ahead declaration optimization strategy based on the acquired power demand information of the user and the day-ahead power price forecast data obtained based on the power price forecast model; The strategy analysis module is used to select a target electricity day-ahead declaration optimization strategy from the at least one electricity day-ahead declaration optimization strategy based on an analysis result obtained by analyzing each electricity day-ahead declaration optimization strategy in the at least one electricity day-ahead declaration optimization strategy.

[0013] Optionally, before the strategy generation module, the device further includes: A data acquisition submodule, used to acquire the power market data of the target area; A data preprocessing submodule, used for preprocessing the electricity market data to obtain processed electricity market data; A model optimization submodule, used to construct a multi-layer neural network model based on the processed power market data to obtain a power price prediction model; The day-ahead electricity price forecast data generating submodule is used to input the processed data into the electricity price forecast model to obtain the day-ahead electricity price forecast data.

[0014] Optionally, the data preprocessing submodule includes: A data analysis unit, used to perform correlation analysis on the data affecting the real-time electricity price in the electricity market data, remove irrelevant data, and obtain a processed data set; The data format conversion unit is used to pre-process the processed data set, convert the processed data set into data in a format suitable for input of the electricity price prediction model, and use the data in the format as the processed electricity market data.

[0015] Optionally, the electricity market data includes: power generation side data, grid side data market transaction data, power sales side data, market transaction data and policy and regulation data, wherein the power generation side data includes power generation data, power generation cost data and unit operation data; the grid side data includes transmission and distribution data, grid operation data and grid construction data; the power sales side data includes power sales data, power sales data and user data; the market transaction data includes spot market transaction data, medium- and long-term transaction data and ancillary service market transaction data; the policy and regulation data includes power market policy documents and regulatory data.

[0016] Optionally, the initial electricity price prediction model is obtained through the following training steps: Acquire a training sample set, wherein each training sample in the training sample set includes: sample historical power data and corresponding day-ahead power price forecast data; The sample historical power data of the training sample set in the training sample set is taken as input, and the day-ahead power price prediction data corresponding to the input sample historical power data is taken as expected output, so as to train and obtain the initial power price prediction model.

[0017] Optionally, the sample historical power data is obtained by the following steps: Collecting raw power data, wherein the raw power data includes historical power transaction price data, power production and consumption data, long-term weather forecast data, historical load data and new energy wind power output data, the weather forecast data includes temperature data, wind speed data and light intensity data, the power production and consumption data includes thermal power generation data, wind power generation data, photovoltaic power generation data, power transportation loss data and power exchange data of various regions; The original power data is averaged to obtain the processed original power data, and the processed original power data is used as the sample historical power data, wherein the weather forecast data at least includes average sunshine, average temperature and average wind speed.

[0018] Optionally, the day-ahead electricity price forecast data is obtained through the following steps: Converting the preprocessed raw power data into an adjacency matrix topology graph, and performing a graph convolution operation on the adjacency matrix to obtain embedding vectors of different dimensions; Splicing the embedding vectors of different dimensions with the preprocessed original power data to obtain spliced ​​data, and obtaining a weighted vector according to the spliced ​​data; The power transaction price data to be predicted is obtained according to the weighted vector, and the power transaction price data to be predicted is used as the day-ahead electricity price prediction data.

[0019] In addition, to achieve the above-mentioned purpose, the present invention also provides an optimization strategy generation device for day-ahead declarations in the electricity spot market, characterized in that the optimization strategy generation device for day-ahead declarations in the electricity spot market comprises: a memory, a processor, and an optimization strategy generation program for day-ahead declarations in the electricity spot market stored in the memory and executable on the processor, and when the optimization strategy generation program for day-ahead declarations in the electricity spot market is executed by the processor, the steps of the optimization strategy generation method for day-ahead declarations in the electricity spot market as described in any one of the above-mentioned methods are implemented.

[0020] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, characterized in that a program for generating an optimization strategy for day-ahead declarations in the electricity spot market is stored on the computer-readable storage medium, and when the program for generating an optimization strategy for day-ahead declarations in the electricity spot market is executed by a processor, the steps of the method for generating an optimization strategy for day-ahead declarations in the electricity spot market as described in any one of the above-mentioned methods are implemented.

[0021] Compared with the prior art, the present invention has the following beneficial effects: The acquired operation boundary data and historical transaction data of the power spot market in the target area are preprocessed to obtain processed data; based on the processed data and the historical electricity price forecast data, the pre-trained initial power price forecast model is optimized to obtain the power price forecast model; according to the acquired user power demand information and the day-ahead power price forecast data obtained based on the power price forecast model, at least one power day-ahead declaration optimization strategy is generated; based on the analysis results obtained by analyzing each power day-ahead declaration optimization strategy in the at least one power day-ahead declaration optimization strategy, a target power day-ahead declaration optimization strategy is selected from the at least one power day-ahead declaration optimization strategy. The optimal declaration plan is provided to users on the premise of meeting market rules and constraints. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of an embodiment of a method for generating an optimization strategy for day-ahead declaration in a power spot market according to an embodiment of the present invention; Figure 2 It is a schematic diagram of an embodiment of a device for generating an optimization strategy for day-ahead declaration in a power spot market according to an embodiment of the present invention; Figure 3 It is a schematic diagram of an embodiment of a device for generating a day-ahead declaration optimization strategy for the electricity spot market in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The embodiments of the present invention provide a method, device and equipment for generating an optimization strategy for day-ahead declaration in a power spot market, which are used to optimize the day-ahead declaration strategies of power market participants.

[0024] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1An embodiment of the method for generating an optimization strategy for day-ahead declaration in the power spot market in the embodiment of the present invention includes: Step 101 , pre-processing the acquired operation boundary data and historical transaction data of the power spot market in the target area to obtain processed data.

[0026] It is understandable that the execution subject of the present invention may be the method for generating the optimization strategy for the day-ahead declaration of the power spot market, or may be a terminal or a server, which is not specifically limited here. The embodiment of the present invention is described by taking the server as the execution subject as an example.

[0027] The above-mentioned execution entity can pre-process the acquired operation boundary data and historical transaction data of the electricity spot market in the target area. The above-mentioned pre-processing can be to interpolate the missing values ​​of the operation boundary data and historical transaction data of the electricity spot market to form multiple data streams; superimpose and aggregate multiple data streams to form a unified processed data corresponding to a certain moment including real-time electricity price, power load, weather, temperature, economy, and society.

[0028] The above-mentioned electricity spot market is a market that conducts transactions based on the balance of electricity supply and demand in a short period of time; the operating boundary data of the above-mentioned electricity spot market may include those related to the power generation side, the power grid side and the load side; the above-mentioned power generation side includes the unit capacity limit: the maximum and minimum power generation capacity of different types of power generation units (such as thermal power generation units, hydropower generation units, wind power generation units, solar power generation units, etc.). Unit ramp rate: indicates the ability of the unit to increase or decrease the power generation power per unit time. Power generation cost range: includes fixed costs (such as equipment depreciation, employee wages, etc.) and variable costs (such as fuel costs, maintenance costs, etc.). The above-mentioned power grid side includes transmission and distribution capacity limits: transmission lines and substations of different voltage levels have their rated transmission capacities. Power grid security constraints: including upper and lower limits of voltage, upper and lower limits of frequency, etc.

[0029] The above-mentioned load-side correlation includes user load characteristics: the load curves of different types of users (such as industrial users, commercial users, and residential users) have different characteristics. The load of industrial users may remain at a high level and relatively stable during the production period, while the load of residential users will fluctuate significantly during the peak hours in the morning and evening. Taking a city as an example, the daily load peak of its residential users may occur between 7 and 9 pm, with a load of 500MW, while the trough period may be between 3 and 5 am, with a load drop to 200MW. Load growth forecast: Forecast future load growth based on factors such as regional economic development, population growth, and industrial structure adjustment. For example, a new industrial park is expected to increase its industrial output value by 10% annually in the next five years, and the corresponding power load is expected to increase by 8% to 10% annually. These forecast data provide forward-looking information for the long-term planning of the power market and the formulation of day-ahead declaration strategies.

[0030] The electricity spot market achieves optimal allocation of power resources and improves the operating efficiency and reliability of the power system through a market-based trading mechanism.

[0031] The above historical transaction data includes historical electricity under different transaction types and electricity classified by transaction subjects. For example, in medium- and long-term transactions, the specific values ​​of bilaterally negotiated transaction electricity in each year, quarter, and month, as well as statistical data on centralized bidding transaction electricity. In the spot market, the day-ahead market and the real-time market have hourly transaction electricity data in different time periods in the past (such as the past few months, quarters, and years). The total amount of electricity successfully sold by a single power generation company in market transactions over the past period of time can be used to analyze the changing trend of its market share. The actual transaction electricity of different types of electricity users (such as large industrial users, general industrial and commercial users, residential users, etc.) in each trading cycle is used to study changes in user electricity consumption behavior.

[0032] Step 102: Based on the processed data and the historical electricity price forecast data, the pre-trained initial electricity price forecast model is optimized to obtain an electricity price forecast model.

[0033] The above-mentioned execution entity can optimize the pre-trained initial electricity price prediction model based on the processed data and historical electricity price prediction data to obtain an electricity price prediction model.

[0034] As an example, the above-mentioned executive entity combines machine learning technology with power market data analysis to achieve trend prediction of electricity spot market electricity prices by building a multi-layer neural network model. By analyzing the operating boundary data and historical transaction data of the electricity spot market and combining it with electricity price forecast data, a corresponding mathematical programming model is established to optimize the day-ahead declaration strategy of electricity market participants.

[0035] Step 103: Generate at least one day-ahead electricity declaration optimization strategy based on the acquired user's electricity demand information and the day-ahead electricity price forecast data obtained based on the electricity price forecast model.

[0036] The above-mentioned execution entity can generate at least one day-ahead electricity declaration optimization strategy based on the acquired user's electricity demand information and the day-ahead electricity price forecast data obtained based on the electricity price forecast model. The power demand information of the above-mentioned users may include self-information, market information and external environment information; the above-mentioned self-information includes: power generation enterprises, power sales companies and power users; the above-mentioned power generation enterprises include cost information and unit information, unit information: including unit type, installed capacity, unit operating characteristics (such as minimum output, maximum output, ramp rate, start-up time, downtime, etc.); cost information: fixed costs and variable costs such as fuel cost, equipment maintenance cost, and labor cost; the above-mentioned market information includes historical transaction data: data on transaction volume, transaction price, market clearing price, etc. in the day-ahead market and real-time market in the past period of time, in order to analyze the changing trend of market supply and demand; demand forecast: forecast data on future power demand through market research, economic data, etc., including demand changes in different regions and industries; supply capacity of other power generation enterprises: information affecting market supply such as unit maintenance plans and new unit commissioning plans of other power generation enterprises; price trend: historical fluctuations in spot market prices, such as cyclical changes in prices, the frequency of high and low prices, etc.

[0037] The above-mentioned external environmental information includes electricity market policies: such as adjustments to market trading rules, renewable energy policies (such as green certificate trading, renewable energy subsidies, etc.), energy transformation-related policies, etc.; environmental protection policies: such as carbon emission restrictions, pollutant emission standards, etc., which affect the operating costs and market competitiveness of power generation companies.

[0038] Optionally, before generating at least one electricity day-ahead declaration optimization strategy based on the acquired user electricity demand information and the day-ahead electricity price forecast data obtained based on the electricity price forecast model, the method also includes: obtaining electricity market data of the target area; preprocessing the electricity market data to obtain processed electricity market data; constructing a multi-layer neural network model based on the processed electricity market data to obtain an electricity price forecast model; and inputting the processed data into the electricity price forecast model to obtain the day-ahead electricity price forecast data.

[0039] The execution entity may obtain the electricity market data of the target area through a wired connection or a wireless connection; the target area may be a pre-set area such as a city or a development zone; It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future.

[0040] Optionally, the preprocessing of the electricity market data to obtain processed electricity market data includes: performing a correlation analysis on the data in the electricity market data that affects the real-time electricity price, eliminating irrelevant data, and obtaining a processed data set; preprocessing the processed data set to convert the processed data set into data in a format suitable for input to the electricity price prediction model, and using the data in the format as the processed electricity market data.

[0041] Optionally, the electricity market data includes: power generation side data, grid side data market transaction data, power sales side data, market transaction data and policy and regulation data, wherein the power generation side data includes power generation data, power generation cost data and unit operation data; the grid side data includes transmission and distribution data, grid operation data and grid construction data; the power sales side data includes power sales data, power sales data and user data; the market transaction data includes spot market transaction data, medium- and long-term transaction data and ancillary service market transaction data; the policy and regulation data includes power market policy documents and regulatory data.

[0042] As an example, power generation data: the total power generation of each power generation company, including the power generation of different types of power generation units (such as thermal power, hydropower, wind power, photovoltaic, etc.); power generation divided by time period, such as daily power generation, monthly power generation, annual power generation, etc. Power generation cost data: fuel costs, including the purchase price and consumption of fuels such as coal and natural gas for thermal power companies; equipment maintenance costs, including regular maintenance and fault repair costs; labor costs, including wages, benefits and other expenditures of power generation company employees; unit operation data: basic information such as unit capacity, type, technical parameters, etc.; unit start and stop time, operating hours, load rate and other operating status data; unit reliability indicators, such as availability, forced outage rate, etc.; transmission and distribution data: transmission and distribution of different voltage levels; power transmission and flow between regions; power grid operation data: grid frequency, voltage and other operating parameters; line load rate, transformer load rate and other equipment operating status data; power grid fault information, including fault type, occurrence time, impact range, etc.; power grid construction data: scale, investment, construction progress and other information of new transmission and transformation projects; status of power grid upgrade and transformation projects; Electricity sales data: total electricity sales of each electricity sales company; electricity sales of different types of users (industrial users, commercial users, residential users, etc.). Electricity sales price data: electricity sales prices of electricity sales companies to users, including prices for different time periods and different types of users; electricity purchase prices with power generation companies; user data: basic information such as the number, type, and electricity consumption characteristics of users; electricity consumption data such as user electricity consumption and load curves. Spot market transaction data: transaction volume and transaction prices of the day-ahead market and real-time market; market clearing results, including winning power generation companies, winning volume and prices. Medium- and long-term transaction data: volume, price, and information on both parties of medium- and long-term transactions such as bilateral negotiation transactions and centralized bidding transactions; contract execution, including executed volume and unexecuted volume. Ancillary service market transaction data: transaction volume and price of ancillary services such as frequency regulation and standby; information on market entities providing ancillary services. Electricity market policy documents: policy documents issued by the national and local governments on electricity market reform, electricity price policy, and renewable energy development. Regulatory data: regulatory reports and penalty decisions of the power regulatory authorities on the electricity market; compliance data of market entities.

[0043] Optionally, the initial electricity price prediction model is obtained through the following training steps: obtaining a training sample set, wherein each training sample in the training sample set includes: sample historical electricity data and corresponding day-ahead electricity price prediction data; taking the sample historical electricity data of the training sample set in the training sample set as input, and taking the day-ahead electricity price prediction data corresponding to the input sample historical electricity data as expected output, and training to obtain the initial electricity price prediction model.

[0044] The execution subject may obtain the training sample set through a wired connection or a wireless connection, wherein each training sample in the training sample set includes: sample historical power data and corresponding day-ahead power price forecast data; Input the sample historical electricity data of at least one training sample in the training sample set into the initial convolutional neural network model respectively, and obtain the day-ahead electricity price prediction data corresponding to each sample historical electricity data in the above at least one training sample; compare the day-ahead electricity price prediction data corresponding to each sample historical electricity data in the above at least one training sample with the corresponding sample day-ahead electricity price prediction data; determine the prediction accuracy of the above initial convolutional neural network model according to the comparison result; determine whether the above prediction accuracy is greater than the preset accuracy threshold; in response to determining that the above accuracy is greater than the above preset accuracy threshold, use the above initial convolutional neural network model as the trained initial electricity price prediction model; in response to determining that the above accuracy is not greater than the above preset accuracy threshold, adjust the parameters of the above initial convolutional neural network model, use unused training samples to form a training sample set, use the adjusted initial convolutional neural network model as the initial convolutional neural network model, and perform the above training steps again.

[0045] Convolutional Neural Networks (CNN) are a type of feedforward neural network that includes convolutional calculations and has a deep structure. It is one of the representative algorithms of deep learning. Convolutional neural networks have representation learning capabilities and can perform shift-invariant classification on input information according to their hierarchical structure.

[0046] Optionally, the sample historical power data is obtained through the following steps: collecting original power data, wherein the original power data includes historical power transaction price data, power production and consumption data, long-term weather forecast data, historical load data and new energy wind power output data, the weather forecast data includes temperature data, wind speed data and light intensity data, and the power production and consumption data includes thermal power generation data, wind power generation data, photovoltaic power generation data, power transportation loss data and power exchange data in various regions; averaging the original power data to obtain the processed original power data, and using the processed original power data as the sample historical power data, wherein the weather forecast data includes at least average light, average temperature and average wind speed.

[0047] Optionally, the day-ahead electricity price forecast data is obtained through the following steps: converting the preprocessed original electricity data into an adjacency matrix topology graph, and performing a graph convolution operation on the adjacency matrix to obtain embedding vectors of different dimensions; splicing the embedding vectors of different dimensions with the preprocessed original electricity data to obtain spliced ​​data, and obtaining a weighted vector based on the spliced ​​data; obtaining the electricity trading price data to be predicted based on the weighted vector, and using the electricity trading price data to be predicted as the day-ahead electricity price forecast data.

[0048] Step 104 : based on the analysis result obtained by analyzing each of the at least one electricity day-ahead declaration optimization strategy, select a target electricity day-ahead declaration optimization strategy from the at least one electricity day-ahead declaration optimization strategy.

[0049] The execution subject may select a target power day declaration optimization strategy from the at least one power day declaration optimization strategy based on an analysis result obtained by analyzing each power day declaration optimization strategy in the at least one power day declaration optimization strategy; as an example, the analysis result may be a price corresponding to each power day declaration optimization strategy, or a carbon emission corresponding to each power day declaration optimization strategy, or a utilization rate corresponding to each power day declaration optimization strategy. The execution subject may make a selection based on any one of the analysis results, or may make a selection based on an average weighted calculation of multiple analysis results.

[0050] In the embodiment of the present invention, the acquired operation boundary data and historical transaction data of the power spot market in the target area are preprocessed to obtain processed data; based on the processed data and the historical electricity price forecast data, the pre-trained initial power price forecast model is optimized to obtain the power price forecast model; according to the acquired user power demand information and the day-ahead power price forecast data obtained based on the power price forecast model, at least one power day-ahead declaration optimization strategy is generated; based on the analysis results obtained by analyzing each power day-ahead declaration optimization strategy in the at least one power day-ahead declaration optimization strategy, a target power day-ahead declaration optimization strategy is selected from the at least one power day-ahead declaration optimization strategy. This realizes providing users with the best declaration plan under the premise of meeting market rules and constraints.

[0051] The above describes the method for generating the optimization strategy for the day-ahead declaration of the power spot market in the embodiment of the present invention. The following describes the device 200 for generating the optimization strategy for the day-ahead declaration of the power spot market in the embodiment of the present invention. Figure 2 In one embodiment of the present invention, an embodiment of a device for generating an optimization strategy for day-ahead declaration in a power spot market includes: A preprocessing module 201 is used to preprocess the acquired operation boundary data and historical transaction data of the power spot market in the target area to obtain processed data; A model optimization module 202 is used to optimize the pre-trained initial power price prediction model based on the processed data and the historical power price prediction data to obtain a power price prediction model; A strategy generation module 203 is used to generate at least one power day-ahead declaration optimization strategy based on the acquired power demand information of the user and the day-ahead power price forecast data obtained based on the power price forecast model; The strategy analysis module 204 is used to select a target electricity day-ahead declaration optimization strategy from the at least one electricity day-ahead declaration optimization strategy based on an analysis result obtained by analyzing each electricity day-ahead declaration optimization strategy in the at least one electricity day-ahead declaration optimization strategy.

[0052] The functional implementation of each module in the above-mentioned power spot market day-ahead declaration optimization strategy generation device corresponds to the steps in the above-mentioned power spot market day-ahead declaration optimization strategy generation method embodiment, and its functions and implementation processes will not be repeated here one by one.

[0053] Optionally, before the strategy generation module 203, the device further includes: A data acquisition submodule, used to acquire the power market data of the target area; A data preprocessing submodule, used for preprocessing the electricity market data to obtain processed electricity market data; A model optimization submodule, used to construct a multi-layer neural network model based on the processed power market data to obtain a power price prediction model; The day-ahead electricity price forecast data generating submodule is used to input the processed data into the electricity price forecast model to obtain the day-ahead electricity price forecast data.

[0054] Optionally, the data preprocessing submodule includes: A data analysis unit, used to perform correlation analysis on the data affecting the real-time electricity price in the electricity market data, remove irrelevant data, and obtain a processed data set; The data format conversion unit is used to pre-process the processed data set, convert the processed data set into data in a format suitable for input of the electricity price prediction model, and use the data in the format as the processed electricity market data.

[0055] Optionally, the electricity market data includes: power generation side data, grid side data market transaction data, power sales side data, market transaction data and policy and regulation data, wherein the power generation side data includes power generation data, power generation cost data and unit operation data; the grid side data includes transmission and distribution data, grid operation data and grid construction data; the power sales side data includes power sales data, power sales data and user data; the market transaction data includes spot market transaction data, medium- and long-term transaction data and ancillary service market transaction data; the policy and regulation data includes power market policy documents and regulatory data.

[0056] Optionally, the initial electricity price prediction model is obtained through the following training steps: Acquire a training sample set, wherein each training sample in the training sample set includes: sample historical power data and corresponding day-ahead power price forecast data; The sample historical power data of the training sample set in the training sample set is taken as input, and the day-ahead power price prediction data corresponding to the input sample historical power data is taken as expected output, so as to train and obtain the initial power price prediction model.

[0057] Optionally, the sample historical power data is obtained by the following steps: Collecting raw power data, wherein the raw power data includes historical power transaction price data, power production and consumption data, long-term weather forecast data, historical load data and new energy wind power output data, the weather forecast data includes temperature data, wind speed data and light intensity data, the power production and consumption data includes thermal power generation data, wind power generation data, photovoltaic power generation data, power transportation loss data and power exchange data of various regions; The original power data is averaged to obtain the processed original power data, and the processed original power data is used as the sample historical power data, wherein the weather forecast data at least includes average sunshine, average temperature and average wind speed.

[0058] Optionally, the day-ahead electricity price forecast data is obtained through the following steps: Converting the preprocessed raw power data into an adjacency matrix topology graph, and performing a graph convolution operation on the adjacency matrix to obtain embedding vectors of different dimensions; Splicing the embedding vectors of different dimensions with the preprocessed original power data to obtain spliced ​​data, and obtaining a weighted vector according to the spliced ​​data; The power transaction price data to be predicted is obtained according to the weighted vector, and the power transaction price data to be predicted is used as the day-ahead electricity price prediction data.

[0059] above Figure 2 The device for generating an optimization strategy for day-ahead declaration in the electricity spot market in an embodiment of the present invention is described in detail from the perspective of modular functional entities. The device for generating an optimization strategy for day-ahead declaration in the electricity spot market in an embodiment of the present invention is described in detail from the perspective of hardware processing.

[0060] Figure 3 It is a structural diagram of a device for generating an optimization strategy for day-ahead declaration of a power spot market provided by an embodiment of the present invention. The device 300 for generating an optimization strategy for day-ahead declaration of a power spot market may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage devices) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 may be temporary storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of computer program operations in the device 300 for generating an optimization strategy for day-ahead declaration of a power spot market. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of computer program operations in the storage medium 330 on the device 300 for generating an optimization strategy for day-ahead declaration of a power spot market.

[0061] The power spot market day-ahead declaration optimization strategy generation device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will appreciate that Figure 3 The structure of the device for generating an optimization strategy for day-ahead declarations in the electricity spot market shown does not constitute a limitation of the device for generating an optimization strategy for day-ahead declarations in the electricity spot market, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0062] The present invention also provides a device for generating an optimization strategy for day-ahead declarations in the electricity spot market, comprising: a memory and at least one processor, wherein a computer program is stored in the memory, and the memory and the at least one processor are interconnected via a line; the at least one processor calls the computer program in the memory so that the device for generating an optimization strategy for day-ahead declarations in the electricity spot market executes the steps in the above-mentioned method for generating an optimization strategy for day-ahead declarations in the electricity spot market.

[0063] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. A computer program is stored in the computer-readable storage medium. When the computer program runs on a computer, the computer executes the steps of a method for generating a day-ahead declaration optimization strategy for the electricity spot market.

[0064] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0065] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several computer programs to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0066] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

[0067] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed, but such modifications are protected by the patent law as long as they are within the scope of the claims of the present application.

Claims

1. A method for generating an optimization strategy for day-ahead declaration in the electricity spot market, characterized in that: include: Preprocessing the acquired operation boundary data and historical transaction data of the power spot market in the target area to obtain processed data; Based on the processed data and the historical electricity price forecast data, optimizing the pre-trained initial electricity price forecast model to obtain an electricity price forecast model; Generate at least one day-ahead electricity declaration optimization strategy based on the acquired user power demand information and the day-ahead electricity price forecast data obtained based on the power price forecast model; Based on the analysis result obtained by analyzing each of the at least one optimization strategy for electricity day-ahead declaration, a target electricity day-ahead declaration optimization strategy is selected from the at least one optimization strategy for electricity day-ahead declaration.

2. The method for generating an optimization strategy for day-ahead declaration in the electricity spot market according to claim 1, characterized in that: Before generating at least one power day-ahead declaration optimization strategy based on the acquired power demand information of the user and the day-ahead power price forecast data obtained based on the power price forecast model, the method further includes: Acquiring electricity market data of the target area; Preprocessing the electricity market data to obtain processed electricity market data; Based on the processed power market data, a multi-layer neural network model is constructed to obtain a power price prediction model; The processed data is input into the electricity price prediction model to obtain the day-ahead electricity price prediction data.

3. The method for generating an optimization strategy for day-ahead declaration in the electricity spot market according to claim 2, characterized in that: The preprocessing of the power market data to obtain processed power market data includes: Performing correlation analysis on the data affecting the real-time electricity price in the electricity market data, eliminating irrelevant data, and obtaining a processed data set; The processed data set is preprocessed to convert the processed data set into data in a format suitable for input into the electricity price prediction model, and the data in the format is used as the processed electricity market data.

4. The method for generating an optimization strategy for day-ahead declaration in the electricity spot market according to claim 1, characterized in that: The electricity market data includes: power generation side data, grid side data market transaction data, power sales side data, market transaction data and policy and regulation data, wherein the power generation side data includes power generation data, power generation cost data and unit operation data; the grid side data includes transmission and distribution data, grid operation data and grid construction data; the power sales side data includes power sales data, power sales data and user data; the market transaction data includes spot market transaction data, medium- and long-term transaction data and ancillary service market transaction data; the policy and regulation data includes power market policy documents and supervision data.

5. The method for generating an optimization strategy for day-ahead declaration in the electricity spot market according to claim 1, characterized in that: The initial electricity price prediction model is obtained through the following training steps: Acquire a training sample set, wherein each training sample in the training sample set includes: sample historical power data and corresponding day-ahead power price forecast data; The sample historical power data in each training sample in the training sample set is taken as input, and the day-ahead power price prediction data corresponding to the input sample historical power data is taken as expected output, so as to train and obtain the initial power price prediction model.

6. A method for generating an optimization strategy for day-ahead declaration in the electricity spot market according to claim 5, characterized in that: The sample historical power data is obtained by the following steps: Collecting raw power data, wherein the raw power data includes historical power transaction price data, power production and consumption data, long-term weather forecast data, historical load data and new energy wind power output data, the weather forecast data includes temperature data, wind speed data and light intensity data, the power production and consumption data includes thermal power generation data, wind power generation data, photovoltaic power generation data, power transportation loss data and power exchange data of various regions; The original power data is averaged to obtain the processed original power data, and the processed original power data is used as the sample historical power data, wherein the weather forecast data at least includes average sunshine, average temperature and average wind speed.

7. A method for generating an optimization strategy for day-ahead declaration in the electricity spot market according to claim 6, characterized in that: The day-ahead electricity price forecast data is obtained by the following steps: Converting the preprocessed raw power data into an adjacency matrix topology graph, and performing a graph convolution operation on the adjacency matrix to obtain embedding vectors of different dimensions; Splicing the embedding vectors of different dimensions with the preprocessed original power data to obtain spliced ​​data, and obtaining a weighted vector according to the spliced ​​data; The power transaction price data to be predicted is obtained according to the weighted vector, and the power transaction price data to be predicted is used as the day-ahead electricity price prediction data.

8. A device for generating an optimization strategy for day-ahead declaration in the electricity spot market, characterized in that: include: A preprocessing module is used to preprocess the acquired operation boundary data and historical transaction data of the power spot market in the target area to obtain processed data; A model optimization module, used to optimize the pre-trained initial electricity price prediction model based on the processed data and the historical electricity price prediction data to obtain an electricity price prediction model; A strategy generation module, used to generate at least one power day-ahead declaration optimization strategy based on the acquired power demand information of the user and the day-ahead power price forecast data obtained based on the power price forecast model; The strategy analysis module is used to select a target electricity day-ahead declaration optimization strategy from the at least one electricity day-ahead declaration optimization strategy based on an analysis result obtained by analyzing each electricity day-ahead declaration optimization strategy in the at least one electricity day-ahead declaration optimization strategy.

9. A device for generating an optimization strategy for day-ahead declaration in the electricity spot market, characterized in that: The device for generating an optimization strategy for day-ahead declarations in the electricity spot market comprises: a memory, a processor and a program for generating an optimization strategy for day-ahead declarations in the electricity spot market stored in the memory and executable on the processor. When the program for generating an optimization strategy for day-ahead declarations in the electricity spot market is executed by the processor, the steps of the method for generating an optimization strategy for day-ahead declarations in the electricity spot market are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program for generating an optimization strategy for day-ahead declarations in the electricity spot market. When the program for generating an optimization strategy for day-ahead declarations in the electricity spot market is executed by a processor, the steps of the method for generating an optimization strategy for day-ahead declarations in the electricity spot market as described in any one of claims 1 to 7 are implemented.

Images