Inter-provincial spot transaction volume price declaration behavior prediction method and system

By classifying and modeling the inter-provincial spot transaction data, combining RNN and SVM algorithms to predict the power demand and electricity price declaration range, and considering the capacity limit of the transmission channel, the low subjectivity and accuracy of volume and price prediction in the existing technology are solved. The problem of this has achieved a more scientific and reliable power purchase strategy, improving the success rate and cost optimization of power purchase.

CN119090544APending Publication Date: 2024-12-06STATE GRID BEIJING ELECTRIC POWER CO +1
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411202643.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing inter-provincial spot power purchase transaction volume and price prediction methods have problems such as strong subjectivity, low accuracy and large deviations, and the transmission channel capacity limit has not been fully considered, resulting in uncertainty in the success rate and cost of power purchase.

Method used

A method for predicting the volume and price declaration behavior of spot transactions between provinces is proposed. By classifying historical data, the power supply guarantee and cost reduction data sets are trained, and the power demand and electricity price are predicted using RNN recurrent neural network and SVM support vector machine algorithm. The declaration range is combined with the capacity limit of the transmission channel, and the power price declaration strategy is generated, and the verification and aggregation is carried out according to the inter-provincial spot trading rules to generate the quantity and price declaration prediction results.

Benefits of technology

It improves the stability and accuracy of electricity price prediction, enhances the scientificity and reliability of the quantity and price declaration decisions of the power purchase entity, and ensures the optimization of the success rate and cost of power purchase.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119090544A_ABST
    Figure CN119090544A_ABST
Patent Text Reader

Abstract

The invention provides an inter-provincial spot transaction volume price declaration behavior prediction method and system. The method comprises the following steps: classifying historical data; according to the power supply guaranteeing type data set and the cost reduction type data set, training to obtain various models; studying and judging electric energy supply and demand conditions of the to-be-predicted electricity purchasing main body; generating a power price declaration strategy of the power purchasing node according to the data corresponding to the power purchasing node; verifying the electricity price declaration strategy of the electricity purchasing node according to the inter-provincial spot transaction declaration rule; and if the check of the spot transaction declaration rule is satisfied, aggregating the electricity price declaration strategies of the electricity purchasing nodes according to a time sequence, and generating a quantity price declaration prediction result of the inter-provincial spot electricity purchasing subject. According to the invention, by fully considering the capacity limit of the power transmission channel and the dynamic change of the market, the accuracy and reliability of the prediction result of the price declaration behavior can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of electricity spot market transactions, and in particular to a method and system for predicting inter-provincial spot transaction quantity and price declaration behavior. Background Art

[0002] Inter-provincial spot trading in the power market is an important way to achieve optimal allocation of power resources, balance the interests of enterprises of different types of power generation, and guide the flexible development of regulatory resources and the construction of transmission channels in the system. With the full promotion of inter-provincial spot trading in the power market, the scale and complexity of the inter-provincial spot power purchase transaction market are increasing. As the electrification level of industry, commerce, residents and agriculture increases year by year, the power consumption gap is gradually increasing. In order to ensure the balance between supply and demand, higher requirements are put forward for the quantity and price declaration decisions of power purchase entities.

[0003] However, although the existing inter-provincial spot power purchase transaction volume and price declaration uses various intelligent algorithms to assist decision-making, there are still the following problems: First, the current intelligent algorithm cannot accurately separate the main influencing factors of the current electricity price, and the final declaration strategy still needs to be adjusted based on expert experience, which has strong subjectivity, low accuracy, and large deviation. Second, for local power grids with multiple power purchase entities in the same node (such as a node containing three power purchase entities in City A, City B, and City C), under the premise of limited capacity of cross-node transmission channels, each power purchase entity has a competitive relationship. The power purchase declaration strategy that does not consider the competitive relationship will greatly reduce its own power purchase success rate. Summary of the invention

[0004] Based on this, the purpose of this application is to propose a method and system for predicting inter-provincial spot transaction quantity and price declaration behavior, aiming to solve the problems of strong subjectivity, low accuracy and large deviation in the traditional inter-provincial spot power purchase transaction quantity and price prediction results.

[0005] In the first aspect, the present application proposes a method for predicting inter-provincial spot transaction volume and price declaration behavior, the method comprising:

[0006] Classify the historical data to obtain a power supply guarantee data set and a cost reduction data set according to the classification results;

[0007] According to the power supply guarantee type data set and the cost reduction type data set, a power supply guarantee type power demand prediction model, a cost reduction type power demand prediction model, a power supply guarantee type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply guarantee type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model are obtained by training;

[0008] For the prediction of the electricity supply and demand of the electricity purchaser itself, if the electricity purchaser's internal power generation is less than or equal to the power consumption of the electricity purchase node, the electricity purchaser's power purchase target will be classified according to the power gap prediction data;

[0009] The data corresponding to the power purchase node is input into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and the power demand and electricity price declaration interval results of the power purchase node are predicted in turn;

[0010] The predicted power demand of the power purchasing node and the power price declaration interval of the power purchasing node are used as input data, and input into the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, so as to generate the power price declaration strategy of the power purchasing node;

[0011] Verify the electricity price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules;

[0012] If the spot transaction declaration rule verification is met, the electricity price declaration strategy of the power purchasing node will be aggregated according to the time series to generate the quantity and price declaration forecast results of the inter-provincial spot power purchasing entities.

[0013] In summary, according to the above-mentioned inter-provincial spot transaction quantity and price declaration behavior prediction method, the user's historical electricity purchasing nodes are first divided into two categories, and the electricity purchasing targets of competitive electricity purchasing entities are divided into power supply guarantee type and cost reduction type through historical electricity gap data, so as to realize the refinement of the prediction of quantity and price declaration behavior of competitive electricity purchasing entities and improve the prediction accuracy. Then, the multi-step prediction method is used to predict the quantity and price declaration behavior. First, the electricity and electricity price of the competitive electricity purchasing entities are decoupled and predicted. Since the formation mechanism of electricity and electricity price has a weak indirect correlation, the electricity demand and electricity price declaration range are predicted respectively, and then the influencing factors with strong correlation with electricity and electricity price are obtained respectively, and the RNN recurrent neural network algorithm is used to train the prediction model. Then, the decoupled predicted power demand and price declaration interval are coupled and analyzed, and the SVM support vector machine algorithm is used to mine the mapping relationship between the power and price declaration data group and the power demand and price declaration interval, forming the prediction results of the power and price declaration data group of competitive power purchase entities. Finally, the predicted power price declaration data group is verified according to the inter-provincial spot declaration rules, and the power and price declaration strategies of the power purchase nodes are aggregated in time series when the verification is met, thereby generating the quantity and price declaration prediction results of the inter-provincial spot power purchase entities with competitive relationships. This prediction method can guarantee the prediction stability and accuracy of power prices to the greatest extent. In addition, by taking the inter-provincial channel capacity limit as a strong correlation factor for the prediction of power demand and price interval, the accuracy of the prediction of power demand and price interval is further improved.

[0014] Furthermore, the step of classifying the historical data to obtain the power supply guarantee type data set and the cost reduction type data set according to the classification results includes:

[0015] Based on historical electricity shortage data As the basis for classification, When , the power purchasing node is a power supply guarantee type power purchasing node, t represents the power purchasing node 1≤t≤96, t∈Z;

[0016] when When , the power purchasing node is a cost-reducing power purchasing node;

[0017] The data under the power supply guarantee type power purchasing node t are all classified into the power supply guarantee type data set, and the data under the cost reduction type power purchasing node t are all classified into the cost reduction type data set.

[0018] Further, the step of training a power supply type electricity demand prediction model, a cost reduction type electricity demand prediction model, a power supply type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model according to the power supply type data set and the cost reduction type data set includes:

[0019] The historical power generation data contained in the power supply guarantee data set and the cost reduction data set are Historical load data Historical channel capacity limit data Historical electricity deficit data Historical electricity total declared data The data is used as model training data and input into RNN for model training to generate a power supply guarantee type power demand forecasting model and a cost reduction type power demand forecasting model respectively;

[0020] The historical proxy power purchase price data contained in the power supply guarantee data set and the cost reduction data set are Historical temperature data Historical channel capacity limit data Historical clearing price data of major electricity sales provinces Historical declared price range data As the model training data, it is input into RNN for model training, and the prediction model of electricity price declaration interval for power supply guarantee type and the prediction model of electricity price declaration interval for cost reduction type are generated respectively;

[0021] The historical electricity total amount data contained in the power supply guarantee data set and the cost reduction data set are Historical declared price range data Historical electricity deficit data Historical electricity price declaration data group The data is input into SVM as model training data for model training, and a coupling model of power supply guarantee type electricity price declaration strategy and a coupling model of cost reduction type electricity price declaration strategy are generated respectively.

[0022] Furthermore, the step of classifying the power purchase target of the power purchase entity according to the power gap forecast data includes:

[0023] The power generation forecast curve data of the power purchasing entity to be predicted Load forecast curve data Subtract to generate supply and demand curve data in,

[0024] when When , the power purchasing strategy of the tth power purchasing node is are all 0, indicating that there is no electricity purchase behavior at this moment; When , the power purchasing strategy of the tth power purchasing node is predicted to obtain the power gap prediction data

[0025] Furthermore, the step of inputting the data corresponding to the power purchase node into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and sequentially predicting the power demand of the power purchase node and the electricity price declaration interval results includes:

[0026] When the power shortage forecast data It is necessary to adopt the guaranteed power supply type electricity price declaration interval prediction model, and convert the monthly proxy power purchase price data of the t-th power purchase node into Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input the trained guaranteed power supply type electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th power purchasing node

[0027] When the power shortage forecast data When the monthly proxy power purchase price data of the tth power purchase node is Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input into the cost-reduction electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th electricity purchasing node

[0028] Furthermore, the predicted power demand of the power purchasing node and the power purchasing node electricity price declaration interval result are input as input data to the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, and the step of generating the power price declaration strategy of the power purchasing node includes:

[0029] The predicted electricity price declaration interval for the tth electricity purchasing node and power demand Input the trained cost-reduction or power supply guarantee type electricity price coupling declaration strategy model, and output the power purchase strategy of the t-th power purchase node And input the power purchasing strategy into the spot transaction declaration rules for verification.

[0030] Furthermore, the step of verifying the electricity price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules includes:

[0031] The verification rules for spot transaction declaration are as follows: When the spot transaction declaration rule verification is not met, the power price coupling declaration strategy model is returned to regenerate the power purchase strategy until the spot transaction declaration rule verification is met.

[0032] Furthermore, if the spot transaction declaration rule verification is met, the power price declaration strategy of the power purchasing node is aggregated according to the time series, and the step of generating the quantity and price declaration prediction result of the inter-provincial spot power purchasing entity includes:

[0033] Aggregate the power purchasing strategies of 96 power purchasing nodes to obtain the power purchasing strategies of competitive power purchasing entities

[0034]

[0035] In the second aspect, the present application proposes a system for predicting the volume and price declaration behavior of inter-provincial spot transactions, the system comprising:

[0036] A data classification module is used to classify historical data to obtain a power supply guarantee type data set and a cost reduction type data set according to the classification results;

[0037] A model training module, used for training, according to the power supply guarantee type data set and the cost reduction type data set, obtaining a power supply guarantee type power demand prediction model, a cost reduction type power demand prediction model, a power supply guarantee type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply guarantee type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model;

[0038] The target classification module is used to analyze and judge the power supply and demand of the power purchase entity itself. If the power purchase entity's internal power generation is less than or equal to the power consumption of the power purchase node, the power purchase entity's power purchase target is classified according to the power gap forecast data;

[0039] The prediction module is used to input the data corresponding to the power purchase node into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and predict the power demand and electricity price declaration interval results of the power purchase node in turn;

[0040] The declaration strategy generation module is used to input the predicted power demand of the power purchasing node and the power price declaration interval results of the power purchasing node as input data to the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, and generate the power price declaration strategy of the power purchasing node;

[0041] A verification module is used to verify the electricity price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules;

[0042] The strategy aggregation module is used to aggregate the electricity price declaration strategy of the power purchasing node in time series if the spot transaction declaration rule verification is met, and generate the quantity and price declaration forecast results of the inter-provincial spot power purchasing entity.

[0043] In a third aspect, the present application also provides a readable storage medium, which stores one or more programs, which, when executed, implement the above-mentioned method for predicting the volume and price declaration behavior of inter-provincial spot transactions.

[0044] In a fourth aspect, the present application further provides a computer device, the computer device comprising a memory and a processor, wherein:

[0045] The memory is used to store computer programs;

[0046] When the processor is used to execute the computer program stored in the memory, the above-mentioned method for predicting the quantity and price declaration behavior of inter-provincial spot transactions is implemented.

[0047] Additional aspects and advantages of the present application will be given in part in the following description, and in part will become apparent from the following description, or will be understood through the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0049] Figure 1 This is one of the flow charts of a method for predicting inter-provincial spot transaction quantity and price declaration behavior in one embodiment of the present application;

[0050] Figure 2 This is a second flow chart of a method for predicting inter-provincial spot transaction volume and price declaration behavior in one embodiment of the present application;

[0051] Figure 3 This is a schematic diagram of a data structure for predicting a power price declaration strategy of a power purchase entity in one embodiment of the present application;

[0052] Figure 4A schematic diagram of a flow chart of a power demand prediction model in one embodiment of the present application;

[0053] Figure 5 This is a flow chart of a prediction model for an electricity price declaration interval in one embodiment of the present application;

[0054] Figure 6 This is a flow chart of a power price declaration strategy coupling model in one embodiment of the present application;

[0055] Figure 7 This is a schematic diagram of a process for dividing a power supply guarantee type data set and a cost reduction type data set in one embodiment of the present application;

[0056] Figure 8 A schematic diagram of neuron learning of the RNN model in one embodiment of the present application;

[0057] Fig. 9 It is a structural diagram of the inter-provincial spot transaction quantity and price declaration behavior prediction system in one embodiment of the present application.

[0058] The following specific implementation methods will further illustrate the present application in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0059] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0060] The following detailed descriptions are all exemplary descriptions, and are intended to provide further detailed descriptions of the present application. Unless otherwise specified, all technical terms used in the present application have the same meanings as those generally understood by those skilled in the art to which the present application belongs. The terms used in the present application are only for describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present application.

[0061] The applicant found that the existing competitive electricity purchasing entity quantity and price declaration behavior prediction technology has the following shortcomings:

[0062] 1. Strong subjectivity: The traditional competitive power purchaser quantity and price declaration forecast mainly relies on the personal experience and judgment of experts, which makes the forecast results highly subjective. Different experts may make forecasts based on different judgment criteria and experience, resulting in inconsistency and uncertainty in the forecast results.

[0063] 2. Low accuracy: Since traditional methods cannot flexibly analyze the main external factors that currently affect the supply and demand of electricity of competitive power purchasers, the parameter weights of the competitive power purchaser's prediction model cannot be flexibly adjusted, and the accuracy of its prediction results fluctuates greatly. This may lead to large deviations in the declaration strategies formulated by power purchasers when referring to the prediction results of competitors' power purchases, thereby affecting the success rate and cost of power purchases.

[0064] 3. Failure to consider channel capacity limits: In the electricity market, the capacity of the transmission channel is limited, and the traditional quantity and price reporting and forecasting methods often do not fully consider the capacity limits of the transmission channel, resulting in transmission channel congestion, power loss and other problems in actual transactions, resulting in the inability to meet electricity purchase needs, and may also have an adverse impact on the stable operation of the entire electricity market.

[0065] 4. Lack of dynamic adjustment mechanism: The supply and demand relationship and price changes in the electricity market are dynamic, and the traditional quantity and price declaration forecasting methods often lack a dynamic adjustment mechanism and cannot adjust the declaration strategy in time according to market changes. This may cause the electricity purchaser to be at a disadvantage in the market and fail to achieve the optimal electricity purchase effect.

[0066] Based on this, this application proposes a method for predicting quantity and price declaration behavior of inter-provincial spot transactions. By introducing scientific data analysis and model support, and fully considering the capacity limit of the transmission channel and the dynamic changes of the market, the accuracy and reliability of the prediction results of quantity and price declaration behavior can be improved, and an important basis for electricity buyers to formulate more scientific and reasonable declaration strategies can be provided.

[0067] See also Figures 1 to 8 , shown is a flow chart of a method for predicting inter-provincial spot transaction volume and price declaration behavior in an embodiment of the present application, the method comprising steps S101 to S107, wherein:

[0068] Step S101: classify the historical data to obtain a power supply guarantee type data set and a cost reduction type data set according to the classification results;

[0069] It should be noted that this step classifies scenarios through historical data, and takes historical electricity gap data as a benchmark. The power purchasing nodes with historical electricity gaps greater than 0 are defined as power supply guarantee type power purchasing nodes, and the power purchasing nodes with historical electricity gaps less than or equal to 0 are defined as cost reduction type power purchasing nodes. Then, all types of data corresponding to power supply guarantee type power purchasing nodes and cost reduction type power purchasing nodes are divided into power supply guarantee type data sets and cost reduction type data sets, respectively.

[0070] Specifically, in some embodiments, in the process of classifying historical data, historical power shortage data is used as the basis for classification. As the basis for classification, When , the power purchasing node t is a power supply guarantee type power purchasing node, where t represents the power purchasing node, 1≤t≤96, t∈Z, and all data types under the historical power purchasing node t are classified into the power supply guarantee type data set. Otherwise, the power purchasing node t is a cost reduction type power purchasing node, and all data types under the historical power purchasing node t are classified into the cost reduction type data set.

[0071] Step S102: training a power supply type electricity demand prediction model, a cost reduction type electricity demand prediction model, a power supply type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model according to the power supply type data set and the cost reduction type data set;

[0072] In some embodiments, the specific training process of the model is:

[0073] The historical power generation data contained in the power supply guarantee data set and the cost reduction data set are Historical load data Historical channel capacity limit data Historical electricity deficit data Historical electricity total declared data The data is used as model training data and input into RNN for model training to generate a power supply guarantee type power demand forecasting model and a cost reduction type power demand forecasting model respectively;

[0074] The historical proxy power purchase price data contained in the power supply guarantee data set and the cost reduction data set are Historical temperature data Historical channel capacity limit data Historical clearing price data of major electricity sales provinces Historical declared price range data As the model training data, it is input into RNN for model training, and the prediction model of electricity price declaration interval for power supply guarantee type and the prediction model of electricity price declaration interval for cost reduction type are generated respectively;

[0075] The historical electricity total amount data contained in the power supply guarantee data set and the cost reduction data set are Historical declared price range data Historical electricity deficit data Historical electricity price declaration data group The data is input into SVM as model training data for model training, and a coupling model of power supply guarantee type electricity price declaration strategy and a coupling model of cost reduction type electricity price declaration strategy are generated respectively.

[0076] Step S103: Analyze and judge the power supply and demand of the power purchase entity to be predicted. If the power purchase entity's internal power generation is less than or equal to the power consumption of the power purchase node, classify the power purchase entity's power purchase target according to the power gap prediction data;

[0077] It should be pointed out that in the specific research and judgment process, when the power purchasing entity's internal power generation is greater than the power purchasing node's power consumption, that is, when its own power generation cannot be used up, it is prohibited to purchase power according to the inter-provincial spot trading rules. Therefore, when the power purchasing node's own power generation is greater than the power consumption, its power purchasing strategy is 0.

[0078] If the internal power generation of the power purchasing entity is less than or equal to the power purchasing node of the power consumption, it means that the power gap is greater than 0, that is, its own power generation cannot meet the demand. At this time, it means that the total amount of internal power generation of the power purchasing entity and medium- and long-term power purchases cannot meet the electricity demand. The power purchasing entity's power purchasing goal is to ensure power supply, that is, to accept a higher market price to ensure that enough electricity is purchased.

[0079] In addition, when the electricity gap is equal to 0, that is, the internal power generation of the electricity purchasing entity is equal to the power consumption, it means that the total amount of internal power generation of the electricity purchasing entity and the medium- and long-term power purchases can meet the electricity demand. At this time, the electricity purchasing entity's power purchasing goal is to reduce costs, that is, to purchase low-priced electricity on the market at a lower price.

[0080] Specifically, in some embodiments, the power generation forecast curve data of the power purchase entity to be predicted needs to be firstly Load forecast curve data Subtract to generate supply and demand curve data in,

[0081] when When , the power purchasing strategy of the tth power purchasing node is are all 0, indicating that there is no electricity purchase behavior at this moment; When , the power purchasing strategy of the tth power purchasing node is predicted to obtain the power gap prediction data

[0082] Step S104: inputting the data corresponding to the power purchase node into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and predicting the power demand and electricity price declaration interval results of the power purchase node in turn;

[0083] In some embodiments, when judging through the power gap forecast data, it is necessary to adopt the guaranteed power supply type electricity price declaration interval forecast model, and use the proxy power purchase price data of the tth power purchase node in the current month as the forecast model. Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input the trained guaranteed power supply type electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th power purchasing node When the power shortage forecast data The cost-reduction electricity price declaration interval prediction model is needed to input the above prediction data into the cost-reduction electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th electricity purchase node.

[0084]

[0085] Step S105: The predicted power demand of the power purchasing node and the power purchasing node electricity price declaration interval results are used as input data, and input into the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, and the power price declaration strategy of the power purchasing node is generated;

[0086] In this step, the predicted electricity price declaration interval of the t-th power purchasing node is calculated based on the type of the t-th power purchasing node. and power demand Input the trained cost-reduction or power supply guarantee type electricity price coupling declaration strategy model, and output the power purchase strategy of the t-th power purchase node

[0087]

[0088] Step S106: verifying the electricity price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules;

[0089] After obtaining the power purchase strategy, the power purchase strategy is input into the spot transaction declaration rule verification. The spot transaction declaration verification rule is: When the spot transaction declaration rule verification is not met, the power price coupling declaration strategy model is returned to regenerate the power purchase strategy until the spot transaction declaration rule verification is met.

[0090] Step S107: If the spot transaction declaration rule verification is met, the power price declaration strategy of the power purchasing node is aggregated according to the time series to generate the quantity and price declaration forecast results of the inter-provincial spot power purchasing entity.

[0091] Specifically, the power purchasing strategies of the 96 power purchasing nodes that have passed the verification are aggregated to obtain the power purchasing strategies of the competitive power purchasing entities. That is, the quantity and price declaration forecast results of inter-provincial spot electricity purchasing entities are obtained.

[0092] In summary, according to the above-mentioned inter-provincial spot transaction quantity and price declaration behavior prediction method, the user's historical electricity purchasing nodes are first divided into two categories, and the electricity purchasing targets of competitive electricity purchasing entities are divided into power supply guarantee type and cost reduction type through historical electricity gap data, so as to realize the refinement of the prediction of quantity and price declaration behavior of competitive electricity purchasing entities and improve the prediction accuracy. Then, the multi-step prediction method is used to predict the quantity and price declaration behavior. First, the electricity and electricity price of the competitive electricity purchasing entities are decoupled and predicted. Since the formation mechanism of electricity and electricity price has a weak indirect correlation, the electricity demand and electricity price declaration range are predicted respectively, and then the influencing factors with strong correlation with electricity and electricity price are obtained respectively, and the RNN recurrent neural network algorithm is used to train the prediction model. Then, the decoupled predicted power demand and price declaration interval are coupled and analyzed, and the SVM support vector machine algorithm is used to mine the mapping relationship between the power and price declaration data group and the power demand and price declaration interval, forming the prediction results of the power and price declaration data group of competitive power purchase entities. Finally, the predicted power price declaration data group is verified according to the inter-provincial spot declaration rules, and the power and price declaration strategies of the power purchase nodes are aggregated in time series when the verification is met, thereby generating the quantity and price declaration prediction results of the inter-provincial spot power purchase entities with competitive relationships. This prediction method can guarantee the prediction stability and accuracy of power prices to the greatest extent. In addition, by taking the inter-provincial channel capacity limit as a strong correlation factor for the prediction of power demand and price interval, the accuracy of the prediction of power demand and price interval is further improved.

[0093] See also Fig. 9 , shown is a structural schematic diagram of a system for predicting the volume and price declaration behavior of inter-provincial spot transactions in one embodiment of the present application, the system comprising:

[0094] A data classification module 10 is used to classify historical data to obtain a power supply guarantee type data set and a cost reduction type data set according to the classification results;

[0095] Furthermore, in some optional embodiments, the data classification module 10 further includes:

[0096] The power supply guarantee type power purchase node determination unit is used to determine the power supply guarantee type power purchase node based on the historical power shortage data. As the basis for classification, When , the power purchasing node is a power supply guarantee type power purchasing node, t represents the power purchasing node 1≤t≤96, t∈Z;

[0097] Cost-reducing power purchase node determination unit, used when When , the power purchasing node is a cost-reducing power purchasing node;

[0098] The data set generating unit is used to classify the data under the power supply guarantee type power purchasing node t into the power supply guarantee type data set, and classify the data under the cost reduction type power purchasing node t into the cost reduction type data set.

[0099] A model training module 20 is used to train a power supply type electricity demand prediction model, a cost reduction type electricity demand prediction model, a power supply type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model according to the power supply type data set and the cost reduction data set;

[0100] Furthermore, in some optional embodiments, the model training module 20 further includes:

[0101] The demand forecasting model building unit is used to convert the historical power generation data contained in the power supply guarantee data set and the cost reduction data set into Historical load data Historical channel capacity limit data Historical electricity deficit data Historical electricity total declared data The data is used as model training data and input into RNN for model training to generate a power supply guarantee type power demand forecasting model and a cost reduction type power demand forecasting model respectively;

[0102] The electricity price declaration interval prediction module construction unit is used to convert the historical proxy electricity purchase price data contained in the power supply guarantee data set and the cost reduction data set into Historical temperature data Historical channel capacity limit data Historical clearing price data of major electricity sales provinces Historical declared price range data As the model training data, it is input into RNN for model training, and the prediction model of electricity price declaration interval for power supply guarantee type and the prediction model of electricity price declaration interval for cost reduction type are generated respectively;

[0103] The power price declaration strategy coupling model construction unit is used to combine the historical power declaration total amount data contained in the power supply guarantee data set and the cost reduction data set Historical declared price range data Historical electricity deficit data Historical electricity price declaration data group The data is input into SVM as model training data for model training, and a coupling model of power supply guarantee type electricity price declaration strategy and a coupling model of cost reduction type electricity price declaration strategy are generated respectively.

[0104] The target classification module 30 is used to judge the power supply and demand of the power purchase entity to be predicted. If the power purchase entity's internal power generation is less than or equal to the power purchase node of the power consumption, the power purchase entity's power purchase target is classified according to the power gap prediction data;

[0105] Furthermore, in some optional embodiments, the target classification module 30 further includes:

[0106] The supply and demand curve data generation unit is used to generate the power generation forecast curve data of the power purchase entity to be forecasted. Load forecast curve data Subtract to generate supply and demand curve data in,

[0107] Power purchase strategy prediction unit, used for When , the power purchasing strategy of the tth power purchasing node is are all 0, indicating that there is no electricity purchase behavior at this moment; When , the power purchasing strategy of the tth power purchasing node is predicted to obtain the power gap prediction data

[0108]

[0109] The prediction module 40 is used to input the data corresponding to the power purchase node into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and predict the power demand and electricity price declaration interval results of the power purchase node in turn;

[0110] Furthermore, in some optional embodiments, the prediction module 40 further includes:

[0111] The first reporting interval forecast unit is used to report the power shortage forecast data It is necessary to adopt the guaranteed power supply type electricity price declaration interval prediction model, and convert the monthly proxy power purchase price data of the t-th power purchase node into Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input the trained guaranteed power supply type electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th power purchasing node

[0112] The second reporting interval forecast unit is used to report the power shortage forecast data. When the monthly proxy power purchase price data of the tth power purchase node is Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input into the cost-reduction electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th electricity purchasing node

[0113]

[0114] The declaration strategy generation module 50 is used to input the predicted power demand of the power purchasing node and the power price declaration interval results of the power purchasing node as input data to the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, and generate the power price declaration strategy of the power purchasing node;

[0115] Furthermore, in some optional embodiments, the declaration strategy generation module 50 further includes:

[0116] The power purchase strategy generation unit is used to predict the electricity price declaration interval of the tth power purchase node. and power demand Input the trained cost-reduction or power supply guarantee type electricity price coupling declaration strategy model, and output the power purchase strategy of the t-th power purchase node And input the power purchasing strategy into the spot transaction declaration rules for verification.

[0117] A verification module 60 is used to verify the power price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules;

[0118] Furthermore, the inspection module 60 is also used for:

[0119] The verification rules for spot transaction declaration are as follows: When the spot transaction declaration rule verification is not met, the power price coupling declaration strategy model is returned to regenerate the power purchase strategy until the spot transaction declaration rule verification is met.

[0120] The strategy aggregation module 70 is used to aggregate the power price declaration strategies of the power purchasing nodes in time sequence if the spot transaction declaration rule verification is met, and generate the quantity and price declaration prediction results of the inter-provincial spot power purchasing entities.

[0121] Furthermore, the policy aggregation module 70 further includes:

[0122] The strategy aggregation unit is used to aggregate the power purchasing strategies of 96 power purchasing nodes to obtain the power purchasing strategies of competitive power purchasing entities.

[0123] On the other hand, the present application also proposes a readable storage medium on which one or more programs are stored, which, when executed by a processor, implement the above-mentioned method for predicting the quantity and price declaration behavior of inter-provincial spot transactions.

[0124] On the other hand, the present application also proposes a computer device, including a memory and a processor, wherein the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the above-mentioned method for predicting the quantity and price declaration behavior of inter-provincial spot transactions.

[0125] It is known from common technical knowledge that the present application can be implemented by other embodiments that do not deviate from its spirit or essential features. Therefore, the above disclosed embodiments are only illustrative in all respects and are not exclusive. All changes within the scope of the present application or within the scope equivalent to the present application are included in the present application.

[0126] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0127] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0128] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0129] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application rather than to limit it. Although the present application has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present application can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present application should be included in the scope of protection of the claims of the present application.

Claims

1. A method for predicting inter-provincial spot transaction volume and price declaration behavior, characterized in that: The method comprises: Classify the historical data to obtain a power supply guarantee data set and a cost reduction data set according to the classification results; According to the power supply guarantee type data set and the cost reduction type data set, a power supply guarantee type power demand prediction model, a cost reduction type power demand prediction model, a power supply guarantee type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply guarantee type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model are obtained by training; For the prediction of the electricity supply and demand of the electricity purchaser itself, if the electricity purchaser's internal power generation is less than or equal to the power consumption of the electricity purchase node, the electricity purchaser's power purchase target will be classified according to the power gap prediction data; The data corresponding to the power purchase node is input into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and the power demand and electricity price declaration interval results of the power purchase node are predicted in turn; The predicted power demand of the power purchasing node and the power price declaration interval of the power purchasing node are used as input data, and input into the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, so as to generate the power price declaration strategy of the power purchasing node; Verify the electricity price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules; If the spot transaction declaration rule verification is met, the electricity price declaration strategy of the power purchasing node will be aggregated according to the time series to generate the quantity and price declaration forecast results of the inter-provincial spot power purchasing entities.

2. The method for predicting inter-provincial spot transaction volume and price declaration behavior according to claim 1, characterized in that: The step of classifying the historical data to obtain a power supply guarantee type data set and a cost reduction type data set according to the classification result includes: Based on historical electricity shortage data As the basis for classification, When , the power purchasing node is a power supply guarantee type power purchasing node, t represents the power purchasing node 1≤t≤96, t∈Z; when When , the power purchasing node is a cost-reducing power purchasing node; The data under the power supply guarantee type power purchasing node t are all classified into the power supply guarantee type data set, and the data under the cost reduction type power purchasing node t are all classified into the cost reduction type data set.

3. The method for predicting inter-provincial spot transaction volume and price reporting behavior according to claim 2, characterized in that: The step of training a power supply type electricity demand prediction model, a cost reduction type electricity demand prediction model, a power supply type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model according to the power supply type data set and the cost reduction data set comprises: The historical power generation data contained in the power supply guarantee data set and the cost reduction data set are Historical load data Historical channel capacity limit data Historical electricity deficit data Historical electricity total declared data The data is used as model training data and input into RNN for model training to generate a power supply guarantee type power demand forecasting model and a cost reduction type power demand forecasting model respectively; The historical proxy power purchase price data contained in the power supply guarantee data set and the cost reduction data set are Historical temperature data Historical channel capacity limit data Historical clearing price data of major electricity sales provinces Historical declared price range data As the model training data, it is input into RNN for model training, and the prediction model of electricity price declaration interval for power supply guarantee type and the prediction model of electricity price declaration interval for cost reduction type are generated respectively; The historical electricity total amount data contained in the power supply guarantee data set and the cost reduction data set are Historical declared price range data Historical electricity deficit data Historical electricity price declaration data group The data is input into SVM as model training data for model training, and a coupling model of power supply guarantee type electricity price declaration strategy and a coupling model of cost reduction type electricity price declaration strategy are generated respectively.

4. The method for predicting inter-provincial spot transaction volume and price declaration behavior according to claim 3 is characterized in that: The step of classifying the power purchase target of the power purchase entity according to the power gap forecast data includes: The power generation forecast curve data of the power purchasing entity to be predicted Load forecast curve data Subtract to generate supply and demand curve data in, when When , the power purchasing strategy of the tth power purchasing node is are all 0, indicating that there is no electricity purchase behavior at this moment; When , the power purchasing strategy of the tth power purchasing node is predicted to obtain the power gap prediction data 5. The method for predicting inter-provincial spot transaction volume and price declaration behavior according to claim 4 is characterized in that: The step of inputting the data corresponding to the power purchase node into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and predicting the power demand of the power purchase node and the electricity price declaration interval results in turn includes: When the power shortage forecast data It is necessary to adopt the guaranteed power supply type electricity price declaration interval prediction model, and convert the monthly proxy power purchase price data of the t-th power purchase node into Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input the trained guaranteed power supply type electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th power purchasing node When the power shortage forecast data When the monthly proxy power purchase price data of the tth power purchase node is Temperature prediction data Channel capacity limit data Clearing forecast price data for major electricity-selling provinces Input into the cost-reduction electricity price declaration interval prediction model, and output the electricity price declaration interval of the t-th electricity purchasing node 6. The method for predicting inter-provincial spot transaction volume and price declaration behavior according to claim 5, characterized in that: The step of using the predicted power demand of the power purchasing node and the power price declaration interval of the power purchasing node as input data to input into the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, and generating the power price declaration strategy of the power purchasing node includes: The predicted electricity price declaration interval for the tth electricity purchasing node and power demand Input the trained cost-reduction or power supply guarantee type electricity price coupling declaration strategy model, and output the power purchase strategy of the t-th power purchase node And input the power purchasing strategy into the spot transaction declaration rules for verification.

7. The method for predicting inter-provincial spot transaction volume and price declaration behavior according to claim 6 is characterized in that: The step of verifying the power price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules includes: The verification rules for spot transaction declaration are as follows: When the spot transaction declaration rule verification is not met, the power price coupling declaration strategy model is returned to regenerate the power purchase strategy until the spot transaction declaration rule verification is met.

8. The method for predicting inter-provincial spot transaction volume and price declaration behavior according to claim 7, characterized in that: If the spot transaction declaration rule verification is met, the steps of aggregating the power price declaration strategies of the power purchasing nodes according to the time series and generating the quantity and price declaration prediction results of the inter-provincial spot power purchasing entities include: Aggregate the power purchasing strategies of 96 power purchasing nodes to obtain the power purchasing strategies of competitive power purchasing entities 9. A system for predicting the volume and price declaration behavior of inter-provincial spot transactions, characterized in that: The system comprises: A data classification module is used to classify historical data to obtain a power supply guarantee type data set and a cost reduction type data set according to the classification results; A model training module, used for training, according to the power supply guarantee type data set and the cost reduction type data set, obtaining a power supply guarantee type power demand prediction model, a cost reduction type power demand prediction model, a power supply guarantee type electricity price declaration interval prediction model, a cost reduction type electricity price declaration interval prediction model, a power supply guarantee type electricity price declaration strategy coupling model, and a cost reduction type electricity price declaration strategy coupling model; The target classification module is used to analyze and judge the power supply and demand of the power purchase entity itself. If the power purchase entity's internal power generation is less than or equal to the power consumption of the power purchase node, the power purchase entity's power purchase target is classified according to the power gap forecast data; The prediction module is used to input the data corresponding to the power purchase node into the power demand prediction model and the electricity price declaration interval prediction model corresponding to the power purchase target, and predict the power demand and electricity price declaration interval results of the power purchase node in turn; The declaration strategy generation module is used to input the predicted power demand of the power purchasing node and the power price declaration interval results of the power purchasing node as input data to the power price declaration strategy coupling model corresponding to the power purchasing target for prediction, and generate the power price declaration strategy of the power purchasing node; A verification module is used to verify the electricity price declaration strategy of the power purchasing node according to the inter-provincial spot transaction declaration rules; The strategy aggregation module is used to aggregate the electricity price declaration strategy of the power purchasing node in time series if the spot transaction declaration rule verification is met, and generate the quantity and price declaration forecast results of the inter-provincial spot power purchasing entity.

10. A readable storage medium, characterized in that: include: The readable storage medium stores one or more programs, which, when executed by a processor, implement the method for predicting the inter-provincial spot transaction quantity and price declaration behavior as described in any one of claims 1-8.

Citation Information

Cited By

  • A power spot transaction data time series mining method capable of characterizing market fluctuation characteristics

    CN121352906B