Power data processing method and device, electronic equipment and readable storage medium

Abstract

The embodiment of the invention provides a power data processing method and device, electronic equipment and a readable storage medium, and relates to the technical field of computers. According to the embodiment of the invention, the target power data set can be acquired, the basic electricity price is determined based on the non-parameter prediction model, the supply and demand adjustment parameters are determined based on the elastic model, and the basic electricity price is adjusted through the supply and demand adjustment parameters to determine the target electricity price corresponding to the predetermined time node. According to the embodiment of the invention, the semi-parameter structure model at least comprises the non-parameter prediction model and the parameter elastic model, so that by combining the two network models of the semi-parameter structure model, high accuracy and interpretability can be achieved when the target electricity price is determined.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a power data processing method, device, electronic device and readable storage medium. Background Art

[0002] At present, with the increasing demand for electricity, the rationality of power generation planning and power consumption arrangement is becoming more and more important. Among them, electricity price is an important factor affecting the rationality of power generation planning and power consumption arrangement.

[0003] In related technologies, the electricity price at a certain time point in the future is generally predicted by time series prediction methods or machine learning prediction methods. However, since the time series prediction method only has the data of historical electricity prices themselves and does not introduce additional factors, the accuracy of the time series prediction method is low, and the machine learning prediction method is not interpretable. Therefore, how to make the determined electricity price have both high accuracy and interpretability is an urgent problem to be solved. Summary of the invention

[0004] In view of this, embodiments of the present application provide a power data processing method, device, electronic device, and readable storage medium to improve the accuracy and interpretability of determining a target electricity price.

[0005] In a first aspect, a method for processing power data is provided, the method comprising:

[0006] A target power data set is obtained, where the target power data set includes at least historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data.

[0007] The historical electricity price data and the historical electricity supply and demand data are input into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price.

[0008] The historical electricity price data, the historical electricity supply and demand data and the current day electricity supply and demand data are input into the elasticity model in the semi-parametric structural model to determine supply and demand adjustment parameters, wherein the elasticity model is a parameter model, and the elasticity model is used to establish an elastic relationship between electricity price and electricity supply and demand data and to determine the supply and demand adjustment parameters, and the supply and demand adjustment parameters are used to characterize the changes in electricity prices under the influence of the current day electricity supply and demand data.

[0009] The basic electricity price is adjusted by the supply and demand adjustment parameters to determine the target electricity price corresponding to the predetermined time node.

[0010] In some embodiments, the historical electricity supply and demand data includes historical maximum supply and demand values ​​and historical average supply and demand values.

[0011] The inputting of the historical electricity price data, the historical electricity supply and demand data and the current day electricity supply and demand data into the elastic model in the semi-parametric structural model to determine the supply and demand adjustment parameters comprises:

[0012] The historical electricity price data, the historical maximum supply and demand value and the historical average supply and demand value are input into the elasticity model to determine a first intermediate result.

[0013] The historical electricity price data, the historical maximum supply and demand values, and the current day electricity supply and demand data are input into the elasticity model to determine a second intermediate result.

[0014] The difference between the first intermediate result and the second intermediate result is used as the supply and demand adjustment parameter.

[0015] In some embodiments, the method further comprises:

[0016] The target power data set is preprocessed to determine a processing result corresponding to the target power data set, wherein the preprocessing includes at least one of the following: data standardization processing, data cleaning, deletion of abnormal data, supplementation of missing data, and supply and demand data difference processing.

[0017] In some embodiments, the historical electricity supply and demand data includes historical electricity demand data and historical electricity supply data, and the current day electricity supply and demand data includes current day electricity demand data and current day electricity supply data.

[0018] The preprocessing of the target power data set to determine a processing result corresponding to the target power data set includes:

[0019] The supply-demand difference between the historical electricity demand data and the historical electricity supply data, and the supply-demand difference between the current day electricity demand data and the current day electricity supply data are determined.

[0020] Each of the supply-demand differences is determined as a processing result corresponding to the supply-demand data difference processing.

[0021] In some embodiments, the target power data set further includes power usage related data, and the power usage related data includes at least one of weather information, time information, and event information.

[0022] The preprocessing of the target power data set to determine a processing result corresponding to the target power data set includes:

[0023] Perform data standardization processing on the electricity consumption related data, and determine a processing result corresponding to the data standardization processing.

[0024] In some embodiments, inputting the historical electricity price data and the historical electricity supply and demand data into a non-parametric prediction model in a semi-parametric structural model to determine the basic electricity price includes:

[0025] The historical electricity price data, the historical electricity supply and demand data and the electricity consumption related data are input into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price.

[0026] In some embodiments, the historical supply and demand data includes historical load information and historical equipment information, and the current day supply and demand data includes current day load information and current day equipment information.

[0027] In some embodiments, the non-parametric prediction model is constructed based on a distributed gradient boosting library, a deep neural network, a support vector machine, or a random forest.

[0028] In a second aspect, a power data processing device is provided, the device comprising:

[0029] The data set acquisition module is configured to execute acquisition of a target power data set, wherein the target power data set at least includes historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data.

[0030] The basic electricity price determination module is configured to execute a non-parametric prediction model that inputs the historical electricity price data and the historical electricity supply and demand data into a semi-parametric structural model to determine the basic electricity price.

[0031] The supply and demand adjustment parameter determination module is configured to execute the elasticity model that inputs the historical electricity price data, the historical electricity supply and demand data and the current day's electricity supply and demand data into the semi-parametric structural model to determine the supply and demand adjustment parameters, wherein the elasticity model is a parameter model, and the elasticity model is used to establish an elastic relationship between electricity price and electricity supply and demand data and determine the supply and demand adjustment parameters, and the supply and demand adjustment parameters are used to characterize the changes in electricity prices under the influence of the current day's electricity supply and demand data.

[0032] The target electricity price determination module is configured to adjust the basic electricity price through the supply and demand adjustment parameters to determine the target electricity price corresponding to a predetermined time node.

[0033] In some embodiments, the historical electricity supply and demand data includes historical maximum supply and demand values ​​and historical average supply and demand values.

[0034] The supply and demand adjustment parameter determination module is specifically configured to execute:

[0035] The historical electricity price data, the historical maximum supply and demand value and the historical average supply and demand value are input into the elasticity model to determine a first intermediate result.

[0036] The historical electricity price data, the historical maximum supply and demand values, and the current day electricity supply and demand data are input into the elasticity model to determine a second intermediate result.

[0037] The difference between the first intermediate result and the second intermediate result is used as the supply and demand adjustment parameter.

[0038] In some embodiments, the apparatus further comprises:

[0039] The preprocessing module is configured to perform preprocessing on the target power data set and determine the processing result corresponding to the target power data set, wherein the preprocessing includes at least one of the following: data standardization processing, data cleaning, deletion of abnormal data, supplementation of missing data and supply and demand data difference processing.

[0040] In some embodiments, the historical electricity supply and demand data includes historical electricity demand data and historical electricity supply data, and the current day electricity supply and demand data includes current day electricity demand data and current day electricity supply data.

[0041] The preprocessing module is specifically configured to perform:

[0042] The supply-demand difference between the historical electricity demand data and the historical electricity supply data, and the supply-demand difference between the current day electricity demand data and the current day electricity supply data are determined.

[0043] Each of the supply-demand differences is determined as a processing result corresponding to the supply-demand data difference processing.

[0044] In some embodiments, the target power data set further includes power usage related data, and the power usage related data includes at least one of weather information, time information, and event information.

[0045] The preprocessing module is specifically configured to perform:

[0046] Perform data standardization processing on the electricity consumption related data, and determine a processing result corresponding to the data standardization processing.

[0047] In some embodiments, the basic electricity price determination module is specifically configured to execute:

[0048] The historical electricity price data, the historical electricity supply and demand data and the electricity consumption related data are input into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price.

[0049] In some embodiments, the historical supply and demand data includes historical load information and historical equipment information, and the current day supply and demand data includes current day load information and current day equipment information.

[0050] In some embodiments, the non-parametric prediction model is constructed based on a distributed gradient boosting library, a deep neural network, a support vector machine, or a random forest.

[0051] In a third aspect, an embodiment of the present application provides an electronic device, comprising a memory and a processor, wherein the memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method described in the first aspect.

[0052] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the method described in the first aspect is implemented.

[0053] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program / instructions, which, when executed by a processor, implement the method described in the first aspect.

[0054] The semi-parametric structural model of the embodiment of the present application includes at least two parts: a non-parametric prediction model and a parameter elastic model. On the one hand, the embodiment of the present application can determine the basic electricity price based on the non-parametric prediction model to determine a basically accurate electricity price. On the other hand, the embodiment of the present application can establish an elastic relationship between electricity price and electricity supply and demand data based on the elastic model, and determine the supply and demand adjustment parameters according to the elastic relationship. Further, the embodiment of the present application can adjust the basic electricity price through the supply and demand adjustment parameters, determine the target electricity price corresponding to the predetermined time node, and further improve the accuracy of the target electricity price. At the same time, the target electricity price can also be explained by the supply and demand adjustment parameters. Therefore, combined with the two-part network model of the semi-parametric structural model, the embodiment of the present application can have high accuracy and explainability when determining the target electricity price. Further, by having a target electricity price with high accuracy and explainability, the embodiment of the present application can effectively affect the power generation plan and electricity consumption arrangement, thereby achieving a balance between power generation and electricity consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The above and other purposes, features and advantages of the embodiments of the present application will become more apparent through the following description of the embodiments of the present application with reference to the accompanying drawings, in which:

[0056] Figure 1 A flowchart of a power data processing method according to an embodiment of the present application;

[0057] Figure 2 It is a schematic diagram of the principle of the semi-parametric structural model of the embodiment of the present application;

[0058] Figure 3 A flow chart for determining supply and demand adjustment parameters for an embodiment of the present application;

[0059] Figure 4 A flowchart of the preprocessing process of an embodiment of the present application;

[0060] Figure 5 A flowchart of determining a processing result corresponding to the supply and demand data difference processing according to an embodiment of the present application;

[0061] Figure 6 A schematic diagram of the structure of a power data processing device according to an embodiment of the present application;

[0062] Figure 7 A schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0063] The present application is described below based on embodiments, but the present application is not limited to these embodiments. In the detailed description of the present application below, some specific details are described in detail. It is possible for those skilled in the art to fully understand the present application without the description of these details. In order to avoid confusing the essence of the present application, known methods, processes, flows, components and circuits are not described in detail.

[0064] In addition, persons of ordinary skill in the art will appreciate that the drawings provided herein are for illustration purposes and are not necessarily drawn to scale.

[0065] Unless the context clearly requires otherwise, the words "include", "comprising" and the like throughout this application should be interpreted as including rather than exclusive or exhaustive; that is, as meaning "including but not limited to".

[0066] In the description of this application, it should be understood that the terms "first", "second", etc. are used for descriptive purposes only and cannot be understood as indicating or implying relative importance. In addition, in the description of this application, unless otherwise specified, "multiple" means two or more. In addition, the schemes described in this specification and the embodiments, if involving the processing of personal information, will be processed on the premise of having a legal basis (for example, with the consent of the subject of personal information, or necessary for the performance of a contract, etc.), and will only be processed within the prescribed or agreed scope. The user's refusal to process personal information other than the necessary information required for basic functions will not affect the user's use of basic functions.

[0067] As the demand for electricity increases, the rationality of power generation plans and power usage arrangements is becoming more and more important. Among them, electricity price is an important factor affecting the rationality of power generation plans and power usage arrangements. Among them, the electricity price can be the day-ahead node electricity price. By adjusting the day-ahead node electricity price, the power generation plan and power usage arrangement can be affected, thereby achieving the balance between power generation and power usage.

[0068] In related technologies, the electricity price at a certain time point in the future is generally predicted by a time series prediction method or a machine learning prediction method. Among them, the time series prediction method is to make a short-term prediction of the electricity price based on the historical electricity price through a model such as an autoregressive integrated moving average model (ARIMA) or an exponential smoothing model. The machine learning prediction method is to make a short-term prediction of the electricity price through a machine learning model.

[0069] However, the current time series prediction method only has the data of historical electricity prices and does not introduce additional factors. Therefore, the accuracy of the time series prediction method is low, and the machine learning prediction method is not interpretable. Therefore, how to make the determined electricity price have both high accuracy and interpretability is an urgent problem to be solved.

[0070] In order to solve the above problems, an embodiment of the present application provides a power data processing method, which can be applied to electronic devices, wherein the electronic device can be a terminal or a server, the terminal can be a smart phone, a tablet computer or a personal computer (PC), etc., the server can be a single server, a server cluster configured in a distributed manner, or a cloud server.

[0071] Specifically, the power data processing method of the embodiment of the present application can adjust the basic electricity price through the supply and demand adjustment parameters, so as to determine the target electricity price corresponding to the predetermined time node. Among them, the supply and demand adjustment parameters are determined by the electricity supply and demand data. Since the electricity supply and demand data is directly provided by the power grid, it is the most direct, important and accurate factor affecting the electricity price. Moreover, there is a direct causal relationship between the electricity supply and demand data and the electricity price. Therefore, by adjusting the basic electricity price through the supply and demand adjustment parameters, a target electricity price with high accuracy and interpretability can be determined.

[0072] like Figure 1 As shown, the power data processing method of the embodiment of the present application may include the following steps:

[0073] In step S110 , a target power data set is acquired.

[0074] Among them, the target power data set includes at least historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data. Historical electricity price data can be used to characterize the node electricity price in the past predetermined time period. Among them, the predetermined time period can be set according to actual conditions, for example, the predetermined time period can be the past week, month, two months, etc. Historical electricity supply and demand data can be used to characterize the electricity supply and demand information published in the past predetermined time period, and the current day electricity supply and demand data can be used to characterize the electricity supply and demand information published on the same day, wherein the above-mentioned electricity supply and demand information is information that can be legally obtained. For example, historical electricity supply and demand data can be unified dispatching load forecast information, Class A energy, local energy output, West-to-East power transmission, Guangdong-Hong Kong interconnection line power, unit maintenance forecast information, must-start and must-stop unit forecast information, transmission and transformation maintenance forecast information, etc., published in the past week, month, or two months. By adding electricity supply and demand data (historical electricity supply and demand data and current day electricity supply and demand data) to the target power data set, the electricity supply and demand data can affect the final electricity price, and the final electricity price can be explained based on the impact of the electricity supply and demand data. At the same time, by adding electricity supply and demand data to the target power data set, the embodiment of the present application can also increase the data dimension for determining the electricity price, thereby further improving the accuracy of determining the electricity price.

[0075] In step S120, historical electricity price data and historical electricity supply and demand data are input into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price.

[0076] Among them, the semi-parametric structural model may include at least a non-parametric prediction model and an elastic model, the non-parametric prediction model may be a network model constructed based on a machine learning model, and the elastic model may be a parameter model. The non-parametric prediction model may be based at least on historical data (i.e., historical electricity price data and historical electricity supply and demand data) to determine the electricity price (i.e., the basic electricity price) predicted based on historical data. In other words, the basic electricity price of the embodiment of the present application can be used to characterize the electricity price under the basic supply and demand information, and the basic supply and demand information includes at least historical electricity supply and demand data.

[0077] In step S130, historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data are input into an elastic model in a semi-parametric structural model to determine supply and demand adjustment parameters.

[0078] Among them, the elastic model of the embodiment of the present application is a parameter model, which can be used to establish an elastic relationship between electricity prices and electricity supply and demand data and determine supply and demand adjustment parameters, which are used to characterize the changes in electricity prices under the influence of the electricity supply and demand data of the day. Combined with the above-mentioned non-parametric prediction model, since the basic electricity price represents the electricity price under the basic supply and demand information, and the basic supply and demand information does not include the supply and demand information of the day, the embodiment of the present application can process the electricity supply and demand data of the day through a parameter model (that is, an elastic model), so that the supply and demand information of the day can affect the basic electricity price, thereby improving the accuracy and interpretability of the target electricity price.

[0079] That is to say, the elastic model can use the elastic relationship between electricity prices and electricity supply and demand data as a basis to determine the impact of electricity supply and demand data on electricity prices. For example, if electricity demand is positively correlated with electricity prices, then when electricity demand increases, electricity prices will also increase. Furthermore, the supply and demand adjustment parameters at this time can adjust the basic electricity price to a higher value. Therefore, through the impact of electricity supply and demand data, the embodiments of the present application can make the final electricity price explainable.

[0080] For the semi-parametric structural model of the embodiment of the present application, Figure 2 As shown, the semi-parametric structural model 21 of the embodiment of the present application may include a non-parametric prediction model 211 and an elastic model 212. Further, after determining the target power data set 22, the embodiment of the present application may input the data in the target power data set 22 into the non-parametric prediction model 211 and the elastic model 212 in the semi-parametric structural model 21, respectively, and obtain the basic electricity price output by the non-parametric prediction model 211 and the supply and demand adjustment parameters output by the elastic model 212. Further, the semi-parametric structural model 21 may adjust the basic electricity price by the supply and demand adjustment parameters, thereby outputting a target electricity price 23.

[0081] It needs to be explained by Figure 2 It can be seen that since the non-parametric prediction model 211 and the elastic model 212 perform power data processing respectively, there is no fixed execution order between step S120 and step S130 in the embodiment of the present application. That is to say, the embodiment of the present application can execute step S120 first and then step S130, or can execute step S130 first and then step S120, or can execute steps S120 and S130 at the same time.

[0082] In step S140, the basic electricity price is adjusted by the supply and demand adjustment parameters to determine the target electricity price corresponding to the predetermined time node.

[0083] The predetermined time node may be a time node in the future. For example, the predetermined time node may be used to represent a time node such as the second day or the third day. The closer the predetermined time node is to the current day, the higher the accuracy of the target electricity price determined by the semi-parametric structural model.

[0084] That is to say, the semi-parametric structural model of the embodiment of the present application includes at least two parts: a non-parametric prediction model and a parameter elastic model. On the one hand, the embodiment of the present application can determine the basic electricity price based on the non-parametric prediction model to determine a basically accurate electricity price. On the other hand, the embodiment of the present application can establish an elastic relationship between electricity price and electricity supply and demand data based on the elastic model, and determine the supply and demand adjustment parameters according to the elastic relationship. Further, the embodiment of the present application can adjust the basic electricity price through the supply and demand adjustment parameters, determine the target electricity price corresponding to the predetermined time node, and further improve the accuracy of the target electricity price. At the same time, the target electricity price can also be explained by adjusting the supply and demand parameters. Therefore, combined with the two-part network model of the semi-parametric structural model, the embodiment of the present application can have high accuracy and explainability when determining the target electricity price. Further, by having a target electricity price with high accuracy and explainability, the embodiment of the present application can effectively affect the power generation plan and electricity consumption arrangement, thereby achieving a balance between power generation and electricity consumption.

[0085] In an optional implementation, the historical electricity supply and demand data of the embodiment of the present application may include a historical maximum supply and demand value and a historical average supply and demand value.

[0086] Further, such as Figure 3 As shown, the above step S130 may include the following steps:

[0087] In step S131, historical electricity price data, historical maximum supply and demand values, and historical average supply and demand values ​​are input into the elasticity model to determine a first intermediate result.

[0088] The first intermediate result is an intermediate result determined by the elastic model based on historical data (i.e., historical electricity price data, historical maximum supply and demand values, and historical supply and demand average values). The first intermediate result can be used to characterize the electricity price under the influence of the historical supply and demand average values. Specifically, the process of determining the first intermediate result can be characterized by the following formula:

[0089] g(Q0)=E[ln(P / P max )]=αln(Q0 / Q max )+c

[0090] Among them, g(Q0) is used to represent the first intermediate result, Q0 is used to represent the historical supply and demand average, Q max Used to represent the maximum value of historical supply and demand, P maxIt is used to characterize the maximum value of the electricity price data within a predetermined time period in the past, P is used to characterize the target electricity price, and α and c are parameters learned from actual data in the elastic model. This makes the difference between the parameter model (i.e., the elastic model) and the non-parametric model in that a priori fitting form is set, and this parameter can be used to adjust the results of the elastic model. It can be seen from the above formula that the elastic model of the embodiment of the present application can be a nested model of a double log structure.

[0091] In step S132, historical electricity price data, historical maximum supply and demand values, and current day electricity supply and demand data are input into the elasticity model to determine a second intermediate result.

[0092] The second intermediate result is an intermediate result determined by the elastic model based on historical data (i.e., historical electricity price data and historical supply and demand maximums) and the electricity supply and demand data of the day. The second intermediate result can be used to characterize the electricity price under the influence of the electricity supply and demand data of the day. Specifically, the process of determining the second intermediate result can be characterized by the following formula:

[0093] g(Q)=E[ln(P / P max )] = αln(Q / Q max )+c

[0094] Among them, g(Q) is used to represent the second intermediate result, and Q is used to represent the electricity supply and demand data of the day.

[0095] In step S133, the difference between the first intermediate result and the second intermediate result is used as a supply and demand adjustment parameter.

[0096] It should be noted that there is no fixed execution order between step S131 and step S132 in the embodiment of the present application. That is to say, the embodiment of the present application can execute step S131 first and then step S132, or can execute step S132 first and then step S131, or can execute steps S131 and S132 at the same time.

[0097] Since the first intermediate result represents the electricity price under the influence of the historical supply and demand average, and the second intermediate result represents the electricity price under the influence of the electricity supply and demand data of the day, the difference between the two can be used to represent the increase (or decrease) in the impact of the electricity supply and demand data of the day compared to the historical supply and demand average, and the increase (or decrease) in the impact can be used to explain the reason for the increase or decrease in the basic electricity price. Furthermore, by adjusting the basic electricity price by the increase (or decrease) in the impact, the final target electricity price can be made more accurate.

[0098] In an optional implementation, the historical supply and demand data of the embodiment of the present application may include historical load information and historical equipment information, and the current day supply and demand data may include current day load information and current day equipment information. Specifically, the load information (i.e., historical load information and current day load information) may include unified load forecast information, Class A energy, local energy output, West-to-East power transmission, Guangdong-Hong Kong interconnection line power, etc. The equipment information (i.e., historical equipment information and current day equipment information) may include unit maintenance forecast information, must-start or must-stop unit forecast information, transmission and transformation maintenance forecast information, etc.

[0099] Furthermore, historical supply and demand data can be used in non-parametric forecasting models and elastic models to improve the accuracy of determining electricity prices. Daily supply and demand data can be used in elastic models so that the elastic model can output explainable supply and demand adjustment parameters. Combined with the two-part network model of the semi-parametric structural model, the embodiment of the present application can have both high accuracy and explainability when determining the target electricity price.

[0100] In an optional implementation, the target power data set of the embodiment of the present application may also include power consumption related data. Among them, the power consumption related data may include at least one of historical and / or current weather information, time information, and event information. For example, weather information may include information such as weather type and wind speed, time information may include statutory holiday information and rest day information (such as whether it is a weekend or not), and event information may include accident disaster information and natural disaster information. In addition, the power consumption related data of the embodiment of the present application may also include other information, such as thermal coal information, power consumption policy information, etc.

[0101] Furthermore, in combination with the above-mentioned first intermediate result, the second intermediate result and the corresponding formulas, the embodiment of the present application inputs the historical electricity price data, the historical electricity supply and demand data and the electricity-related data into the non-parametric prediction model in the semi-parametric structural model to determine the basic electricity price. Specifically, the process of determining the basic electricity price can be characterized by the following formula:

[0102] h(Q0,x)=E[ln(P / P max )|Q0]

[0103] Among them, h(Q0,x) is used to represent the basic electricity price, and x is used to represent the historical and / or current electricity consumption related data (such as weather information, time information or event information, etc.) in the target power data set.

[0104] It can be seen from the formula corresponding to the basic electricity price that the embodiment of the present application can determine the basic electricity price under the influence of the historical supply and demand average and electricity consumption related data. It should be noted that the embodiment of the present application can also determine the basic electricity price only based on the historical supply and demand average to reduce the amount of calculation and improve the efficiency of power data processing.

[0105] Furthermore, the process of determining the target electricity price in the embodiment of the present application can be represented by the following formula:

[0106] E[ln(P / P max )|Q]=g(Q)+h(Q0,x)-g(Q0)

[0107] Among them, P is used to characterize the target electricity price, and g(Q)-g(Q0) is the supply and demand adjustment parameter of the embodiment of the present application. In combination with the above formulas, it can be seen that the embodiment of the present application can, on the one hand, determine the basic electricity price based on the non-parametric prediction model to determine a basically accurate electricity price. On the other hand, the embodiment of the present application can establish an elastic relationship between electricity price and electricity supply and demand data based on the elastic model, and determine the supply and demand adjustment parameters according to the elastic relationship. Furthermore, the embodiment of the present application can adjust the basic electricity price through the supply and demand adjustment parameters to determine the target electricity price corresponding to the predetermined time node. While further improving the accuracy of the target electricity price, the target electricity price can also be made explainable through the supply and demand adjustment parameters.

[0108] In an optional embodiment, the non-parametric prediction model of the embodiment of the present application can be constructed based on a distributed gradient boosting library (eXtreme Gradient Boosting, XGBoost), a deep neural network (Deep Neural Networks, DNN), a support vector machine (SVM) or a random forest. Among them, XGBoost implements a machine learning algorithm under the Gradient Boosting framework, which can provide parallel tree boosting, can quickly and accurately solve many data science problems, and has the advantages of high efficiency, flexibility and portability. DNN is a technology in the field of machine learning, a framework for deep learning, and a neural network with at least one hidden layer. DNN can provide modeling for complex nonlinear systems, and the extra levels can provide a higher level of abstraction for the model, thereby improving the model's capabilities. SVM is a generalized linear classifier that performs binary classification of data in a supervised learning manner, and its decision boundary is the maximum margin hyperplane for solving learning samples. SVM uses a hinge loss function to calculate empirical risk and adds a regularization term to the solution system to optimize structural risk. It is a classifier with sparsity and robustness. Random forest is a classifier that contains multiple decision trees, and its output category is determined by the majority of the categories output by individual trees. It has the advantages of strong model readability and fast classification speed.

[0109] That is to say, the non-parametric prediction model of the embodiment of the present application can be a network model constructed based on the above-mentioned machine learning models, which can use historical electricity price data and historical electricity supply and demand data (such as historical supply and demand averages) as input parameters of the model, so that the non-parametric prediction model predicts a basically accurate basic electricity price based on the above-mentioned input parameters. In addition, the embodiment of the present application can also add electricity-related data to the input parameters of the model, so that the input parameters of the non-parametric prediction model can add dimensions such as weather, time, and events, thereby further improving the accuracy of the basic electricity price. Moreover, the embodiment of the present application can construct a training set based on historical data to train the non-parametric prediction model, and deploy the trained non-parametric prediction model to a semi-parametric structural model for determining the basic electricity price.

[0110] In an optional implementation, the embodiment of the present application may also pre-process the target power data set, specifically, Figure 4 As shown, the process may include the following steps:

[0111] In step S210, the target power data set is preprocessed to determine a processing result corresponding to the target power data set.

[0112] Among them, the preprocessing of the embodiment of the present application may include at least one of the following: data standardization, data cleaning, deletion of abnormal data, supplementation of missing data and supply and demand data difference processing. Specifically, data standardization can unify the various data in the target power data set into data at the same latitude to facilitate the determination of the target electricity price. Data cleaning can remove invalid data in the target power data set (such as invalid data such as garbled characters and data without practical meaning such as punctuation marks). Deleting abnormal data can delete data that deviates too much from the normal value in the target power data set to avoid interference from abnormal data. Supplementing missing data can fill in the missing data in the target power data set to ensure that the process of determining the target electricity price proceeds smoothly. Supply and demand data difference processing can determine the difference between demand data and supply data in the power supply and demand data, and use the difference as input data for the semi-parametric structural model. By adding power supply and demand data to the target power data set, the embodiment of the present application can increase the data dimension for determining the electricity price, thereby further improving the accuracy of determining the electricity price.

[0113] Furthermore, after preprocessing, the embodiment of the present application can input the processing result corresponding to the target power data set into the semi-parametric structural model to determine the target electricity price corresponding to the predetermined time node. Combining the two-part network model of the semi-parametric structural model, the embodiment of the present application can have high accuracy and interpretability when determining the target electricity price.

[0114] Specifically, in an optional implementation, the historical electricity supply and demand data of the embodiment of the present application may include historical electricity demand data and historical electricity supply data, and the current day electricity supply and demand data may include current day electricity demand data and current day electricity supply data. Among them, the electricity demand data (i.e., historical electricity demand data and current day electricity demand data) is used to characterize the data of electricity consumed by the party that consumes electricity, and the electricity supply data (i.e., historical electricity supply data and current day electricity supply data) is used to characterize the data of electricity generated by the party that generates electricity.

[0115] Further, such as Figure 5 As shown, the above step S210 may include the following steps:

[0116] In step S211, the supply-demand difference between the historical power demand data and the historical power supply data, and the supply-demand difference between the current day power demand data and the current day power supply data are determined.

[0117] In step S212, each supply-demand difference is determined as a processing result corresponding to the supply-demand data difference processing.

[0118] It should be noted that after determining the supply-demand difference between the electricity demand data and the electricity supply data, the embodiment of the present application can further perform data standardization on the supply-demand difference to facilitate determination of the target electricity price. Furthermore, by adding electricity supply and demand data to the target power data set, the embodiment of the present application can increase the data dimension for determining the electricity price, thereby further improving the accuracy of determining the electricity price.

[0119] Furthermore, after preprocessing, the embodiment of the present application can input the processing results corresponding to the target power data set (including the processing results corresponding to the supply and demand data difference processing) into the semi-parametric structural model to determine the target electricity price corresponding to the predetermined time node. Combining the two-part network model of the semi-parametric structural model, the embodiment of the present application can have high accuracy and interpretability when determining the target electricity price.

[0120] In an optional implementation, if the target power data set also includes electricity consumption related data, the embodiment of the present application may also pre-process the electricity consumption related data. Specifically, the embodiment of the present application may perform data standardization processing on the electricity consumption related data and determine the processing results corresponding to the data standardization processing.

[0121] Specifically, the embodiment of the present application can convert weather information, time information, event information or other information mentioned above into standard parameters (such as parameters such as "00" or "01") to determine the processing result corresponding to the data standardization processing. The embodiment of the present application can increase the data dimension for determining the electricity price by adding electricity-related data to the target power data set, thereby further improving the accuracy of determining the electricity price.

[0122] Further, the embodiment of the present application can input the preprocessed target power data set into a pre-built semi-parametric structural model to determine the target electricity price corresponding to the predetermined time node. Specifically, the embodiment of the present application can determine the basic electricity price based on the non-parametric prediction model on the one hand to determine a basically accurate electricity price. On the other hand, the embodiment of the present application can establish an elastic relationship between the electricity price and the electricity supply and demand data based on the elastic model, and determine the supply and demand adjustment parameters according to the elastic relationship. Further, the embodiment of the present application can adjust the basic electricity price through the supply and demand adjustment parameters to determine the target electricity price corresponding to the predetermined time node, and while further improving the accuracy of the target electricity price, the target electricity price can also be explained by the supply and demand adjustment parameters. Therefore, combined with the two-part network model of the semi-parametric structural model, the embodiment of the present application can have high accuracy and explainability when determining the target electricity price. Further, by having a target electricity price with high accuracy and explainability, the embodiment of the present application can effectively affect the power generation plan and power consumption arrangement, thereby achieving a balance between power generation and power consumption.

[0123] Based on the same technical concept, the embodiment of the present application also provides a power data processing device, such as Figure 6 As shown, the device includes: a data set acquisition module 61, a basic electricity price determination module 62, a supply and demand adjustment parameter determination module 63 and a target electricity price determination module 64.

[0124] The data set acquisition module 61 is configured to execute acquisition of a target power data set, wherein the target power data set at least includes historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data.

[0125] The basic electricity price determination module 62 is configured to execute a non-parametric prediction model that inputs the historical electricity price data and the historical electricity supply and demand data into a semi-parametric structural model to determine the basic electricity price.

[0126] The supply and demand adjustment parameter determination module 63 is configured to execute the elasticity model that inputs the historical electricity price data, the historical electricity supply and demand data and the current day's electricity supply and demand data into the semi-parametric structural model to determine the supply and demand adjustment parameters, wherein the elasticity model is a parameter model, and the elasticity model is used to establish an elastic relationship between electricity price and electricity supply and demand data and determine the supply and demand adjustment parameters, and the supply and demand adjustment parameters are used to characterize the changes in electricity prices under the influence of the current day's electricity supply and demand data.

[0127] The target electricity price determination module 64 is configured to adjust the basic electricity price by using the supply and demand adjustment parameters to determine the target electricity price corresponding to a predetermined time node.

[0128] In some embodiments, the historical electricity supply and demand data includes historical maximum supply and demand values ​​and historical average supply and demand values.

[0129] The supply and demand adjustment parameter determination module 63 is specifically configured to execute:

[0130] The historical electricity price data, the historical maximum supply and demand value and the historical average supply and demand value are input into the elasticity model to determine a first intermediate result.

[0131] The historical electricity price data, the historical maximum supply and demand values, and the current day electricity supply and demand data are input into the elasticity model to determine a second intermediate result.

[0132] The difference between the first intermediate result and the second intermediate result is used as the supply and demand adjustment parameter.

[0133] In some embodiments, the apparatus further comprises:

[0134] The preprocessing module is configured to perform preprocessing on the target power data set and determine the processing result corresponding to the target power data set, wherein the preprocessing includes at least one of the following: data standardization processing, data cleaning, deletion of abnormal data, supplementation of missing data and supply and demand data difference processing.

[0135] In some embodiments, the historical electricity supply and demand data includes historical electricity demand data and historical electricity supply data, and the current day electricity supply and demand data includes current day electricity demand data and current day electricity supply data.

[0136] The preprocessing module is specifically configured to perform:

[0137] The supply-demand difference between the historical electricity demand data and the historical electricity supply data, and the supply-demand difference between the current day electricity demand data and the current day electricity supply data are determined.

[0138] Each of the supply-demand differences is determined as a processing result corresponding to the supply-demand data difference processing.

[0139] In some embodiments, the target power data set further includes power usage related data, and the power usage related data includes at least one of weather information, time information, and event information.

[0140] The preprocessing module is specifically configured to perform:

[0141] Perform data standardization processing on the electricity consumption related data, and determine a processing result corresponding to the data standardization processing.

[0142] In some embodiments, the basic electricity price determination module 62 is specifically configured to execute:

[0143] The historical electricity price data, the historical electricity supply and demand data and the electricity consumption related data are input into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price.

[0144] In some embodiments, the historical supply and demand data includes historical load information and historical equipment information, and the current day supply and demand data includes current day load information and current day equipment information.

[0145] In some embodiments, the non-parametric prediction model is constructed based on a distributed gradient boosting library, a deep neural network, a support vector machine, or a random forest.

[0146] The semi-parametric structural model of the embodiment of the present application includes at least two parts: a non-parametric prediction model and a parameter elastic model. On the one hand, the embodiment of the present application can determine the basic electricity price based on the non-parametric prediction model to determine a basically accurate electricity price. On the other hand, the embodiment of the present application can establish an elastic relationship between electricity price and electricity supply and demand data based on the elastic model, and determine the supply and demand adjustment parameters according to the elastic relationship. Further, the embodiment of the present application can adjust the basic electricity price through the supply and demand adjustment parameters, determine the target electricity price corresponding to the predetermined time node, and further improve the accuracy of the target electricity price. At the same time, the target electricity price can also be explained by the supply and demand adjustment parameters. Therefore, combined with the two-part network model of the semi-parametric structural model, the embodiment of the present application can have high accuracy and explainability when determining the target electricity price. Further, by having a target electricity price with high accuracy and explainability, the embodiment of the present application can effectively affect the power generation plan and electricity consumption arrangement, thereby achieving a balance between power generation and electricity consumption.

[0147] Figure 7 Schematic diagram of an electronic device according to an embodiment of the present application. Figure 7 As shown, Figure 7The electronic device shown is a general address query device, which includes a general computer hardware structure, which at least includes a processor 71 and a memory 72. The processor 71 and the memory 72 are connected via a bus 73. The memory 72 is suitable for storing instructions or programs executable by the processor 71. The processor 71 can be an independent microprocessor or a collection of one or more microprocessors. Thus, the processor 71 executes the instructions stored in the memory 72, thereby executing the method flow of the embodiment of the present application as described above to realize the processing of data and the control of other devices. The bus 73 connects the above-mentioned multiple components together, and at the same time connects the above-mentioned components to the display controller 74 and the display device and the input / output (I / O) device 75. The input / output (I / O) device 75 can be a mouse, a keyboard, a modem, a network interface, a touch input device, a somatosensory input device, a printer, and other devices known in the art. Typically, the input / output device 75 is connected to the system via an input / output (I / O) controller 76.

[0148] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, devices (equipment) or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present application may adopt a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0149] The present application is described with reference to flowcharts of methods, apparatuses (devices) and computer program products according to embodiments of the present application. It should be understood that each process in the flowchart can be implemented by computer program instructions.

[0150] These computer program instructions may 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 including an instruction device that implements the process Figure 1 A function specified in a process or multiple processes.

[0151] These computer program instructions may also be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce the instructions for implementing the process Figure 1 A device that specifies functions in a process or multiple processes.

[0152] Another embodiment of the present application relates to a non-volatile storage medium for storing a computer-readable program, wherein the computer-readable program is used for a computer to execute part or all of the above method embodiments.

[0153] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by specifying relevant hardware through a program, and the program is stored in a storage medium, including several instructions for a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0154] Another embodiment of the present application relates to a computer program product, including a computer program / instruction, which can implement part or all of the above method embodiments when the computer program / instruction is executed by a processor.

[0155] That is, those skilled in the art can understand that the embodiments of the present application can specify relevant hardware (including the processor itself) through the processor executing a computer program product (computer program / instructions), thereby implementing all or part of the steps in the above-mentioned embodiment method.

[0156] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for processing power data, characterized in that: The method comprises: Acquire a target power data set, wherein the target power data set includes at least historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data; Inputting the historical electricity price data and the historical electricity supply and demand data into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price; Inputting the historical electricity price data, the historical electricity supply and demand data and the current day electricity supply and demand data into the elastic model in the semi-parametric structural model to determine supply and demand adjustment parameters, wherein the elastic model is a parameter model, the elastic model is used to establish an elastic relationship between electricity price and electricity supply and demand data and determine the supply and demand adjustment parameters, and the supply and demand adjustment parameters are used to characterize the change in electricity price under the influence of the current day electricity supply and demand data; and The basic electricity price is adjusted by the supply and demand adjustment parameters to determine the target electricity price corresponding to the predetermined time node.

2. The method according to claim 1, characterized in that The historical electricity supply and demand data include the historical maximum supply and demand and the historical average supply and demand; The inputting of the historical electricity price data, the historical electricity supply and demand data and the current day electricity supply and demand data into the elastic model in the semi-parametric structural model to determine the supply and demand adjustment parameters comprises: Inputting the historical electricity price data, the historical maximum supply and demand value and the historical average supply and demand into the elasticity model to determine a first intermediate result; Inputting the historical electricity price data, the historical maximum supply and demand values, and the current day electricity supply and demand data into the elasticity model to determine a second intermediate result; and The difference between the first intermediate result and the second intermediate result is used as the supply and demand adjustment parameter.

3. The method according to claim 1, characterized in that The method further comprises: The target power data set is preprocessed to determine a processing result corresponding to the target power data set, wherein the preprocessing includes at least one of the following: data standardization processing, data cleaning, deletion of abnormal data, supplementation of missing data, and supply and demand data difference processing.

4. The method according to claim 3, characterized in that: The historical electricity supply and demand data includes historical electricity demand data and historical electricity supply data, and the current day electricity supply and demand data includes current day electricity demand data and current day electricity supply data; The preprocessing of the target power data set to determine a processing result corresponding to the target power data set includes: Determining the supply-demand difference between the historical electricity demand data and the historical electricity supply data, and the supply-demand difference between the current day electricity demand data and the current day electricity supply data; as well as Each of the supply-demand differences is determined as a processing result corresponding to the supply-demand data difference processing.

5. The method according to claim 3, characterized in that: The target power data set further includes power consumption related data, and the power consumption related data includes at least one of weather information, time information and event information; The preprocessing of the target power data set to determine a processing result corresponding to the target power data set includes: Perform data standardization processing on the electricity consumption related data, and determine a processing result corresponding to the data standardization processing.

6. The method according to claim 5, characterized in that The inputting of the historical electricity price data and the historical electricity supply and demand data into the non-parametric prediction model in the semi-parametric structural model to determine the basic electricity price includes: The historical electricity price data, the historical electricity supply and demand data and the electricity consumption related data are input into a non-parametric prediction model in a semi-parametric structural model to determine a basic electricity price.

7. The method according to claim 1, characterized in that The historical supply and demand data includes historical load information and historical equipment information, and the current day supply and demand data includes current day load information and current day equipment information.

8. The method according to claim 1, characterized in that The non-parametric prediction model is constructed based on a distributed gradient boosting library, a deep neural network, a support vector machine or a random forest.

9. A power data processing device, characterized in that: The device comprises: A data set acquisition module is configured to execute acquisition of a target power data set, wherein the target power data set at least includes historical electricity price data, historical electricity supply and demand data, and current day electricity supply and demand data; A basic electricity price determination module is configured to execute a non-parametric prediction model that inputs the historical electricity price data and the historical electricity supply and demand data into a semi-parametric structural model to determine a basic electricity price; a supply and demand adjustment parameter determination module, configured to execute an elastic model that inputs the historical electricity price data, the historical electricity supply and demand data, and the current day electricity supply and demand data into the semi-parametric structural model to determine supply and demand adjustment parameters, wherein the elastic model is a parameter model, the elastic model is used to establish an elastic relationship between electricity price and electricity supply and demand data and determine the supply and demand adjustment parameters, and the supply and demand adjustment parameters are used to characterize changes in electricity prices under the influence of the current day electricity supply and demand data; and The target electricity price determination module is configured to adjust the basic electricity price through the supply and demand adjustment parameters to determine the target electricity price corresponding to a predetermined time node.

10. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

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

Images