Day-ahead and intraday two-stage regulation based dynamic price generation method and system

Abstract

The application relates to the technical field of optimization of power distribution networks and demand response, and particularly discloses a dynamic electricity price generation method and system based on two-stage regulation in a day-ahead and day-in phase, which comprises the following steps: in the day-ahead phase, power and voltage prediction values are obtained by prediction based on historical voltage and power data and meteorological data; voltage risk index values and photovoltaic consumption index values are calculated based on the power and voltage prediction values, preliminary pricing of demand response is carried out, and coarse-granularity interval electricity prices are generated; in the day-in phase, current actual photovoltaic consumption gaps are obtained by analysis based on voltage data and load power data; the coarse-granularity interval electricity prices are corrected based on the current actual photovoltaic consumption gaps, and current real-time electricity prices are obtained. According to the characteristic tracking analysis of the power and voltage lifting curve, the method is mapped to a price adjustment parameter model and forms real-time prices, users change the size of the load according to the real-time prices, demand response is executed, and photovoltaic in the power distribution network is consumed on site.

Description

Technical Field

[0001] This invention relates to the field of distribution network optimization and demand response technology, specifically to a dynamic electricity price generation method and system based on day-ahead and intraday two-stage regulation. Background Technology

[0002] With the widespread application of distributed photovoltaic (PV) power generation in distribution areas, the intermittency and volatility of its output pose challenges to voltage and load balance. Increased PV output leads to significant voltage spikes at nodes within the distribution area, potentially causing voltage exceedances and impacting grid safety. Traditional regulation methods, such as transformer tap adjustment or reactive power compensation device switching, suffer from slow speed, low accuracy, and high cost, making them ill-suited to adapting to real-time changes in PV output. Demand response, a key smart grid technology, effectively incentivizes users to adjust their electricity consumption to meet grid demands; however, common incentive-based demand response relies on substantial subsidies, resulting in high costs and unsustainability. Price-based demand response, on the other hand, leverages electricity price signals to guide users in flexibly changing their load. Therefore, a price-based demand response method based on real-time PV output and voltage variations in distribution areas is urgently needed to achieve local PV consumption and ensure stable voltage and reliable grid operation. Summary of the Invention

[0003] To address the aforementioned problems, the first objective of this invention is to provide a dynamic electricity price generation method based on day-ahead and intraday two-stage regulation. This method maps the characteristics of power and voltage boost curves to a price adjustment parameter model to form a real-time price. Users adjust the load size according to the real-time price to complete demand response execution, thereby consuming photovoltaic power in the local area.

[0004] The second objective of this invention is to provide a dynamic electricity price generation system based on two-stage regulation between day and day.

[0005] The first technical solution adopted in this invention is: a dynamic electricity price generation method based on two-stage day-ahead and intraday regulation, comprising the following steps: S100: During the day-ahead phase, acquire historical voltage and power data and meteorological data of transformer outlets and important nodes in any region; make predictions based on the historical voltage and power data and meteorological data to obtain predicted values ​​of power and voltage; calculate voltage risk index value and photovoltaic absorption index value based on the predicted values ​​of power and voltage; and perform preliminary demand response pricing based on the voltage risk index value and photovoltaic absorption index value to generate coarse-grained range electricity prices. S200: During the daytime phase, real-time voltage and load power data of transformer outlets and user-responsive resource nodes in the region are collected, and the current actual photovoltaic absorption gap is obtained based on the voltage and load power data; and the coarse-grained range electricity price is corrected based on the current actual photovoltaic absorption gap to obtain the current real-time electricity price.

[0006] Preferably, step S100, which involves making a prediction based on the historical voltage and power data and meteorological data, includes: A Transformer deep spatiotemporal prediction model is constructed. The historical voltage and power data and meteorological data are input into the Transformer deep spatiotemporal prediction model to obtain power prediction values ​​and voltage prediction values.

[0007] Preferably, the construction of the Transformer deep spatiotemporal prediction model includes: acquiring historical voltage and power data and meteorological data to generate a training set; and training the Transformer model based on the training set to obtain the Transformer deep spatiotemporal prediction model.

[0008] Preferably, the calculation of voltage risk index value and photovoltaic power consumption index value in step S100 includes: The voltage risk index and photovoltaic absorption index are evaluated based on the predicted values ​​of power and voltage to obtain the voltage risk index value and photovoltaic absorption index value.

[0009] Preferably, step S100 includes: generating a coarse-grained range electricity price based on the voltage risk index value and the photovoltaic absorption index value through a user response uncertainty model.

[0010] Preferably, the coarse-grained range electricity price in step S100 is expressed by the following formula: In the formula, This refers to the current day's coarse-grained range electricity price unit group; For the 96 elements corresponding to the daily range electricity price; for t The price ranges of each level within the 15-minute time period at which the time is located belong to the set of price ranges corresponding to each level. These are the upper and lower limits of the price range corresponding to Level 1; Levels The upper and lower limits of the corresponding price range; The highest level The corresponding price range.

[0011] Preferably, step S200 includes: Users execute response actions based on the midpoint of the coarse-grained range electricity price, and collect voltage and load power data of transformer outlets and user-responsive resource nodes in the area in real time. The voltage and load power data are tracked and analyzed to obtain the power of each node and the output power of each photovoltaic power generation device; and the current actual photovoltaic consumption gap is obtained based on the power of each node and the output power of each photovoltaic power generation device.

[0012] Preferably, in step S200, the coarse-grained range electricity price is corrected based on the current actual photovoltaic absorption gap using the following formula to obtain the current real-time electricity price: In the formula, for t The DR price responded by the user at the start of the corresponding 15-minute time period; Levels The upper and lower limits of the corresponding price range; For users based The current actual photovoltaic consumption gap after the implementation of response actions; for t Time Node The power; for t Photovoltaic power generation devices at all times j ; output power; n , M These are the number of nodes and the number of photovoltaic power generation devices, respectively. This refers to the current real-time electricity price; The actual value is determined based on the current size of the actual photovoltaic absorption gap within the range. Generate proportionally; These represent the lower limit of the price range corresponding to level h-1 and the upper limit of the price range corresponding to level h+1, respectively.

[0013] Preferably, step S200 further includes: acquiring data on the user's response actions based on the real-time electricity price, and calculating the update criterion value of the Transformer deep spatiotemporal prediction model and the user response uncertainty model based on the data on the user's response actions based on the real-time electricity price; And based on the update criterion value of the Transformer deep spatiotemporal prediction model and the user response uncertainty model, determine whether to update the Transformer deep spatiotemporal prediction model and the user response uncertainty model in step S100.

[0014] The second technical solution adopted in this invention is: a dynamic electricity price generation system based on two-stage regulation of day-ahead and intraday pricing, including a coarse-grained interval electricity price generation module and a current real-time electricity price generation module; The coarse-grained interval electricity price generation module is used to: acquire historical voltage and power data and meteorological data of transformer outlets and important nodes in any region during the day-ahead phase; make predictions based on the historical voltage and power data and meteorological data to obtain predicted values ​​of power and voltage; calculate voltage risk index value and photovoltaic absorption index value based on the predicted values ​​of power and voltage; and perform preliminary demand response pricing based on the voltage risk index value and photovoltaic absorption index value to generate a coarse-grained interval electricity price. The current real-time electricity price generation module is used to: collect voltage data and load power data of transformer outlets and user-responsive resource nodes in the area in real time during the day, and analyze the voltage data and load power data to obtain the current actual photovoltaic absorption gap; and correct the coarse-grained interval electricity price based on the current actual photovoltaic absorption gap to obtain the current real-time electricity price.

[0015] The beneficial effects of the above technical solution are as follows: (1) The dynamic electricity price generation method based on day-ahead and intraday two-stage regulation disclosed in this invention achieves local consumption of photovoltaic power in the distribution network environment with high photovoltaic penetration rate through price-based demand response, balances photovoltaic output and local load, optimizes voltage in the distribution area and improves energy utilization efficiency; the DR price generation method designed based on this can more effectively guide the supply and demand matching in the distribution area and promote the safe and stable operation of the power grid.

[0016] (2) Based on the characteristic tracking analysis of power and voltage boost curves, the present invention maps to a price adjustment parameter model and forms a real-time price. Users change the load size according to the real-time price to complete the demand response execution and thus consume the photovoltaic power in the local area. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a dynamic electricity price generation method based on two-stage day-ahead and intraday regulation, as provided in one embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following detailed description of the embodiments and the accompanying drawings are used to illustrate the principles of the present invention by way of example, but should not be used to limit the scope of the present invention. That is, the present invention is not limited to the described preferred embodiments, and the scope of the present invention is defined by the claims.

[0019] In the description of this invention, it should be noted that, unless otherwise stated, "a plurality of" means two or more; the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance; those skilled in the art can understand the specific meaning of the above terms in this invention as appropriate.

[0020] Example 1 like Figure 1 As shown, one embodiment of the present invention provides a dynamic electricity price generation method based on two-stage day-ahead and intraday regulation, comprising the following steps: S100: During the day-ahead phase, acquire historical voltage and power data and meteorological data of transformer outlets and important nodes in any region; make predictions based on the historical voltage and power data and meteorological data to obtain predicted values ​​of power and voltage; calculate voltage risk index value and photovoltaic absorption index value based on the predicted values ​​of power and voltage; and perform preliminary demand response pricing based on the voltage risk index value and photovoltaic absorption index value to generate coarse-grained range electricity prices. (1) During the day-ahead phase, acquire historical voltage and power data and meteorological data for any region; make predictions based on the historical voltage and power data and meteorological data to obtain predicted values ​​of power and voltage; calculate voltage risk index value and photovoltaic absorption index value based on the predicted values ​​of power and voltage; Historical voltage and power data, and meteorological data for transformer outlets and key nodes within any region can be obtained, including: For a given distribution network with photovoltaic power generation devices, meteorological data, transformer outlet data, and voltage and power data of important nodes (important nodes are nodes that can characterize the electrical quantity level of a certain distribution network area selected by voltage sensitivity matrix or correlation analysis) are collected and stored based on various sensors; historical voltage and power data and meteorological data are obtained through the stored voltage and power data.

[0021] Meteorological data acquisition includes meteorological data collected through locally deployed micro weather stations, wind speed sensors, irradiance meters and other monitoring equipment, as well as meteorological data obtained through other relevant platforms or institutions (such as official / third-party meteorological service APIs); meteorological data includes, for example, wind speed, irradiance, temperature and humidity.

[0022] Based on the historical voltage and power data and meteorological data, predictions are made to obtain power and voltage forecasts, including: A Transformer deep spatiotemporal prediction model is constructed. The historical voltage and power data and meteorological data are input into the Transformer deep spatiotemporal prediction model to predict the operation of the transformer area in the next 24 hours and obtain the predicted power and voltage values.

[0023] The construction of the Transformer deep spatiotemporal prediction model includes: acquiring historical voltage and power data and meteorological data to generate a training set; and training the Transformer model based on the training set to obtain the Transformer deep spatiotemporal prediction model.

[0024] The predicted power and voltage values ​​are expressed by the following formulas: In the formula, and The future outputs of the Transformer deep spatiotemporal prediction model are respectively Time's up Power prediction and voltage prediction at any given time; F The Transformer deep spatiotemporal prediction model obtained through training; This is a directed graph representing the spatial relationships between nodes in a transformer substation. ， in, S It is a set of nodes, representing n Distribution network status (voltage, power) of each sub-region; E It is a set of edges, representing the physical connection between different sub-regions; It is an adjacency matrix constructed based on the parameterized Euclidean distance between different nodes; All are inputs to the Transformer deep spatiotemporal prediction model; They represent from Time's up Power sequence data and voltage sequence data at any given time; Indicates the future Time's up Meteorological data at any given time; Indicates any given moment, Indicates the length of the input history sequence. This indicates the predicted length of the future sequence.

[0025] The power series data, voltage series data, and meteorological data at any given time are represented by the following formula: In the formula, for Power sequence data at any given time; for Voltage sequence data at any given time; for Meteorological data at any given time; They are t The power and voltage at the transformer output at all times; They are respectively t Time Node Power and key nodes The voltage; express t Photovoltaic power generation devices at all times j ; output power; They are respectively Light intensity and surface temperature at any given time; n , M , N These are the number of nodes, the number of photovoltaic power generation devices, and the number of important nodes, respectively.

[0026] Based on power and voltage forecasts, the potential voltage risk and photovoltaic (PV) grid integration status of the region are assessed, yielding voltage risk index values ​​and PV grid integration index values, including: Based on power and voltage forecasts, the Voltage Risk Index (VRI) and Photovoltaic Accommodation Index (PAI) are assessed to obtain their values, reflecting the intensity of distribution network operation and supply-demand matching in different time periods within a 24-hour period. The Max-min normalization method is then used to normalize the range of VRI and PAI values. Inside.

[0027] Voltage Risk Index (VRI) is a comprehensive parameter that measures the potential harm caused by voltage anomalies in a power system to equipment or the power grid. It is usually assessed by combining factors such as voltage deviation, fluctuation frequency and duration. Photovoltaic Integration Index (PAI) is a core parameter that measures the effective utilization of electricity in a photovoltaic power generation system. Its calculation and optimization involve multiple dimensions of factors such as grid acceptance capacity, energy storage configuration and policy regulations.

[0028] The voltage risk index and the photovoltaic absorption index are calculated using the following formulas: In the formula, for Voltage risk index value at any given time; for The photovoltaic power consumption index value at any given time; N Number of important nodes; This is the window length for that time period; As an important node k exist t The weight of time (can be set as load importance or load power ratio; if there is no prior information, it can be set to 1). for t Important milestones The voltage; Rated voltage (per unit, typically 1.0 pu); The maximum allowable relative deviation threshold is defined. The greater the distance beyond the threshold, the greater the risk increases quadratically, which can highlight severely stressful scenarios. n , M These are the number of nodes and the number of photovoltaic power generation devices, respectively. for t Time Node The power; for t Photovoltaic power generation devices at all times j ; output power; To evaluate the window length.

[0029] (2) Based on the voltage risk index value and the photovoltaic absorption index value, conduct preliminary demand response (DR) pricing to generate coarse-grained range electricity prices; The obtained voltage risk index value and photovoltaic consumption index value are mapped to the user electricity consumption response target (i.e. the desired user electricity consumption response amount), and then a coarse-grained interval electricity price is generated through the User response uncertainty model (URUM), that is, the interval pricing at the 15-minute level.

[0030] The User Response Uncertainty Model (URUM), used to reflect the uncertainty of user responses, is expressed by the following formula: In the formula, To take into account photovoltaic power consumption and voltage risks, the acceptance level of user response DR within the distribution area is generated based on the improved Sigmoid function; This represents the current average benefit. Steepness and skewness, respectively, together determine the shape of the function; This is a cutoff parameter used to characterize the viscous resistance to a user changing their behavior. These are the actual prices of demand response (DR) and different users. The expected price; for t The estimated user response volume within the designated area at any given time, i.e., the user electricity response target; This represents the maximum response capacity within the user's capabilities within the designated area. for t Always consider photovoltaic power consumption and voltage risks, and the acceptance level of DR by users in the distribution area; The operation is defined to represent random fluctuations within a certain range.

[0031] Most users are sensitive to price levels, but the degree of sensitivity varies. At the same time, users' acceptance is related to psychological changes, unexpected events, etc., which leads to uncertainty in users' response behavior. Therefore, user response volume is estimated based on the above user response uncertainty model.

[0032] Coarse-grained range electricity prices are expressed by the following formula: In the formula, This refers to the current day's coarse-grained range electricity price unit group; For the 96 elements corresponding to the daily range electricity price; for t The price ranges of each level within the 15-minute time period at which the time is located belong to the set of price ranges corresponding to each level. These are the upper and lower limits of the price range corresponding to Level 1; Levels The upper and lower limits of the corresponding price range; The highest level The corresponding price ranges. Each level reflects how much the electricity price influences user response volume.

[0033] That is, based on the aforementioned voltage risk index value, photovoltaic absorption index value, and URUM, this invention will output the future voltage risk index value from the Transformer deep spatiotemporal prediction model. Time's up The power and voltage forecast values ​​at any given time are evaluated to obtain voltage risk index values ​​and photovoltaic absorption index values. Time periods are then divided (e.g., a single day is divided into 96 time periods by default). By arranging the target to be responded to at different scales and the price range levels in ascending order, the corresponding price range for each time period is obtained. In other words, the user electricity response targets for different time periods are combined with URUM to generate ζ levels of coarse-grained range electricity prices, thereby obtaining the coarse-grained range electricity prices for users in each time period.

[0034] S200: During the daytime phase, real-time voltage and load power data of transformer outlets and user-responsive resource nodes in the region are collected, and the current actual photovoltaic absorption gap is obtained based on the voltage and load power data; and the coarse-grained range electricity price is corrected based on the current actual photovoltaic absorption gap to obtain the current real-time electricity price.

[0035] During the intraday phase, for a given distribution network area containing photovoltaic (PV) power generation devices, real-time voltage and load power data are collected from the transformer outlets and user-responsive resource nodes (user-responsive resource nodes are response nodes mainly affected by electricity prices) in that area. Based on the voltage and load power data, the current actual PV absorption gap is analyzed, including: Use the midpoint of the coarse-grained range electricity price for users (Right now t The system executes response actions based on the DR price at the start of a 15-minute time period (corresponding to the user's response price), and collects in real time the voltage and load power data of the transformer outlet and user-responsive resource nodes in the area. It tracks and analyzes the voltage and load power data (e.g., existing technology load monitoring and decomposition, photovoltaic output-node voltage mapping relationship) to obtain the power of each node and the output power of each photovoltaic power generation device. Based on the power of each node and the output power of each photovoltaic power generation device, it obtains the current actual photovoltaic consumption gap.

[0036] The midpoint of the coarse-grained range electricity price is represented by the following formula: In the formula, for t The DR price responded by the user at the start of the corresponding 15-minute time period; Levels The upper and lower limits of the corresponding price range.

[0037] Based on the current actual photovoltaic absorption gap, the coarse-grained range electricity price is corrected using the following formula to obtain the current real-time electricity price (that is, the supply-demand mismatch is mapped a second time to a 5-minute price signal): In the formula, for t The DR price responded by the user at the start of the corresponding 15-minute time period; Levels The upper and lower limits of the corresponding price range; For users based The current actual photovoltaic consumption gap after the implementation of response actions; for t Time Node The power; for t Photovoltaic power generation devices at all times j ; output power; n , M These are the number of nodes and the number of photovoltaic power generation devices, respectively. The current real-time electricity price (i.e., the granular real-time electricity price corresponding to each 15-minute time period). It is a triple, where each element corresponds to the real-time electricity price for 5 minutes. The actual value is determined based on the current size of the actual photovoltaic absorption gap within the range. Generate proportionally; These represent the lower limit of the price range corresponding to level h-1 and the upper limit of the price range corresponding to level h+1, respectively.

[0038] Furthermore, in one embodiment, step S200 further includes: acquiring data on user response actions based on the corrected real-time electricity price; calculating update criterion values ​​for the Transformer deep spatiotemporal prediction model and the user response uncertainty model based on the data on user response actions based on the real-time electricity price; and determining whether to update the Transformer deep spatiotemporal prediction model and the user response uncertainty model in step S100 based on the update criterion values ​​of the Transformer deep spatiotemporal prediction model and the user response uncertainty model. The data on user response actions based on the corrected real-time electricity price includes, for example, the predicted and actual values ​​of each prediction variable (power and voltage at each point) handled by the Transformer deep spatiotemporal prediction model at each time; the real-time response price at each time; and the actual user response amount, etc.

[0039] The following formulas are used to determine whether to update the Transformer deep spatiotemporal prediction model and the user response uncertainty model: In the formula, These are the update criterion values ​​for the Transformer deep spatiotemporal prediction model and the user response uncertainty model, respectively. S The variable representing the prediction, For predictor variables within the evaluation window b The actual value of each sample; For predictor variables within the evaluation window b The model output for each sample; The distance between the user's actual response and the response predicted by the uncertainty model of the user's response; , These are the historical update criteria values ​​for the Transformer deep spatiotemporal prediction model and the user response uncertainty model, respectively, obtained from the previous evaluation window; For users, an acceptable upper limit for the error of the uncertainty model is recommended to be 7% to 10%. The acceptable upper limit of error for the Transformer depth spatiotemporal prediction model is recommended to be 5% to 8%. These are the evaluation window length, the number of samples within the window, and the performance degradation coefficient, with the performance degradation coefficient set to an empirical value of 0.95 to 0.99. n , M These are the number of nodes and the number of photovoltaic power generation devices, respectively. The actual power response of the user at node k at time t; The predicted response value for the user response uncertainty model at node k at time t can be obtained by inverse URUM solution from the real-time response price at time t and the maximum response capacity of the user at node k.

[0040] If the update criterion value of the Transformer depth spatiotemporal prediction model Greater than the acceptable error upper limit of the Transformer depth spatiotemporal prediction model If the condition is met, the Transformer depth spatiotemporal prediction model is updated; or the update criterion value of the Transformer depth spatiotemporal prediction model is updated. The update criterion history value of the Transformer deep spatiotemporal prediction model obtained in the previous evaluation window is less than the previous evaluation window. Then the Transformer depth spatiotemporal prediction model is updated.

[0041] If the user responds to the updated criterion value of the uncertainty model Greater than the upper limit of acceptable error for user response uncertainty models If so, the user response uncertainty model is updated; or the update criterion value of the user response uncertainty model is updated. The updated criterion historical value of the user response uncertainty model obtained in the previous evaluation window is less than the previous evaluation window. Then the user response uncertainty model will be updated.

[0042] Example 2 An embodiment of the present invention provides a dynamic electricity price generation system based on two-stage regulation of day-ahead and intraday pricing, including a coarse-grained interval electricity price generation module and a current real-time electricity price generation module; The coarse-grained interval electricity price generation module is used to: acquire historical voltage and power data and meteorological data of transformer outlets and important nodes in any region during the day-ahead phase; make predictions based on the historical voltage and power data and meteorological data to obtain predicted values ​​of power and voltage; calculate voltage risk index value and photovoltaic absorption index value based on the predicted values ​​of power and voltage; and perform preliminary demand response pricing based on the voltage risk index value and photovoltaic absorption index value to generate a coarse-grained interval electricity price. The current real-time electricity price generation module is used to: collect voltage and load power data of transformer outlets and user-responsive resource nodes in the region in real time during the day, and analyze the voltage and load power data to obtain the current actual photovoltaic absorption gap; and correct the coarse-grained interval electricity price based on the current actual photovoltaic absorption gap to obtain the current real-time electricity price.

[0043] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0044] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A dynamic electricity price generation method based on two-stage day-ahead and intraday regulation, characterized in that, Includes the following steps: S100: During the day-ahead phase, acquire historical voltage and power data, as well as meteorological data, for transformer outlets and key nodes within any region; Based on the historical voltage and power data and meteorological data, predictions are made to obtain predicted values ​​for power and voltage; based on the predicted values ​​for power and voltage, voltage risk index values ​​and photovoltaic absorption index values ​​are calculated; and based on the voltage risk index values ​​and photovoltaic absorption index values, preliminary demand response pricing is performed to generate coarse-grained range electricity prices. S200: During the daytime phase, real-time voltage and load power data of transformer outlets and user-responsive resource nodes in the region are collected, and the current actual photovoltaic absorption gap is obtained based on the voltage and load power data; and the coarse-grained range electricity price is corrected based on the current actual photovoltaic absorption gap to obtain the current real-time electricity price.

2. The dynamic electricity price generation method according to claim 1, characterized in that, Step S100 involves making a prediction based on the historical voltage and power data and meteorological data, including: A Transformer deep spatiotemporal prediction model is constructed. The historical voltage and power data and meteorological data are input into the Transformer deep spatiotemporal prediction model to obtain power prediction values ​​and voltage prediction values.

3. The dynamic electricity price generation method according to claim 2, characterized in that, The construction of the Transformer deep spatiotemporal prediction model includes: acquiring historical voltage and power data and meteorological data to generate a training set; training the Transformer model based on the training set to obtain the Transformer deep spatiotemporal prediction model.

4. The dynamic electricity price generation method according to claim 1, characterized in that, Step S100 includes calculating the voltage risk index value and the photovoltaic absorption index value, which includes: The voltage risk index and photovoltaic absorption index are evaluated based on the predicted values ​​of power and voltage to obtain the voltage risk index value and photovoltaic absorption index value.

5. The dynamic electricity price generation method according to claim 1, characterized in that, Step S100 includes: generating a coarse-grained range electricity price based on the voltage risk index value and the photovoltaic absorption index value through a user response uncertainty model.

6. The dynamic electricity price generation method according to claim 1, characterized in that, The coarse-grained range electricity price mentioned in step S100 is expressed by the following formula: In the formula, This refers to the current day's coarse-grained range electricity price unit group; The 96 elements correspond to the daily interval electricity price; for t The price ranges of each level within the 15-minute time period at which the time is located belong to the set of price ranges corresponding to each level. These are the upper and lower limits of the price range corresponding to Level 1; Levels The upper and lower limits of the corresponding price range; The highest level The corresponding price range.

7. The dynamic electricity price generation method according to claim 1, characterized in that, Step S200 includes: Users execute response actions based on the midpoint of the coarse-grained range electricity price, and collect voltage and load power data of transformer outlets and user-responsive resource nodes in the area in real time. The voltage and load power data are tracked and analyzed to obtain the power of each node and the output power of each photovoltaic power generation device; and the current actual photovoltaic consumption gap is obtained based on the power of each node and the output power of each photovoltaic power generation device.

8. The dynamic electricity price generation method according to claim 1, characterized in that, In step S200, based on the current actual photovoltaic absorption gap, the coarse-grained range electricity price is corrected using the following formula to obtain the current real-time electricity price: In the formula, for t The DR price responded by the user at the start of the corresponding 15-minute time period; Levels The upper and lower limits of the corresponding price range; For users based The current actual photovoltaic consumption gap after the implementation of response actions; for t Time Node The power; for t Photovoltaic power generation devices at all times j ; output power; n , M These are the number of nodes and the number of photovoltaic power generation devices, respectively. This refers to the current real-time electricity price; The actual value is determined based on the current size of the actual photovoltaic absorption gap within the range. Generate proportionally; These represent the lower limit of the price range corresponding to level h-1 and the upper limit of the price range corresponding to level h+1, respectively.

9. The dynamic electricity price generation method according to claim 1, characterized in that, Step S200 further includes: obtaining data on the user's response actions based on the real-time electricity price, and calculating the update criterion values ​​of the Transformer deep spatiotemporal prediction model and the user response uncertainty model based on the data on the user's response actions based on the real-time electricity price; And based on the update criterion value of the Transformer deep spatiotemporal prediction model and the user response uncertainty model, determine whether to update the Transformer deep spatiotemporal prediction model and the user response uncertainty model in step S100.

10. A dynamic electricity price generation system based on day-ahead and intraday two-stage regulation, characterized in that, This includes a coarse-grained interval electricity price generation module and a current real-time electricity price generation module; The coarse-grained interval electricity price generation module is used to: acquire historical voltage and power data and meteorological data of transformer outlets and important nodes in any region during the day-ahead phase; make predictions based on the historical voltage and power data and meteorological data to obtain predicted values ​​of power and voltage; calculate voltage risk index value and photovoltaic absorption index value based on the predicted values ​​of power and voltage; and perform preliminary demand response pricing based on the voltage risk index value and photovoltaic absorption index value to generate a coarse-grained interval electricity price. The current real-time electricity price generation module is used to: collect voltage data and load power data of transformer outlets and user-responsive resource nodes in the area in real time during the day, and analyze the voltage data and load power data to obtain the current actual photovoltaic absorption gap; and correct the coarse-grained interval electricity price based on the current actual photovoltaic absorption gap to obtain the current real-time electricity price.

Images