A new energy surplus electricity hierarchical consumption method based on a virtual power plant

Abstract

The application is suitable for the technical field of new energy consumption, and provides a new energy surplus power consumption grading method based on a virtual power plant, which comprises the following steps: uploading a response feature vector by a load side terminal, uploading an adjustment feature matrix by an energy storage cluster terminal, and uploading a transaction feature tensor by a demand response aggregation terminal; a virtual power plant central controller generates an interruption instruction package based on the duration of a new energy output prediction deviation, and sends the interruption instruction package to the load side terminal, generates a charging and discharging parameter set and sends the charging and discharging parameter set to the energy storage cluster terminal, and generates a transaction contract frame and sends the transaction contract frame to the demand response aggregation terminal; the load side terminal, the energy storage cluster terminal and the demand response aggregation terminal respectively feed back a consumption confirmation signal to the central controller after performing corresponding operations; and the central controller generates a surplus power consumption state result based on the received consumption confirmation signal. The method effectively improves the consumption capacity of new energy surplus power, thereby meeting the real-time balance demand of a high penetration rate new energy scene.

Description

Technical Field

[0001] This invention relates to the field of new energy consumption technology, and in particular to a method for the graded consumption of surplus new energy power based on a virtual power plant. Background Technology

[0002] With the large-scale grid connection of new energy sources such as wind and solar power, the randomness and volatility of their output have led to an increasingly prominent problem of surplus power. During periods of abundant wind and solar resources, the actual output of new energy often far exceeds the predicted value and the grid's absorption capacity, which can lead to wind and solar curtailment and even grid overload risks. Existing technologies mainly rely on a unified dispatch command mode and a centralized main grid peak-shaving mode. The unified dispatch command mode involves the grid dispatch center broadcasting fixed absorption commands to all adjustable resources, ignoring the differences in response time among load, energy storage, and trading resources. This results in second-level fluctuations being responded to by hour-level resources, causing frequent equipment operation and wear. In addition, the centralized main grid peak-shaving mode relies on long-distance transmission of surplus power from the main grid. Due to high transmission losses and high response delays, it is difficult to meet the real-time balancing needs of high-penetration new energy scenarios.

[0003] In view of this, a hierarchical consumption method for surplus renewable energy power based on virtual power plants is proposed. Summary of the Invention

[0004] This invention provides a method for the graded consumption of surplus renewable energy based on virtual power plants, which is used to solve the problem of difficulty in meeting the real-time balance requirements of renewable energy scenarios with high penetration rates.

[0005] This invention provides a method for the tiered consumption of surplus renewable energy based on a virtual power plant, executed by a virtual power plant central controller in coordination with load-side terminals, energy storage cluster terminals, and demand response aggregation terminals, including:

[0006] The load-side terminal uploads a response feature vector to the virtual power plant central controller; the energy storage cluster terminal uploads a regulation feature matrix to the virtual power plant central controller; and the demand response aggregation terminal uploads a transaction feature tensor to the virtual power plant central controller.

[0007] The virtual power plant central controller executes the following based on the duration of the new energy output prediction deviation:

[0008] When the duration is within a first preset time range, an interrupt command packet is generated based on the response feature vector and sent to the load-side terminal.

[0009] When the duration is within the second preset time range, a set of charging and discharging parameters is generated based on the adjustment feature matrix and sent to the energy storage cluster terminal.

[0010] When the duration is within the third preset time range, a transaction contract frame is generated based on the transaction feature tensor and sent to the demand response aggregation terminal;

[0011] Wherein, the first preset time range Second preset time range The third preset time range; the duration is the cumulative time during which the difference between the actual power and the predicted power continuously exceeds a set threshold.

[0012] After executing the operations corresponding to the interruption instruction package, the charging and discharging parameter set, or the transaction contract frame, the load-side terminal, the energy storage cluster terminal, and the demand response aggregation terminal respectively, send a consumption confirmation signal back to the central controller.

[0013] The central controller generates a result indicating the status of surplus power consumption based on the received power consumption confirmation signal.

[0014] Furthermore, the response feature vector includes the maximum interruptible power, the upper limit of response delay, and the amount of change in switching state.

[0015] Furthermore, the step of generating an interrupt command packet based on the response feature vector and sending it to the load-side terminal includes:

[0016] Based on the current surplus power value, the maximum interruptible power, and the change in switch state, the total amount of load to be interrupted is calculated using the switch state adaptive coefficient.

[0017] The predefined interrupt operation instruction code, the total amount of interruptible load, and the transaction identifier generated based on the current timestamp are encapsulated into an interrupt instruction package;

[0018] The interrupt command packet is sent to the load-side terminal via the first communication protocol.

[0019] Furthermore, the adjustment feature matrix includes average state of charge, maximum charging power, maximum discharging power, remaining rechargeable capacity, remaining discharging capacity, and power adjustment response time.

[0020] Furthermore, the step of generating a set of charging and discharging parameters based on the adjustment feature matrix and sending it to the energy storage cluster terminal includes:

[0021] The charging and discharging direction is determined based on the sign of the power value of the current surplus power. When the power value is positive, a charging command code is generated, and a reference power is determined in conjunction with the maximum charging power. When the power value is negative, a discharging command code is generated, and a reference power is determined in conjunction with the maximum discharging power.

[0022] The power hold-up duration is calculated based on the remaining rechargeable capacity or the remaining dischargeable capacity, the reference power, and the duration of the new energy output prediction deviation.

[0023] Using the average state of charge in the adjustment feature matrix as the center value, a safe boundary for the state of charge within a preset range is generated;

[0024] The charging command code, discharging command code, reference power, power hold duration, and state of charge safety boundary are encapsulated into a charging and discharging parameter set;

[0025] The charging and discharging parameter set is sent to the energy storage cluster terminal through the second communication protocol.

[0026] Furthermore, the transaction feature tensor includes parameters for electricity volume, cost, and time window.

[0027] Furthermore, the step of generating a transaction contract frame based on the transaction feature tensor and sending it to the demand response aggregation terminal includes:

[0028] The product of the duration of the new energy output forecast deviation and the average power is taken as the total electricity to be traded;

[0029] Based on the total amount of electricity to be traded and the lower and upper limits of the tradable electricity for each trading period in the trading feature tensor, select the trading period with the highest electricity matching degree and obtain the corresponding trading period identifier;

[0030] Compare the unit power transmission cost coefficients of all inter-regional power transmission channels in the transaction feature tensor, and select the channel identifier corresponding to the minimum value as the lowest cost channel identifier;

[0031] The predefined transaction instruction code, the total electricity to be traded, the selected transaction time period identifier, the lowest cost channel identifier, and the current timestamp are encapsulated into a transaction contract frame;

[0032] The transaction contract frame is sent to the demand response aggregation terminal via a third communication protocol.

[0033] Furthermore, after executing the operations corresponding to the interruption instruction package, the charge / discharge parameter set, or the transaction contract frame, the load-side terminal, the energy storage cluster terminal, and the demand response aggregation terminal respectively, send a consumption confirmation signal back to the central controller, including:

[0034] The load-side terminal feedback confirmation signal includes the actual amount of interrupted load;

[0035] The energy storage cluster terminal feeds back the power consumption confirmation signal, which includes the adjustment amount of charging and discharging power.

[0036] The consumption confirmation signal fed back by the demand response aggregation terminal includes a transaction electricity certificate.

[0037] Furthermore, based on the received power consumption confirmation signal, the central controller generates a surplus power consumption status result, including:

[0038] The actual total power consumption is obtained by summing the actual interrupted load amount reported by the load-side terminal, the charging and discharging power adjustment amount reported by the energy storage cluster terminal, and the transaction power certificate reported by the demand response aggregation terminal.

[0039] Construct a structured consumption status result that includes the actual total consumption, the consumption components of each terminal, and the timestamp.

[0040] Furthermore, the structured consumption status result includes a total consumption power field, a load consumption component field, an energy storage consumption component field, a transaction consumption component field, and a timestamp field.

[0041] As can be seen from the above technical solutions, the present invention has the following advantages:

[0042] This invention generates a dynamic profile of resource capacity by uploading response feature vectors, adjustment feature matrices, and transaction feature tensors through load-side terminals, energy storage cluster terminals, and demand response aggregation terminals, respectively. The central controller calculates the duration of continuous over-threshold deviation of renewable energy output prediction and selects the consumption terminal according to the level to which the duration belongs. After the terminal executes, it feeds back a consumption confirmation signal, and the central controller aggregates and generates a structured consumption status result. This effectively improves the consumption capacity of surplus renewable energy, thereby meeting the real-time balance requirements of high-penetration renewable energy scenarios. Attached Figure Description

[0043] Figure 1 This is a schematic flowchart of an embodiment of a method for the graded consumption of surplus renewable energy based on a virtual power plant in this invention. Detailed Implementation

[0044] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “corresponding to,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0045] Example 1

[0046] Please see Figure 1 The method provided in this application includes the following steps:

[0047] 101. The load-side terminal uploads the response feature vector to the virtual power plant central controller;

[0048] In this embodiment, the response feature vector includes the maximum interruptible power, the upper limit of response delay, and the amount of switching state change. The mathematical expression for the response feature vector is: , Maximum interruptible power indicates the maximum power interruption capability that the load-side terminal can provide in the next control cycle; The response delay limit represents the maximum allowed time from receiving an instruction to completing its execution at the load-side terminal. This represents the change in load switching state between the current moment and the previous cycle, calculated by statistically analyzing the switching state changes of all interruptible loads. The calculation formula is: ,in This represents the total number of interruptible loads. , For the first A load in and The on / off state at any given time, 0 for off and 1 for on.

[0049] 102. The energy storage cluster terminal uploads the regulation characteristic matrix to the central controller of the virtual power plant;

[0050] In this embodiment, the adjustment feature matrix includes average state of charge, maximum charging power, maximum discharging power, remaining rechargeable capacity, remaining discharging capacity, and power adjustment response time. The mathematical expression for the adjustment feature matrix is: ,in, The average state of charge (SBC) represents the average energy storage level of the terminals in an energy storage cluster. Maximum charging power, representing the maximum charging power currently available to the energy storage cluster terminal. Maximum discharge power, representing the maximum discharge power currently available at the energy storage cluster terminal. The remaining rechargeable capacity represents the amount of chargeable energy remaining when charging to the maximum state of charge limit using the current maximum charging power. The remaining discharge capacity represents the amount of dischargeable capacity remaining after discharging to the state-of-charge limit at the current maximum discharge power. Power adjustment response time represents the time required for the energy storage converter to complete power adjustment.

[0051] 103. The demand response aggregation terminal uploads the transaction feature tensor to the virtual power plant central controller;

[0052] In this embodiment, the transaction feature tensor includes parameters for electricity volume, cost, and time window. The mathematical expression for the transaction feature tensor is: ,in For the power dimension matrix, ,in and They represent the first The lower and upper limits of tradable electricity volume for each trading session; As a cost dimension vector, , Indicates the first One power transmission channel in The unit power transmission cost coefficient for a given time period, where ; The time window dimension vector, ,in and They represent the first The start and end times of each trading session.

[0053] 104. The virtual power plant central controller executes the following based on the duration of the new energy output prediction deviation:

[0054] When the duration is within the first preset time range, an interruption instruction packet is generated based on the response feature vector and sent to the load-side terminal; when the duration is within the second preset time range, a charging and discharging parameter set is generated based on the adjustment feature matrix and sent to the energy storage cluster terminal; when the duration is within the third preset time range, a transaction contract frame is generated based on the transaction feature tensor and sent to the demand response aggregation terminal.

[0055] Among them, the first preset time range Second preset time range The third preset time range; the duration is the cumulative time during which the difference between the actual power and the predicted power continuously exceeds a set threshold.

[0056] In this embodiment, the first preset time range is a second-level response range, set to a duration ≤ 1 minute. This range corresponds to scenarios where the duration of renewable energy output prediction deviation is short, such as instantaneous fluctuations caused by sudden changes in wind speed or rapid cloud movement. In such cases, rapid response resources are needed to absorb the excess energy within a second-level timescale. The second preset time range is a minute-level adjustment range, set to 1 minute < duration ≤ 60 minutes. This range corresponds to scenarios where the duration of renewable energy output prediction deviation is moderate, such as continuous wind speed decreases or partially cloudy conditions. In such cases, energy storage devices need to adjust charging and discharging within a minute-level timescale. The third preset time range is an hour-level trading range, set to a duration > 60 minutes. This range corresponds to scenarios where the duration of renewable energy output prediction deviation is long, such as continuous overcapacity in wind or solar power caused by large-scale weather system changes. In such cases, demand response aggregators need to participate in cross-regional electricity trading to absorb the excess energy within an hour-level timescale. The aforementioned preset time range is divided according to the response characteristics of different resources in the power system: interruptible resources on the load side (such as industrial loads, air conditioning, etc.) can respond quickly (in seconds), but should not be interrupted for a long time (to avoid affecting production or comfort); energy storage devices (such as battery energy storage) can be regulated for charging and discharging on a time scale of minutes, but are limited by capacity and the number of charging and discharging cycles, and are not suitable for regulation for several hours; electricity market transactions usually require hours to complete transaction matching and settlement, and are suitable for handling surplus electricity on a long time scale.

[0057] The duration is the cumulative time during which the difference between the actual power and the predicted power continuously exceeds a set threshold. Specifically, timing begins when the absolute difference between the actual power and the predicted power exceeds the set threshold; timing continues as long as the difference remains above the threshold; when the difference falls below the threshold, timing stops and is reset to zero. This cumulative time is the duration. The threshold is typically set proportionally based on the rated capacity of the renewable energy unit (such as a wind farm or photovoltaic power station). The proportionality coefficient The value range is 5% to 15%. This refers to the rated power. For example, for a wind farm with a rated capacity of 100MW, if we take... =10%, then the threshold =10MW. When the difference between the actual power and the predicted power exceeds 10MW continuously, the accumulation period begins. Through a time-scale hierarchical strategy, the predicted deviation of new energy output is dynamically allocated to the most suitable terminal resources for consumption, thereby optimizing resource allocation, reducing equipment losses, reducing the peak-shaving pressure on the main grid, and improving the local consumption rate of surplus electricity.

[0058] In this embodiment, when the duration is within a first preset time range, an interrupt command packet is generated based on the response feature vector and sent to the load-side terminal, including the following steps:

[0059] 1. Based on the current surplus power, maximum interruptible power, and changes in switch status, calculate the total load to be interrupted using an adaptive coefficient for switch status;

[0060] Formula for calculating the total load that needs to be interrupted:

[0061]

[0062] in: This represents the power value of the current surplus electricity. For the adaptive coefficient of the switching state, , For maximum interruptible power, This refers to the change in switch state. This represents the total number of interruptible loads.

[0063] 2. Encapsulate the predefined interrupt operation instruction code, the total interrupt load, and the transaction identifier generated based on the current timestamp into an interrupt instruction package;

[0064] The predefined interrupt operation instruction code is 0. 01 (fixed value, corresponding to load interruption operation), where the instruction code mapping table (stored in the central controller) is shown in the table below:

[0065]

[0066] The generated transaction identifier is: ,in This is the current timestamp. For a secure hash algorithm, This is a unique identifier for the load-side terminal.

[0067] 3. The interrupt command packet is sent to the load-side terminal via the first communication protocol.

[0068] The first communication protocol here is a low-latency communication protocol. When the duration is within the first preset time range (≤1 minute), based on the response feature vector (such as maximum interruptible power, response delay, etc.) reported by the load-side terminal, an interrupt command packet (containing the amount of load to be interrupted, command code, etc.) is generated and sent to the load-side terminal through the low-latency communication protocol to trigger a second-level load interruption.

[0069] In this embodiment, when the duration is within a second preset time range, a charging and discharging parameter set is generated based on the adjustment feature matrix and sent to the energy storage cluster terminal, including the following steps:

[0070] 1. Determine the charging / discharging direction based on the sign of the power value of the current surplus power; when the power value is positive, generate a charging command code and determine the reference power based on the maximum charging power; when the power value is negative, generate a discharging command code and determine the reference power based on the maximum discharging power.

[0071] If there is excess power, a charging instruction code 0x10 is generated. Power deficiency was detected, generating discharge command code 0x11. Reference power was calculated. .

[0072] 2. Calculate the power hold-up duration based on the remaining rechargeable capacity or the remaining dischargeable capacity, the reference power, and the duration of the predicted deviation of the new energy output;

[0073] Take the remaining chargeable amount when charging. During discharge, the remaining dischargeable quantity is taken. ; Calculation time: , For the current adjustable energy capacity, Duration of prediction deviation for new energy output.

[0074] 3. Using the average state of charge in the adjustment feature matrix as the center value, generate a safe boundary for the state of charge within a preset range;

[0075] Read the average state of charge from the adjustment feature matrix Calculate the safe range: Lower limit protection prevents over-discharge, such as Discharging is prohibited when the charge level is less than 0.2 ohms; upper limit protection prevents overcharging. Charging is prohibited when the value is greater than 0.9.

[0076] 4. Encapsulate the charging command code, discharging command code, reference power, power hold duration, and state-of-charge safety boundary into a charging and discharging parameter set;

[0077] Charge / discharge command code (0) 10 indicates charging, 0 indicates charging. 11 represents discharge), the calculated reference power value, power hold-up time, and state-of-charge safety boundary (a range consisting of the center value ±10%) are combined in a fixed order into a binary data packet. The safety boundary is converted into two 32-bit floating-point numbers (lower limit first, upper limit second), and all numerical fields are arranged in big-endian byte order to form a set of charging and discharging parameters that can be directly parsed by the energy storage cluster terminal.

[0078] 5. The charging and discharging parameter set is sent to the energy storage cluster terminal via the second communication protocol.

[0079] The second communication protocol here is a reliable communication protocol. When the duration is within the second preset time range (1~60 minutes), based on the adjustment feature matrix (such as SOC, maximum charge and discharge power, etc.) reported by the energy storage cluster terminal, a set of charge and discharge parameters (including reference power, hold time, etc.) is generated and sent to the energy storage cluster terminal through the reliable communication protocol to start minute-level charge and discharge adjustment.

[0080] In this embodiment, when the duration is within a third preset time range, a transaction contract frame is generated based on the transaction feature tensor and sent to the demand response aggregation terminal, including the following steps:

[0081] 1. The product of the duration of the new energy output forecast deviation and the average power is taken as the total electricity to be traded;

[0082] Calculate average power :

[0083] The amount of electricity to be traded is:

[0084] in: This refers to the initial time when the deviation first exceeds the set threshold. The termination point is when the deviation falls back below the set threshold. for Real-time output power of new energy sources for Real-time forecast of new energy power output.

[0085] 2. Based on the total amount of electricity to be traded and the lower and upper limits of the tradable electricity for each trading period in the trading feature tensor, select the trading period with the highest electricity matching degree and obtain the corresponding trading period identifier;

[0086] Calculate battery matching degree:

[0087]

[0088] in: , For the first Duration of each trading session For the first The end time of each trading session. For the first The start time of each trading session. For time period Average tradable capacity, For the first The minimum tradable electricity volume for each trading session For the first The maximum amount of electricity that can be traded during each trading session.

[0089] 3. Compare the unit power transmission cost coefficients of all inter-regional power transmission channels in the transaction feature tensor, and select the channel identifier corresponding to the minimum value as the lowest cost channel identifier;

[0090] Cost coefficient comparison:

[0091]

[0092] in: For the first Cost per unit of electricity transmitted through a single channel.

[0093] 4. Encapsulate the predefined transaction instruction code, total electricity to be traded, transaction time period identifier, lowest cost channel identifier, and current timestamp into a transaction contract frame;

[0094] The predefined transaction instruction code (fixed value 0) 20) The total electricity to be traded, the optimal trading time period identifier (2-byte integer number), the lowest cost channel identifier (2-byte integer number), and the current timestamp are concatenated in the order specified in the protocol. The time period identifier and channel identifier are extracted from the transaction feature tensor, the timestamp adopts the UTC time zone format, and all fields are Base64 encoded to form a standardized transaction contract frame.

[0095] 5. The transaction contract frame is sent to the demand response aggregation terminal via a third communication protocol.

[0096] The third communication protocol here is a secure communication protocol. When the duration is within the third preset time range (>60 minutes), a transaction contract frame (containing the transaction amount, time period identifier, etc.) is generated based on the transaction feature tensor (such as the tradable electricity range, transmission cost, etc.) reported by the demand response aggregation terminal. This frame is then sent to the demand response aggregation terminal through the secure communication protocol to initiate hourly power transactions.

[0097] 105. After the load-side terminal executes the operation corresponding to the interrupt instruction packet, it sends a confirmation signal for acceptance to the central controller.

[0098] 106. After the energy storage cluster terminal executes the operation corresponding to the charging and discharging parameter set, it sends a consumption confirmation signal back to the central controller;

[0099] 107. After executing the operation corresponding to the transaction contract frame, the demand response aggregation terminal sends a confirmation signal to the central controller.

[0100] In steps 105-107, the absorption confirmation signal fed back by the load-side terminal includes the actual interrupted load amount; the absorption confirmation signal fed back by the energy storage cluster terminal includes the charging and discharging power adjustment amount; and the absorption confirmation signal fed back by the demand response aggregation terminal includes the transaction power certificate.

[0101] Specifically, upon receiving an interruption command packet, the load-side terminal parses the total load to be interrupted. Then, based on a preset load priority strategy (e.g., cutting off non-critical loads first), it selects specific load devices for shutdown and monitors the actual interrupted load in real time to ensure the required interruption amount is met. The energy storage cluster terminal parses the charging and discharging parameter set to obtain the charging and discharging command code (determining whether it's charging or discharging), reference power, power hold-up duration, and SOC safety boundary. Then, the terminal sends the reference power command to the PCS (Power Conversion System) of each energy storage unit, controlling it to charge and discharge at the reference power within a specified time, while simultaneously monitoring the charging and discharging power adjustment in real time and ensuring the SOC remains within the safety boundary during charging and discharging. The demand response aggregation terminal parses the transaction contract frame to obtain the total electricity to be traded, the transaction period identifier, and the channel identifier. Then, the terminal submits an electricity transaction order through the power trading platform interface according to the specified transaction period and channel, and obtains the transaction electricity certificate upon successful transaction.

[0102] After completing the load interruption operation within seconds, the load-side terminal reports the actual interrupted load to the central controller; after completing the charging and discharging power adjustment within minutes, the energy storage cluster terminal reports the actual charging and discharging power adjustment; and after completing the power transaction within hours, the demand response aggregation terminal reports the transaction power certificate.

[0103] 108. The central controller generates the surplus power consumption status result based on the received consumption confirmation signal.

[0104] In this embodiment, after receiving the rejection confirmation signal, the central controller executes the following:

[0105] 1. Sum the actual interrupted load amount reported by the load-side terminal, the charging and discharging power adjustment amount reported by the energy storage cluster terminal, and the transaction power voucher reported by the demand response aggregation terminal to obtain the actual total power consumption;

[0106] 2. Construct a structured consumption status result that includes the actual total consumption, the consumption components of each terminal, and timestamps.

[0107] The structured consumption status results include the total consumed electricity field, the load consumption component field, the energy storage consumption component field, the transaction consumption component field, and the timestamp field.

[0108] Specifically, after receiving the actual interrupted load (representing the power value of load shedding) from the load-side terminal, the charging and discharging power adjustment (representing the absolute value of energy storage charging and discharging power, i.e., the amount of power absorbed) from the energy storage cluster terminal, and the transaction electricity voucher (representing the electricity absorbed through market transactions) from the demand response aggregation terminal, the central controller first performs an algebraic summation of these three values ​​to obtain the actual total absorbed electricity. Then, it constructs a structured absorption status result, which includes five fields: total absorbed electricity field (recording the actual total absorbed electricity value), load absorption component field (recording the actual interrupted load), energy storage absorption component field (recording the absolute value of the charging and discharging power adjustment), transaction absorption component field (recording the transaction electricity voucher), and timestamp field (recording the time the status result was generated), thus forming a complete surplus electricity absorption status record, which is used to report to the power grid dispatching system or as a basis for internal optimization analysis of the virtual power plant.

[0109] It is understood that those skilled in the art can combine various implementation methods in the above embodiments under the guidance of the above examples to obtain technical solutions with multiple implementation methods.

[0110] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for hierarchical consumption of new energy surplus power based on a virtual power plant, characterized in that, The virtual power plant central controller cooperates with a load side terminal, an energy storage cluster terminal and a demand response aggregation terminal to perform, including: The load side terminal uploads a response feature vector to the virtual power plant central controller; the energy storage cluster terminal uploads an adjustment feature matrix to the virtual power plant central controller; and the demand response aggregation terminal uploads a transaction feature tensor to the virtual power plant central controller. The virtual power plant central controller executes based on the duration of the new energy output prediction deviation: When the duration is in a first preset time range, an interrupt instruction package is generated based on the response feature vector and is sent to the load side terminal; When the duration is in a second preset time range, a charging and discharging parameter set is generated based on the adjustment feature matrix and is sent to the energy storage cluster terminal; When the duration is in a third preset time range, a transaction contract frame is generated based on the transaction feature tensor and is sent to the demand response aggregation terminal; The first preset time range The second preset time range The third preset time range; the duration is the cumulative time that the difference between the actual power and the predicted power continuously exceeds the set threshold value After the load side terminal, the energy storage cluster terminal and the demand response aggregation terminal perform the corresponding operations of the interrupt instruction package, the charging and discharging parameter set or the transaction contract frame, they respectively feed back the consumption confirmation signal to the central controller; The central controller generates a surplus power consumption state result based on the received consumption confirmation signal.

2. The method of claim 1, wherein the method is characterized in that, The response feature vector includes maximum interruptable power, response delay upper limit and switch state change amount.

3. The method of claim 2, wherein the method further comprises: The interrupt instruction package is generated based on the response feature vector and is sent to the load side terminal, including: Based on the power value of the current surplus power, the maximum interruptable power and the switch state change amount, the total amount of interrupted load is calculated through the switch state adaptive coefficient; The predefined interrupt operation instruction code, the total amount of interrupted load and the transaction identifier generated based on the current timestamp are packaged into an interrupt instruction package; The interrupt instruction package is sent to the load side terminal through the first communication protocol.

4. The method of claim 1, wherein the method is characterized in that, The adjustment feature matrix includes average state of charge, maximum charging power, maximum discharging power, remaining chargeable amount, remaining dischargeable amount and power adjustment response time.

5. The method of claim 4, wherein the method further comprises: The charging and discharging parameter set is generated based on the adjustment feature matrix and is sent to the energy storage cluster terminal, including: Determine the charging and discharging direction according to the sign of the power value of the current surplus power; when the power value is positive, generate a charging instruction code and determine a reference power in combination with the maximum charging power; when the power value is negative, generate a discharging instruction code and determine a reference power in combination with the maximum discharging power; Based on the remaining chargeable amount or the remaining dischargeable amount, the reference power and the duration of the new energy output prediction deviation, the power retention time is calculated; Taking the average state of charge in the adjustment feature matrix as the center value, the state of charge safety boundary in the preset range is generated; The charging instruction code, the discharging instruction code, the reference power, the power retention time and the state of charge safety boundary are packaged into a charging and discharging parameter set; The charging and discharging parameter set is sent to the energy storage cluster terminal through the second communication protocol.

6. The method of claim 1, wherein the method further comprises: The transaction feature tensor includes power dimension parameters, cost dimension parameters and time window dimension parameters.

7. The method of claim 6, wherein the method further comprises: The transaction contract frame is generated based on the transaction feature tensor and is sent to the demand response aggregation terminal, including: A product of a duration of a new energy output prediction deviation and an average power is taken as total electricity to be traded; According to the total electricity to be traded and lower and upper limit values of tradable electricity of each trading period in a transaction feature tensor, a trading period with a highest electricity matching degree is selected and a corresponding trading period identifier is obtained; Unit electricity transmission cost coefficients of all cross-regional power transmission channels in the transaction feature tensor are compared, and an identifier of a channel corresponding to a minimum value is taken as a lowest cost channel identifier; A predefined transaction instruction code, the total electricity to be traded, the selected trading period identifier, the lowest cost channel identifier and a current timestamp are packaged into a transaction contract frame; The transaction contract frame is sent to a demand response aggregation terminal through a third communication protocol.

8. The method of claim 1, wherein the method further comprises: The load side terminal, the energy storage cluster terminal and the demand response aggregation terminal respectively feed back consumption confirmation signals to the central controller after performing operations corresponding to the interruption instruction package, the charging and discharging parameter set or the transaction contract frame, and the operations include: The consumption confirmation signal fed back by the load side terminal contains an actual interrupted load amount; The consumption confirmation signal fed back by the energy storage cluster terminal contains a charging and discharging power adjustment amount; The consumption confirmation signal fed back by the demand response aggregation terminal contains a transaction electricity amount certificate.

9. The method of claim 1, wherein the method further comprises: The central controller generates a surplus electricity consumption state result based on the received consumption confirmation signals, and the operations include: The actual interrupted load amount fed back by the load side terminal, the charging and discharging power adjustment amount fed back by the energy storage cluster terminal and the transaction electricity amount certificate fed back by the demand response aggregation terminal are summed to obtain an actual total consumption electricity amount; A structured consumption state result containing the actual total consumption electricity amount, consumption components of each terminal and a timestamp is constructed.

10. The method of claim 9, wherein the method further comprises: The structured consumption state result includes a total consumption electricity amount field, a load consumption component field, an energy storage consumption component field, a transaction consumption component field and a timestamp field.

Images