Optimal scheduling methods for demand response resources, computer equipment and media

By establishing objective functions and constraints, and combining predicted electricity prices and load fluctuation characteristics to optimize the dispatch capacity of demand response resources, the problem of inaccurate dispatch capacity values ​​in existing technologies has been solved. This enables precise dispatch of the power grid during periods of high electricity load and peak electricity prices, thereby improving the dispatch accuracy and economy of the power system.

CN123121689APending Publication Date: 2026-11-17SHENZHEN POWER SUPPLY BUREAU
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611254305.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-18
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In existing technologies, the capacity value for demand response resources is not accurately determined, resulting in inaccurate scheduling by the power grid during periods of high electricity load and peak electricity prices.

Method used

By establishing an objective function and constraints, and combining predicted electricity prices, total load, unit call cost, load fluctuation auxiliary variables, and initial startup state variables, the call capacity of scheduling demand response resources is optimized. A mixed-integer linear programming model is used to solve the problem, ensuring that the economically optimal call capacity allocation is achieved while satisfying the system safety boundary.

Benefits of technology

It improves the accuracy of demand response resource allocation capacity values, ensuring more precise grid dispatch during periods of high electricity load and peak electricity prices, and reducing the uncertainty risk of the power system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123121689A_ABST
    Figure CN123121689A_ABST
Patent Text Reader

Abstract

This application relates to an optimized scheduling method, computer equipment, and medium for demand response resources. The method includes: establishing an objective function and constraints based on the unit call cost of each demand response resource, the predicted electricity price for each time period D+1, the predicted total load for each time period D+1, the call capacity variable for each demand response resource in each time period D+1, the load fluctuation auxiliary variable for each time period D+1, the load deficit variable for each time period D+1, and the initial startup state variable for each demand response resource in each time period D+1; solving the objective function under the constraints to obtain the call capacity value, load fluctuation auxiliary value, load deficit value, and startup state value for each demand response resource in each time period D+1. This method can improve the accuracy of the determined call capacity value of the demand response resource.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of resource scheduling and optimization technology, and in particular to an optimized scheduling method, computer equipment, and medium for demand response resources. Background Technology

[0002] As the penetration rate of new energy sources continues to increase, the peak-valley difference in power system load is widening. The grid side needs to accurately predict peak electricity price periods and high load periods within the 24-hour period specified in D+1, in order to optimize the scheduling of demand response resources. Here, D represents the current day (today), and D+1 represents the day after today (tomorrow).

[0003] In related technologies, the peak electricity price period and high electricity load period within 24 hours in D+1 are often determined based on experience. Based on the peak electricity price period and high electricity load period within 24 hours in D+1, the call capacity value of the demand response resource is determined according to the priority of the callable resources, so as to realize the call of the callable resources.

[0004] However, the above solution has the problem that the defined demand response resource call capacity value is inaccurate. Summary of the Invention

[0005] Based on this, this application provides an optimized scheduling method, computer device, and medium for demand response resources, which can improve the accuracy of the determined call capacity value of demand response resources.

[0006] Firstly, this application provides an optimized scheduling method for demand response resources, the method comprising:

[0007] Based on historical electricity supply and demand price statistics for the target area and power supply load impact data for each time period in D+1, at least one demand response resource is identified.

[0008] The objective function is established based on the unit call cost of each demand response resource, the predicted electricity price of each time period in D+1, the predicted total load of each time period in D+1, the call capacity variable of each demand response resource in D+1, the load fluctuation auxiliary variable of each time period in D+1, the shortfall load variable of each time period in D+1, and the initial start-up state variable of each demand response resource in D+1.

[0009] Establish constraints for the call capacity variable, load fluctuation auxiliary variable, shortfall load variable, and initial startup state variable of each demand response resource in each time period of D+1.

[0010] Solve the objective function under the constraints to obtain the call capacity value, load fluctuation auxiliary value, deficit load value, and start-up status value of each demand response resource in each time period of D+1.

[0011] In some embodiments, at least one demand response resource is determined based on historical electricity supply and demand price statistics for the target area and power supply load impact data for the target area in each time period of D+1, including:

[0012] Based on historical electricity supply and demand price statistics and power supply load impact data, the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1 are determined.

[0013] Based on the predicted electricity price, predicted total load, predicted renewable energy output, and rated aggregate capacity of each available resource in each period of D+1, determine the baseline aggregate capacity of each available resource in each period of D+1.

[0014] Based on the baseline aggregate capacity of each callable resource in each time period of D+1, at least one demand response resource is determined from a variety of callable resources.

[0015] In some embodiments, the baseline aggregate capacity of each callable resource in each period of D+1 is determined based on the predicted electricity price, predicted total load, predicted renewable energy output, and the rated aggregate capacity of each callable resource in each period of D+1, including:

[0016] Based on the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1, the comprehensive risk score for each period of D+1 is determined.

[0017] Based on the comprehensive risk score of each time period in D+1 and the rated aggregate capacity of each callable resource in each time period in D+1, determine the baseline aggregate capacity of each callable resource in each time period in D+1.

[0018] In some embodiments, a comprehensive risk score for each period of D+1 is determined based on the predicted electricity price, predicted total load, and predicted renewable energy output for each period, including:

[0019] Based on the predicted electricity prices for each period of D+1, determine the price risk score for each period of D+1;

[0020] Based on the predicted total load for each time period of D+1, determine the total load risk score for each time period of D+1;

[0021] Based on the predicted total load and new energy output for each period of D+1, determine the predicted net load for each period of D+1.

[0022] Based on the predicted net load for each period of D+1, determine the net load risk score for each period of D+1.

[0023] The price risk score, total load risk score, and net load risk score for each period in D+1 are weighted and summed to obtain the comprehensive risk score for each period in D+1.

[0024] In some embodiments, the baseline aggregate capacity of each callable resource in each period of D+1 is determined based on the comprehensive risk score of each period in D+1 and the rated aggregate capacity of each callable resource in each period of D+1, including:

[0025] Determine the demand response call load for each period of D+1 based on the maximum callable load for a single period, the comprehensive risk score for each period of D+1, and the predicted total load for each period of D+1.

[0026] The baseline aggregate capacity of each callable resource in each time period of D+1 is determined based on the rated aggregate capacity of each callable resource in each time period of D+1, the proportion of the load of each callable resource to the load of all callable resources, the predicted total load in each time period of D+1, and the availability coefficient of each callable resource in each time period of D+1.

[0027] In some embodiments, at least one demand response resource is determined from a plurality of callable resources based on the baseline aggregate capacity of each callable resource in each time period of D+1, including:

[0028] Based on the baseline aggregate capacity of each callable resource in each time period of D+1, the maximum callable ratio of each callable resource in each time period of D+1, and the demand response call load of each callable resource in each time period of D+1, determine the capacity coverage of each callable resource in each time period of D+1.

[0029] Based on the capacity coverage of each callable resource in each time period of D+1, the availability coefficient of each callable resource in each time period of D+1, the historical response reliability parameter of each callable resource, the normalized unit call cost of each callable resource, and the call constraint mismatch penalty parameter of each callable resource in each time period of D+1, determine the call adaptation parameters of each callable resource in each time period of D+1.

[0030] Based on the call adaptation parameters and preset call adaptation thresholds of each callable resource in each time period of D+1, at least one demand response resource is determined from a variety of callable resources.

[0031] In some embodiments, an objective function is established based on the unit call cost of each demand response resource, the predicted electricity price for each time period of D+1, the predicted total load for each time period of D+1, the call capacity variable for each demand response resource for each time period of D+1, the load fluctuation auxiliary variable for each time period of D+1, the deficit load variable for each time period of D+1, and the initial startup state variable for each demand response resource for each time period of D+1, including:

[0032] Based on the predicted electricity price for each period of D+1, the predicted total load for each period of D+1, and the call capacity variables for each demand response resource for each period of D+1, determine the electricity purchase cost item for D+1.

[0033] Based on the unit call cost of each demand response resource and the call capacity variable of each demand response resource in each time period of D+1, determine the D+1 resource call cost item;

[0034] The load fluctuation penalty item for D+1 is determined based on the fluctuation penalty coefficient and the auxiliary variables of load fluctuation for each period of D+1.

[0035] Based on the demand non-metance penalty coefficient and the deficit load variables for each period of D+1, determine the D+1 deficit load penalty item;

[0036] Based on the resource activation penalty coefficient and the initial activation state variables of each time period in D+1 for each demand response resource, determine the D+1 resource activation penalty item;

[0037] The objective function is determined based on the D+1 electricity purchase cost item, the D+1 resource allocation cost item, the D+1 load fluctuation penalty item, the D+1 shortfall load penalty item, and the D+1 resource activation penalty item.

[0038] In some embodiments, the constraints include at least one of the following:

[0039] The call capacity variable for each type of demand response resource in D+1 time period is between the minimum callable capacity value and the maximum callable capacity value;

[0040] The initial startup status variable of each time period of D+1 for each demand response resource is greater than or equal to the difference between the call status value of each time period and the call status value of the corresponding previous time period; the initial startup status variable of each time period of D+1 for each demand response resource is less than or equal to the call status value of each time period; the initial startup status variable of each time period of D+1 for each demand response resource is less than or equal to the difference between 1 and the call status value of the corresponding previous time period.

[0041] The sum of the total call capacity variable and the deficit load variable of all demand response resources in each time period of D+1 is greater than or equal to the demand response call load in each time period of D+1.

[0042] The call capacity variable for each demand response resource in each time period of D+1 is less than or equal to the sum of the demand response call load and the additional load in each time period of D+1.

[0043] When the call status value is 1 in any time period, the call status value between any time period and the target time period is 1, and the duration between the target time period and any time period is the shortest call duration for the resource.

[0044] The initial startup status value is less than or equal to 1 within a preset number of consecutive time periods;

[0045] The number of times the initial startup status variable of each demand response resource in D+1 is 1 is less than or equal to the maximum number of calls for each demand response resource;

[0046] The total call capacity variable for all demand response resources in each time period of D+1 is between the minimum and maximum declared capacity values.

[0047] The auxiliary variables for load fluctuations in each time period of D+1 are greater than or equal to the absolute value of the difference between the corrected load of each time period and the corrected load of the corresponding previous time period.

[0048] In a second aspect, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method of any one of the first aspects.

[0049] Thirdly, this application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any one of the first aspects.

[0050] In the technical solution provided in this application, the predicted electricity price, predicted total load, unit dispatch cost and dispatch capacity variables, load fluctuation auxiliary variables, shortage load variables, and initial start-up state variables for each time period of D+1 are all incorporated into a unified objective function and constraint system. This allows the variables to mutually balance and correct each other during the solution process. The load fluctuation auxiliary variable can reflect the trend characteristics of net load changes in adjacent time periods, and the shortage load variable can capture the risk of power shortage caused by fluctuations in new energy output or a sharp increase in load. Both of these variables, along with the dispatch capacity variable, are subject to constraints, prompting the solution process to seek the most economically optimal dispatch capacity allocation while satisfying the system safety boundary. Since the solution process is simultaneously guided by the electricity price signal and constrained by the load fluctuation characteristics, the obtained dispatch capacity value is no longer determined based on fixed time period rules or static thresholds, but is dynamically adapted to the actual prediction scenarios of each time period of D+1. This helps to improve the accuracy of the determined dispatch capacity value of the demand response resources. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 A flowchart illustrating an optimized scheduling method for demand response resources provided in some embodiments;

[0053] Figure 2 A flowchart illustrating a method for determining at least one demand response resource provided in some embodiments;

[0054] Figure 3 A flowchart illustrating a method for determining the baseline aggregate capacity of each callable resource in each time period of D+1, provided for some embodiments;

[0055] Figure 4 A flowchart illustrating a method for selecting at least one demand response resource as provided in some embodiments;

[0056] Figure 5 A flowchart illustrating the method for establishing the objective function provided in some embodiments;

[0057] Figure 6 A schematic diagram of load curves before and after demand response resource optimization provided for some embodiments;

[0058] Figure 7 A schematic diagram of the structure of an optimized scheduling device for demand response resources provided in some embodiments;

[0059] Figure 8 A schematic diagram of the structure of a computer device provided for some embodiments. Detailed Implementation

[0060] The embodiments of the technical solution of this application will now be described in detail with reference to the accompanying drawings. These embodiments are only used to more clearly illustrate the technical solution of this application and are therefore merely examples, and should not be used to limit the scope of protection of this application.

[0061] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0062] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, "multiple groups" means two or more, and "each" means each of the multiple, unless otherwise explicitly defined.

[0063] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0064] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0065] Unless otherwise specified, the order of execution steps in the embodiments of this application is not limited. It should also be noted that any step in the embodiments of this application can be executed independently, that is, the execution of any step in the above embodiments can be performed without depending on the execution of other steps.

[0066] The various steps in the methods described in this application embodiment can be applied to a computer device or a processor, and the processor can be applied to a computer device. Exemplarily, a computer device may include at least one or a combination of at least two of the following: a server, a mobile phone, a tablet computer, a computer with transceiver capabilities, a handheld computer, a desktop computer, a personal digital assistant, a portable media player, a smart speaker, a navigation device, a smartwatch, smart glasses, a smart necklace, and other wearable devices, a pedometer, a digital TV, a virtual reality (VR) device, an augmented reality (AR) device, or devices used in industrial control.

[0067] Figure 1 A flowchart illustrating an optimized scheduling method for demand response resources provided in some embodiments, such as... Figure 1 As shown, the method includes the following steps:

[0068] S101. Based on the historical electricity supply and demand price statistics of the target area and the power supply load impact data of the target area in each time period of D+1, determine at least one demand response resource.

[0069] In this embodiment, the method can be executed at any point in the last period of each day. For any given day, it can be divided into 96 periods, each lasting 15 minutes.

[0070] The target region can be any area where resources can be invoked in response to demand. For example, the target region could be a province, a city, or a district / county.

[0071] Demand response resources refer to a collection of flexible electricity resources on the power user side that have the ability to regulate their electricity consumption and can adjust their own power consumption according to the grid dispatch instructions in order to participate in the grid supply and demand balance regulation. They are the callable adjustment carriers for the grid to fill the power supply gap during peak, high price and insufficient output of new energy sources.

[0072] Electricity supply and demand price represents renewable energy output, total grid load, and electricity price. Historical electricity supply and demand price statistics include historical renewable energy output statistics, historical total grid load statistics, and historical electricity price statistics. These statistics include at least one of the following: renewable energy output statistics for each time period on the current day (denoted by D), renewable energy output statistics for each time period from the first historical day to the current day, renewable energy output statistics for each time period from the second historical day to the current day, and renewable energy output statistics for each time period from the third historical day to the current day. The first historical day is D-2, the second historical day is D-6, and the third historical day is D-13. The statistical data in this application embodiment may include at least one of the following: average, maximum, minimum, median, mode, etc. For example, the renewable energy output statistics for a time period are the average of all renewable energy output obtained for that time period.

[0073] Power supply load impact data represents data affecting power supply and / or load. For example, power supply load impact data includes at least one of the following: power supply side operation data, meteorological data, calendar data, and fuel cost data. For example, power supply side operation data includes at least one of the following: local thermal power output, output of different types of power generation, and tie-line power, etc. For example, meteorological data may include at least one of the following: temperature, wind speed, wind direction, precipitation, and irradiance. For example, calendar data may include at least one of the following: date, day of the week, whether it is a weekend, whether it is a holiday, whether it is a few days before a holiday, whether it is a few days after a holiday, etc. For example, fuel cost data may include coal price data for different calorific value grades.

[0074] In some embodiments, S101 may include: determining the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1 based on historical electricity supply and demand price statistics for the target area and power supply load impact data for the target area in each period of D+1; and determining at least one demand response resource based on the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1.

[0075] For example, the statistical data of historical renewable energy output and the power supply load impact data of the target area in each time period of D+1 are vectorized to obtain a first vector; the statistical data of historical total grid load and the power supply load impact data of the target area in each time period of D+1 are vectorized to obtain a second vector; and the statistical data of historical electricity prices and the power supply load impact data of the target area in each time period of D+1 are vectorized to obtain a third vector. Furthermore, an electricity price prediction model, a total load prediction model, and a renewable energy output prediction model can be trained; the first vector is input into the electricity price prediction model to output the predicted electricity price for each time period of D+1; the second vector is input into the total load prediction model to output the predicted total load for each time period of D+1; and the third vector is input into the renewable energy output prediction model to output the predicted renewable energy output for each time period of D+1.

[0076] For example, any one of the electricity price forecasting model, total load forecasting model, and new energy output forecasting model can adopt a gradient boosting tree model, random forest model, XGBoost model, LightGBM model, long short-term memory network model, temporal convolutional network model, Transformer model, Ridge regression model, etc.

[0077] S102. Establish the objective function based on the unit call cost of each demand response resource, the predicted electricity price of each time period in D+1, the predicted total load of each time period in D+1, the call capacity variable of each demand response resource in D+1, the load fluctuation auxiliary variable of each time period in D+1, the shortfall load variable of each time period in D+1, and the initial start-up state variable of each demand response resource in D+1.

[0078] Initial startup status variables can refer to startup indicator variables.

[0079] Unit call cost refers to the comprehensive economic cost per unit of adjustable power capacity and per unit of response time for a single type of demand response resource.

[0080] The call capacity variable is a decision variable characterizing the call capacity of demand response resources during a corresponding time period. The load fluctuation auxiliary variable is a linearized auxiliary variable characterizing the magnitude of load changes between adjacent time periods, used to quantify the degree of load fluctuation. The load shortfall variable is a slack variable characterizing the unmet portion of demand response call demand, used to measure the demand gap. The initial startup status variable is a binary decision variable characterizing the transition of demand response resources from an outage state to a call state during a corresponding time period. For example, a value of 1 for the initial startup status variable indicates that startup has just occurred, and a value of 0 indicates that startup has not occurred.

[0081] S103. Establish constraints for the call capacity variable, load fluctuation auxiliary variable, shortfall load variable, and initial startup state variable of each demand response resource in each time period of D+1.

[0082] The constraints may include the call capacity variable, load fluctuation auxiliary variable, deficit load variable, and initial startup state variable for each demand response resource in each time period of D+1.

[0083] S104. Solve the objective function under the constraints to obtain the call capacity value, load fluctuation auxiliary value, deficit load value, and start-up status value of each demand response resource in each time period of D+1.

[0084] Among them, the call capacity value, load fluctuation auxiliary value, shortage load value, and start-up status value of each demand response resource in each D+1 time period are respectively the values ​​of the call capacity variable, load fluctuation auxiliary variable, shortage load variable, and initial start-up status variable of each demand response resource in each D+1 time period.

[0085] For example, the objective function under the constraints can be a mixed-integer linear programming model, which can be solved using any method for solving mixed-integer linear programming models. For example, CPLEX, Gurobi, CBC, or other mixed-integer programming solvers can be used.

[0086] In the technical solution provided in this application, the predicted electricity price, predicted total load, unit dispatch cost and dispatch capacity variables, load fluctuation auxiliary variables, shortage load variables, and initial start-up state variables for each time period of D+1 are all incorporated into a unified objective function and constraint system. This allows the variables to mutually balance and correct each other during the solution process. The load fluctuation auxiliary variable can reflect the trend characteristics of net load changes in adjacent time periods, and the shortage load variable can capture the risk of power shortage caused by fluctuations in new energy output or a sharp increase in load. Both of these variables, along with the dispatch capacity variable, are subject to constraints, prompting the solution process to seek the most economically optimal dispatch capacity allocation while satisfying the system safety boundary. Since the solution process is simultaneously guided by the electricity price signal and constrained by the load fluctuation characteristics, the obtained dispatch capacity value is no longer determined based on fixed time period rules or static thresholds, but is dynamically adapted to the actual prediction scenarios of each time period of D+1. This helps to improve the accuracy of the determined dispatch capacity value of the demand response resources.

[0087] In some embodiments, an optimized scheduling method for demand response resources may include the following steps: determining the predicted electricity price, predicted total load, and predicted renewable energy output for each time period D+1 based on historical electricity supply and demand price statistics for a target area and power supply load impact data for the target area in each time period D+1; determining the comprehensive risk score for each time period D+1 based on the predicted electricity price, predicted total load, and predicted renewable energy output for each time period D+1; determining at least one demand response resource from a variety of callable resources based on the comprehensive risk score for each time period D+1 and the rated aggregate capacity of each callable resource in each time period D+1; and determining the unit call cost of each demand response resource, the predicted electricity price for each time period D+1, the predicted total load for each time period D+1, and the predicted renewable energy output for each demand response resource. A target function is established by defining the call capacity variable, load fluctuation auxiliary variable, and load shortfall variable for each time period (D+1) of the response resource, as well as the initial startup state variable for each time period (D+1) of the demand response resource. Constraints are then established for these variables. The objective function under these constraints is solved to obtain the call capacity value, load fluctuation auxiliary value, load shortfall value, and startup state value for each time period (D+1) of the demand response resource.

[0088] For example, determining the at least one demand response resource from a plurality of callable resources based on the comprehensive risk score for each time period of D+1 and the rated aggregate capacity of each callable resource for each time period of D+1 includes: determining the baseline aggregate capacity of each callable resource for each time period of D+1 based on the comprehensive risk score for each time period of D+1 and the rated aggregate capacity of each callable resource for each time period of D+1; and determining the at least one demand response resource from a plurality of callable resources based on the baseline aggregate capacity of each callable resource for each time period of D+1.

[0089] It should be noted that the explanations of the same steps in the following text can be the same.

[0090] Figure 2 A flowchart illustrating a method for determining at least one demand response resource provided in some embodiments, such as... Figure 2 As shown, this method is an exemplary interpretation of S101, which includes steps S1011-S1013.

[0091] S1011. Based on historical electricity supply and demand price statistics and power supply load impact data, determine the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1.

[0092] The predicted electricity price, predicted total load, and predicted renewable energy output for each time period in D+1 can be referred to the description in the above embodiment, and will not be repeated here.

[0093] For example, given the predicted electricity price, predicted total load, and predicted renewable energy output for each time period of D+1, a comprehensive risk score for each time period of D+1 can be determined based on these factors. This allows for the determination of the demand response load for each time period of D+1 based on the maximum callable load for a single time period, the comprehensive risk score for each time period of D+1, and the predicted total load for each time period of D+1. Furthermore, based on the rated aggregate capacity of each callable resource in each time period of D+1, the proportion of the load of each callable resource to the total load of all callable resources, the predicted total load for each time period of D+1, and the... The availability coefficient of each callable resource in each time period of D+1 is described, and the baseline aggregate capacity of each callable resource in each time period of D+1 is determined. Based on the baseline aggregate capacity of each callable resource in each time period of D+1, at least one demand response resource is determined from multiple callable resources. Given the demand response resource, an objective function and constraints can be established, and the objective function under the constraints is solved to obtain the call capacity value, load fluctuation auxiliary value, deficit load value, and start-up status value of each demand response resource in each time period of D+1. The methods for determining the comprehensive risk score, demand response call load, baseline aggregate capacity, and at least one demand response resource will be explained below and will not be repeated here.

[0094] S1012. Based on the predicted electricity price, predicted total load, predicted renewable energy output, and rated aggregate capacity of each available resource in each period of D+1, determine the baseline aggregate capacity of each available resource in each period of D+1.

[0095] The callable resources are all user-side flexible load clusters within the target area that possess the hardware conditions for power regulation and have completed demand response registration. For example, callable resources may include at least one of the following: industrial interruptible load resources in at least one sub-region, commercial air conditioning flexible load resources in at least one sub-region, and electric vehicle flexible charging load resources in at least one sub-region. Wherein, at least one sub-region is a sub-region within the target area.

[0096] The rated aggregated capacity is the theoretical maximum callable power output that can be achieved after aggregating and summing the rated aggregated capacity of each callable resource. The baseline aggregated capacity is the actual maximum callable power output obtained by adjusting the rated aggregated capacity of each callable resource.

[0097] S1013. Based on the baseline aggregated capacity of each callable resource in each time period of D+1, determine at least one demand response resource from a variety of callable resources.

[0098] In the technical solution provided in this application, predicted electricity prices, predicted total load, and predicted renewable energy output jointly participate in the determination process of the baseline aggregated capacity of each callable resource. This process uses the rated aggregated capacity as the upper limit reference for the adjustable range of each resource, so that the baseline aggregated capacity can reflect the theoretical callable capacity of different callable resources under the combined effects of electricity price guidance, load level, and renewable energy output fluctuations. Based on this, demand response resources are screened from multiple callable resources according to the baseline aggregated capacity, so that the screening results are related to the actual supply and demand matching demand under the predicted scenarios of each time period in D+1. Compared with the method of directly screening based on historical time period averages or fixed resource priorities, this solution helps to reduce the adaptation deviation between demand response resources and the actual callable scenarios of D+1 by introducing multi-dimensional prediction information of each time period in D+1, thereby improving the accuracy of the determined at least one demand response resource.

[0099] Figure 3 A flowchart illustrating the method for determining the baseline aggregate capacity of each callable resource in each time period of D+1, as provided in some embodiments, is shown below. Figure 3 As shown, this method is an exemplary interpretation of S1012, which includes steps S1012A to S1012B.

[0100] S1012A. Based on the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1, determine the comprehensive risk score for each period of D+1.

[0101] In some embodiments, S1012A may include: determining the price risk score for each time period of D+1 based on the predicted electricity price for each time period; determining the total load risk score for each time period of D+1 based on the predicted total load for each time period of D+1; determining the predicted net load for each time period of D+1 based on the predicted total load and renewable energy output for each time period of D+1; determining the net load risk score for each time period of D+1 based on the predicted net load for each time period of D+1; and performing a weighted summation of the price risk score, the total load risk score, and the net load risk score for each time period of D+1 to obtain the comprehensive risk score for each time period of D+1.

[0102] In the above scheme, the predicted electricity price, predicted total load, and predicted renewable energy output correspond to three independent dimensions for calculation: price risk score, total load risk score, and net load risk score, respectively. The net load risk score is determined based on the difference between the predicted total load and the predicted renewable energy output, which reflects the actual power supply pressure that the grid needs to bear after fluctuations in renewable energy output. By weighted summing of the risk scores of the above three dimensions, the comprehensive risk score can simultaneously characterize the superposition degree of the three types of risks—high electricity prices, peak load, and sharp increase in net load—in each time period of D+1. Since the weights of each dimension in the weighted summation can be adjusted according to the actual operational needs of the target area, the comprehensive risk score has configurable sensitivity to different types of risks. This scheme compresses multi-dimensional day-ahead forecast information into a single comprehensive risk index, which helps to provide a unified quantitative basis for identifying the time periods for subsequent demand response resource allocation.

[0103] The following explains the formulas for determining the price risk score, total load risk score, and net load risk score:

[0104] Price risk score for time period t (any time period in D+1) pass Determined; among them, This represents the predicted electricity price for time period t in D+1; This indicates the price risk trigger threshold, compared to a pre-set value; This indicates the upper limit of price risk, which is a pre-set value. It represents a very small positive number.

[0105] Total load risk score for time period D+1 pass Determined; among them, This represents the predicted total load for time period t in D+1; This represents the total load risk threshold, which is a preset value. This indicates the maximum risk level for the total load, which is a preset value. It represents a very small positive number.

[0106] For example, determining the predicted net load for each period of D+1 based on the predicted total load and renewable energy output for each period of D+1 may include: determining the difference between the predicted total load for each period of D+1 and the renewable energy output for each period of D+1 as the predicted net load for each period of D+1.

[0107] Net load risk score for time period t (D+1) pass Determined; among them, This represents the predicted net load for time period t in D+1; This indicates the net load risk trigger threshold, which is a preset value. This indicates the upper limit of net load risk, which is a preset value. It represents a very small positive number.

[0108] The comprehensive risk score for time period t in D+1 It can be done Determined; among them, , , These represent the price risk weight, load risk weight, and net load risk weight, respectively, all of which are pre-set values. For example, w p =0.35、w L =0.35、w n =0.30.

[0109] S1012B. Based on the comprehensive risk score of each time period in D+1 and the rated aggregate capacity of each callable resource in each time period in D+1, determine the baseline aggregate capacity of each callable resource in each time period in D+1.

[0110] In some embodiments, S1012B may include: determining the demand response call load for each period of D+1 based on the maximum callable load for a single period, the comprehensive risk score for each period of D+1, and the predicted total load for each period of D+1; and determining the baseline aggregate capacity for each callable resource in each period of D+1 based on the rated aggregate capacity for each callable resource in each period of D+1, the proportion of the load of each callable resource to the load of all callable resources, the predicted total load for each period of D+1, and the availability coefficient for each callable resource in each period of D+1.

[0111] In the above scheme, the determination of demand response call load introduces the maximum callable load for a single time period as an upper limit constraint, and uses the comprehensive risk score and predicted total load of each time period in D+1 as the adjustment basis, so that the demand response call load has differentiated values ​​under different comprehensive risk levels. Based on this, the baseline aggregated capacity of each callable resource is obtained through the joint calculation of four types of parameters: rated aggregated capacity, load percentage, predicted total load, and availability coefficient. The rated aggregated capacity reflects the inherent adjustable capacity upper limit of each type of resource, the load percentage reflects the relative scale of each type of resource among all callable resources, and the availability coefficient quantitatively characterizes whether each type of resource is available for call in each time period in D+1. After the above four types of parameters are used in the calculation, the resulting baseline aggregated capacity can reflect the theoretical callable level of each type of callable resource in each time period in D+1 under the joint constraints of its own physical characteristics, relative scale, and availability status, which helps to improve the matching degree between the baseline aggregated capacity and the actual callable capacity of each type of resource.

[0112] Demand response load refers to the target adjustment power required by the grid side to mitigate the pressure of high prices, high loads, and high net loads.

[0113] For example, if the overall risk score is less than a preset overall risk threshold at any time during period D+1. In the case of a situation where the demand response call load for any period of D+1 is determined to be 0; and the comprehensive risk score for any period of D+1 is greater than or equal to the preset comprehensive risk threshold. In this case, the demand response call load for each period of D+1 is determined based on the maximum callable load for a single period, the comprehensive risk score for each period of D+1, and the predicted total load for each period of D+1.

[0114] Demand response call load for time period D+1 (t-th time) It can be determined in the following ways: ;in, This represents the overall risk score for time period t in D+1; This represents the maximum callable load for a single time period, and the load for different time periods. same; This represents the risk mapping coefficient, which is a pre-set value. For example, The value is a decimal between 0.03 and 0.08; This represents the predicted total load for time period t in the D+1 period.

[0115] In some embodiments, with resource ledger data available, the baseline aggregated capacity B of the r-th callable resource in the t-th time period D+1 is... r,t Determined in the following manner: B r,t =B r base ×a r,t Among them, B r base This represents the rated aggregate capacity for the t-th time period recorded in the resource ledger; a r,t This represents the availability coefficient of the r-th callable resource in the t-th time period of D+1, and its value is a decimal of 0, 1, or 0-1.

[0116] In some embodiments, without resource ledger data, the baseline aggregated capacity B of the r-th callable resource in the t-th time period of D+1. r,t Determined in the following manner: B r,t =ω r ×L̂ t ×a r,t , where ω r This represents the proportion of the load of the r-th callable resource to the load of all callable resources; a represents the predicted total load for time period t in D+1; r,t This represents the availability coefficient of the r-th callable resource in the t-th time period of D+1, and its value is a decimal of 0, 1, or 0-1.

[0117] In some embodiments, where some callable resources have resource ledger data and some callable resources do not, the baseline aggregated capacity B of the r-th callable resource in the t-th time period D+1 is... r,t Determined in the following ways: .

[0118] In the technical solution provided in this application, predicted electricity price, predicted total load, and predicted renewable energy output are first involved in the determination of a comprehensive risk score. This score quantifies the degree of superposition of three types of risks—high electricity price, peak load, and sharp increase in net load—during each period of D+1. Based on this, the baseline aggregated capacity of each callable resource is adjusted using the rated aggregated capacity as the physical upper limit, combined with the comprehensive risk score for each period of D+1. This process allows the value of the baseline aggregated capacity to dynamically change with the difference in the comprehensive risk level of each period; that is, the same callable resource can correspond to different baseline aggregated capacities in different risk periods. Compared with methods that use fixed-period rules or uniform proportions to determine the baseline aggregated capacity, this solution, by introducing a comprehensive risk score reflecting the combined impact of electricity price, load, and renewable energy output into the baseline calculation, helps to establish a correlation between the baseline aggregated capacity and the actual call demand in each period of D+1, thereby improving the distinguishability and scenario adaptability of the determined baseline aggregated capacity across different periods.

[0119] Figure 4 A flowchart illustrating a method for selecting at least one demand response resource provided in some embodiments, such as Figure 4 As shown, this method is an exemplary interpretation of S1013, which includes steps S1013A-S1013C.

[0120] S1013A. Determine the capacity coverage of each callable resource in each time period of D+1 based on the baseline aggregate capacity of each callable resource in each time period of D+1, the maximum callable ratio of each callable resource in each time period of D+1, and the demand response call load of each callable resource in each time period of D+1.

[0121] For example, the capacity coverage of the r-th callable resource in the t-th time period of D+1. It can be determined in the following ways: ;in, This represents the maximum callable proportion of the r-th type of resource in the t-th time period of D+1. This represents the baseline aggregate capacity of the r-th callable resource in the t-th time period of D+1. This represents the demand response call load during time period t in period D+1. It represents a very small positive number.

[0122] S1013B. Based on the capacity coverage of each callable resource in each time period of D+1, the availability coefficient of each callable resource in each time period of D+1, the historical response reliability parameter of each callable resource, the normalized unit call cost of each callable resource, and the call constraint mismatch penalty parameter of each callable resource in each time period of D+1, determine the call adaptation parameters of each callable resource in each time period of D+1.

[0123] For example, the invocation adaptation parameters of the r-th callable resource in the t-th time period of D+1. It can be determined in the following ways: Where λ1 to λ5 are the fitness weight coefficients, which can be preset values; a represents the capacity coverage of the r-th callable resource in the t-th time period of D+1. r,t This represents the availability coefficient of the r-th callable resource in the t-th time period of D+1, and its value is a decimal among 0, 1, or 0-1; The parameter representing the historical response reliability of the r-th available resource is determined based on historical scheduling data. Represents the normalized unit call cost of the r-th callable resource; This represents the penalty parameter for mismatched call constraints of the r-th callable resource in the t-th time period of D+1. , This represents the unit cost of calling the r-th available resource. This represents the minimum unit cost of calling the r-th available resource. This represents the maximum unit cost of calling the r-th available resource. It represents a very small positive number.

[0124] In some implementations, q r It can be obtained from the ratio of the actual historical response capacity to the planned call capacity, the percentage of on-time responses, or the execution deviation. When historical execution records are missing, default values ​​can be set according to the resource type.

[0125] In some implementations... This is used to characterize whether the r-th type of callable resource has insufficient duration, insufficient startup interval, reaches the daily call limit, or fails to meet the minimum application capacity during time period t. When the parameters of the r-th type of callable resource during time period t all meet the constraints in the embodiments of this application, m can be set... r,t =0, otherwise set it to a positive value or remove the resource directly.

[0126] S1013C. Based on the call adaptation parameters and preset call adaptation thresholds of each callable resource in each time period of D+1, determine at least one demand response resource from a variety of callable resources.

[0127] In some embodiments, S1013C may include: determining the callable resources corresponding to call adaptation parameters that are greater than or equal to a preset call adaptation threshold as demand response resources, thereby obtaining at least one demand response resource.

[0128] In other embodiments, S1013C may include: determining the callable resources corresponding to call adaptation parameters that are greater than or equal to a preset call adaptation threshold as candidate call resources; and determining the candidate call resources as demand response resources when the availability coefficient of the candidate call resources is greater than 0 in the t-th time period of D+1, thereby obtaining at least one demand response resource.

[0129] For example, the preset call adaptation threshold can be set based on the number of callable resources, the call demand, or historical scheduling experience.

[0130] In some embodiments, at least one demand response resource It can be determined in the following ways: ;in, It represents the set of all callable resources. This represents the call adaptation parameters for the r-th callable resource in the t-th time period of D+1. Indicates the preset call adaptation threshold, a r,t This represents the availability coefficient of the r-th callable resource in the t-th time period of D+1, and its value is a decimal of 0, 1, or 0-1.

[0131] The technical solution provided in this application calculates capacity coverage based on the baseline aggregated capacity, maximum callable ratio, and demand response call load for each time period. It further combines availability coefficients, historical response reliability parameters, normalized unit call cost, and call constraint mismatch penalty parameters to obtain call adaptation parameters. Then, based on a preset call adaptation threshold, it filters and determines demand response resources from a variety of callable resources. This overcomes the limitations of filtering based solely on capacity, completing a comprehensive resource evaluation from multiple dimensions: adjustment capability, time period adaptability, fulfillment reliability, economic cost, and operational constraint matching. This method considers the multi-dimensional attributes of various resources. The selected demand response resources not only possess corresponding capacity support capabilities but also adapt to the corresponding time period's operational scenario, offering reasonable call costs and stable response performance. Simultaneously, it eliminates callable resources with insufficient adaptability, reducing the scale of decision variables in subsequent optimization models and lowering the computational load. This provides a more reliable resource foundation for subsequent call optimization, helping to improve the feasibility and overall operational efficiency of the day-ahead demand response scheduling scheme.

[0132] Figure 5 A flowchart illustrating the method for establishing the objective function provided in some embodiments, such as... Figure 5 As shown, this method is an exemplary interpretation of S102, which includes steps S1021-S1026.

[0133] S1021. Determine the electricity purchase cost item D+1 based on the predicted electricity price for each time period of D+1, the predicted total load for each time period of D+1, and the call capacity variables for each demand response resource for each time period of D+1.

[0134] For example, the D+1 electricity purchase cost item can be represented as ,in, ; This represents the predicted electricity price for time period t in time zone D+1. This represents the corrected load for time period t in D+1. Indicates the duration of a single time period, for example, in the embodiments of this application. =0.25h, This represents the predicted total load for time period t in D+1. This represents the call capacity variable for the r-th callable resource in the t-th time period of D+1.

[0135] S1022. Determine the D+1 resource call cost item based on the unit call cost of each demand response resource and the call capacity variable of each demand response resource in each time period of D+1.

[0136] For example, the resource call cost item D+1 can be represented as Where represents the unit call cost of the r-th available resource. This represents the call capacity variable for the r-th callable resource in the t-th time period of D+1. Indicates the duration of a single time period, for example, in the embodiments of this application. =0.25h.

[0137] S1023. Determine the load fluctuation penalty item for D+1 based on the fluctuation penalty coefficient and the auxiliary variables of load fluctuation for each period of D+1.

[0138] For example, the D+1 load fluctuation penalty term can be expressed as ;in, This represents the fluctuation penalty coefficient, which is a preset value. This represents the auxiliary variable for load fluctuation in time period t of D+1.

[0139] S1024. Determine the D+1 load shortfall penalty item based on the demand shortfall penalty coefficient and the load shortfall variables for each time period of D+1.

[0140] For example, the D+1 deficit load penalty term can be expressed as ;in, This indicates the penalty coefficient for unmet demand; it is a pre-set value. This represents the deficit load variable in time period t of D+1.

[0141] S1025. Determine the D+1 resource startup penalty item based on the resource startup penalty coefficient and the initial startup state variables of each time period of D+1 for each demand response resource.

[0142] For example, the D+1 resource startup penalty can be represented as ;in, This represents the resource startup penalty coefficient, which is a pre-set value. This represents the initial startup state variable of the r-th callable resource in the t-th time period of D+1, and its value can be 0 or 1.

[0143] S1026. Determine the objective function based on the D+1 electricity purchase cost item, the D+1 resource allocation cost item, the D+1 load fluctuation penalty item, the D+1 shortfall load penalty item, and the D+1 resource activation penalty item.

[0144] For example, the sum of the D+1 electricity purchase cost, D+1 resource mobilization cost, D+1 load fluctuation penalty, D+1 shortfall load penalty, and D+1 resource activation penalty is determined as the target value, and the objective function is obtained by minimizing the target value.

[0145] For example, the target value is obtained by weighted summation of the D+1 electricity purchase cost, D+1 resource allocation cost, D+1 load fluctuation penalty, D+1 shortfall load penalty, and D+1 resource activation penalty. The objective function is obtained by minimizing the target value.

[0146] For example, the objective function can be expressed as ;in, This represents the target value.

[0147] In the technical solution provided in this application, the electricity purchase cost, resource dispatch cost, load fluctuation penalty, load shortfall penalty, and resource activation penalty correspond to five categories of economic and safety indicators faced by the power grid side in each time period D+1: electricity purchase expenditure, resource dispatch expenditure, load fluctuation risk, power shortage risk, and resource activation cost. After weighted combination of these five categories within a unified objective function framework, the final objective function can simultaneously quantify the comprehensive costs of both economic costs and safety risks, avoiding the limitation of focusing solely on economic optimization and potentially ignoring the increased load fluctuations or escalating load shortfall risks. The introduction of the load fluctuation penalty and load shortfall penalty ensures that the solution process, while reducing electricity purchase and dispatch costs, also considers the smoothness of load changes between time periods and the control level of power shortage risk. This objective function unifies multi-dimensional optimization objectives into a single quantitative expression, helping to provide a clear and comprehensive optimization direction for the subsequent solution process under constraints.

[0148] After interpreting the objective function, the constraints are then interpreted:

[0149] The constraints in the embodiments of this application include at least one of the following:

[0150] The call capacity variable for each type of demand response resource in D+1 time period is between the minimum callable capacity value and the maximum callable capacity value;

[0151] The initial startup status variable of each time period of D+1 for each demand response resource is greater than or equal to the difference between the call status value of each time period and the call status value of the corresponding previous time period; the initial startup status variable of each time period of D+1 for each demand response resource is less than or equal to the call status value of each time period; the initial startup status variable of each time period of D+1 for each demand response resource is less than or equal to the difference between 1 and the call status value of the corresponding previous time period.

[0152] The sum of the total call capacity variable and the deficit load variable of all demand response resources in each time period of D+1 is greater than or equal to the demand response call load in each time period of D+1.

[0153] The call capacity variable for each demand response resource in each time period of D+1 is less than or equal to the sum of the demand response call load and the additional load in each time period of D+1.

[0154] When the call status value is 1 in any time period, the call status value between any time period and the target time period is 1, and the duration between the target time period and any time period is the shortest call duration for the resource.

[0155] The initial startup status value is less than or equal to 1 within a preset number of consecutive time periods;

[0156] The number of times the initial startup status variable of each demand response resource in D+1 is 1 is less than or equal to the maximum number of calls for each demand response resource;

[0157] The total call capacity variable for all demand response resources in each time period of D+1 is between the minimum and maximum declared capacity values.

[0158] The auxiliary variables for load fluctuations in each time period of D+1 are greater than or equal to the absolute value of the difference between the corrected load of each time period and the corrected load of the corresponding previous time period.

[0159] For example, the call capacity variable for each time period of D+1 for each demand response resource lies between the minimum callable capacity value and the maximum callable capacity value, which can be represented as: ;in, ρ represents the call capacity variable for the r-th callable resource in the t-th time period of D+1. r min and ρ r max Let represent the minimum and maximum adjustable proportions of the r-th type of resource, respectively. This represents the baseline aggregate capacity of the r-th callable resource in the t-th time period of D+1. This represents the call status value of the r-th callable resource in the t-th time period of D+1, and the value can be 0 or 1.

[0160] For example, the initial startup state variable of each time period of D+1 for each demand response resource is greater than or equal to the difference between the call state value of each time period and the call state value of the corresponding previous time period; the initial startup state variable of each time period of D+1 for each demand response resource is less than or equal to the call state value of each time period; and the initial startup state variable of each time period of D+1 for each demand response resource is less than or equal to the difference between 1 and the call state value of the corresponding previous time period, which can be expressed as follows: ;in, This variable represents the initial startup state of the r-th callable resource in the t-th time period of D+1. It takes the value 1 when switching from the shutdown state to the callable state, and otherwise takes the value 0. This represents the call status value of the r-th callable resource in the t-th time period of D+1, and the value can be 0 or 1. This represents the invocation status value of the r-th callable resource in the (t-1)-th time period of D+1, and can take the value 0 or 1. For example, the first scheduling time period can be based on y... r,1 z r,1 deal with.

[0161] For example, if the sum of the total call capacity variable and the deficit load variable for all demand response resources in each time period of D+1 is greater than or equal to the demand response call load for each time period of D+1, it can be expressed as: ;in, This represents the call capacity variable for the r-th callable resource in the t-th time period of D+1; This represents the deficit load variable in the t-th time period of D+1; This represents the demand response call load during the t-th time period. For example, This can represent the demand shortfall that is not met by the actual call capacity in the t-th time period; when the total call capacity of various resources in that time period is insufficient to meet the call demand. At that time, the insufficient part shall be provided by This indicates that when the total call capacity is not less than the call demand, the following can be taken: =0.

[0162] For example, the call capacity variable for each time period of D+1 for each demand response resource is less than or equal to the sum of the demand response call load and the additional load for each time period of D+1, which can be expressed as follows: ;in, This represents the call capacity variable for the r-th callable resource in the t-th time period of D+1. This represents the demand response call load during the t-th time period. This represents the additional load during the t-th time period. When D... t req Less than the minimum aggregate declaration capacity P min In such cases, it is possible to choose not to generate a valid invocation instruction and to relax the variable ξ by addressing the unmet invocation requirement. t Record the shortfall for that period; in implementations that allow for economical requisition, D can be set. t extra Not less than P min -D t req This enables the model to generate call instructions that meet the minimum application capacity requirements.

[0163] When the call status value is 1 in any time period, the call status value between any time period and the target time period is 1. The duration between the target time period and any time period is the minimum call duration of the resource. If the tail time period of D-1 cannot meet the minimum duration, it can be prohibited to start in the corresponding tail time period, or the upper bound of the summation can be truncated to the last time period of the target day.

[0164] By setting the initial startup status value to be less than or equal to 1 for a preset number of consecutive time periods, the same resource can be prevented from starting repeatedly within the set startup interval.

[0165] By limiting the number of times a resource can be called, the number of times its initial startup state variable is 1 in each time period of D+1 is less than or equal to the maximum number of times it can be called. For example, an industrial load can only be shut down once a day because each shutdown results in huge losses.

[0166] The total call capacity variable for all demand response resources in each time period of D+1 can be determined as follows: The value between the minimum and maximum requested capacity can be determined by the following method: , ;in, This is the minimum declaration capacity value (i.e., the aggregated minimum declaration capacity). For a sufficiently large constant, This indicates whether a valid call instruction is formed in the t-th time period of D+1. The value is 1 if yes and 0 if no. This represents the total call capacity variable for all demand response resources in the t-th time period of D+1. This constraint guarantees that only when s t When s = 1, the total call capacity must reach the minimum declared capacity; when s t When the value is 0, the total call capacity is 0, which avoids the forced call of resources when there is no call requirement.

[0167] The auxiliary variable for load fluctuation in each time period of D+1 is greater than or equal to the absolute value of the difference between the corrected load of each time period and the corrected load of the corresponding previous time period, which can be expressed as: ;in, Let represent the auxiliary variable for load fluctuation in the t-th time period of D+1, which is the absolute value of load changes between adjacent time periods. This represents the corrected load for time period t in D+1. This represents the corrected postload of D+1 in the (t-1)th time period.

[0168] In some embodiments, Used to characterize the capacity allocation of various resources in different time periods. Used to indicate whether a resource is in a calling state. Used to indicate whether a resource is started during the corresponding time period. Used to characterize unmet call requests and to evaluate request fulfillment rate and unmet request penalty. It is used to characterize the corrected load results after demand response is invoked and for evaluating electricity purchase costs, peak shaving, and smoothness.

[0169] Figure 6 This diagram illustrates load curves before and after demand response resource optimization in some embodiments. The horizontal axis represents time in hours, and the vertical axis represents load in MW. Curve A shows the load curve before demand response resource optimization (i.e., the load without using the method in the embodiments of this application), and curve B shows the load curve after demand response resource optimization (i.e., the load using the method in the embodiments of this application). It can be seen that the method in the embodiments of this application can reduce the load on the grid side.

[0170] Based on the same concept, this application also provides an optimized scheduling apparatus for demand response resources to implement the optimized scheduling method for demand response resources described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method. Therefore, the specific limitations in one or more embodiments of the optimized scheduling apparatus for demand response resources provided below can be found in the limitations of the optimized scheduling method for demand response resources described above, and will not be repeated here.

[0171] In one exemplary embodiment, Figure 7 A schematic diagram of the structure of an optimized scheduling device for demand response resources provided in some embodiments, such as... Figure 7 As shown, the demand response resource optimization scheduling device 700 includes:

[0172] The demand response resource determination module 701 is used to determine at least one demand response resource based on the historical electricity supply and demand price statistics of the target area and the power supply load impact data of the target area in each time period of D+1.

[0173] The function establishment module 702 is used to establish an objective function based on the unit call cost of each demand response resource, the predicted electricity price of each time period in D+1, the predicted total load of each time period in D+1, the call capacity variable of each demand response resource in D+1, the load fluctuation auxiliary variable of each time period in D+1, the shortfall load variable of each time period in D+1, and the initial start-up state variable of each demand response resource in D+1.

[0174] The constraint establishment module 703 is used to establish the constraints for the call capacity variable, load fluctuation auxiliary variable, shortfall load variable and initial start-up state variable of each demand response resource in each time period of D+1.

[0175] The solver module 704 is used to solve the objective function under constraints to obtain the call capacity value, load fluctuation auxiliary value, deficit load value and start-up status value of each demand response resource in each time period of D+1.

[0176] The descriptions of the above device embodiments are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0177] Each module in the aforementioned demand response resource optimization scheduling device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can invoke and execute the operations corresponding to each module.

[0178] In one exemplary embodiment, Figure 8 This is a schematic diagram of the structure of a computer device provided in some embodiments. The computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor of this computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be implemented through Wireless Fidelity (WIFI), mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements an optimized scheduling method for demand-response resources. The display unit of the computer device is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0179] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0180] For example, a computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method provided in any of the above embodiments.

[0181] In one embodiment, a computer-readable storage medium is provided, wherein a computer program, when executed by a processor, implements the steps of the method provided in any of the above embodiments.

[0182] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method provided in any of the above embodiments.

[0183] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the methods described above.

[0184] The processor, functional modules, or functional units in any embodiment of this application may include an integration of one or more of the following: a general-purpose processor, an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field-programmable gate array (FPGA), a central processing unit (CPU), a graphics processing unit (GPU), an embedded neural network processing unit (NPU), a controller, a microcontroller, a microprocessor, a programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, a quantum computing-based data processing logic unit, an artificial intelligence (AI) processor, etc.

[0185] The memory or computer-readable storage medium in any embodiment of this application may include at least one of non-volatile memory and volatile memory. Non-volatile memory includes integration of one or more of the following: Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Ferromagnetic Random Access Memory (FRAM), Flash Memory, Magnetic Surface Memory, Optical Disc, Compact Disc Read-Only Memory (CD-ROM), Magnetic Tape, Floppy Disk, Flash Memory, Optical Memory, High-Density Embedded Non-Volatile Memory, Resistive Random Access Memory (ReRAM), Magnetoresistive Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), Graphene Memory, Volatile Memory, etc. Volatile memory includes one or more of the following: Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0186] The acquisition, transmission, storage, use, and processing of data in this application comply with relevant national laws and regulations. It should be noted that certain software, components, models, and other existing industry solutions may be mentioned in the embodiments of this application. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0187] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0188] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An optimized scheduling method for demand response resources, characterized in that, The method includes: Based on historical electricity supply and demand price statistics for the target area and power supply load impact data for the target area in each time period of D+1, at least one demand response resource is determined; An objective function is established based on the unit call cost of each demand response resource, the predicted electricity price for each time period of D+1, the predicted total load for each time period of D+1, the call capacity variable for each demand response resource for each time period of D+1, the load fluctuation auxiliary variable for each time period of D+1, the shortfall load variable for each time period of D+1, and the initial start-up state variable for each demand response resource for each time period of D+1. Establish constraints on the call capacity variable, load fluctuation auxiliary variable, shortfall load variable, and initial startup state variable of each demand response resource in each time period of D+1 for each demand response resource. Solving the objective function under the constraints yields the call capacity value, load fluctuation auxiliary value, deficit load value, and start-up status value for each demand response resource in time period D+1.

2. The method according to claim 1, characterized in that, The determination of at least one demand response resource based on historical electricity supply and demand price statistics for the target area and power supply load impact data for each time period of D+1 includes: Based on the historical electricity supply and demand price statistics and the power supply load impact data, the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1 are determined. Based on the predicted electricity price, predicted total load, predicted renewable energy output, and rated aggregate capacity of each available resource in each period of D+1, determine the baseline aggregate capacity of each available resource in each period of D+1. Based on the baseline aggregate capacity of each callable resource in each time period of D+1, at least one demand response resource is determined from a variety of callable resources.

3. The method according to claim 2, characterized in that, The determination of the baseline aggregate capacity of each available resource in each period of D+1, based on the predicted electricity price, predicted total load, predicted renewable energy output, and the rated aggregate capacity of each available resource in each period of D+1, includes: Based on the predicted electricity price, predicted total load, and predicted renewable energy output for each period of D+1, the comprehensive risk score for each period of D+1 is determined. Based on the comprehensive risk score of each time period in D+1 and the rated aggregate capacity of each callable resource in each time period in D+1, the baseline aggregate capacity of each callable resource in each time period in D+1 is determined.

4. The method according to claim 3, characterized in that, The determination of the comprehensive risk score for each period of D+1 based on the predicted electricity price, predicted total load, and predicted renewable energy output for each period includes: Based on the predicted electricity prices for each period of D+1, determine the price risk score for each period of D+1; Based on the predicted total load for each time period of D+1, determine the total load risk score for each time period of D+1; Based on the predicted total load and new energy output for each period of D+1, determine the predicted net load for each period of D+1. Based on the predicted net load for each period of D+1, determine the net load risk score for each period of D+1. The price risk score, total load risk score, and net load risk score for each period in D+1 are weighted and summed to obtain the comprehensive risk score for each period in D+1.

5. The method according to claim 3, characterized in that, The step of determining the baseline aggregate capacity of each callable resource in each period of D+1 based on the comprehensive risk score of each period of D+1 and the rated aggregate capacity of each callable resource in each period of D+1 includes: Determine the demand response call load for each period of D+1 based on the maximum callable load for a single period, the comprehensive risk score for each period of D+1, and the predicted total load for each period of D+1. Based on the rated aggregate capacity of each callable resource in each time period of D+1, the proportion of the load of each callable resource to the load of all callable resources, the predicted total load in each time period of D+1, and the availability coefficient of each callable resource in each time period of D+1, the baseline aggregate capacity of each callable resource in each time period of D+1 is determined.

6. The method according to claim 2, characterized in that, The step of determining the at least one demand response resource from multiple callable resources based on the baseline aggregated capacity of each callable resource in each time period of D+1 includes: Based on the baseline aggregate capacity of each callable resource in each time period of D+1, the maximum callable ratio of each callable resource in each time period of D+1, and the demand response call load of each callable resource in each time period of D+1, determine the capacity coverage of each callable resource in each time period of D+1. Based on the capacity coverage of each callable resource in each time period of D+1, the availability coefficient of each callable resource in each time period of D+1, the historical response reliability parameter of each callable resource, the normalized unit call cost of each callable resource, and the call constraint mismatch penalty parameter of each callable resource in each time period of D+1, the call adaptation parameters of each callable resource in each time period of D+1 are determined. Based on the call adaptation parameters and preset call adaptation thresholds of each callable resource in each time period of D+1, at least one demand response resource is determined from the multiple callable resources.

7. The method according to any one of claims 1-6, characterized in that, The objective function is established based on the unit dispatch cost of each demand response resource, the predicted electricity price for each time period of D+1, the predicted total load for each time period of D+1, the dispatch capacity variable for each demand response resource for each time period of D+1, the load fluctuation auxiliary variable for each time period of D+1, the deficit load variable for each time period of D+1, and the initial startup state variable for each demand response resource for each time period of D+1, including: Based on the predicted electricity price for each time period of D+1, the predicted total load for each time period of D+1, and the call capacity variables for each demand response resource for each time period of D+1, determine the electricity purchase cost item for D+1. Based on the unit call cost of each demand response resource and the call capacity variable of each demand response resource in each time period of D+1, determine the D+1 resource call cost item; The load fluctuation penalty item for D+1 is determined based on the fluctuation penalty coefficient and the auxiliary variables of load fluctuation for each period of D+1. Based on the demand non-metance penalty coefficient and the deficit load variables for each period of D+1, determine the D+1 deficit load penalty item; Based on the resource activation penalty coefficient and the initial activation state variables of each time period of D+1 for each type of demand response resource, determine the D+1 resource activation penalty item; The objective function is determined based on the D+1 electricity purchase cost item, the D+1 resource allocation cost item, the D+1 load fluctuation penalty item, the D+1 shortfall load penalty item, and the D+1 resource activation penalty item.

8. The method according to any one of claims 1-6, characterized in that, The constraints include at least one of the following: The call capacity variable for each type of demand response resource in each time period of D+1 is between the minimum callable capacity value and the maximum callable capacity value; The initial startup status variable of each time period of D+1 for each demand response resource is greater than or equal to the difference between the call status value of each time period and the call status value of the corresponding previous time period; the initial startup status variable of each time period of D+1 for each demand response resource is less than or equal to the call status value of each time period; the initial startup status variable of each time period of D+1 for each demand response resource is less than or equal to the difference between 1 and the call status value of the corresponding previous time period. The sum of the total call capacity variable and the deficit load variable of all demand response resources in each time period of D+1 is greater than or equal to the demand response call load in each time period of D+1. The call capacity variable for each demand response resource in each time period of D+1 is less than or equal to the sum of the demand response call load in each time period of D+1 and the additional load in each time period of D+1. When the call status value is 1 in any time period, the call status value between the time period and the target time period is 1, and the duration between the target time period and the time period is the shortest call duration for the resource. The initial startup status value is less than or equal to 1 within a preset number of consecutive time periods; The number of times the initial startup status variable of each demand response resource in each time period of D+1 is 1 is less than or equal to the maximum number of times each demand response resource is invoked; The total call capacity variable for all demand response resources in each time period of D+1 is between the minimum and maximum declared capacity values. The auxiliary variables for load fluctuations in each time period of D+1 are greater than or equal to the absolute value of the difference between the corrected load of each time period and the corrected load of the corresponding previous time period.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.