Flexible load scheduling method and device, electronic equipment and storage medium
By clustering data and building models for flexible loads, the problem of lacking reasonable scheduling methods in existing technologies has been solved. This has enabled the smoothing of peak-valley differences in the power system and the minimization of load scheduling costs, thereby improving the system's scheduling flexibility and economy, and enhancing its capacity to absorb renewable energy.
Patent Information
- Application Number
- CN202511726318.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-24
- Publication Date
- 2026-02-10
AI Technical Summary
The lack of reasonable flexible load dispatching methods in existing technologies has led to prominent peak-shaving problems in the power system, increased dispatching costs, and difficulty in effectively absorbing renewable energy.
By acquiring load data of flexible loads, clustering is performed to construct an adjustable aggregation model of movable and delayed loads. The objective function is constructed and solved with the goal of minimizing the peak-valley difference smoothing cost and load scheduling cost of the power system, so as to achieve reasonable scheduling of flexible loads.
It has achieved peak-valley difference smoothing and load dispatching cost minimization in the power system, improved the dispatching flexibility and economy of the power system, and enhanced the ability to absorb renewable energy.
Smart Images
Figure CN121507774A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of power distribution scheduling, in particular to a flexible load scheduling method and device, electronic equipment and storage medium. BACKGROUND
[0002] With large-scale access of uncertain power generation resources such as wind power to the power system, the general scheduling mode is difficult to meet the requirements of renewable energy consumption of the power system under the new situation, and the power system peak shaving problem is increasingly prominent. In order to cope with the inherent uncertainty and intermittency of wind power output, the power system needs to reserve a large amount of power generation standby to meet the load demand, which not only reduces the power generation efficiency of conventional generating units, but also inevitably increases the power system scheduling cost. Based on this, part of the demand side load can be involved in the economic scheduling of the power system, and the load can be used as a kind of schedulable resource. These loads are called flexible loads. The participation of flexible loads can exert the initiative and flexibility of the load, and the scheduling of the power system is more flexible, economical and reasonable. As one of the important means to alleviate the contradiction between power supply and demand, the scheduling of demand side flexible load has been gradually included in the economic scheduling problem of the power system in recent years. However, most of the existing literature is too simple in the classification and modeling of flexible loads, and lacks a reasonable flexible load scheduling method. SUMMARY
[0003] The present application provides a flexible load scheduling method, device, electronic equipment and storage medium, which can solve the problem of lack of reasonable flexible load scheduling method in the prior art.
[0004] In order to solve the above technical problems, the present application provides a flexible load scheduling method, comprising: obtaining load data of flexible load; wherein the load data includes available time, power demand and working time of movable load and delayable load; clustering the load data to obtain a plurality of clustering clusters of flexible load with similar available time, power demand and working time; According to the load data and the clustering cluster, a first adjustable aggregation model corresponding to the movable load and a second adjustable aggregation model corresponding to the delayable load are constructed; According to the first adjustable aggregation model and the second adjustable aggregation model, a target function corresponding to the flexible load scheduling and a corresponding constraint condition are constructed, with the minimum peak-valley difference suppression cost and load scheduling cost of the power system as the target; Under the constraint of the constraint condition, the target function is solved to obtain the scheduling amount of the movable load and the delayable load at each time when the peak-valley difference suppression cost and the load scheduling cost of the power system are minimized, and then the movable load and the delayable load are scheduled according to the scheduling amount.
[0005] As a preferred embodiment, the step of constructing a first adjustable aggregation model corresponding to movable loads and a second adjustable aggregation model corresponding to delayed loads based on the load data and the clusters includes: Based on the load data, a first individual load model for a single movable load and a second individual load model for a single deferred load are constructed. Based on the cluster and the first individual load model, a first discretized aggregation model of multiple movable loads within the same cluster is constructed; based on the cluster and the second individual load model, a second discretized aggregation model of multiple deferred loads within the same cluster is constructed. The discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0006] As a preferred embodiment, the step of converting the discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load, and converting the discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayable load, includes: Based on the power demand of the flexible load in each cluster, the maximum power demand and minimum power demand corresponding to the flexible load are obtained. Then, based on the maximum power demand and the minimum power demand, the power adjustment range of the flexible load in each cluster in the first discretized aggregation model and the second discretized aggregation model is obtained. Based on the power adjustment range, the discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0007] As a preferred embodiment, the objective function is: ; ; ; ; in, It is the weighted sum of the cost of peak-valley difference smoothing and the cost of load scheduling; This represents the degree of peak and trough fluctuations in the load curve; for Total load power after time period optimization; and These are the weighting coefficients for peak-valley difference mitigation costs and load scheduling costs, respectively. The scheduling cost of mobile loads; The scheduling cost of deferred load; for Available mobile load dispatch volume for a given time period; for Delayable load scheduling amount for a given time period; and They are respectively The unit scheduling cost of mobile load and the unit scheduling cost of delayed load for a given time period.
[0008] Based on the above embodiments, another embodiment of the present invention provides a flexible load scheduling device, including: a load data acquisition module, a load data clustering module, an adjustable aggregation model construction module, an objective function construction module, and a load scheduling module; The load data acquisition module is used to acquire load data of flexible loads; wherein, the load data includes: available time, power demand, and operating time of movable loads and delayable loads; The load data clustering module is used to cluster the load data to obtain several clusters of flexible loads with similar available time, power requirements and operating time. The adjustable aggregation model construction module is used to construct a first adjustable aggregation model corresponding to the movable load and a second adjustable aggregation model corresponding to the delayed load based on the load data and the clusters. The objective function construction module is used to construct a corresponding objective function and corresponding constraints for flexible load scheduling based on the first adjustable aggregation model and the second adjustable aggregation model, with the goal of minimizing the peak-valley difference smoothing cost and load scheduling cost of the power system. The load scheduling module is used to solve the objective function under the constraints of the constraints, to obtain the scheduling amount of mobile load and deferred load at each moment when the peak-valley difference smoothing cost and load scheduling cost of the power system are minimized, and then to perform load scheduling on mobile load and deferred load according to the scheduling amount.
[0009] As a preferred embodiment, the step of constructing a first adjustable aggregation model corresponding to movable loads and a second adjustable aggregation model corresponding to delayed loads based on the load data and the clusters includes: Based on the load data, a first individual load model for a single movable load and a second individual load model for a single deferred load are constructed. Based on the cluster and the first individual load model, a first discretized aggregation model of multiple movable loads within the same cluster is constructed; based on the cluster and the second individual load model, a second discretized aggregation model of multiple deferred loads within the same cluster is constructed. The discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0010] As a preferred embodiment, the step of converting the discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load, and converting the discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayable load, includes: Based on the power demand of the flexible load in each cluster, the maximum power demand and minimum power demand corresponding to the flexible load are obtained. Then, based on the maximum power demand and the minimum power demand, the power adjustment range of the flexible load in each cluster in the first discretized aggregation model and the second discretized aggregation model is obtained. Based on the power adjustment range, the discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0011] As a preferred embodiment, the objective function is: ; ; ; ; in, It is the weighted sum of the cost of peak-valley difference smoothing and the cost of load scheduling; This represents the degree of peak and trough fluctuations in the load curve; for Total load power after time period optimization; and These are the weighting coefficients for peak-valley difference mitigation costs and load scheduling costs, respectively. The scheduling cost of mobile loads; The scheduling cost of deferred load; for Available mobile load dispatch volume for a given time period; for Delayable load scheduling amount for a given time period; and They are respectively The unit scheduling cost of mobile load and the unit scheduling cost of delayed load for a given time period.
[0012] Based on the above embodiments, another embodiment of the present invention provides an electronic device, the device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the flexible load scheduling method described in the above embodiments of the invention.
[0013] Based on the above embodiments, another embodiment of the present invention provides a storage medium, the storage medium including a stored computer program, wherein, when the computer program is running, it controls the device where the storage medium is located to execute the flexible load scheduling method described in the above embodiments of the invention.
[0014] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention provides a flexible load scheduling method, which acquires load data of flexible loads. The load data includes the available time, power demand, and operating time of mobile and deferred loads. The load data is clustered to obtain several clusters of flexible loads with similar available time, power demand, and operating time. Based on the load data and the clusters, a first adjustable aggregation model corresponding to mobile loads and a second adjustable aggregation model corresponding to deferred loads are constructed. Based on the first and second adjustable aggregation models, a corresponding objective function and corresponding constraints for flexible load scheduling are constructed, aiming to minimize the peak-valley difference smoothing cost and load scheduling cost of the power system. Under the constraints, the objective function is solved to obtain the scheduling amount of mobile and deferred loads at each moment when the peak-valley difference smoothing cost and load scheduling cost of the power system are minimized. Load scheduling of mobile and deferred loads is then performed based on the scheduling amount. Through this invention, an adjustable aggregation model can be constructed based on the load data of flexible loads. Then, an objective function for flexible load scheduling can be constructed based on the adjustable aggregation model. Solving the objective function yields a scheduling scheme for flexible loads. Attached Figure Description
[0015] Figure 1 This is a flowchart illustrating a flexible load scheduling method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the scheduling volume of various loads at each moment; Figure 3 This is a schematic diagram of the structure of a flexible load scheduling device provided in an embodiment of the present invention. Detailed Implementation
[0016] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0017] 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.
[0018] 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, unless otherwise explicitly defined.
[0019] 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.
[0020] 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.
[0021] In the description of the embodiments of this application, the terms "multiple" and "several" refer to two or more (including two), similarly, "multiple groups" refer to two or more (including two groups), and "multiple pieces" refer to two or more (including two pieces).
[0022] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.
[0023] Example 1 Please refer to Figure 1 To address the lack of reasonable flexible load scheduling methods in existing technologies, this invention provides a flowchart of a flexible load scheduling method. First, the K-means++ algorithm is used for flexible load clustering to obtain a flexible load aggregation model. This aggregation model is then used for parameter identification in flexible load clustering to obtain core parameters such as the equivalent availability time and operating time range of the flexible load cluster. Second, individual load models for mobile and deferred loads are established. A discretized aggregation model for mobile and deferred loads is constructed using the algebraic sum of the rated power of each load, yielding adjustable boundaries. Then, to make the aggregation model suitable for optimized scheduling, the discretized aggregation model is reconstructed into a continuously adjustable aggregation model. Finally, based on the aggregation model, with the goal of minimizing the peak-valley difference smoothing cost and load scheduling cost of the power system, a corresponding objective function for flexible load scheduling is constructed. Load scheduling is then performed based on the solution to the objective function. The specific steps include: S1. Obtain load data for flexible loads; wherein, the load data includes: available time, power demand, and operating time of movable loads and delayable loads; In a specific embodiment, for step S1, after obtaining the load data of the flexible load, the load data can be grouped according to the similar availability time of each other, and after grouping, a load curve is constructed according to the power demand of each flexible load in the corresponding time interval. The curve has the following attributes: availability time, power demand, and working time requirement.
[0024] Specifically, based on the overlap, duration, and time offset of available load time, loads with similar available time periods can be grouped together to unify scheduling time constraints. Based on the power demand, total running time, and start-stop restrictions of the loads within the group, the rated power is superimposed within the merged time window to generate an initial load curve, which is then extended into a continuously adjustable curve through adjustable power range and time elastic constraints (such as minimum continuous running time).
[0025] S2. Cluster the load data to obtain several clusters of flexible loads with similar available time, power requirements and operating time; In a specific embodiment, for step S2, the k-means++ clustering algorithm can be used to cluster the load curves constructed in step S1. The following is the process of using k-means++ clustering: Suppose that at any given time, let Indicates from (The i-th load) to the nearest selected center curve distance, , The time number for the center curve.
[0026] Step 1: Construct a multi-dimensional feature vector to characterize the load characteristics using features such as power fluctuation, time entropy, and frequency domain characteristics, and randomly and uniformly select an initial center point within the load. .
[0027] Step 2: At the initial center point Based on, Choose the next one based on the principle of minimum , This is a set of all load curves, serving as the initial basis for parameter classification.
[0028] Step 3: Repeat step 2, select A central curve.
[0029] Step 4: Use Euclidean distance as the criterion for parameter clustering, and group the individual clusters... Assigned to the nearest center curve, forming Each cluster completes the initial classification of the parameters.
[0030] Step 5: Based on the initial center curve classification described above, the average value of all load curves within the same cluster is taken as the cluster average. The new center curve.
[0031] Step 6: Repeat steps 5 and 6 until the sum of the distances from each curve in the cluster to the center no longer decreases or a fixed number of iterations is reached.
[0032] Step 7: Based on Step 6, the new center curve has been determined, and the parameters are... Assign them to the cluster containing the nearest central curve using Euclidean distance.
[0033] Step 8: Based on Step 7, update the center curve in each cluster to the mean of all parameters in that cluster, thus completing the adaptive update of the center curve.
[0034] Step 9: Output the final cluster and its center curve.
[0035] Through the above steps, the flexible loads in the clusters we obtain have similar availability time, operating time, and power curves, and the availability time, operating time, and power curves of the clusters are uniformly represented by the cluster centers.
[0036] S3. Based on the load data and the clusters, construct a first adjustable aggregation model corresponding to the movable load and a second adjustable aggregation model corresponding to the delayed load; Preferably, the step of constructing a first adjustable aggregation model corresponding to a movable load and a second adjustable aggregation model corresponding to a delayed load based on the load data and the clusters includes: constructing a first individual load model for a single movable load and a second individual load model for a single delayed load based on the load data; constructing a first discretized aggregation model for multiple movable loads within the same cluster based on the clusters and the first individual load model; constructing a second discretized aggregation model for multiple delayed loads within the same cluster based on the clusters and the second individual load model; converting the discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load; and converting the discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0037] Preferably, the step of converting the discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the mobile load, and converting the discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the deferred load, includes: obtaining the maximum power demand and minimum power demand corresponding to the flexible load based on the power demand of the flexible load in each cluster; then obtaining the power adjustment range of the flexible load in each cluster of the first and second discretized aggregation models based on the maximum power demand and the minimum power demand; and converting the discrete first discretized aggregation model into a continuously adjustable aggregation model based on the power adjustment range to obtain the first adjustable aggregation model corresponding to the mobile load, and converting the discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the deferred load.
[0038] In a specific embodiment, the specific steps for step S3, constructing the adjustable aggregation model corresponding to the movable load and the delayable load, are as follows: 1. Establish individual load models for movable and deferred loads: Mobile loads are loads that can be moved entirely over time, but their operating duration and power requirements are fixed. The load's operating time can be moved entirely within its available time range, but it cannot be interrupted or operated in segments. Typical examples include household appliances such as washing machines and dishwashers, which need to complete their work within a specific time, but the specific start-up time can be flexibly adjusted.
[0039] set up Indicates a single load Demand response (kW), single load Working hours are Behavioral attributes correspond to available time. The first unit load model of a single movable load can then be represented as follows: (1) (2) (3) Formula (1) ensures load The power is equal to the rated power required for its operation, (2) ensuring sufficient working time. Formula (3) meets the minimum operating time requirement required for the load to complete its operation. Where: Let i be the rated power of load i, used to describe its power demand during operation; The state of the i-th movable load at time t; , Start / stop time of movable load; The number of movable loads; For the i-th movable load at t .
[0040] Delayable loads are loads that can be interrupted and resumed in time. Their operating time and power demand can be flexibly adjusted, as long as the total energy demand is met before the final deadline. Typical examples include electric vehicle charging and battery storage, where the charging process can be interrupted and resumed. For each delayable load... Within a given time range Internal demand response as well as arrival and departure times The second unit load model for a single deferred load is represented by the following formula: (4) (5) Formula (4) constrains the adjustable power range and time range of the deferred load, while formula (5) expresses the calculation method for calculating the deferred load adjustable energy based on the adjustable power of the deferred load. Where... Indicates load In time Power requirements; Indicates load The rated power, i.e., the maximum power requirement when the load is running; and Representing loads respectively The arrival and departure times define the available time range of the load; Indicates load The total energy requirement, which is the total amount of work that needs to be done throughout the entire available time; Indicates a time interval used to calculate total energy demand; Indicates load The feasible power set is the set of all possible power demands that satisfy the constraints. This represents the set of all individual loads.
[0041] 2. Construct a discretized aggregation model for movable and deferred loads: The feasible power set of the i-th movable load This is expressed in the following formula (6): (6) in, It is made up of all time within The power vector formed by this.
[0042] Based on the load models shown in formulas (1)-(3) and (6), an aggregation model for the above load types can be established. For a set It aggregates A set of movable loads, where the aggregate power at a single moment equals the sum of the power of each individual load, includes... A cluster of movable loads The first discretization aggregation model is represented as follows: (7) (8) Equation (7) represents the set of mobile loads, and Equation (8) represents the discrete range of the adjustable capacity of the mobile load cluster, where: For set power set, for The power vector formed; For mobile load clusters Power at each time point after aggregation; Indicates the load i at time Power requirements; Indicates the load i at time The running state is a binary variable; This represents the rated power of load i, i.e., the fixed power requirement when the load is running. This represents the running time of load i, i.e., the total time the load needs to run. and Cluster The time range, i.e. the period during which the load can operate; Cluster The total energy requirement, that is, the minimum energy required throughout the entire available time; This is a collection of clusters. Clusters are distinguished by whether different loads possess the properties of being portable and deferred. k-means++ first selects points farthest from other points as initial centers (e.g., selecting the load with the highest power), and then selects the next center by weighting the squared distance. Portable loads (e.g., electric vehicles) and deferred loads (e.g., washing machines) are clustered according to available time, power curves, etc., and the centers are iteratively updated to eventually form clusters. The power, available time, operating time, and power curve of each cluster of loads are classified into the same category.
[0043] Delayable load The feasible power set can be represented as Constrained by formula (4-5). For sets ,polymerization For each load, its aggregate power should be equal to the sum of the powers of all individual loads in the set. The corresponding second discretized aggregate model can be characterized by the following equation: (9) (10) Equation (9) represents the set of delayable loads, and Equation (10) represents the adjustable capacity range of the delayable load cluster. Where: Represents a set The aggregated power set is the sum of the power demands of all loads in the set; It is made up of all time within The power vector formed; Cluster The aggregation power requirement is a collection The sum of the power demands of all loads in the area. For delayed load clusters, Cluster The total energy requirement, that is, the minimum energy required throughout the entire available time. and Representing loads respectively Arrival and departure times.
[0044] 3. Construct an adjustable aggregation model for scheduling-oriented movable and delayed loads: The above discretized aggregation model consists of the algebraic sum of the rated power of individual loads. The aggregation model is inherently discrete, which is equivalent to obtaining the adjustable boundary of the aggregation model, but without a power adjustment range. To make the aggregation model applicable to various scheduling algorithms, it needs to be converted into an aggregation model that includes an adjustable power range. The first adjustable aggregation model for mobile loads can be specifically represented by the following formula: (11) Equation (11) is the cluster adjustable power continuous range model obtained by combining the cluster parameters obtained by clustering with the physical model of Equation (8). Where: Cluster In time The aggregation power requirement is a continuous variable; Cluster In time The running status is a binary variable, where 1 indicates running and 0 indicates stopped; A set of clusters; and Representing clusters The minimum power lower limit and maximum power upper limit, obtained from the clustering algorithm described above, are used to constrain... Scope; Cluster The available time range, i.e. the period during which the load can operate; Cluster The total energy requirement, that is, the minimum energy required throughout the entire available time; Cluster The actual running time is equal to all time points. The sum; Cluster The maximum allowed runtime.
[0045] Scheduling delayed load clusters requires adjustments across discontinuous power ranges to meet load demands before the latest departure time. Reconfigured parameters. Limited by cluster Maximum power requirement during arrival and departure times. Here, The lower bound is zero because the demand response load is interruptible and can be serviced within the next time interval. Total energy demand must be met during arrival and departure times. The second adjustable aggregation model for deferred loads can be specifically characterized by the following equation: (12) Equation (12) is the continuous range model of adjustable power of the cluster obtained by combining the cluster parameters obtained by clustering with the physical model of Equation (10). Where: Cluster In time The reconfigured power requirements Cluster The maximum power requirement is the sum of the maximum power of all loads within the cluster. This represents the set of all demand response load clusters. Cluster The total energy requirement, that is, the minimum energy required throughout the entire available time.
[0046] Formulas (8) to (11) transform the first discrete aggregation model of mobile load (based on binary switching control of rated power) into a continuously adjustable aggregation model, enabling it to adapt to the power adjustment range requirements of the scheduling algorithm while retaining core constraints such as load running time and total energy demand.
[0047] The principle behind transforming a discrete, discretized aggregation model into a continuously adjustable aggregation model is as follows: (1) Variable type expansion: Formula (8): The power of a single load is a discrete value: ,in Polymerization power It is a simple algebraic sum of the rated power of each load, and is essentially discrete and unadjustable.
[0048] Formula (11): Introducing continuous variables It allows adjustment within the upper and lower power limits: Power range and The distribution of rated power is determined by clustering algorithms (such as the load distribution within a cluster), enabling an expansion from discrete to continuous.
[0049] (2) Adjustment of runtime constraints: Formula (8): Strictly constrain the running time of a single load: It cannot be adjusted.
[0050] Formula (11): Cluster-level runtime It can be dynamically adjusted (but not exceeding the maximum allowed value). ): It allows the scheduling algorithm to flexibly allocate start and stop periods within the total running time.
[0051] (3) Energy demand constraint retained: Commonality: Total energy demand Must meet: and Ensure that the basic electricity demand of the load remains unchanged.
[0052] Connection: Both are based on total energy demand. To constrain the core, and retain the time window Inoperability restrictions ( (Power is 0 at this time).
[0053] Formula (11) introduces upper and lower power limits. and This extends the discrete rated power aggregation model into a continuously adjustable model, providing power adjustment space for scheduling algorithms (such as peak shaving and valley filling, and demand response).
[0054] Formulas (10) to (12) transform the fixed power demand model of the deferred load (constrained by rated power and time window) into a continuously adjustable model, allowing dynamic adjustment within the power range while satisfying total energy demand and time availability constraints.
[0055] Conversion process and differences: (1) Power variable extension: Formula (10): The power of a single load is a fixed value: ,and In the time window Internally adjustable. Aggregation power. It is a simple summation of the power of each load, but the power regulation range is not explicitly defined.
[0056] Formula (12): Introducing cluster-level power upper and lower limits and Constrained polymerization power: The power range is determined by the clustering results (e.g., power distribution based on intra-cluster loads) and supports continuous adjustment.
[0057] (2) Time window constraint retention: Commonality: External power is 0 during the time window. .
[0058] Total energy demand must be met: .
[0059] (3) Improved scheduling flexibility: Formula (10): Only a single load is allowed to regulate within the rated power range, but cluster-level regulation capability is not defined.
[0060] Formula (12): Through cluster-level power upper and lower limits and This allows scheduling algorithms to dynamically allocate power at the aggregation level, such as reducing total power during peak periods. Increase power during off-peak hours ( ), while ensuring total energy constant.
[0061] Both use time windows and total energy demand As a constraint basis, Formula (12) integrates the dispersed individual load regulation capabilities into a flexible regulation capability at the aggregation level by defining the cluster-level power regulation range, which is more suitable for power grid dispatching algorithms (such as economic dispatch and safety-constrained unit combination).
[0062] S4. Based on the first adjustable aggregation model and the second adjustable aggregation model, with the goal of minimizing the peak-valley difference smoothing cost and load dispatching cost of the power system, construct the corresponding objective function and corresponding constraints for flexible load dispatching. Preferably, the objective function is: ; ; ; ; in, It is the weighted sum of the cost of peak-valley difference smoothing and the cost of load scheduling; This represents the degree of peak and trough fluctuations in the load curve; for Total load power after time period optimization; and These are the weighting coefficients for peak-valley difference mitigation costs and load scheduling costs, respectively. The scheduling cost of mobile loads; The scheduling cost of deferred load; for Available mobile load dispatch volume for a given time period; for Delayable load scheduling amount for a given time period; and They are respectively The unit scheduling cost of mobile load and the unit scheduling cost of delayed load for a given time period.
[0063] In a specific embodiment, for step S4, after constructing the adjustable aggregation model corresponding to the movable load and the delayable load through step S3 above, the objective function for the corresponding flexible load scheduling is constructed through the following specific steps: Specifically, taking a smart building system in a commercial park as an example, the flexible load can be an air conditioning unit (delayable load) and an electric vehicle charging pile cluster (mobile load). In order to respond to the grid's time-of-use pricing strategy and with the goal of minimizing the system's electricity cost, the constraint set includes the physical constraints of equipment operation in the system and the flexible load power range constraints of equations (11) and (12) of this invention. The system load Adjusting from flexible load to post-response load : .
[0064] The constructed objective function is: To achieve a balance between peak-valley smoothing and economic cost in flexible load scheduling, a comprehensive scheduling objective function is constructed. This function comprehensively considers the cost of peak-valley smoothing and the economic cost of scheduling. By reasonably setting weighting coefficients, it prioritizes peak-valley optimization while also taking into account the economic efficiency of the scheduling process. The mathematical expression is as follows: (13) In the formula: To comprehensively consider scheduling costs, the weighted sum of peak-valley difference mitigation costs and load scheduling economic costs needs to be minimized; Quantifying the peak-valley fluctuations of the load curve is a core physical indicator for the stable operation of the power grid. for Total load power after time period optimization (unit: kW); and These are the peak-valley difference weighting coefficient and the scheduling cost weighting coefficient, respectively. For the cost of dispatching transferable load (SL); It can reduce load (DL) scheduling costs.
[0065] S5. Under the constraints of the above constraints, the objective function is solved to obtain the dispatch amount of mobile load and deferred load at each time when the peak-valley difference smoothing cost and load dispatch cost of the power system are minimized. Then, load dispatch is performed on mobile load and deferred load according to the dispatch amount.
[0066] In a specific embodiment, for step S5, after constructing the objective function and corresponding constraints for the corresponding flexible load scheduling, since the above typical optimization scheduling model belongs to mixed integer linear programming (MILP) or nonlinear programming (NLP) optimization problems, mature commercial solvers (such as CPLEX, Gurobi) or heuristic algorithms can be applied to solve it efficiently, thereby obtaining the system's optimized scheduling scheme.
[0067] Specifically, typical optimization scheduling models belong to mixed-integer linear programming (MILP) or nonlinear programming (NLP) optimization problems. These can be efficiently solved using mature commercial solvers (such as CPLEX, Gurobi) or heuristic algorithms, thereby obtaining the optimal scheduling scheme for the system. This yields the scheduling amount for each type of load at each time step, allowing for load scheduling of movable and deferred loads based on these scheduling amounts. Please refer to [link / reference]. Figure 2 This is a schematic diagram showing the scheduling volume of various loads at each moment. Figure 2 (1) is a schematic diagram of the scheduling amount of mobile load at each time, and (2) is a schematic diagram of the scheduling amount of deferred load at each time.
[0068] Therefore, the present invention provides a flexible load scheduling method, which enables the reasonable scheduling of flexible loads and has the following effects: Parameter extraction and model accuracy improvement: By using a data-driven load clustering algorithm, key parameters of the aggregation model are extracted, breaking through the simplification assumptions of the traditional algebraic superposition model, and significantly improving the matching degree between the aggregation model and the actual load operation characteristics.
[0069] Computational efficiency optimization: Improved clustering algorithms (such as K-means++) are used to group and reduce the dimensionality of diverse flexible loads, thereby improving the efficiency of extracting key parameters.
[0070] Coordination of Discreteness and Continuity: Construct a continuous time-domain aggregation model based on load clustering characteristics to achieve a unified expression of discretized start-stop characteristics and continuous power regulation capabilities. This model is suitable for optimizing scheduling algorithms and avoids the segmented processing losses of traditional discrete models.
[0071] Example 2 Please refer to Figure 3 This is a schematic diagram of a flexible load scheduling device according to an embodiment of the present invention. The device includes: a load data acquisition module, a load data clustering module, an adjustable aggregation model construction module, an objective function construction module, and a load scheduling module. The load data acquisition module is used to acquire load data of flexible loads; wherein, the load data includes: available time, power demand, and operating time of movable loads and delayable loads; The load data clustering module is used to cluster the load data to obtain several clusters of flexible loads with similar available time, power requirements and operating time. The adjustable aggregation model construction module is used to construct a first adjustable aggregation model corresponding to the movable load and a second adjustable aggregation model corresponding to the delayed load based on the load data and the clusters. The objective function construction module is used to construct a corresponding objective function and corresponding constraints for flexible load scheduling based on the first adjustable aggregation model and the second adjustable aggregation model, with the goal of minimizing the peak-valley difference smoothing cost and load scheduling cost of the power system. The load scheduling module is used to solve the objective function under the constraints of the constraints, to obtain the scheduling amount of mobile load and deferred load at each moment when the peak-valley difference smoothing cost and load scheduling cost of the power system are minimized, and then to perform load scheduling on mobile load and deferred load according to the scheduling amount.
[0072] As a preferred embodiment, the step of constructing a first adjustable aggregation model corresponding to movable loads and a second adjustable aggregation model corresponding to delayed loads based on the load data and the clusters includes: Based on the load data, a first individual load model for a single movable load and a second individual load model for a single deferred load are constructed. Based on the cluster and the first individual load model, a first discretized aggregation model of multiple movable loads within the same cluster is constructed; based on the cluster and the second individual load model, a second discretized aggregation model of multiple deferred loads within the same cluster is constructed. The discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0073] As a preferred embodiment, the step of converting the discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load, and converting the discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayable load, includes: Based on the power demand of the flexible load in each cluster, the maximum power demand and minimum power demand corresponding to the flexible load are obtained. Then, based on the maximum power demand and the minimum power demand, the power adjustment range of the flexible load in each cluster in the first discretized aggregation model and the second discretized aggregation model is obtained. Based on the power adjustment range, the discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
[0074] As a preferred embodiment, the objective function is: ; ; ; ; in, It is the weighted sum of the cost of peak-valley difference smoothing and the cost of load scheduling; This represents the degree of peak and trough fluctuations in the load curve; for Total load power after time period optimization; and These are the weighting coefficients for peak-valley difference mitigation costs and load scheduling costs, respectively. The scheduling cost of mobile loads; The scheduling cost of deferred load; for Available mobile load dispatch volume for a given time period; for Delayable load scheduling amount for a given time period; and They are respectively The unit scheduling cost of mobile load and the unit scheduling cost of delayed load for a given time period.
[0075] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.
[0076] Those skilled in the art will clearly understand that, for convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.
[0077] Example 3 Accordingly, embodiments of the present invention provide an electronic device, the device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the flexible load scheduling method described in the above embodiments of the invention.
[0078] The electronic device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The device may include, but is not limited to, a processor and a memory.
[0079] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the device, connecting various parts of the device via various interfaces and lines.
[0080] Example 4 Accordingly, embodiments of the present invention provide a storage medium, the storage medium including a stored computer program, wherein, when the computer program is running, it controls the device where the storage medium is located to execute the flexible load scheduling method described in the above embodiments of the invention.
[0081] The memory can be used to store the computer program. The processor implements various functions of the device by running or executing the computer program stored in the memory and calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc.; the data storage area may store data created based on the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
[0082] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.
[0083] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A flexible load dispatching method, characterized in that, include: Acquire load data for flexible loads; wherein the load data includes: available time, power demand, and operating time of movable and deferred loads; The load data is clustered to obtain several clusters of flexible loads with similar availability, power requirements, and operating time; Based on the load data and the clusters, a first adjustable aggregation model corresponding to movable loads and a second adjustable aggregation model corresponding to delayed loads are constructed. Based on the first adjustable aggregation model and the second adjustable aggregation model, with the goal of minimizing the peak-valley difference smoothing cost and load dispatching cost of the power system, a corresponding objective function for flexible load dispatching and corresponding constraints are constructed. Under the constraints of the above conditions, the objective function is solved to obtain the dispatch amount of mobile load and deferred load at each time when the peak-valley difference smoothing cost and load dispatch cost of the power system are minimized. Then, load dispatch is performed on mobile load and deferred load according to the dispatch amount.
2. The flexible load scheduling method as described in claim 1, characterized in that, The step of constructing a first adjustable aggregation model corresponding to movable loads and a second adjustable aggregation model corresponding to delayed loads based on the load data and the clusters includes: Based on the load data, a first individual load model for a single movable load and a second individual load model for a single deferred load are constructed. Based on the cluster and the first individual load model, a first discretized aggregation model of multiple movable loads within the same cluster is constructed; based on the cluster and the second individual load model, a second discretized aggregation model of multiple deferred loads within the same cluster is constructed. The discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
3. The flexible load scheduling method as described in claim 2, characterized in that, The process of transforming a discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain a first adjustable aggregation model corresponding to movable loads, and transforming a discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain a second adjustable aggregation model corresponding to deferred loads, includes: Based on the power demand of the flexible load in each cluster, the maximum power demand and minimum power demand corresponding to the flexible load are obtained. Then, based on the maximum power demand and the minimum power demand, the power adjustment range of the flexible load in each cluster in the first discretized aggregation model and the second discretized aggregation model is obtained. Based on the power adjustment range, the discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
4. The flexible load scheduling method as described in claim 1, characterized in that, The objective function is: ； ； ； ； in, It is the weighted sum of the cost of peak-valley difference smoothing and the cost of load scheduling; This represents the degree of peak and trough fluctuations in the load curve; for Total load power after time period optimization; and These are the weighting coefficients for peak-valley difference mitigation costs and load scheduling costs, respectively. The scheduling cost of mobile loads; The scheduling cost of deferred load; for Available mobile load dispatch volume for a given time period; for Delayable load scheduling amount for a given time period; and They are respectively The unit scheduling cost of mobile load and the unit scheduling cost of delayed load for a given time period.
5. A flexible load dispatching device, characterized in that, include: The system includes a load data acquisition module, a load data clustering module, an adjustable aggregation model construction module, an objective function construction module, and a load scheduling module. The load data acquisition module is used to acquire load data of flexible loads; wherein, the load data includes: available time, power demand, and operating time of movable loads and delayable loads; The load data clustering module is used to cluster the load data to obtain several clusters of flexible loads with similar available time, power requirements and operating time. The adjustable aggregation model construction module is used to construct a first adjustable aggregation model corresponding to the movable load and a second adjustable aggregation model corresponding to the delayed load based on the load data and the clusters. The objective function construction module is used to construct a corresponding objective function and corresponding constraints for flexible load scheduling based on the first adjustable aggregation model and the second adjustable aggregation model, with the goal of minimizing the peak-valley difference smoothing cost and load scheduling cost of the power system. The load scheduling module is used to solve the objective function under the constraints of the constraints, to obtain the scheduling amount of mobile load and deferred load at each moment when the peak-valley difference smoothing cost and load scheduling cost of the power system are minimized, and then to perform load scheduling on mobile load and deferred load according to the scheduling amount.
6. The flexible load dispatching device as described in claim 5, characterized in that, The step of constructing a first adjustable aggregation model corresponding to movable loads and a second adjustable aggregation model corresponding to delayed loads based on the load data and the clusters includes: Based on the load data, a first individual load model for a single movable load and a second individual load model for a single deferred load are constructed. Based on the cluster and the first individual load model, a first discretized aggregation model of multiple movable loads within the same cluster is constructed; based on the cluster and the second individual load model, a second discretized aggregation model of multiple deferred loads within the same cluster is constructed. The discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
7. The flexible load dispatching device as described in claim 6, characterized in that, The process of transforming a discrete first discretized aggregation model into a continuously adjustable aggregation model to obtain a first adjustable aggregation model corresponding to movable loads, and transforming a discrete second discretized aggregation model into a continuously adjustable aggregation model to obtain a second adjustable aggregation model corresponding to deferred loads, includes: Based on the power demand of the flexible load in each cluster, the maximum power demand and minimum power demand corresponding to the flexible load are obtained. Then, based on the maximum power demand and the minimum power demand, the power adjustment range of the flexible load in each cluster in the first discretized aggregation model and the second discretized aggregation model is obtained. Based on the power adjustment range, the discrete first discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the first adjustable aggregation model corresponding to the movable load. The discrete second discretized aggregation model is transformed into a continuously adjustable aggregation model to obtain the second adjustable aggregation model corresponding to the delayed load.
8. The flexible load dispatching device as described in claim 5, characterized in that, The objective function is: ； ； ； ； in, It is the weighted sum of the cost of peak-valley difference smoothing and the cost of load scheduling; This represents the degree of peak and trough fluctuations in the load curve; for Total load power after time period optimization; and These are the weighting coefficients for peak-valley difference mitigation costs and load scheduling costs, respectively. The scheduling cost of mobile loads; The scheduling cost of deferred load; for Available mobile load dispatch volume for a given time period; for Delayable load scheduling amount for a given time period; and They are respectively The unit scheduling cost of mobile load and the unit scheduling cost of delayed load for a given time period.
9. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the flexible load scheduling method as described in any one of claims 1 to 4.
10. A storage medium, characterized in that, The storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the storage medium is located to perform the flexible load scheduling method as described in any one of claims 1 to 4.