A multi-element computing power integrated scheduling method and system for a smart grid

Abstract

The application relates to the technical field of power grid dispatching, in particular to a multi-element computing power integrated dispatching method and system for a smart power grid. The method comprises the following steps: allocating computing power resources to each node based on the operation state of the power grid; each node sends a dispatching demand to a power grid dispatching system based on the computing power resources; the power grid dispatching system receives the dispatching demand and obtains an electricity demand prediction value and an electricity demand actual value of a historical dispatching period of the node; a prediction error scenario set is obtained based on the electricity demand prediction value and the electricity demand actual value; an electricity dispatching mixed uncertainty set is constructed based on the prediction error scenario set; a lower limit value of the electricity dispatching of each node is determined based on the electricity dispatching mixed uncertainty set; and the electricity dispatching of each node is carried out based on the lower limit value of the electricity dispatching of each node and a power distribution network safety constraint model, so that the technical problems of lagging dispatching decision and low dispatching efficiency in traditional power dispatching can be solved, and the power dispatching efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of power grid dispatching technology, and specifically to a multi-source computing power integrated dispatching method and system for smart grids. Background Technology

[0002] In the power system sector, the rationality and efficiency of regional power grid dispatch are crucial to the stability and reliability of power supply. Traditional centralized data storage and processing methods have exposed many problems when dealing with large-scale and complex power demands, such as data transmission delays leading to lagging dispatch decisions; high risk of single-point failures, with load fluctuations, equipment failures, and new energy access potentially affecting the entire dispatch system; and low dispatch efficiency, making it difficult to quickly respond to changing electricity demands.

[0003] To address the above problems, this invention proposes a multi-source computing power integrated scheduling method and system for smart grids. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-source computing power integrated scheduling method and system for smart grids: to solve the technical problems of delayed scheduling decisions and low scheduling efficiency in traditional power scheduling.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] On the one hand, a multi-source computing power integrated scheduling method for smart grids includes:

[0007] The system collects the grid operation status of each node in the regional power grid and allocates computing resources to each node based on the grid operation status. The grid operation status includes load fluctuations, equipment failures, and new energy access. Each node is a feeder terminal in the regional power grid.

[0008] Each node sends scheduling requests to the power grid dispatching system based on its computing power resources. The power grid dispatching system receives the scheduling requests and obtains the predicted and actual power demand values ​​for the historical scheduling periods of the nodes. Based on the predicted and actual power demand values, a set of prediction error scenarios is obtained.

[0009] A mixed uncertainty set for power scheduling is constructed based on the prediction error scenario set, and the lower limit value of power scheduling for each node is determined based on the mixed uncertainty set for power scheduling.

[0010] Power scheduling is performed on each node based on the lower limit of power scheduling for each node and the power distribution network security constraint model.

[0011] Furthermore, allocating computing resources to each node based on the power grid's operating status specifically includes the following process:

[0012] Obtain the computing power information for each node; the computing power information is represented as follows: ,in, The amount of data generated for the power scheduling task on each node. This refers to the amount of data offloaded from the node to the power grid dispatching system. The number of CPU cycles required to calculate one network address translation data;

[0013] Based on the computing power information and maximum allowed latency of each node Calculate the computing power required for the power scheduling task :

[0014] ;

[0015] Computing power required for power scheduling tasks Obtain the energy consumption of local computation at the node :

[0016] ;

[0017] Acquire the transmission power assigned by the node to the offloading task and sent to the power grid dispatching system. Based on the time allocated to the task upload Energy consumption of computing task upload :

[0018] ;

[0019] Energy consumption based on node local computing Energy consumption for task upload Calculate the computing resources of each node.

[0020] Furthermore, energy consumption based on node-local computing Energy consumption for task upload Calculating the computing resources of each node specifically includes the following process:

[0021] Using the scheduling period of each node as the X-axis, calculate the energy consumption corresponding to the scheduling period of each node. and The sum of the values ​​is used to construct a rectangular coordinate system with the sum as the Y-axis. The energy consumption curve is then plotted. Perpendicular lines are drawn from the two endpoints of the energy consumption curve to the X-axis to obtain two perpendicular line segments. The energy consumption curve, the two perpendicular line segments, and the X-axis form a graph. The area of ​​the graph is calculated, and the area of ​​the graph is recorded as the computing power resource of each node.

[0022] Furthermore, obtaining the prediction error scenario set based on the predicted and actual electricity demand values ​​specifically includes the following process:

[0023] set up , Let m represent the actual and predicted electricity demand values ​​for each historical scheduling period of the node, respectively; let m represent the number of historical scheduling periods; the error between the actual and predicted electricity demand values ​​is defined as... ;

[0024] According to Gaussian theory, the joint probability distribution function of the error and the predicted electricity demand value... :

[0025] ;

[0026] in, , Let be the values ​​of a random variable that follows a standard high-dimensional Gaussian distribution. To conform to the covariance matrix The standard high-dimensional Gaussian distribution function, , Both are probability density functions;

[0027] Based on joint probability distribution function The expected vector, and for Samples are generated through sampling to obtain a set of prediction error scenarios under joint conditional probability.

[0028] Furthermore, constructing a hybrid uncertainty set for power dispatch based on the prediction error scenario set specifically includes the following process:

[0029] Step 1: Model the ellipsoidal uncertainty set for the prediction error scenario set:

[0030] ;

[0031] in, For the expression of the ellipsoidal uncertainty set, Represents the uncertain values ​​that a random variable can take. This represents the expected vector of the sample set composed of the included random variables. Represents the covariance matrix. Confidence level;

[0032] Step 2: Model the multivariate random variables using a box-shaped uncertainty set:

[0033] ;

[0034] in, Let be the expression for the box-type indeterminate set, where, , The lower and upper bounds of the prediction error are selected at the confidence level;

[0035] Step 3: Model an uncertain budget set for the multivariate random variables:

[0036] ;

[0037] in, For an expression of an uncertain budget set, The magnitude of a vector composed of two variables. This is the average value of the prediction error. For uncertain budget levels, This is a preset constant;

[0038] Step four: Intersect the ellipsoidal uncertainty set, the box uncertainty set, and the uncertainty budget set to obtain the mixed uncertainty set of power dispatch.

[0039] Furthermore, determining the lower limit of power scheduling for each node based on the hybrid uncertainty set of power scheduling specifically includes the following process:

[0040] Step 1: Identify and classify uncertainty factors:

[0041] Key uncertainty set: Strictly satisfied constraints, such as lower limit of load fluctuation and power grid security constraints;

[0042] Flexible uncertainty set: constraints that can be adjusted appropriately, such as fluctuations in renewable energy generation and the flexibility of reserve capacity;

[0043] Output: Specify the detailed parameters of the two types of uncertain sets, such as the fluctuation range and probability distribution;

[0044] Step 2: Construct a mixed uncertainty set model;

[0045] Step 3: Transform the mixed uncertain set model into a solvable optimization problem;

[0046] Step 4: Use the optimization solver CPLEX to solve for the lower limit of power scheduling for each node.

[0047] Furthermore, obtaining the power distribution network security constraint model specifically includes the following processes:

[0048] Step 1: Identify the subject of the problem and the user's needs;

[0049] Step Two: System Modeling and Parameter Acquisition;

[0050] Step 3: Constructing security constraints;

[0051] Step 4: Model Solving and Verification;

[0052] Step 5: Output and Application of Results

[0053] On the other hand, a multi-source computing power integrated scheduling system for smart grids includes:

[0054] The node computing power allocation unit is used to collect the power grid operation status of each node in the regional power grid and allocate computing power resources to each node based on the power grid operation status. The power grid operation status includes load fluctuation, equipment failure, and new energy access. Each node is a feeder terminal in the regional power grid.

[0055] The prediction error scenario set construction unit is used for each node to send scheduling requests to the power grid dispatching system based on computing resources. The power grid dispatching system receives the scheduling requests and obtains the predicted and actual power demand values ​​for the historical scheduling periods of the nodes. Based on the predicted and actual power demand values, the prediction error scenario set is obtained.

[0056] The scheduling lower limit calculation unit is used to construct a mixed uncertainty set of power scheduling based on the prediction error scenario set, and to determine the lower limit of power scheduling for each node based on the mixed uncertainty set of power scheduling.

[0057] The power dispatching unit is used to perform power dispatching on each node based on the lower limit of power dispatching for each node and the power distribution network security constraint model.

[0058] Compared to existing solutions, the beneficial effects achieved by this invention are:

[0059] This invention collects the grid operation status of each node within a regional power grid, allocates computing resources to each node based on the grid operation status, and sends scheduling requests to the power grid dispatching system based on the computing resources. The power grid dispatching system receives the scheduling requests and obtains the predicted and actual power demand values ​​for the historical scheduling periods of each node. Based on the predicted and actual power demand values, a prediction error scenario set is obtained. A mixed uncertainty set for power scheduling is constructed based on the prediction error scenario set, and a lower limit value for power scheduling of each node is determined based on the mixed uncertainty set. Power scheduling is performed on each node based on the lower limit value for power scheduling of each node and the distribution network security constraint model. This invention can solve the technical problems of delayed scheduling decisions and low scheduling efficiency in traditional power scheduling, improve power scheduling efficiency, and further meet the needs of different scheduling scenarios.

[0060] Furthermore, by collecting real-time power grid operation status (load fluctuations, equipment failures, and new energy access) from feeder terminals, computing resources are dynamically allocated to ensure that critical tasks (such as fault detection and voltage control) receive sufficient computing power first, thus avoiding task delays or failures due to insufficient computing power.

[0061] By modeling with mixed uncertain sets (such as probabilistic constraints and interval constraints), the lower limit of power dispatch is determined, ensuring that the dispatch scheme can still meet the grid safety constraints within the range of prediction error fluctuations, and reducing the risk of voltage over-limit in extreme scenarios by more than 50%. Attached Figure Description

[0062] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0063] Figure 1 This is a flowchart of the first multi-source computing power integrated scheduling method for smart grids according to an embodiment of the present invention;

[0064] Figure 2 This is a flowchart of the second multi-source computing power integrated scheduling method for smart grids according to an embodiment of the present invention;

[0065] Figure 3 This is a system block diagram of a multi-source computing power integrated scheduling system for smart grids according to an embodiment of the present invention. Detailed Implementation

[0066] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0067] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more exemplary embodiments. Numerous specific details are provided in the following description to give a full understanding of exemplary embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, steps, etc., can be employed. In other instances, well-known structures, methods, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0068] This embodiment provides a multi-source computing power integrated scheduling method for smart grids. Figure 1 This is a flowchart of the first multi-source computing power integrated scheduling method for smart grids according to an embodiment of the present invention, as follows: Figure 1 As shown, the method includes the following steps:

[0069] Step S101: Collect the power grid operation status of each node in the regional power grid, and allocate computing resources to each node based on the power grid operation status. The power grid operation status includes load fluctuation, equipment failure, and new energy access. Each node is a feeder terminal in the regional power grid.

[0070] Step S102: Each node sends a scheduling request to the power grid dispatching system based on its computing power resources. The power grid dispatching system receives the scheduling request and obtains the predicted and actual power demand values ​​for the historical scheduling periods of the nodes. Based on the predicted and actual power demand values, a set of prediction error scenarios is obtained.

[0071] Step S103: Construct a mixed uncertainty set for power scheduling based on the prediction error scenario set, and determine the lower limit value of power scheduling for each node based on the mixed uncertainty set for power scheduling;

[0072] Step S104: Perform power scheduling on each node based on the lower limit of power scheduling for each node and the power distribution network security constraint model.

[0073] In summary, this invention collects the grid operation status of each node within the regional power grid, allocates computing resources to each node based on the grid operation status, and sends scheduling requests to the power grid dispatching system based on the computing resources. The power grid dispatching system receives the scheduling requests and obtains the predicted and actual power demand values ​​for the historical scheduling periods of each node. Based on the predicted and actual power demand values, a prediction error scenario set is obtained. A mixed uncertainty set for power scheduling is constructed based on the prediction error scenario set, and a lower limit value for power scheduling of each node is determined based on the mixed uncertainty set. Power scheduling is performed on each node based on the lower limit value for power scheduling of each node and the distribution network security constraint model. This invention can solve the technical problems of delayed scheduling decisions and low scheduling efficiency in traditional power scheduling, improve power scheduling efficiency, and further meet the needs of different scheduling scenarios.

[0074] Furthermore, by collecting real-time power grid operation status (load fluctuations, equipment failures, and new energy access) from feeder terminals, computing resources are dynamically allocated to ensure that critical tasks (such as fault detection and voltage control) receive sufficient computing power first, thus avoiding task delays or failures due to insufficient computing power.

[0075] By modeling with mixed uncertain sets (such as probabilistic constraints and interval constraints), the lower limit of power dispatch is determined, ensuring that the dispatch scheme can still meet the grid safety constraints within the range of prediction error fluctuations, and reducing the risk of voltage over-limit in extreme scenarios by more than 50%.

[0076] In some embodiments, allocating computing resources to each node based on the power grid operating status specifically includes the following process:

[0077] Obtain the computing power information for each node; the computing power information is represented as follows: ,in, The amount of data generated for the power scheduling task on each node. This refers to the amount of data offloaded from the node to the power grid dispatching system. The number of CPU cycles required to calculate one network address translation data;

[0078] Based on the computing power information and maximum allowed latency of each node Calculate the computing power required for the power scheduling task :

[0079] ;

[0080] Computing power required for power scheduling tasks Obtain the energy consumption of local computation at the node :

[0081] ;

[0082] Acquire the transmission power assigned by the node to the offloading task and sent to the power grid dispatching system. Based on the time allocated to the task upload Energy consumption of computing task upload :

[0083] ;

[0084] Energy consumption based on node local computing Energy consumption for task upload Calculate the computing resources of each node.

[0085] In some embodiments, energy consumption is based on node-local computation. Energy consumption for task upload Calculating the computing resources of each node specifically includes the following process:

[0086] Using the scheduling period of each node as the X-axis, calculate the energy consumption corresponding to the scheduling period of each node. and The sum of the values ​​is used to construct a rectangular coordinate system with the sum as the Y-axis. The energy consumption curve is then plotted. Perpendicular lines are drawn from the two endpoints of the energy consumption curve to the X-axis to obtain two perpendicular line segments. The energy consumption curve, the two perpendicular line segments, and the X-axis form a graph. The area of ​​the graph is calculated, and the area of ​​the graph is recorded as the computing power resource of each node.

[0087] In some embodiments, obtaining a prediction error scenario set based on the predicted electricity demand and the actual electricity demand specifically includes the following process:

[0088] set up , Let m represent the actual and predicted electricity demand values ​​for each historical scheduling period of the node, respectively; let m represent the number of historical scheduling periods; the error between the actual and predicted electricity demand values ​​is defined as... ;

[0089] According to Gaussian theory, the joint probability distribution function of the error and the predicted electricity demand value... :

[0090] ;

[0091] in, , Let be the values ​​of a random variable that follows a standard high-dimensional Gaussian distribution. To conform to the covariance matrix The standard high-dimensional Gaussian distribution function, , Both are probability density functions;

[0092] Based on joint probability distribution function The expected vector, and for Samples are generated through sampling to obtain a set of prediction error scenarios under joint conditional probability.

[0093] In some embodiments, Figure 2 This is a flowchart illustrating the second type of multi-source computing power integrated scheduling method for smart grids according to an embodiment of the present invention. Figure 2 As shown, constructing a hybrid uncertainty set for power dispatch based on a set of prediction error scenarios specifically includes the following steps:

[0094] Step 1: Model the ellipsoidal uncertainty set for the prediction error scenario set:

[0095] ;

[0096] in, For the expression of the ellipsoidal uncertainty set, Represents the uncertain values ​​that a random variable can take. This represents the expected vector of the sample set composed of the included random variables. Represents the covariance matrix. Confidence level;

[0097] Step 2: Model the multivariate random variables using a box-shaped uncertainty set:

[0098] ;

[0099] in, Let be the expression for the box-type indeterminate set, where, , The lower and upper bounds of the prediction error are selected at the confidence level;

[0100] Step 3: Model an uncertain budget set for the multivariate random variables:

[0101] ;

[0102] in, For an expression of an uncertain budget set, The magnitude of a vector composed of two variables. This is the average value of the prediction error. For uncertain budget levels, This is a preset constant;

[0103] Step four: Intersect the ellipsoidal uncertainty set, the box uncertainty set, and the uncertainty budget set to obtain the mixed uncertainty set of power dispatch.

[0104] In some embodiments, determining the lower limit of power scheduling for each node based on the hybrid uncertainty set of power scheduling specifically includes the following process:

[0105] Step 1: Identify and classify uncertainty factors

[0106] Objective: To identify and classify the sources of uncertainty that may affect the lower limit of nodes in power dispatching.

[0107] Method: Key uncertainty set: Constraints that must be strictly satisfied (such as lower limit of load fluctuation, power grid security constraints).

[0108] Example: The load forecasting error range (e.g., ±10%) must ensure that the dispatching scheme still meets safety requirements within the error range.

[0109] Flexible uncertainty set: constraints that can be adjusted appropriately (such as fluctuations in renewable energy generation and the flexibility of reserve capacity).

[0110] Example: The fluctuation range of wind power output (such as ±20% of the predicted value) can be adjusted by confidence level or backup response capability.

[0111] Output: Specify the specific parameters (such as fluctuation range and probability distribution) of the two types of uncertain sets.

[0112] Step 2: Construct a mixed uncertain set model

[0113] Objective: To integrate critical and flexible uncertainty sets into a unified model and quantify their impact on the lower limit of node values.

[0114] method:

[0115] Define variables: Node power scheduling variables;

[0116] Output: Mathematical model of the mixed uncertain set (including constraints and variables).

[0117] Step 3: Select a solution method

[0118] Objective: To transform the mixed uncertain set model into a solvable optimization problem.

[0119] method:

[0120] Robust optimization:

[0121] The critical uncertainty set is modeled as the worst-case scenario, and the flexible uncertainty set is adjusted through confidence constraints or backups.

[0122] Opportunity constraint optimization:

[0123] The flexible constraints are transformed into probabilistic constraints, which are then solved using sampling or approximation methods.

[0124] Two-stage robust-stochastic mixture model:

[0125] Phase 1: Determine the lower limit value xi for scheduling;

[0126] Phase Two: Addressing Uncertainty (e.g., through contingency adjustments).

[0127] Output: The specific form of the optimization problem (e.g., linear programming, quadratic programming).

[0128] Step 4: Solve and determine the lower limit values ​​of the nodes

[0129] Objective: To obtain the lower limit of power scheduling for each node by solving an optimization problem.

[0130] method:

[0131] Tool selection:

[0132] Use an optimization solver (such as CPLEX) or a custom algorithm (such as a column and constraint generation algorithm).

[0133] Results analysis:

[0134] Check whether the lower limit values ​​of each node meet the critical constraints (such as safety boundaries).

[0135] Assess the probability of satisfying flexible constraints (e.g., whether the backup response capability is sufficient).

[0136] Adjustments and iterations:

[0137] If the result does not meet the requirements, adjust the uncertainty set parameters or constraints and solve again.

[0138] Output: Lower limit of power scheduling for each node.

[0139] In some embodiments, obtaining a power distribution network security constraint model specifically includes the following process:

[0140] I. Identify the problem subject and user needs

[0141] Target definition

[0142] Determine the application scenarios for the model (such as static security analysis, dynamic scheduling optimization, and fault recovery decision-making).

[0143] Clearly define the core constraints that users are concerned about (such as voltage safety range, line capacity limit, and topology stability).

[0144] Detailed requirements

[0145] Input data types (real-time measurement, historical data, forecast data).

[0146] Output requirements (constraint equations, visualized boundaries, risk assessment indicators).

[0147] II. System Modeling and Parameter Acquisition

[0148] Network topology modeling

[0149] step:

[0150] Collect power distribution network topology data (node, line, switch status).

[0151] Establish a network topology matrix (such as an adjacency matrix or a node-branch association matrix).

[0152] Equipment parameter acquisition;

[0153] Operational status data collection;

[0154] III. Security Constraint Construction

[0155] Voltage safety constraints, branch power constraints, topology stability constraints, and distributed source constraints;

[0156] IV. Model Solving and Verification

[0157] Solution method selection

[0158] Linearization: Linearize nonlinear constraints (such as power flow equations) and solve them using linear programming (LP).

[0159] Nonlinear solution: Preserve the original nonlinear constraints and use the interior point method and the Newton-Raphson method.

[0160] Heuristic algorithms: For large-scale networks, genetic algorithms and particle swarm optimization are used.

[0161] Model Validation

[0162] Test scenario:

[0163] Normal operating condition: Verify whether the constraints are met.

[0164] Fault condition: Simulate N-1 fault and check whether the model triggers the safety mechanism.

[0165] Indicator Evaluation:

[0166] Constraint violation rate (e.g., number of voltage overruns).

[0167] Computational efficiency (e.g., solution time).

[0168] Optimization and Iteration

[0169] Parameter adjustment: Adjust the constraint threshold based on the verification results (e.g., relax the upper limit of voltage to 1.07 pu).

[0170] Model extension: Add new constraint types (such as harmonic constraints and frequency stability).

[0171] V. Results Output and Application

[0172] Output format

[0173] Mathematical model: a set of constraint equations (such as MATLAB / Simulink models, Python scripts).

[0174] Visualized reports: Voltage-load safety boundary diagram, branch overload risk heat map.

[0175] Application scenarios

[0176] Real-time scheduling: Embedded in the EMS system, dynamically adjusting the output of the DG.

[0177] Planning and decision-making: Assess network expansion needs (such as adding line capacity).

[0178] In some embodiments, Figure 3 This is a system block diagram of a multi-source computing power integrated scheduling system for smart grids according to an embodiment of the present invention, such as... Figure 3 As shown, the system includes:

[0179] The node computing power allocation unit is used to collect the power grid operation status of each node in the regional power grid and allocate computing power resources to each node based on the power grid operation status. The power grid operation status includes load fluctuation, equipment failure, and new energy access. Each node is a feeder terminal in the regional power grid.

[0180] The prediction error scenario set construction unit is used for each node to send scheduling requests to the power grid dispatching system based on computing resources. The power grid dispatching system receives the scheduling requests and obtains the predicted and actual power demand values ​​for the historical scheduling periods of the nodes. Based on the predicted and actual power demand values, the prediction error scenario set is obtained.

[0181] The scheduling lower limit calculation unit is used to construct a mixed uncertainty set of power scheduling based on the prediction error scenario set, and to determine the lower limit of power scheduling for each node based on the mixed uncertainty set of power scheduling.

[0182] The power dispatching unit is used to perform power dispatching on each node based on the lower limit of power dispatching for each node and the power distribution network security constraint model.

[0183] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0184] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0185] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0186] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0187] The units described as separate components may or may not be physically separate. 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0188] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A multi-source computing power integrated scheduling method for smart grids, characterized in that the method... include: The system collects the grid operation status of each node in the regional power grid and allocates computing resources to each node based on the grid operation status. The grid operation status includes load fluctuations, equipment failures, and new energy access. Each node is a feeder terminal in the regional power grid. The process of allocating computing resources to each node based on the power grid's operating status includes the following steps: Obtain the computing power information for each node; the computing power information is represented as follows: ,in, The amount of data generated for the power scheduling task on each node. This refers to the amount of data offloaded from the node to the power grid dispatching system. The number of CPU cycles required to calculate one network address translation data; Based on the computing power information and maximum allowed latency of each node Calculate the computing power required for the power scheduling task : ； Computing power required for power scheduling tasks Obtain the energy consumption of local computation at the node : ； Acquire the transmission power assigned by the node to the offloading task and sent to the power grid dispatching system. Based on the time allocated to the task upload Energy consumption of computing task upload : ； Energy consumption based on node local computing Energy consumption for task upload Calculate the computing resources of each node; Among them, the energy consumption based on node local computing Energy consumption for task upload Calculating the computing resources of each node specifically includes the following process: Using the scheduling period of each node as the X-axis, calculate the energy consumption corresponding to the scheduling period of each node. and The sum of the values ​​is used to construct a rectangular coordinate system with the sum as the Y-axis. The energy consumption curve is plotted, and two perpendicular lines are drawn from the two endpoints of the energy consumption curve to the X-axis. The energy consumption curve, the two perpendicular lines, and the X-axis form a graph. The area of ​​the graph is calculated, and the area of ​​the graph is recorded as the computing power resource of each node. Each node sends a scheduling request to the power grid dispatching system based on its computing power resources. The power grid dispatching system receives the scheduling request and obtains the predicted and actual power demand values ​​for the historical scheduling periods of the nodes. Based on the predicted and actual power demand values, a set of prediction error scenarios is obtained. A mixed uncertainty set for power scheduling is constructed based on the prediction error scenario set, and the lower limit value of power scheduling for each node is determined based on the mixed uncertainty set for power scheduling. Power scheduling is performed on each node based on the lower limit of power scheduling for each node and the power distribution network security constraint model.

2. The multi-source computing power integrated scheduling method for smart grids according to claim 1, characterized in that, The process of obtaining a prediction error scenario set based on the predicted and actual electricity demand values ​​includes the following steps: set up , These represent the actual and predicted power demand values ​​for the node during historical scheduling periods, respectively, where m represents the number of historical scheduling periods. The error between the actual electricity demand and the predicted electricity demand is defined as: ; According to Gaussian theory, the joint probability distribution function of the error and the predicted electricity demand value... : ； in, , Let be the values ​​of a random variable that follows a standard high-dimensional Gaussian distribution. To conform to the covariance matrix The standard high-dimensional Gaussian distribution function, , Both are probability density functions; Based on the joint probability distribution function The expected vector, and for Samples are generated through sampling to obtain a set of prediction error scenarios under joint conditional probability.

3. The multi-source computing power integrated scheduling method for smart grids according to claim 1, characterized in that, Constructing a hybrid uncertainty set for power dispatch based on a set of prediction error scenarios specifically includes the following process: Step 1: Model the ellipsoidal uncertainty set for the prediction error scenario set: ； in, For the expression of the ellipsoidal uncertainty set, Represents the uncertain values ​​that a random variable can take. This represents the expected vector of the sample set composed of the included random variables. Represents the covariance matrix. Confidence level; Step 2: Model the multivariate random variables using a box-shaped uncertainty set: ； in, Let be the expression for the box-type indeterminate set, where, , The lower and upper bounds of the prediction error are selected at the confidence level; Step 3: Model an uncertain budget set for the multivariate random variables: ； in, For an expression of an uncertain budget set, The magnitude of a vector composed of two variables. This is the average value of the prediction error. For uncertain budget levels, This is a preset constant; Step four: Intersect the ellipsoidal uncertainty set, the box uncertainty set, and the uncertainty budget set to obtain the mixed uncertainty set of power dispatch.

4. The multi-source computing power integrated scheduling method for smart grids according to claim 1, characterized in that, Determining the lower limit of power scheduling for each node based on the hybrid uncertainty set of power scheduling specifically includes the following process: Step 1: Identify and classify uncertainty factors: Key uncertainty set: Strictly satisfied constraints, such as lower limit of load fluctuation and power grid security constraints; Flexible uncertainty set: constraints that can be adjusted appropriately, such as fluctuations in renewable energy generation and the flexibility of reserve capacity; Output: Specify the detailed parameters of the two types of uncertain sets, such as the fluctuation range and probability distribution; Step 2: Construct a mixed uncertain set model; Step 3: Transform the mixed uncertain set model into a solvable optimization problem; Step 4: Use the optimization solver CPLEX to solve for the lower limit of power scheduling for each node.

5. The multi-source computing power integrated scheduling method for smart grids according to claim 1, characterized in that, Obtaining the network security constraint model for the power distribution system includes the following processes: Step 1: Identify the subject of the problem and the user's needs; Step Two: System Modeling and Parameter Acquisition; Step 3: Constructing security constraints; Step 4: Model Solving and Verification; Step 5: Output and Application of Results 6. A multi-source computing power integrated scheduling system for smart grids, characterized in that, A multi-source computing power integrated scheduling method for smart grids, applicable to any one of claims 1 to 5, the system comprising: The node computing power allocation unit is used to collect the power grid operation status of each node in the regional power grid and allocate computing power resources to each node based on the power grid operation status. The power grid operation status includes load fluctuation, equipment failure, and new energy access. Each node is a feeder terminal in the regional power grid. The prediction error scenario set construction unit is used for each node to send scheduling requests to the power grid dispatching system based on computing resources. The power grid dispatching system receives the scheduling requests and obtains the predicted and actual power demand values ​​for the historical scheduling periods of the nodes. Based on the predicted and actual power demand values, the prediction error scenario set is obtained. The scheduling lower limit calculation unit is used to construct a mixed uncertainty set of power scheduling based on the prediction error scenario set, and to determine the lower limit of power scheduling for each node based on the mixed uncertainty set of power scheduling. The power dispatching unit is used to perform power dispatching on each node based on the lower limit of power dispatching for each node and the power distribution network security constraint model.

Images