Unit commitment scheduling method based on constraint space-time diagram neural network and related device

Abstract

The invention discloses a unit combination scheduling method based on a constraint space-time diagram neural network and a related device. The method comprises the steps of collecting bus load data and power system load data of a power grid at each moment every day; establishing a unit combination problem by taking each day as a unit, and solving to obtain corresponding optimal unit power under constraint conditions; using a sliding window to divide the bus load and the corresponding optimal unit power according to time periods; based on the expanded training data set, taking the divided bus load as input data, taking the corresponding optimal unit power as a data label, and training a pre-established space-time diagram neural network model; and dividing bus load data needing to be predicted according to time periods, and inputting the trained time-space diagram neural network model to obtain a unit combination scheduling scheme. According to the method, the solving precision of a unit commitment scheme based on data driving is improved, the time consumption and the calculation amount for calling a solver are reduced, a linear constraint layer and a post-processing method are constructed, and the reliability of a prediction result is perfected.

Description

Technical Field

[0001] The present invention belongs to the field of power system dispatching, and in particular relates to a unit combination dispatching method based on a constrained spatiotemporal graph neural network and a related device. Background Art

[0002] Unit commitment (UC) is a key issue in power system scheduling to formulate power generation plans, that is, to arrange the start and stop and output of the units in advance according to the load forecast results, so as to minimize the cost of power system operation. In order to ensure the safe operation of the power system, the safety constraints of the power system must be considered in the unit commitment, such as the ramp rate of the generator, the upper and lower limits of the output, and the load balance constraints at each moment.

[0003] At present, the mainstream algorithms for solving UC problems include heuristic methods, mixed integer linear programming, Lagrangian relaxation, machine learning and deep learning. Heuristic methods are algorithms that simulate evolution or physical phenomena. Heuristic algorithms have a small amount of computation and are relatively simple to implement, but they cannot guarantee the global optimality of the solution. Mixed integer linear programming is the mainstream algorithm for solving UC problems. Commonly used solution methods include branch and bound method and cutting plane method. Mixed integer linear programming has high solution accuracy. Commercial solvers such as Gurobi and Cplex are based on this method, but this method also has disadvantages such as long solution time and high consumption of computing resources. Lagrangian relaxation method iteratively solves the original problem and updates the dual variables through the dual ascent method, and obtains a set of feasible solutions by reducing the dual gap, but it cannot guarantee the optimality of the results. In recent years, data-driven algorithms, such as machine learning methods and deep learning methods, have also been widely used in UC problems. The mainstream method is to use machine learning algorithms to predict the start and stop of units and then solve the unit power. This method can greatly improve the solution efficiency while obtaining a high-quality initial solution. However, existing data-driven algorithms break the connection between nodes and cannot directly establish a mapping from load to approximate solution. Data-driven algorithms usually focus on the data itself, but do not pay enough attention to the real physical connection between grid nodes, making it difficult to learn the complex mapping relationship between bus load and unit output. At the same time, most of the existing data-driven algorithms for solving unit combination problems require the help of solvers to complete economic scheduling, which is still time-consuming. Summary of the invention

[0004] The purpose of the present invention is to address the problems in the above-mentioned prior art and provide a unit combination scheduling method and related devices based on constrained spatiotemporal graph neural network, improve the accuracy of solving unit combination schemes based on data-driven, reduce the time and amount of calculation for calling the solver, construct a linear constraint layer and post-processing method, and improve the reliability of the prediction results of the spatiotemporal graph neural network.

[0005] In order to achieve the above object, the present invention has the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a unit commitment scheduling method based on a constrained spatiotemporal graph neural network, comprising:

[0007] Collect bus load data and power system load data of the power grid at various times of the day to establish a training data set;

[0008] Construct the unit combination problem on a daily basis and solve it under the constraints to obtain the corresponding optimal unit power;

[0009] Use a sliding window to divide the bus load and the corresponding optimal unit power according to time periods to expand the training data set;

[0010] Based on the expanded training data set, the divided bus load is used as input data, and the corresponding optimal unit power is used as a data label to train a pre-established spatiotemporal graph neural network model; the spatiotemporal graph neural network model adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints;

[0011] The bus load data that needs to be predicted is divided into time periods and input into the trained space-time graph neural network model. The predicted power output by the space-time graph neural network model is restricted, including filtering the prediction results that are less than the minimum power and adjusting the prediction results that do not meet the start-stop requirements, to ensure the availability of the predicted power and obtain the unit combination scheduling plan.

[0012] As a preferred solution, the step of constructing the unit commitment problem on a daily basis includes:

[0013] Assume there are N nodes, N B busbar nodes, N G The total number of decision-making periods per day in the power system is N. T ; The objective function of the unit commitment problem is constructed as follows:

[0014]

[0015] In the formula, g and t represent the sequence number of the unit and time period respectively; N G and N T Represent the total number of units and time periods respectively; p g,t is the output of unit g in time period t; a g , b g , c gare the coefficients of the quadratic term, linear term and constant term of the operating cost function of unit g; SU g is the startup cost of unit g; y g,t , z g,t It is an integer variable indicating whether the unit g has started or stopped in the time period t. The value of 1 indicates that the unit g has started or stopped in the time period t, and the value of 0 indicates that the unit g has not started or stopped.

[0016] As a preferred solution, in the step of solving the corresponding optimal unit power under the constraint conditions, the constraint conditions include:

[0017] Power balance constraints:

[0018]

[0019] Where b is the node number; N B is the total number of busbar nodes; D b,t is the load of node b in time period t;

[0020] Unit output upper and lower limit constraints:

[0021]

[0022] In the formula, and are the minimum and maximum output of unit g respectively; u g,t is an integer variable indicating the on / off status of unit g in time period t. A value of 1 indicates that unit g is turned on in time period t, and a value of 0 indicates that unit g is turned off in time period t.

[0023] Unit climbing constraints:

[0024] -RD g ≤p g,t -p g,t-1 ≤RU g ,g=1,…,N G ,T=2,…,N T

[0025] Where, RD g and RU g They are the climbing rate limits for the decrease and increase of the unit g output respectively;

[0026] Minimum start and stop time constraints:

[0027]

[0028] y g,t +z g,t ≤1,g=1,…,N G ,t=1,…,N T

[0029] In the formula, TU g and TD g are the minimum start and stop time of unit g respectively;

[0030] Safety constraints on active power flow in base-state branches:

[0031]

[0032] Where l is the branch number; N L is the total number of branches; is the active power flow of branch l in time period t under the base state; T is the power transfer distribution factor; g∈G(b) is the set of units on node b; F l max is the long-term allowable current-carrying power of branch l.

[0033] As a preferred solution, the steps of training a pre-established spatiotemporal graph neural network model based on the expanded training data set, taking the divided bus load as input data, and taking the corresponding optimal unit power as a data label include:

[0034] Based on the expanded training data set, the graph neural connection relationship is obtained from the real power grid topology structure;

[0035] The load D of all nodes in the power grid within a period of time F is input into the spatiotemporal graph convolutional network without a linear constraint layer, and the input spatial features are extracted through the graph convolution layer of the spatiotemporal graph convolutional network; the graph convolution layer extracts and aggregates the load features of each node, learns the spatial features between the data, and expands the time series length of the data;

[0036] The graph data processed by the graph convolution layer is converted into time series data, and then the causal expansion time convolution is used to extract the time series features of each node. Subsequently, the fully connected neural network layer is used to aggregate the time series features between different nodes, and the format of the data is converted to output the predicted power P of the thermal power unit within a period of time F; the neural network updates the loss function between the predicted power and the actual unit power scheduling data, repeatedly executes the above steps, and completes the training of the spatiotemporal graph neural network model.

[0037] As a preferred solution, the step of projecting the predicted power of the neural network into a feasible domain that meets the specified linear constraints includes:

[0038] make represents the input value of the constrained activation layer, that is, the predicted power value of the spatiotemporal graph neural network model; represents the output of the constrained activation layer, that is, the power value that meets the linear constraint; where N GRepresents the number of units, F represents the time series length; Convex feasible sets are constructed for all linear constraints in the unit combination problem. The convex feasible set Ω is the intersection of a set of constraints in the form of m inequality constraints, and the mathematical expression is as follows:

[0039] Ω={x|h i (x)≤0,i=1,...,m}

[0040] Ensure that each constraint h i (x)≤0 is convex, that is, it conforms to the following mathematical expression:

[0041]

[0042] Define a fixed point p in the convex feasible set Ω, that is, p∈Ω, then any point x in the set Ω is expressed as x=p+α·r, and the proportional factor α≥0 is calculated; is a vector from point p; for any p and r, the scaling factor α has an upper limit The mathematical expression is as follows:

[0043]

[0044] At this time, the line segment Belongs to the set Ω, upper limit The meaning of is the intersection of ray r and a constraint;

[0045] For the point z outside the constrained feasible domain that appears in the spatiotemporal graph neural network, by specifying the ray r(z) = zp, the scaling factor The calculation is performed through the following mathematical expression, that is, projecting the point z outside the feasible domain along the ray from z to p to the point g inside the feasible domain:

[0046]

[0047] As a preferred solution, the predicted power output by the spatiotemporal graph neural network model is limited, including filtering the predicted results less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements. The steps of ensuring the availability of the predicted power include:

[0048] Set the threshold δ, first check and adjust the output power of the spatiotemporal graph neural network model for each unit, set the prediction result whose difference between the minimum power and the prediction result is less than the threshold δ as the minimum power of the corresponding unit, and set the prediction result whose difference between the minimum power and the prediction result is greater than the threshold δ as 0;

[0049] Check and adjust the minimum start and stop time of each unit in the forecast results;

[0050] The prediction results are re-passed through the linear constraint activation layer with branch active power flow safety constraints added to ensure that the output power meets the branch active power flow safety constraints.

[0051] As a preferred solution, the step of checking and adjusting the minimum start and stop time of each unit in the prediction result includes:

[0052] Record the duration of each start and stop of each unit;

[0053] Check the duration of each start and stop of each unit. For the start and stop that violate the minimum start and stop time constraints, compare the start or stop time with the stop or start time before and after it in chronological order;

[0054] When the start and stop time of a constraint violation is shorter than the start and stop time on both sides, adjust the corresponding start and stop.

[0055] In a second aspect, a unit commitment scheduling system based on a constrained spatiotemporal graph neural network is provided, comprising:

[0056] The data collection module is used to collect the bus load data and power system load data of the power grid at various times of the day and establish a training data set;

[0057] The optimal unit power solving module is used to construct the unit combination problem on a daily basis and solve the corresponding optimal unit power under the constraints;

[0058] A training data set expansion module is used to use a sliding window to divide the bus load and the corresponding optimal unit power according to time periods to expand the training data set;

[0059] The spatiotemporal graph neural network model training module is used to train the pre-established spatiotemporal graph neural network model based on the expanded training data set, using the divided bus load as input data and the corresponding optimal unit power as a data label; the spatiotemporal graph neural network model adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints;

[0060] The predicted power post-processing output module is used to divide the bus load data that needs to be predicted according to time periods, input the trained space-time graph neural network model, and limit the predicted power output by the space-time graph neural network model, including filtering the prediction results that are less than the minimum power and adjusting the prediction results that do not meet the start-stop requirements, to ensure the availability of the predicted power and obtain the unit combination scheduling plan.

[0061] As a preferred solution, the steps of constructing the unit combination problem on a daily basis in the optimal unit power solving module include:

[0062] Assume there are N nodes, N B busbar nodes, N G The total number of decision-making periods per day in the power system is N. T ; The objective function of the unit commitment problem is constructed as follows:

[0063]

[0064] In the formula, g and t represent the sequence number of the unit and time period respectively; N G and N T Represent the total number of units and time periods respectively; p g,t is the output of unit g in time period t; a g , b g , c g are the coefficients of the quadratic term, linear term and constant term of the operating cost function of unit g; SU g is the startup cost of unit g; y g,t , z g,t It is an integer variable indicating whether the unit g has started or stopped in the time period t. The value of 1 indicates that the unit g has started or stopped in the time period t, and the value of 0 indicates that the unit g has not started or stopped.

[0065] As a preferred solution, when the optimal unit power solving module solves the corresponding optimal unit power under constraint conditions, the constraint conditions include:

[0066] Power balance constraints:

[0067]

[0068] Where b is the node number; N B is the total number of busbar nodes; D b,t is the load of node b in time period t;

[0069] Unit output upper and lower limit constraints:

[0070]

[0071] In the formula, and are the minimum and maximum output of unit g respectively; u g,t is an integer variable indicating the on / off status of unit g in time period t. A value of 1 indicates that unit g is turned on in time period t, and a value of 0 indicates that unit g is turned off in time period t.

[0072] Unit climbing constraints:

[0073] -RD g ≤p g,t-p g,t-1 ≤RU g ,g=1,…,N G ,T=2,…,N T

[0074] Where, RD g and RU g They are the climbing rate limits for the decrease and increase of the unit g output respectively;

[0075] Minimum start and stop time constraints:

[0076]

[0077] y g,t +z g,t ≤1,g=1,…,N G ,t=1,…,N T

[0078] In the formula, TU g and TD g are the minimum start and stop time of unit g respectively;

[0079] Safety constraints on active power flow in base-state branches:

[0080]

[0081] Where l is the branch number; N L is the total number of branches; is the active power flow of branch l in time period t under the base state; T is the power transfer distribution factor; g∈G(b) is the set of units on node b; F l max is the long-term allowable current-carrying power of branch l.

[0082] As a preferred solution, the space-time graph neural network model training module uses the divided bus load as input data and the corresponding optimal unit power as a data label, and the steps of training the pre-established space-time graph neural network model include:

[0083] Based on the expanded training data set, the graph neural connection relationship is obtained from the real power grid topology structure;

[0084] The load D of all nodes in the power grid within a period of time F is input into the spatiotemporal graph convolutional network without a linear constraint layer, and the input spatial features are extracted through the graph convolution layer of the spatiotemporal graph convolutional network; the graph convolution layer extracts and aggregates the load features of each node, learns the spatial features between the data, and expands the time series length of the data;

[0085] The graph data processed by the graph convolution layer is converted into time series data, and then the causal expansion time convolution is used to extract the time series features of each node. Subsequently, the fully connected neural network layer is used to aggregate the time series features between different nodes, and the format of the data is converted to output the predicted power P of the thermal power unit within a period of time F; the neural network updates the loss function between the predicted power and the actual unit power scheduling data, repeatedly executes the above steps, and completes the training of the spatiotemporal graph neural network model.

[0086] As a preferred solution, the step of the spatiotemporal graph neural network model training module projecting the predicted power of the neural network into a feasible domain that meets the specified linear constraints includes:

[0087] make represents the input value of the constrained activation layer, that is, the predicted power value of the spatiotemporal graph neural network model; represents the output of the constrained activation layer, that is, the power value that meets the linear constraint; where N G Represents the number of units, F represents the time series length; Convex feasible sets are constructed for all linear constraints in the unit combination problem. The convex feasible set Ω is the intersection of a set of constraints in the form of m inequality constraints, and the mathematical expression is as follows:

[0088] Ω={x|h i (x)≤0,i=1,...,m}

[0089] Ensure that each constraint h i (x)≤0 is convex, that is, it conforms to the following mathematical expression:

[0090]

[0091] Define a fixed point p in the convex feasible set Ω, that is, p∈Ω, then any point x in the set Ω is expressed as x=p+α·r, and the proportional factor α≥0 is calculated; is a vector from point p; for any p and r, the scaling factor α has an upper limit The mathematical expression is as follows:

[0092]

[0093] At this time, the line segment Belongs to the set Ω, upper limit The meaning of is the intersection of ray r and a constraint;

[0094] For a point z outside the constrained feasible domain that appears in the spatiotemporal graph neural network, by specifying the ray r(z) = zp, the scaling factor The calculation is performed through the following mathematical expression, that is, projecting the point z outside the feasible domain along the ray from z to p to the point g inside the feasible domain:

[0095]

[0096] As a preferred solution, the predicted power post-processing output module limits the predicted power output by the spatiotemporal graph neural network model, including filtering the predicted results less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements. When ensuring the availability of the predicted power, a threshold δ is set, and the output power of the spatiotemporal graph neural network model is first checked and adjusted for each unit. The predicted result whose difference between the minimum power and the predicted result is less than the threshold δ is set as the minimum power of the corresponding unit, and the predicted result whose difference between the minimum power and the predicted result is greater than the threshold δ is set to 0;

[0097] Check and adjust the minimum start and stop time of each unit in the forecast results;

[0098] The prediction results are re-passed through the linear constraint activation layer with branch active power flow safety constraints added to ensure that the output power meets the branch active power flow safety constraints.

[0099] As a preferred solution, the step of the predicted power post-processing output module checking and adjusting the minimum start and stop time of each unit in the prediction result includes:

[0100] Record the duration of each start and stop of each unit;

[0101] Check the duration of each start and stop of each unit. For the start and stop that violate the minimum start and stop time constraints, compare the start or stop time with the stop or start time before and after it in chronological order;

[0102] When the start and stop time of a constraint violation is shorter than the start and stop time on both sides, adjust the corresponding start and stop.

[0103] In a third aspect, an embodiment of the present invention provides an electronic device, comprising a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the unit combination scheduling method based on a constrained spatiotemporal graph neural network.

[0104] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the unit combination scheduling method based on the constrained spatiotemporal graph neural network is implemented.

[0105] Compared with the prior art, the first aspect of the present invention has at least the following beneficial effects:

[0106] The unit combination dispatching method based on the constrained spatiotemporal graph neural network of the present invention collects bus load data and power system load data of the power grid at each time of the day, divides the bus load and the corresponding optimal unit power according to time periods, uses the divided bus load as input data, and uses the corresponding optimal unit power as a data label to train a pre-established spatiotemporal graph neural network model. The spatiotemporal graph neural network model is a framework that combines graph neural networks and time series convolutional networks. It can simultaneously learn the physical topology of the power grid and the time series information of historical data, and directly use the bus load to solve the unit power dispatching. In addition, in view of the problem that the data-driven methods currently applied to the power system do not have output hard constraints, the spatiotemporal graph neural network model of the present invention adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints. In addition, the present invention designs a post-processing process, that is, to limit the predicted power output by the spatiotemporal graph neural network model, including filtering the predicted results that are less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements, to ensure the availability of the predicted power, and to improve the reliability of the spatiotemporal graph neural network prediction results. The present invention uses realistic node connection relationships and edge weights to convert the physical power grid topology into graph neural modeling. The graph neural network uses the physical power grid model to learn network topology information during the training process, and achieves higher prediction accuracy with fewer samples. Many existing studies use neural network methods for auxiliary decision-making of unit commitment problems, but the present invention constructs a mapping relationship between known load inputs and generator unit power decisions, and uses the bus load of each node at each moment of the power grid to directly predict the output of the generator unit, which simplifies the calculation process and improves the computational efficiency. The present invention combines the temporal convolutional network with the graph neural network, and simultaneously extracts the spatiotemporal characteristics in the data, thereby improving the prediction accuracy and model robustness. Traditional data-driven algorithms cannot guarantee any constraints during the training process, but the present invention combines data-driven with model-driven methods, and introduces a linear constraint layer in the spatiotemporal graph neural network model, so that the neural network can add linear constraints to the results. The addition of the linear constraint layer simplifies the constraint verification and post-processing of the prediction results, and helps to find feasible solutions that meet the constraints. Combining the above characteristics, the unit combination scheduling method based on the constrained spatiotemporal graph neural network of the present invention can quickly and accurately solve the unit combination problem, and automatically check and adjust the feasibility and accuracy of the solution results. The method of the present invention not only significantly reduces the solution time of power system scheduling, but also greatly improves the level of intelligence in the field of power system scheduling.

[0107] It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0108] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0109] Figure 1 A schematic diagram of the spatiotemporal graph neural network model architecture established for an embodiment of the present invention;

[0110] Figure 2 This is a flow chart of a unit commitment scheduling method based on a constrained spatiotemporal graph neural network according to an embodiment of the present invention;

[0111] Figure 3 Schematic diagram of the working principle of the linear constraint layer according to an embodiment of the present invention. DETAILED DESCRIPTION

[0112] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0113] In order to get rid of the problem that the data-driven unit combination solution algorithm is not accurate enough, and the call of the solver is time-consuming and computationally intensive, the embodiment of the present invention proposes a unit combination scheduling method based on a constrained space-time graph neural network, and constructs a framework that combines graph neural networks and time series convolutional networks, namely, the space-time graph convolutional network. The space-time graph neural network model of the present invention can simultaneously learn the physical topology of the power grid and the time series information of historical data, and directly use the bus load to solve the unit power scheduling. In addition, the data-driven methods currently applied to power systems do not have output hard constraints. The present invention innovatively constructs a linear constraint layer and designs a post-processing method to improve the reliability of the prediction results of the space-time graph neural network.

[0114] See also Figure 2 The unit commitment scheduling method based on the constrained spatiotemporal graph neural network in the embodiment of the present invention mainly includes:

[0115] S1. Collect bus load data and power system load data of the power grid at each time of the day to establish a training data set;

[0116] S2. Construct the unit combination problem on a daily basis and solve it under the constraints to obtain the corresponding optimal unit power;

[0117] S3, using a sliding window, divide the bus load and the corresponding optimal unit power according to time periods to expand the training data set;

[0118] S4. Based on the expanded training data set, the divided bus load is used as input data, and the corresponding optimal unit power is used as a data label to train a pre-established spatiotemporal graph neural network model; the spatiotemporal graph neural network model adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints;

[0119] S5. The bus load data that needs to be predicted is divided into time periods and input into the trained space-time graph neural network model. The predicted power output by the space-time graph neural network model is limited, including filtering the prediction results that are less than the minimum power and adjusting the prediction results that do not meet the start-stop requirements, to ensure the availability of the predicted power and obtain the unit combination scheduling plan.

[0120] The embodiment of the present invention takes the IEEE30-BUS system and the IEEE118-BUS system as examples to verify the effectiveness of the method.

[0121] In a possible implementation, step S1 collects bus load data and power system load data of the power grid at each time of the day, organizes the data format, generates training samples, and establishes a training data set.

[0122] In a possible implementation manner, the step of constructing the unit commitment problem on a daily basis in step S2 includes:

[0123] Assume there are N nodes, N B busbar nodes, N G The total number of decision-making periods per day in the power system is N. T ; The objective function of the unit commitment problem is constructed as follows:

[0124]

[0125] In the formula, g and t represent the sequence number of the unit and time period respectively; N G and N T Represent the total number of units and time periods respectively; p g,t is the output of unit g in time period t; a g , b g , c g are the coefficients of the quadratic term, linear term and constant term of the operating cost function of unit g; SU g is the startup cost of unit g; y g,t , z g,tIt is an integer variable indicating whether the unit g has started or stopped in the time period t. The value of 1 indicates that the unit g has started or stopped in the time period t, and the value of 0 indicates that the unit g has not started or stopped.

[0126] Furthermore, in the step of solving the corresponding optimal unit power under the constraint conditions, the constraint conditions include:

[0127] Power balance constraints:

[0128]

[0129] Where b is the node number; N B is the total number of busbar nodes; D b,t is the load of node b in time period t;

[0130] Unit output upper and lower limit constraints:

[0131]

[0132] In the formula, and are the minimum and maximum output of unit g respectively; u g,t is an integer variable indicating the on / off status of unit g in time period t. A value of 1 indicates that unit g is turned on in time period t, and a value of 0 indicates that unit g is turned off in time period t.

[0133] Unit climbing constraints:

[0134] -RD g ≤p g,t -p g,t-1 ≤RU g ,g=1,…,N G ,T=2,…,N T

[0135] Where, RD g and RU g They are the climbing rate limits for the decrease and increase of the unit g output respectively;

[0136] Minimum start and stop time constraints:

[0137]

[0138] y g,t +z g,t ≤1,g=1,…,N G ,t=1,…,N T

[0139] In the formula, TU g and TD g are the minimum start and stop time of unit g respectively;

[0140] Safety constraints on active power flow in base-state branches:

[0141]

[0142] Where l is the branch number; N L is the total number of branches; is the active power flow of branch l in time period t under the base state; T is the power transfer distribution factor; g∈G(b) is the set of units on node b; F l max is the long-term allowable current-carrying power of branch l.

[0143] Furthermore, in a possible implementation manner, step S2 of the present invention calls the Gurobi solver to solve the unit commitment problem constructed every day and record the optimal unit power dispatch.

[0144] In a possible implementation, the present invention uses a sliding window to construct a new data set in step S3. Specifically, the width of the sliding window is defined as F, and the step length is 1. Then, the present invention starts from the first moment of each day and moves 1 time period at a time until the end of the data set is reached. For each time period t (from 1 to T-F+1, with a step length of s), the present invention takes the load of the F time periods in the window [D b,t ,D b,t+1 ,...,D b,t+F-1 ] and unit output [P g,t ,P g,t+1 ,...,P g,t+F-1 ] as a new sample. The sliding window length can be 6, 12, or 24;

[0145] like Figure 1 As shown, in a possible implementation manner, step S4 of the present invention uses the divided bus load as input data and the corresponding optimal unit power as a data label, and the steps of training the pre-established spatiotemporal graph neural network model include:

[0146] 1) Based on the expanded training data set, the graph neural connection relationship is obtained from the real power grid topology structure;

[0147] 2) The load D of all nodes in the power grid within a period of time F is input into the spatiotemporal graph convolutional network without a linear constraint layer, and the input spatial features are extracted through the graph convolution layer of the spatiotemporal graph convolutional network; the graph convolution layer extracts and aggregates the load features of each node, learns the spatial features between the data, and expands the time series length of the data;

[0148] 3) The graph data processed by the graph convolution layer is converted into time series data, and the causal expansion time convolution is used to extract the time series features of each node. Then, the fully connected neural network layer is used to aggregate the time series features between different nodes, and the format of the data is converted to output the predicted power P of the thermal power unit within a period of time F; the neural network updates the loss function between the predicted power and the actual unit power scheduling data, repeatedly executes steps 2) and 3), and completes the training of the spatiotemporal graph neural network model.

[0149] Further, such as Figure 3 As shown, in a possible implementation manner, step S4 of the embodiment of the present invention adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints, including:

[0150] make represents the input value of the constrained activation layer, that is, the predicted power value of the spatiotemporal graph neural network model; represents the output of the constrained activation layer, that is, the power value that meets the linear constraint; where N G Represents the number of units, F represents the time series length; Convex feasible sets are constructed for all linear constraints in the unit combination problem. The convex feasible set ... is the intersection of a set of constraints in the form of m inequality constraints, and the mathematical expression is as follows:

[0151] Ω={x|h i (x)≤0,i=1,...,m}

[0152] Ensure that each constraint h i (x)≤0 is convex, that is, it conforms to the following mathematical expression:

[0153]

[0154] Define a fixed point p in the convex feasible set Ω, that is, p∈Ω, then any point x in the set Ω is expressed as x=p+α·r, and the proportional factor α≥0 is calculated; is a vector from point p; for any p and r, the scaling factor α has an upper limit The mathematical expression is as follows:

[0155]

[0156] At this time, the line segment Belongs to the set Ω, upper limit The meaning of is the intersection of ray r and a constraint;

[0157] For the point z outside the constrained feasible domain that appears in the spatiotemporal graph neural network, by specifying the ray r(z) = zp, the scaling factor The calculation is performed through the following mathematical expression, that is, projecting the point z outside the feasible domain along the ray from z to p to the point g inside the feasible domain:

[0158]

[0159] In a possible implementation, the step S5 of limiting the predicted power output by the spatiotemporal graph neural network model includes filtering the predicted results less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements. The step of ensuring the availability of the predicted power includes:

[0160] Set the threshold δ, first check and adjust the output power of the spatiotemporal graph neural network model for each unit, set the prediction result whose difference between the minimum power and the prediction result is less than the threshold δ as the minimum power of the corresponding unit, and set the prediction result whose difference between the minimum power and the prediction result is greater than the threshold δ as 0;

[0161] Check and adjust the minimum start and stop time of each unit in the forecast results;

[0162] The prediction results are re-passed through the linear constraint activation layer with branch active power flow safety constraints added to ensure that the output power meets the branch active power flow safety constraints.

[0163] Furthermore, the step of checking and adjusting the minimum start and stop time of each unit in the prediction result includes:

[0164] Record the duration of each start and stop of each unit;

[0165] Check the duration of each start and stop of each unit. For the start and stop that violate the minimum start and stop time constraint, compare the start (stop) time with the stop (start) time before and after it in chronological order;

[0166] When the start and stop time of a constraint violation is shorter than the start and stop time on both sides, adjust the corresponding start and stop.

[0167] The unit combination dispatching method based on the constrained spatiotemporal graph neural network in the embodiment of the present invention can quickly and accurately solve the unit combination problem, and automatically check and adjust the feasibility and accuracy of the solution results. The present invention not only significantly reduces the solution time of power system dispatching, but also greatly improves the intelligence level in the field of power system dispatching.

[0168] Another embodiment of the present invention further provides a unit commitment scheduling system based on a constrained spatiotemporal graph neural network, comprising:

[0169] The data collection module is used to collect the bus load data and power system load data of the power grid at various times of the day and establish a training data set;

[0170] The optimal unit power solving module is used to construct the unit combination problem on a daily basis and solve the corresponding optimal unit power under the constraints;

[0171] A training data set expansion module is used to use a sliding window to divide the bus load and the corresponding optimal unit power according to time periods to expand the training data set;

[0172] The spatiotemporal graph neural network model training module is used to train the pre-established spatiotemporal graph neural network model based on the expanded training data set, using the divided bus load as input data and the corresponding optimal unit power as a data label; the spatiotemporal graph neural network model adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints;

[0173] The predicted power post-processing output module is used to divide the bus load data that needs to be predicted according to time periods, input the trained space-time graph neural network model, and limit the predicted power output by the space-time graph neural network model, including filtering the prediction results that are less than the minimum power and adjusting the prediction results that do not meet the start-stop requirements, to ensure the availability of the predicted power and obtain the unit combination scheduling plan.

[0174] In a possible implementation, the optimal unit power solving module constructs the unit commitment problem on a daily basis, including:

[0175] Assume there are N nodes, N B busbar nodes, N G The total number of decision-making periods per day in the power system is N. T ; The objective function of the unit commitment problem is constructed as follows:

[0176]

[0177] In the formula, g and t represent the sequence number of the unit and time period respectively; N G and N T Represent the total number of units and time periods respectively; p g,t is the output of unit g in time period t; a g , b g , c g are the coefficients of the quadratic term, linear term and constant term of the operating cost function of unit g; SU g is the startup cost of unit g; y g,t , zg,t It is an integer variable indicating whether the unit g has started or stopped in the time period t. The value of 1 indicates that the unit g has started or stopped in the time period t, and the value of 0 indicates that the unit g has not started or stopped.

[0178] In a possible implementation manner, when the optimal unit power solving module solves the corresponding optimal unit power under constraint conditions, the constraint conditions include:

[0179] Power balance constraints:

[0180]

[0181] Where b is the node number; N B is the total number of busbar nodes; D b,t is the load of node b in time period t;

[0182] Unit output upper and lower limit constraints:

[0183]

[0184] In the formula, and are the minimum and maximum output of unit g respectively; u g,t is an integer variable indicating the on / off status of unit g in time period t. A value of 1 indicates that unit g is turned on in time period t, and a value of 0 indicates that unit g is turned off in time period t.

[0185] Unit climbing constraints:

[0186] -RD g ≤p g,t -p g,t-1 ≤RU g ,g=1,…,N G ,T=2,…,N T

[0187] Where, RD g and RU g They are the climbing rate limits for the decrease and increase of the unit g output respectively;

[0188] Minimum start and stop time constraints:

[0189]

[0190] y g,t +z g,t ≤1,g=1,…,N G ,t=1,…,N T

[0191] In the formula, TU g and TD gare the minimum start and stop time of unit g respectively;

[0192] Safety constraints on active power flow in base-state branches:

[0193]

[0194]

[0195] Where l is the branch number; N L is the total number of branches; is the active power flow of branch l in time period t under the base state; T is the power transfer distribution factor; g∈G(b) is the set of units on node b; F l max is the long-term allowable current-carrying power of branch l.

[0196] In a possible implementation, the space-time graph neural network model training module uses the divided bus load as input data and the corresponding optimal unit power as a data label. The steps of training the space-time graph neural network model include:

[0197] Based on the expanded training data set, the graph neural connection relationship is obtained from the real power grid topology structure;

[0198] The load D of all nodes in the power grid within a period of time F is input into the spatiotemporal graph convolutional network without a linear constraint layer, and the input spatial features are extracted through the graph convolution layer of the spatiotemporal graph convolutional network; the graph convolution layer extracts and aggregates the load features of each node, learns the spatial features between the data, and expands the time series length of the data;

[0199] The graph data processed by the graph convolution layer is converted into time series data, and then the causal expansion time convolution is used to extract the time series features of each node. Subsequently, the fully connected neural network layer is used to aggregate the time series features between different nodes, and the format of the data is converted to output the predicted power P of the thermal power unit within a period of time F; the neural network updates the loss function between the predicted power and the actual unit power scheduling data, repeatedly executes the above steps, and completes the training of the spatiotemporal graph neural network model.

[0200] In a possible implementation, the step of the spatiotemporal graph neural network model training module projecting the predicted power of the neural network into a feasible domain that meets the specified linear constraints includes:

[0201] make represents the input value of the constrained activation layer, that is, the predicted power value of the spatiotemporal graph neural network model; represents the output of the constrained activation layer, that is, the power value that meets the linear constraint; where N GRepresents the number of units, F represents the time series length; Convex feasible sets are constructed for all linear constraints in the unit combination problem. The convex feasible set Ω is the intersection of a set of constraints in the form of m inequality constraints, and the mathematical expression is as follows:

[0202] Ω={x|h i (x)≤0,i=1,...,m}

[0203] Ensure that each constraint h i (x)≤0 is convex, that is, it conforms to the following mathematical expression:

[0204]

[0205] Define a fixed point p in the convex feasible set Ω, that is, p∈Ω, then any point x in the set Ω is expressed as x=p+α·r, and the proportional factor α≥0 is calculated; is a vector from point p; for any p and r, the scaling factor α has an upper limit The mathematical expression is as follows:

[0206]

[0207] At this time, the line segment Belongs to the set Ω, upper limit The meaning of is the intersection of ray r and a constraint;

[0208] For the point z outside the constrained feasible domain that appears in the spatiotemporal graph neural network, by specifying the ray r(z) = zp, the scaling factor The calculation is performed through the following mathematical expression, that is, projecting the point z outside the feasible domain along the ray from z to p to the point g inside the feasible domain:

[0209]

[0210] In a possible implementation, the predicted power post-processing output module limits the predicted power output by the spatiotemporal graph neural network model, including filtering the predicted results that are less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements. When the availability of the predicted power is ensured, a threshold δ is set, and the output power of the spatiotemporal graph neural network model is first checked and adjusted for each unit, and the predicted result whose difference between the minimum power and the predicted result is less than the threshold δ is set as the minimum power of the corresponding unit, and the predicted result whose difference between the minimum power and the predicted result is greater than the threshold δ is set to 0;

[0211] Check and adjust the minimum start and stop time of each unit in the forecast results;

[0212] The prediction results are re-passed through the linear constraint activation layer with branch active power flow safety constraints added to ensure that the output power meets the branch active power flow safety constraints.

[0213] Further, the step of the predicted power post-processing output module checking and adjusting the minimum start and stop time of each unit in the predicted result includes:

[0214] Record the duration of each start and stop of each unit;

[0215] Check the duration of each start and stop of each unit. For the start and stop that violate the minimum start and stop time constraints, compare the start or stop time with the stop or start time before and after it in chronological order;

[0216] When the start and stop time of a constraint violation is shorter than the start and stop time on both sides, adjust the corresponding start and stop.

[0217] Another embodiment of the present invention further proposes an electronic device, including a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement the unit combination scheduling method based on the constrained spatiotemporal graph neural network.

[0218] Another embodiment of the present invention further proposes a computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by a processor, the unit combination scheduling method based on the constrained spatiotemporal graph neural network is implemented.

[0219] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electric carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals. For ease of explanation, the above content only shows the part related to the embodiment of the present invention. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The computer-readable storage medium is non-temporary and can be stored in a storage device formed by various electronic devices, which can realize the execution process recorded in the method of the embodiment of the present invention.

[0220] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0221] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0222] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0223] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0224] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A unit commitment scheduling method based on constrained spatiotemporal graph neural network, characterized in that: include: Collect bus load data and power system load data of the power grid at various times of the day to establish a training data set; Construct the unit combination problem on a daily basis and solve it under the constraints to obtain the corresponding optimal unit power; Use a sliding window to divide the bus load and the corresponding optimal unit power according to time periods to expand the training data set; Based on the expanded training data set, the divided bus load is used as input data, and the corresponding optimal unit power is used as a data label to train a pre-established spatiotemporal graph neural network model; the spatiotemporal graph neural network model adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints; The bus load data that needs to be predicted is divided into time periods and input into the trained space-time graph neural network model. The predicted power output by the space-time graph neural network model is restricted, including filtering the prediction results that are less than the minimum power and adjusting the prediction results that do not meet the start-stop requirements, to ensure the availability of the predicted power and obtain the unit combination scheduling plan.

2. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 1 is characterized in that: The steps of constructing the unit commitment problem on a daily basis include: Assume there are N nodes, N B busbar nodes, N G The total number of decision-making periods per day in the power system is N. T ; The objective function of the unit commitment problem is constructed as follows: In the formula, g and t represent the sequence number of the unit and time period respectively; N G and N T Represent the total number of units and time periods respectively; p g,t is the output of unit g in time period t; a g , b g , c g are the coefficients of the quadratic term, linear term and constant term of the operating cost function of unit g; SU g is the startup cost of unit g; y g,t , z g,t It is an integer variable indicating whether the unit g has started or stopped in the time period t. The value of 1 indicates that the unit g has started or stopped in the time period t, and the value of 0 indicates that the unit g has not started or stopped.

3. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 2 is characterized in that: In the step of solving the corresponding optimal unit power under the constraint conditions, the constraint conditions include: Power balance constraints: Where b is the node number; N B is the total number of busbar nodes; D b,t is the load of node b in time period t; Unit output upper and lower limit constraints: In the formula, and are the minimum and maximum output of unit g respectively; u g,t is an integer variable indicating the on / off status of unit g in time period t. A value of 1 indicates that unit g is turned on in time period t, and a value of 0 indicates that unit g is turned off in time period t. Unit climbing constraints: -RD g ≤p g,t -p g,t-1 ≤RU g ,g=1,…,N G ,T=2,…,N T Where, RD g and RU g They are the climbing rate limits for the decrease and increase of the unit g output respectively; Minimum start and stop time constraints: y g,t +z g,t ≤1,g=1,…,N G ,t=1,…,N T In the formula, TU g and TD g are the minimum start and stop time of unit g respectively; Safety constraints on active power flow in base-state branches: Where l is the branch number; N L is the total number of branches; is the active power flow of branch l in time period t under the base state; T is the power transfer distribution factor; g∈G(b) is the set of units on node b; F l max is the long-term allowable current-carrying power of branch l.

4. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 1 is characterized in that: The steps of training the pre-established spatiotemporal graph neural network model based on the expanded training data set, taking the divided bus load as input data, and taking the corresponding optimal unit power as a data label include: Based on the expanded training data set, the graph neural connection relationship is obtained from the real power grid topology structure; The load D of all nodes in the power grid within a period of time F is input into the spatiotemporal graph convolutional network without a linear constraint layer, and the input spatial features are extracted through the graph convolution layer of the spatiotemporal graph convolutional network; the graph convolution layer extracts and aggregates the load features of each node, learns the spatial features between the data, and expands the time series length of the data; The graph data processed by the graph convolution layer is converted into time series data, and then the causal expansion time convolution is used to extract the time series features of each node. Subsequently, the fully connected neural network layer is used to aggregate the time series features between different nodes, and the format of the data is converted to output the predicted power P of the thermal power unit within a period of time F; the neural network updates the loss function between the predicted power and the actual unit power scheduling data, repeatedly executes the above steps, and completes the training of the spatiotemporal graph neural network model.

5. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 1 is characterized in that: The step of projecting the predicted power of the neural network into a feasible domain that meets the specified linear constraints includes: make represents the input value of the constrained activation layer, that is, the predicted power value of the spatiotemporal graph neural network model; represents the output of the constrained activation layer, that is, the power value that meets the linear constraint; where N G Represents the number of units, F represents the time series length; Convex feasible sets are constructed for all linear constraints in the unit combination problem. The convex feasible set Ω is the intersection of a set of constraints in the form of m inequality constraints, and the mathematical expression is as follows: Ω={x|h i (x)≤0,i=1,...,m} Ensure that each constraint h i (x)≤0 is convex, that is, it conforms to the following mathematical expression: Define a fixed point p in the convex feasible set Ω, that is, p∈Ω, then any point x in the set Ω is expressed as x=p+α·r, and the proportional factor α≥0 is calculated; is a vector from point p; for any p and r, the scaling factor α has an upper limit The mathematical expression is as follows: At this time, the line segment Belongs to the set Ω, upper limit The meaning of is the intersection of ray r and a constraint; For a point z outside the constrained feasible domain that appears in the spatiotemporal graph neural network, by specifying the ray r(z) = zp, the scaling factor The calculation is performed through the following mathematical expression, that is, projecting the point z outside the feasible domain along the ray from z to p to the point g inside the feasible domain:

6. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 1 is characterized in that: The step of limiting the predicted power output by the spatiotemporal graph neural network model, including filtering the predicted results less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements, and ensuring the availability of the predicted power includes: Set the threshold δ, first check and adjust the output power of the spatiotemporal graph neural network model for each unit, set the prediction result whose difference between the minimum power and the prediction result is less than the threshold v as the minimum power of the corresponding unit, and set the prediction result whose difference between the minimum power and the prediction result is greater than the threshold δ as 0; Check and adjust the minimum start and stop time of each unit in the forecast results; The prediction results are re-passed through the linear constraint activation layer with branch active power flow safety constraints added to ensure that the output power meets the branch active power flow safety constraints.

7. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 6 is characterized in that: The step of checking and adjusting the minimum start and stop time of each unit in the prediction result comprises: Record the duration of each start and stop of each unit; Check the duration of each start and stop of each unit. For the start and stop that violate the minimum start and stop time constraints, compare the start or stop time with the stop or start time before and after it in chronological order; When the start and stop time of a constraint violation is shorter than the start and stop time on both sides, adjust the corresponding start and stop.

8. A unit commitment dispatching system based on constrained spatiotemporal graph neural network, characterized in that: include: The data collection module is used to collect the bus load data and power system load data of the power grid at various times of the day and establish a training data set; The optimal unit power solving module is used to construct the unit combination problem on a daily basis and solve the corresponding optimal unit power under the constraints; A training data set expansion module is used to use a sliding window to divide the bus load and the corresponding optimal unit power according to time periods to expand the training data set; The spatiotemporal graph neural network model training module is used to train the pre-established spatiotemporal graph neural network model based on the expanded training data set, using the divided bus load as input data and the corresponding optimal unit power as a data label; the spatiotemporal graph neural network model adds a linear constraint layer to the end of the spatiotemporal graph convolution, uses the linear constraint layer to linearly constrain the output result of the fully connected layer, and projects the predicted power of the neural network into a feasible domain that meets the specified linear constraints; The predicted power post-processing output module is used to divide the bus load data that needs to be predicted according to time periods, input the trained space-time graph neural network model, and limit the predicted power output by the space-time graph neural network model, including filtering the prediction results that are less than the minimum power and adjusting the prediction results that do not meet the start-stop requirements, to ensure the availability of the predicted power and obtain the unit combination scheduling plan.

9. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 8 is characterized in that: The steps of constructing the unit combination problem in units of daily units in the optimal unit power solving module include: Assume there are N nodes, N B busbar nodes, N G The total number of decision-making periods per day in the power system is N. T ; The objective function of the unit commitment problem is constructed as follows: In the formula, g and t represent the sequence number of the unit and time period respectively; N G and N T Represent the total number of units and time periods respectively; p g,t is the output of unit g in time period t; a g , b g , c g are the coefficients of the quadratic term, linear term and constant term of the operating cost function of unit g; SU g is the startup cost of unit g; y g,t , z g,t It is an integer variable indicating whether the unit g has started or stopped in the time period t. The value of 1 indicates that the unit g has started or stopped in the time period t, and the value of 0 indicates that the unit g has not started or stopped.

10. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 9 is characterized in that: When the optimal unit power solving module solves the corresponding optimal unit power under the constraint conditions, the constraint conditions include: Power balance constraints: Where b is the node number; N B is the total number of busbar nodes; D b,t is the load of node b in time period t; Unit output upper and lower limit constraints: In the formula, and are the minimum and maximum output of unit g respectively; u g,t is an integer variable indicating the on / off status of unit g in time period t. A value of 1 indicates that unit g is turned on in time period t, and a value of 0 indicates that unit g is turned off in time period t. Unit climbing constraints: -RD g ≤p g,t -p g,t-1 ≤RU g ,g=1,…,N G ,T=2,…,N T Where, RD g and RU g They are the climbing rate limits for the decrease and increase of the unit g output respectively; Minimum start and stop time constraints: y g,t +z g,t ≤1,g=1,…,N G ,t=1,…,N T In the formula, TU g and TD g are the minimum start and stop time of unit g respectively; Safety constraints on active power flow in base-state branches: Where l is the branch number; N L is the total number of branches; is the active power flow of branch l in time period t under the base state; T is the power transfer distribution factor; g∈G(b) is the set of units on node b; F l max is the long-term allowable current-carrying power of branch l.

11. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 8, characterized in that: The spatiotemporal graph neural network model training module uses the divided bus load as input data and the corresponding optimal unit power as a data label. The steps of training the pre-established spatiotemporal graph neural network model include: Based on the expanded training data set, the graph neural connection relationship is obtained from the real power grid topology structure; The load D of all nodes in the power grid within a period of time F is input into the spatiotemporal graph convolutional network without a linear constraint layer, and the input spatial features are extracted through the graph convolution layer of the spatiotemporal graph convolutional network; the graph convolution layer extracts and aggregates the load features of each node, learns the spatial features between the data, and expands the time series length of the data; The graph data processed by the graph convolution layer is converted into time series data, and then the causal expansion time convolution is used to extract the time series features of each node. Subsequently, the fully connected neural network layer is used to aggregate the time series features between different nodes, and the format of the data is converted to output the predicted power P of the thermal power unit within a period of time F; the neural network updates the loss function between the predicted power and the actual unit power scheduling data, repeatedly executes the above steps, and completes the training of the spatiotemporal graph neural network model.

12. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 8, characterized in that: The step of the spatiotemporal graph neural network model training module projecting the predicted power of the neural network into a feasible domain that meets the specified linear constraints includes: make represents the input value of the constrained activation layer, that is, the predicted power value of the spatiotemporal graph neural network model; represents the output of the constrained activation layer, that is, the power value that meets the linear constraint; where N G Represents the number of units, F represents the time series length; Convex feasible sets are constructed for all linear constraints in the unit combination problem. The convex feasible set Ω is the intersection of a set of constraints in the form of m inequality constraints, and the mathematical expression is as follows: Ensure that each constraint h i (x)≤0 is convex, that is, it conforms to the following mathematical expression: Define a fixed point p in the convex feasible set Ω, that is, p∈Ω, then any point x in the set Ω is expressed as x=p+α·r, and the proportional factor α≥0 is calculated; is a vector from point p; for any p and r, the scaling factor α has an upper limit The mathematical expression is as follows: At this time, the line segment Belongs to the set Ω, upper limit The meaning of is the intersection of ray r and a constraint; For a point z outside the constrained feasible domain that appears in the spatiotemporal graph neural network, by specifying the ray r(z) = zp, the scaling factor The calculation is performed through the following mathematical expression, that is, projecting the point z outside the feasible domain along the ray from z to p to the point g inside the feasible domain:

13. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 8, characterized in that: The predicted power post-processing output module limits the predicted power output by the spatiotemporal graph neural network model, including filtering the predicted results that are less than the minimum power and adjusting the predicted results that do not meet the start-stop requirements. When ensuring the availability of the predicted power, a threshold δ is set, and the output power of the spatiotemporal graph neural network model is first checked and adjusted for each unit. The predicted result whose difference between the minimum power and the predicted result is less than the threshold δ is set as the minimum power of the corresponding unit, and the predicted result whose difference between the minimum power and the predicted result is greater than the threshold δ is set to 0; Check and adjust the minimum start and stop time of each unit in the forecast results; The prediction results are re-passed through the linear constraint activation layer with branch active power flow safety constraints added to ensure that the output power meets the branch active power flow safety constraints.

14. The unit commitment scheduling method based on constrained spatiotemporal graph neural network according to claim 13, characterized in that: The step of the predicted power post-processing output module checking and adjusting the minimum start and stop time of each unit in the prediction result includes: Record the duration of each start and stop of each unit; Check the duration of each start and stop of each unit. For the start and stop that violate the minimum start and stop time constraints, compare the start or stop time with the stop or start time before and after it in chronological order; When the start and stop time of a constraint violation is shorter than the start and stop time on both sides, adjust the corresponding start and stop.

15. An electronic device, characterized in that: It includes a processor and a memory, wherein the processor is used to execute a computer program stored in the memory to implement a unit combination scheduling method based on a constrained spatiotemporal graph neural network as described in any one of claims 1 to 7.

16. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and when the at least one instruction is executed by the processor, it implements the unit combination scheduling method based on the constrained spatiotemporal graph neural network as described in any one of claims 1 to 7.

Images