A method and system for selecting a typical day for power system planning based on operation simulation

Abstract

The present invention relates to the technical field of power systems. Specifically, it relates to a method and system for selecting typical days for power system planning based on operation simulation, which mainly includes constructing the residual load curve of the operating days, obtaining the classification labels and solution difficulties of each operating day based on the Mamba model, selecting the operating day with the greatest solution difficulty among various types of operating days as the typical day, solving the unit commitment start-stop results of various typical days, and solving the unit output of the operating days based on the start-stop results of the typical day units. The present invention classifies the operating days based on the historical unit start-stop results, avoiding aggregating operating days with similar load characteristics but different unit start-stops or large load differences but the same unit start-stops, preventing the drawbacks of large clustering errors or rough clustering results. By autonomously learning the historical operation simulation data through the Mamba model, the typical day label vector is directly output. Avoiding the artificial selection of important information such as load curve characteristics and the number of typical days, the results are more objective.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and more particularly, to a method and system for selecting typical days for power system planning based on operation simulation. Background Art

[0002] Power planning models are basic tools for power system operation analysis. On the premise of meeting the constraints of system security and stability, by simulating different operation scenarios, predicting possible system faults and overload situations, and reasonably allocating power resources, the safe and stable operation of the power system can be ensured while achieving economic maximization.

[0003] However, power planning models often simulate with a time span of months or even years, containing a large number of integer variables and complex constraints, resulting in extremely long calculation times for direct solution. With the advancement of the global energy transition, the proportion of new energy in the power system has increased significantly. New energy has strong volatility and uncertainty, making the constraint conditions of power planning models more complex and further increasing the calculation difficulty. To improve the calculation efficiency, power planning models often select typical days that can represent the annual operation scenarios for solution. How to select an appropriate number of typical days containing effective information is crucial for the efficiency of operation simulation.

[0004] Existing domestic and foreign research mainly selects typical days through three methods: clustering method, sampling method, and heuristic scenario reduction method. The clustering method classifies the daily load curves through an algorithm and obtains the load curve of the class center as the typical daily load curve. The sampling method captures the load curves with higher occurrence probabilities in different scenarios as the typical daily load curves by randomly selecting sample points. The heuristic scenario reduction method selects representative load curves as the typical daily load curves through experience and simple approximate calculations.

[0005] The clustering method ignores the different contributions of different points on the daily load curve to the characteristics of the load curve, and may not be able to accurately aggregate load curves with similar key characteristics. At the same time, the number of clusters depends on manual selection, which has strong subjectivity and may lead to large errors.

[0006] When the sample size is small or the sampling process is not fine enough in the sampling method, the selected typical daily load curve may not have sufficient representativeness. At the same time, the sampling method may ignore some extreme load situations, which may have a greater impact on the operation and planning of the power system.

[0007] The heuristic scenario reduction method mainly relies on experience and simple approximate calculations, and may not accurately evaluate high-proportion new energy load curves with strong volatility, lacking a certain degree of generalization. At the same time, when the system scale becomes larger or the situation is complex, the stability of the heuristic algorithm may be poor, resulting in large errors. Summary of the Invention

[0008] The object of the present invention is to provide a method and system for selecting typical days in power system planning based on operation simulation to solve the above problems of the prior art.

[0009] The present invention is realized through the following technical solutions:

[0010] In the first aspect, a method for selecting typical days in power system planning based on operation simulation includes:

[0011] Obtain historical operation day data and construct a residual load curve for historical operation days;

[0012] Solve the unit commitment according to the residual load curves of historical operation days to obtain the unit on / off results for each historical operation day, and classify and label the unit on / off results of historical operation days to obtain the classification labels for each historical operation day;

[0013] Output the solution difficulty for the classification labels of each historical operation day based on the Mamba model and the classification labels of each historical operation day, and obtain the operation day with the greatest solution difficulty among various types of operation days as the typical day;

[0014] Solve the unit commitment on / off results for various typical days, establish a unit commitment model under the condition of full absorption of new energy output, solve the output results for each time period of each operation day based on the unit commitment on / off results for various typical days and the unit commitment model, and output the output results.

[0015] Preferably, the output of the solution difficulty for the classification labels of each historical operation day based on the Mamba model and the classification labels of each historical operation day includes:

[0016] Construct a time series feature matrix for all historical operation days, where the first dimension of the time series feature matrix is the load curve, the second dimension is the wind power output curve, and the third dimension is the photovoltaic power output curve;

[0017] Construct a label vector for all historical operation days, where the first dimension of the label vector is the classification label and the second dimension is the solution time;

[0018] Use the time series feature matrix of historical operation days as the input and the label vector of historical operation days as the output to train the Mamba model, obtain the solution time in the label vector of each historical operation day through the trained Mamba model, and take the one with the longest solution time as the operation day with the greatest solution difficulty.

[0019] Preferably, the establishment of the unit commitment model includes:

[0020] Establish the objective function of the unit commitment model and set the constraint conditions of the objective function;

[0021] The constraint conditions include load balance constraint, system positive reserve constraint, system negative reserve constraint, upper and lower limits of unit output constraint, unit ramp rate constraint, and minimum continuous on - off time constraint of the unit.

[0022] Preferably, the objective function for establishing the unit commitment model includes:

[0023] A functional equation is constructed with the goal of minimizing the total operating cost of the system, and the functional equation includes:

[0024]

[0025] In the formula, is the thermal power output cost of unit i at time t in a typical day, N is the total number of units, is the start - up cost of unit i at time t in a typical day, and T is the number of 24 time periods in a day. is the start - up cost of unit i at time t in a typical day, and T is the number of 24 time periods in a day. is the start - up cost of unit i at time t in a typical day, and T is the number of 24 time periods in a day.

[0026] Preferably, the load balance constraint includes:

[0027]

[0028] In the formula: is the output of thermal power unit in a typical day at time , is the total system load in a typical day , is the new - energy output in a typical day .

[0029] Preferably, the system positive reserve constraint includes:

[0030]

[0031] The system negative reserve constraint includes:

[0032]

[0033] In the formula, is the start - stop state of thermal power unit in a typical day at time , that is, the unit commitment start - stop result of the typical day, is the maximum output of thermal power unit in a typical day at time , is the maximum output of thermal power unit in a typical day The minimum output during the time period is the positive reserve capacity of the system during the time period of a typical day during the time period ; and the negative reserve capacity of the system during the time period of a typical day during the time period is the negative reserve capacity of the system during the time period

[0034] Preferably, the upper and lower limits of the unit output constraint include:

[0035] .

[0036] Preferably, the unit ramp rate constraint includes:

[0037]

[0038] ;

[0039]

[0040] ;

[0041] In the formula, is the maximum upward ramp rate of the thermal power unit ; is the maximum downward ramp rate of the thermal power unit ; is the maximum output of the thermal power unit in a typical day during the time period ; is the start / stop state of the thermal power unit in a typical day during the time period .

[0042] Preferably, the minimum continuous start / stop time constraint of the unit includes:

[0043]

[0044] In the formula: is the continuous running time of the thermal power unit in a typical day during the time period ; is the continuous shutdown time of the thermal power unit in a typical day during the time period ; is the minimum continuous running time of the unit ; and

[0045] In a second aspect, the present invention also provides a system for selecting typical days for power system planning based on operation simulation, including:

[0046] A data processing module, configured to obtain historical operation day data and construct a residual load curve for historical operation days; solve unit commitment based on the residual load curves of historical operation days to obtain the unit start-stop results for each operation day, and classify and label the historical unit start-stop results to obtain the classification labels for each operation day;

[0047] A model operation model, configured to output the classification labels and solution difficulties for each operation day based on the Mamba model and the classification labels of each operation day, and obtain the operation day with the greatest solution difficulty among various types of operation days as the typical day, solve the unit commitment start-stop results for various typical days, establish a unit commitment model on the condition of full consumption of new energy output, solve the output results for each time period of each operation day based on the unit commitment start-stop results of various typical days and the unit commitment model, and output the output results;

[0048] A main control module, connected to the data processing module and the model operation model, for executing the method for selecting typical days for power system planning based on operation simulation as described above.

[0049] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0050] 1. The present invention classifies operation days based on historical unit start-stop results, avoiding aggregating operation days with similar load characteristics but different unit start-stops or large load differences but the same unit start-stops, and preventing the disadvantages of large clustering errors or rough clustering results.

[0051] 2. The present invention autonomously learns historical operation simulation data through the Mamba model and directly outputs the typical day label vector. It avoids artificial selection of important information such as load curve characteristics and the number of typical days, and the result is more objective and has stronger accuracy.

[0052] 3. The present invention reflects the difficult-to-quantify unit commitment solution difficulty through the easily obtained unit commitment solution time, greatly reducing the model training difficulty under the condition of ensuring data validity and enhancing the practicability. Selecting the unit start-stop of the operation day with the greatest solution difficulty among typical days as the typical day result ensures that the typical day can take into account the most extreme load scenarios, thereby enhancing the robustness of the system.

[0053] 4. The present invention significantly reduces the number of model variables and calculation time by selecting typical days to calculate unit start-stop results. Then, it uses the unit start-stop results to solve the unit output for each time period of each operation day, reduces the unit complexity, and obtains more accurate model simulation results. Description of the Drawings

[0054] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0055] Figure 1 is a schematic flowchart of the present invention;

[0056] Figure 2 is a schematic structural diagram of the system of the present invention. Detailed implementation manners

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Usually, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0058] The division of modules in this application is a logical division. In actual implementation, there may be other division methods. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0059] The independently described modules or sub-modules can be physically separated or not: they can be implemented by software or hardware, and some of the modules or sub-modules can be implemented by software, and the functions of these modules or sub-modules are called by the processor through the software, and other parts of the modules or sub-modules are implemented by hardware, for example, through a hardware circuit. In addition, some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application.

[0060] Please refer to Figure 1 - Figure 2 , a method for selecting a typical day for power system planning based on operation simulation provided by the present invention, includes:

[0061] S101: Obtain historical operation day data and construct a remaining load curve for the historical operation days;

[0062] In this embodiment, each operation day has a load curve , a wind power output curve , and a photovoltaic power output curve . Assuming this, define the remaining load curve for each operation day .

[0063] S102: Solve the unit commitment according to the remaining load curve of the historical operating days, obtain the unit start-stop results of each historical operating day, classify and label the historical unit start-stop results, and obtain the classification labels of each historical operating day;

[0064] Classify the historical unit start-stop results. Set the operating day labels to be the same when the start-stop results of all units in each time period are the same. Set the label for the first type of unit start-stop result to 1, the second type to 2, the third type to 3, and so on, until all the unit start-stop results that have occurred in the historical operating days are marked.

[0065] S103: Output the solution difficulty regarding the classification labels of each operating day based on the Mamba model and the classification labels of each operating day, and obtain the operating day with the greatest solution difficulty among various types of operating days as the typical day;

[0066] Among them, since it is difficult to accurately quantify the solution difficulty of the unit commitment for each operating day, and the solution time can better reflect the solution difficulty. This patent uses the solution time of the historical operating days to train the ability of the Mamba model to predict the solution difficulty of the operating day.

[0067] Secondly, since operation simulation needs to ensure that the system can operate stably under the most challenging scenarios, this patent selects the operating day with the greatest solution difficulty in each type of operating day as the typical day of this type of operating day, and uses the remaining load curve of this typical day for production simulation, so as to ensure that the load conditions of the most extreme operating days can be taken into account, and the robustness of the planning scheme is enhanced.

[0068] S104: Solve the unit start-stop results of various typical day unit commitments, establish a unit commitment model on the condition of full consumption of new energy output, solve and obtain the output results of each time period of each operating day based on the unit start-stop results of various typical day unit commitments and the unit commitment model, and output the output results.

[0069] Among them, the unit commitment (UC) problem is usually a mixed integer programming problem (MIP), and solvers such as CPLEX and Gurobi can be used for solving. The goal of UC is to determine the start-stop states of units in each time period.

[0070] Economic dispatch (ED) further optimizes the output of each unit on the basis of the unit start-stop states determined by UC to minimize the generation cost. ED is usually a linear programming problem (LP) and can be solved using the same solver.

[0071] Common tools include MATLAB combined with Yalmip and CPLEX / Gurobi. For example, decision variables (such as unit status u and output p) are defined in MATLAB, the objective function and constraint conditions are set, and then the solver is called to obtain the optimal solution. After the solution is completed, the unit output results for each time period of each operating day (i.e., the output plan of the unit for each time period) are obtained. These results can be used to guide the actual operation to ensure that the system achieves the optimal economy while meeting the load demand.

[0072] In view of the requirement that the power planning model needs to select appropriate typical days to improve the efficiency of operation simulation calculation, the present invention establishes a method for selecting typical days for power system planning based on historical operation simulation data. The present invention regards operating days with the same unit start-stop in each time period as the same type of operating day, which is represented by a classification label. The operating day with the greatest solution difficulty in this type of operating day is selected as the typical day for unit commitment solution to ensure the robustness of the typical day. The unit start-stop results of the typical day are used as the unit start-stop results of all operating days in this type. Based on the start-stop results, the security-constrained economic dispatch model is solved using the residual load curves of each operating day to obtain the unit output conditions for each time period of each operating day, which are used as the final results of operation simulation.

[0073] The classification label and solution difficulty of the operating day are output by the Mamba model. The Mamba model can predict the classification label and solution difficulty of each operating day based on the time series feature matrix of each operating day by autonomously learning the operation simulation data of historical operating days.

[0074] An exemplary implementation of the present invention, the solution difficulty regarding the classification label of each operating day based on the Mamba model output includes:

[0075] Construct the time series feature matrix of all historical operating days. The first dimension of the time series feature matrix is the load curve, the second dimension is the wind power output curve, and the third dimension is the photovoltaic power output curve; construct the label vector of all historical operating days. The first dimension of the label vector is the classification label, and the second dimension is the solution time; use the time series feature matrix of historical operating days as the input and the label vector of historical operating days as the output to train the Mamba model. The solution time in the label vector of each historical operating day is obtained through the trained Mamba model, and the operating day with the longest solution time is taken as the operating day with the greatest solution difficulty.

[0076] An exemplary implementation of the present invention, establishing a unit commitment model includes:

[0077] Establish the objective function of the unit commitment model and set the constraint conditions of the objective function;

[0078] The constraint conditions include load balance constraint, system positive reserve constraint, system negative reserve constraint, upper and lower limits of unit output constraint, unit ramp rate constraint, and minimum continuous on-off time constraint of units.

[0079] Specifically, the objective function for establishing the unit commitment model includes:

[0080] A functional equation is constructed with the goal of minimizing the total system operation cost, and the functional equation includes:

[0081]

[0082] In the formula, is the thermal power generation cost of unit i at time t in the typical day, N is the total number of units, is the start-up cost of unit i at time t in the typical day, and T is the number of 24 time periods in a day.

[0083] The load balance constraint includes:

[0084]

[0085] In the formula: is the output of thermal power unit at time in the typical day, is the total system load in the typical day, is the new energy output in the typical day.

[0086] The system positive reserve constraint includes:

[0087]

[0088] The system negative reserve constraint includes:

[0089]

[0090] In the formula, is the start-stop state of thermal power unit at time in the typical day, that is, the unit commitment start-stop result of the typical day, is the maximum output of thermal power unit at time in the typical day, is the minimum output of thermal power unit at time ​​​​​​​​​For a typical day The medium period of the system's positive reserve capacity, For a typical day The medium period of the system's negative reserve capacity.

[0091] The upper and lower limits of unit output constraints include:

[0092] .

[0093] The unit ramp rate constraints include:

[0094]

[0095] ;

[0096]

[0097] ;

[0098] In the formula, is the maximum upward ramp rate of the thermal power unit ; is the maximum downward ramp rate of the thermal power unit , is for a typical day in which the thermal power unit at the time period the maximum output, is for a typical day in which the thermal power unit at the time period the start-stop state.

[0099] The unit minimum continuous on-off time constraints include:

[0100]

[0101] In the formula: is for a typical day in which the thermal power unit at the time period the continuous running time; is for a typical day in which the thermal power unit at the time period the continuous shutdown time, is the minimum continuous running time of the unit, is the minimum continuous shutdown time of the unit.

[0102] In the second aspect, the present invention also provides a system for selecting a typical day for power system planning based on operation simulation, including:

[0103] A data processing module, configured to obtain historical daily operation data and construct a historical daily residual load curve; solve unit commitment according to the historical daily residual load curve to obtain the unit start-stop results for each operation day, and classify and label the historical unit start-stop results to obtain the classification labels for each operation day;

[0104] A model operation model, configured to output the classification labels and solution difficulties for each operation day based on the Mamba model and the classification labels for each operation day, and obtain the operation day with the greatest solution difficulty among various operation days as a typical day, solve the unit commitment start-stop results for various typical days, establish a unit commitment model on the condition of full consumption of new energy output, solve the output results for each time period of each operation day based on the unit commitment start-stop results for various typical days and the unit commitment model, and output the output results;

[0105] A main control module, connected to the data processing module and the model operation model, for executing a method for selecting a typical day in a power system planning based on operation simulation as described above.

[0106] In addition, in each embodiment of the present invention, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0107] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present invention. The aforementioned storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program codes.

[0108] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for selecting a typical day for power system planning based on operation simulation, characterized in that Including: Obtain historical operation daily data and construct a historical operation daily residual load curve; Solve the unit commitment according to the residual load curve of the historical operation day, obtain the unit start-stop results of each historical operation day, classify and label the unit start-stop results of the historical operation day, and obtain the classification labels of each historical operation day; Based on the Mamba model and the classification labels of each historical operation day, output the solution difficulty of the classification labels of each historical operation day, and obtain the operation day with the greatest solution difficulty among various operation days as the typical day; Solve the unit commitment start-stop results of various typical days, establish a unit commitment model on the condition of full consumption of new energy output, solve the output results of each time period of each operation day based on the unit commitment start-stop results of various typical days and the unit commitment model, and output the output results; The establishment of the unit commitment model includes: Establish the objective function of the unit commitment model and set the constraint conditions of the objective function; The constraint conditions include load balance constraint, system positive reserve constraint, system negative reserve constraint, unit output upper and lower limit constraint, unit ramp rate constraint, and unit minimum continuous start-stop time constraint; The system positive reserve constraint includes: The system negative reserve constraint includes: Wherein, is the typical day in which the thermal power unit is in the start-stop state during the time period , that is, the start-stop result of the unit combination of the typical day is the typical day in which the thermal power unit has the maximum output during the time period , is the typical day in which the thermal power unit has the minimum output during the time period , is the typical day in which the system positive reserve capacity during the time period is is the typical day in which the system negative reserve capacity during the time period is 2. The method for selecting a typical day for power system planning based on operation simulation according to claim 1, characterized in that The output of the solution difficulty of each operation day classification label based on the Mamba model and the classification labels of each historical operation day includes: Construct the time series feature matrix of all historical operation days, where the first dimension of the time series feature matrix is the load curve, the second dimension is the wind power output curve, and the third dimension is the photovoltaic power output curve; Construct the label vector of all historical operation days, where the first dimension of the label vector is the classification label and the second dimension is the solution time; Use the time series feature matrix of the historical operation day as the input and the label vector of the historical operation day as the output to train the Mamba model. Obtain the solution time in the label vector of each historical operation day through the trained Mamba model, and use the historical operation day with the longest solution time as the operation day with the greatest solution difficulty.

3. A method for selecting a typical day for power system planning based on operation simulation according to claim 2, characterized in that, The objective function of the establishment of the unit commitment model includes: Construct a functional formula with the goal of minimizing the total system operation cost, and the functional formula includes: In the formula, is the typical day The thermal power output cost of unit i in period t in, N is the total number of units, is the typical day The start-up cost of unit i in period t in, T is the number of 24 periods in a day.

4. A method for selecting a typical day for power system planning based on operation simulation according to claim 3, characterized in that, The load balance constraint includes: Where: is the typical day the thermal power unit at the time period output is the typical day the total system load is the typical day the new energy output 5. A method for selecting a typical day for power system planning based on operation simulation according to claim 4, characterized in that, The unit output upper and lower limit constraint includes: 。 6. A method for selecting a typical day for power system planning based on operation simulation according to claim 5, characterized in that The unit ramp rate constraint includes: ； ； Wherein, is the maximum upward ramp rate of the thermal power unit ; is the maximum downward ramp rate of the thermal power unit , is the typical day in the thermal power unit in the time period of the maximum output, is the typical day in the thermal power unit in the time period of the start-stop state.

7. A method for selecting a typical day for power system planning based on operation simulation according to claim 6, characterized in that The unit minimum continuous start-stop time constraint includes: Wherein: is the typical day the thermal power unit in the time period the continuously running time is the typical day the thermal power unit in the time period the continuously shutdown time, is the minimum continuous running time of the unit, is the minimum continuous shutdown time of the unit.

8. A typical day selection system for power system planning based on operation simulation, characterized in that Including: A data processing module configured to obtain historical operation daily data and construct a historical operation daily residual load curve; Solve the unit commitment according to the residual load curve of the historical operation day, obtain the unit start-stop results of each historical operation day, classify and label the unit start-stop results of the historical operation day, and obtain the classification labels of each historical operation day; The model operation model is configured to output the solution difficulty regarding the classification labels of each operation day based on the Mamba model and the classification labels of each operation day, and obtain the operation day with the greatest solution difficulty among various types of operation days as the typical day, solve the unit commitment start-stop results of each typical day, establish a unit commitment model on the condition of full consumption of new energy output, solve the output results of each operation day and each time period based on the unit commitment start-stop results of each typical day and the unit commitment model, and output the output results; The main control module, connected to the data processing module and the model operation model, is used to execute a method for selecting typical days in power system planning based on operation simulation according to any one of claims 1-7.

Images