Method, device and equipment for determining unit commitment of power system

Abstract

The present application provides a method, device, and equipment for determining unit commitment in a power system. The method of the present application determines a historical integer solution to the current unit commitment problem by using a small amount of easily obtainable historical unit commitment information. Based on the historical integer solution, a feasibility pump method is used to determine an initial integer feasible solution to the current unit commitment model. This initial integer feasible solution is used as an initial solution to hot-start an integer programming solver, and the integer programming solver is used to determine the optimal solution of the unit commitment. According to the optimal solution of the unit commitment, the unit start-stop and output plan of the power system during the target period is determined, which can improve the solution efficiency of the unit commitment problem in the power system, improve the quality of the obtained unit commitment information, and enhance the performance of the power system.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular, to a method, device, and equipment for determining unit commitment in a power system. Background Art

[0002] Unit commitment in a power system is a core issue in power system dispatching and operation. It reduces the generation cost by optimizing the start-stop plans of each unit within the dispatching period, while meeting the system load demand and other constraint conditions. Unit commitment configuration / optimization in a power system is a core link in power system dispatching and operation. Based on the predicted system load values for future time periods, it obtains the unit on-off patterns and generation output curves through modeling and optimization solving. Unit commitment arranges the production plan of the power system in advance, provides a scientific decision-making basis for the start-stop, adjustment, maintenance, etc. plans of power equipment, and plays an extremely important role in the safe and economic operation of the power system.

[0003] The unit commitment problem in a power system is a large-scale mixed integer programming problem. As the scale of the unit commitment model in the power system continues to increase, the difficulty of solving unit commitment also increases. The optimization efficiency of unit commitment in a large-scale power system is low, unable to meet the requirements of efficiently optimizing unit commitment in a large-scale power system, and affecting the performance of the power system. Summary of the Invention

[0004] The present application provides a method, device, and equipment for determining unit commitment in a power system to solve the problem that the optimization efficiency of unit commitment in a power system is low and affects the performance of the power system.

[0005] On the one hand, the present application provides a method for determining unit commitment in a power system, including:

[0006] Obtain the power load prediction information and system data of the power system in a future target time period, and the unit commitment information configured by the power system in a historical time period;

[0007] Construct a unit commitment model of the power system according to the power load prediction information and system data in the target time period, and the integer feasible solution of the unit commitment model represents the available unit commitment in the power system in the target time period;

[0008] Determine the initial integer feasible solution of the unit commitment model according to the unit commitment information configured in the historical time period;

[0009] Use the initial integer feasible solution as the initial solution of an integer programming solver, hot start the integer programming solver, and use the integer programming solver to determine the optimized unit commitment feasible solution;

[0010] Based on the optimized unit commitment feasible solution, determine the unit commitment information of the power system during the target period, where the unit commitment information includes the on / off status information and output plan of the units.

[0011] On the other hand, the present application provides a device for determining the unit commitment of a power system, including:

[0012] A power system related data acquisition module, configured to acquire the power load prediction information and system data of the power system during a future target period, as well as the unit commitment information configured by the power system during a historical period;

[0013] A unit commitment model construction module, configured to construct a unit commitment model of the power system according to the power load prediction information and system data during the target period, where the integer feasible solution of the unit commitment model represents the available unit commitment of the power system during the target period;

[0014] A feasible solution determination module, configured to determine an initial integer feasible solution of the unit commitment model according to the unit commitment information configured during the historical period;

[0015] A unit commitment feasible solution optimization module, configured to use the initial integer feasible solution as the initial solution of an integer programming solver, hot start the integer programming solver, and use the integer programming solver to determine an optimized unit commitment feasible solution;

[0016] A unit commitment information determination module, configured to determine the unit commitment information of the power system during the target period according to the optimized unit commitment feasible solution, where the unit commitment information includes the on / off status information and output plan of the units.

[0017] On the other hand, the present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0018] The memory stores computer execution instructions;

[0019] The processor executes the computer execution instructions stored in the memory to implement the method described in the first aspect.

[0020] On the other hand, the present application provides a computer-readable storage medium, where computer execution instructions are stored in the computer-readable storage medium, and when the computer execution instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0021] The method, device, and equipment for determining the unit commitment of a power system provided in this application determine the historical integer feasible solutions of the unit commitment by using a small amount of easily obtainable historical unit commitment information. Based on the historical integer feasible solutions, an initial integer feasible solution of the current unit commitment model is determined. This initial integer feasible solution is used as the initial solution to warm-start the integer programming solver, and the integer programming solver is used to determine the optimized unit commitment feasible solution, which can ensure the feasibility of the optimized unit commitment feasible solution during the target period. According to the optimized unit commitment feasible solution, the unit commitment information of the power system during the target period is determined, which can improve the efficiency of determining the unit commitment information of the power system during the target period and improve the quality of the obtained unit commitment information, thereby enhancing the performance of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings herein are incorporated into the specification and form a part of the specification, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0023] Figure 1 It is a schematic diagram of the system framework on which this application is based;

[0024] Figure 2 It is a flowchart of the method for determining the unit commitment of a power system provided in an exemplary embodiment of this application;

[0025] Figure 3 It is a flowchart of the method for determining the unit commitment of a power system provided in another exemplary embodiment of this application;

[0026] Figure 4 It is a flowchart of generating an initial integer feasible solution of the unit commitment model provided in an exemplary embodiment of this application;

[0027] Figure 5 It is a flowchart of solving the unit commitment model in a warm-start manner provided in an exemplary embodiment of this application;

[0028] Figure 6 It is a schematic diagram of the structure of the device for determining the unit commitment of a power system provided in an exemplary embodiment of this application;

[0029] Figure 7 It is a schematic diagram of the structure of an electronic device provided in an exemplary embodiment of this application.

[0030] Through the above accompanying drawings, the clear embodiments of this application have been shown, and there will be more detailed descriptions hereinafter. These drawings and textual descriptions are not intended to limit the scope of the concept of this application in any way, but to illustrate the concept of this application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0032] First, the terms involved in the present application will be explained:

[0033] Unit commitment: Based on the predicted values of the power system load in future time periods, the on-off state combination of the power system generating units is obtained through optimal solution.

[0034] Feasibility pump: A heuristic method that searches for integer feasible solutions by iteratively reducing the distance between the relaxed solution and the integer solution of an integer linear programming problem.

[0035] The unit commitment configuration / optimization of the power system is the core link of the power system dispatching operation. It is based on the predicted values of the system load in future time periods, and through modeling and optimal solution, the on-off mode and power generation output curve of the generating units are obtained. Unit commitment arranges the production plan of the power system in advance, provides a scientific decision-making basis for the start-stop, adjustment, maintenance, etc. plans of power equipment, and plays an extremely important role in the safe and economic operation of the power system.

[0036] Generally, the unit commitment (Security Constrained Unit Commitment, SCUC) problem considering network security constraints is modeled as an SCUC model.

[0037] Exemplarily, if it is necessary to optimize the unit commitment of the power system within the next 24 hours, the target time period can be the next 24 hours. When determining the optimized unit commitment, usually the 24 hours are divided into multiple small time periods. Within each small time period, the generating units in the power system maintain one operating state (continuously on or continuously off) and keep a fixed output. Therefore, it is necessary to determine the unit commitment within each small time period, including the output and operating state of each unit within each small time period t. For example, each time period is divided every 15 minutes, and the total number of time periods T included in the target time period is 96. It is necessary to predict the output and operating state of each unit i within 96 time periods respectively.

[0038] Exemplarily, the objective function of the SCUC model includes the operating cost and start-up cost of the generating units in the power system. The objective is to minimize the cost of the unit commitment of the power system, which can be expressed in the form of the following formula (1):

[0039]

[0040] Among them, N is the total number of units in the power system, T is the total number of time periods considered (i.e., the total number of small time periods included in the target time period), p i，t , α i，t are the output and operating status of unit i at time period t respectively. f i，t (p i，t ) is the operating cost of unit i at time period t, is the start-up cost of unit i. Where i = 1, 2,..., N, t = 1, 2,..., T.

[0041] The constraint conditions of the SCUC model mainly include the following three categories: system constraints, unit constraints, and network security constraints.

[0042] Among them, the system constraints of the SCUC model mainly include:

[0043] 1) System load balance constraint

[0044] For each time period t, the system load balance constraint can be expressed as:

[0045]

[0046] Among them, D t is the predicted value of the system load at time period t, p i，t is the output of unit i at time period t.

[0047] 2) System positive and negative reserve capacity constraints

[0048] To prevent system supply-demand imbalance problems caused by system load prediction deviations, etc., generally, the power system needs to reserve a certain amount of positive and negative reserve capacity. For each time period t, the system positive and negative reserve capacity constraints can be expressed as:

[0049]

[0050]

[0051] Among them, are the maximum and minimum allowable outputs of unit i respectively, are the system positive reserve capacity requirement and negative reserve capacity requirement at time period t respectively. α i，t is the operating status of unit i at time period t.

[0052] 3) System spinning reserve constraint

[0053] For each time period t, the sum of the upward adjustment capabilities and the sum of the downward adjustment capabilities of all unit outputs need to meet the actual operating upward and downward spinning reserve requirements:

[0054]

[0055]

[0056] Among them, are the maximum upward and downward ramping rates of unit i respectively, are the upward regulation spinning reserve requirement and downward regulation spinning reserve requirement in period t respectively. p i，t is the output of unit i at time t.

[0057] The unit constraints of the above SCUC model mainly include:

[0058] 4) Unit output upper and lower limit constraints

[0059] In the power system, the output of the unit should be within its allowable output range. The unit output upper and lower limit constraints can be expressed as:

[0060]

[0061] Among them, are the maximum allowable output and minimum allowable output of unit i respectively, p i，t , α i，t are the output and operating status of unit i at time t respectively.

[0062] 5) Unit ramping constraint

[0063] When the unit ramps up or down, it should meet the ramping rate requirement. The unit ramping constraint can be expressed as:

[0064]

[0065]

[0066] Among them, p i，t , α i，t are the output and operating status of unit i at time t respectively, p i，t-1 , α i，t-1 are the output and operating status of unit i at time t - 1 respectively, are the maximum allowable output and minimum allowable output of unit i respectively.

[0067] 6) Unit minimum continuous on / off time constraint

[0068] Due to the operating characteristics of thermal power units, thermal power units need to meet the minimum continuous on / off time requirements:

[0069]

[0070]

[0071] Among them, α i，t is the operating state of unit i in period t, and α i，t-1 is the operating state of unit i in period t - 1. T i，U and T i，D are the minimum continuous on - time and the minimum continuous off - time of unit i respectively; are the time that the unit has been continuously on and the time that the unit has been continuously off at time t respectively, and can be expressed by the unit operating state as follows:

[0072]

[0073]

[0074] The network security constraints of the above - mentioned SCUC model mainly include:

[0075] 7) Network security constraints

[0076] The network security constraints of the power system mainly include the security constraints of line power flow, which can be specifically expressed as:

[0077]

[0078] In the formula, is the thermal stability limit of line l; G l-i is the generator output power transfer distribution factor of the node where unit i is located to line l; G l-k is the generator output power transfer distribution factor of the node where bus load k is located to line l; D k，t is the predicted value of bus load k in period t. p i，t is the output of unit i in period t.

[0079] The unit commitment problem of the power system (such as the solution of the above - mentioned SCUC model) is a large - scale mixed - integer programming problem. As the scale of the power system unit commitment model continues to increase, the difficulty of solving the unit commitment also increases accordingly.

[0080] Currently, the industry mostly adopts data - based methods for accelerating the solution of unit commitment, using a large amount of historical data (including bus loads, unit bids, optimal solutions, etc.), and adopting large - scale deep learning / reinforcement learning models to predict the values of integer variables. Such methods require a large amount of data, high training costs, and cannot guarantee the feasibility of the results. The optimization efficiency of large - scale power system unit commitment is low and cannot meet the requirements of efficiently optimizing the unit commitment of large - scale power systems.

[0081] To solve the problem of low efficiency in unit commitment optimization in a power system, this application provides a method for optimizing unit commitment in a power system. Based on the system data of the power system in the future target time period, a unit commitment model of the current power system is constructed. And by using the integer feasible solution corresponding to the unit commitment information actually configured in the historical time period in the power system, an integer feasible solution of the current unit commitment model is found as the initial integer feasible solution. The warm start function of the integer programming solver is used to optimize and determine a high-quality integer feasible solution (referred to as the unit commitment feasible solution) for the unit commitment of the power system based on the existing initial integer feasible solution, and the unit commitment of the power system in the target time period is optimized based on the high-quality unit commitment feasible solution. This can ensure the feasibility of the optimized unit commitment, and only by using a small amount of easily obtained unit commitment information of historical configurations, a high-quality unit commitment feasible solution can be obtained, improving the optimization efficiency of the unit commitment in the power system, meeting the requirements for efficient optimization of unit commitment in a large-scale power system, and thus enhancing the performance of the power system.

[0082] Figure 1 FIG. is a schematic diagram of the system framework on which this application is based, as Figure 1 shown. The system architecture includes a power system and an electronic device for determining the unit commitment of the power system. The power system contains multiple generating units. By continuously optimizing the start-stop status and output of the generating units at different time periods, that is, optimizing the unit commitment of the power system at different time periods, the safe and economic operation of the power system can be ensured.

[0083] The electronic device can be a local device or a cloud server (such as a cluster, etc.). The electronic device can obtain the power load prediction information and system data of the power system in the future target time period, as well as the unit commitment information actually configured in the historical time period of the power system. Based on the power load prediction information and system data in the target time period, a unit commitment model of the power system is constructed. According to the unit commitment information configured in the historical time period, the initial integer feasible solution of the unit commitment model is determined; the initial integer feasible solution is used as the initial solution of the integer programming solver, and the integer programming solver is warm started. The optimized unit commitment feasible solution is determined by using the integer programming solver; according to the optimized unit commitment feasible solution, the unit commitment information of the power system in the target time period is determined. The unit commitment information of the power system determined by the electronic device can be used to configure / optimize the unit commitment of the power system in the target time period, or to generate production plan information of the power system, etc.

[0084] Among them, the system data of the power system includes unit configuration, grid data, unit cost, etc.

[0085] The unit configuration includes the number of units in the power system, the maximum allowable output and minimum allowable output of the units, the maximum upward ramp rate and maximum downward ramp rate of the units, the minimum continuous on - time and minimum continuous off - time of the units, the requirements for positive and negative reserve capacities of the system during the target period, the requirements for upward and downward spinning reserves during the target period, etc.

[0086] The unit cost includes: unit operation cost, start - up cost, etc.

[0087] The power grid data includes: power grid line topology, the thermal stability limit of the line, the generator output power transfer distribution factor of any unit i's node to line l, the generator output power transfer distribution factor of the bus load k's node to line l, etc.

[0088] The following uses specific embodiments to elaborate in detail on the technical solution of this application and how the technical solution of this application solves the above - mentioned technical problems. These several specific embodiments below can be combined with each other, and for the same or similar concepts or processes, they may not be repeated in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0089] Figure 2 It is a flowchart of a method for determining the unit commitment of a power system provided for an exemplary embodiment of this application. The method provided in this embodiment can be executed by the above - mentioned electronic device. As Figure 2 shown, the specific steps of this method are as follows:

[0090] Step S201: Obtain the power load prediction information and system data of the power system during the future target period, and the unit commitment information configured by the power system during the historical period.

[0091] Among them, the target period can be a future time period. The purpose of this embodiment is to find the optimal or sub - optimal unit commitment in the power system during the target period to optimize the unit commitment of the power system, thereby optimizing the power system. The specific start and end times of the target period, the length of the small time periods divided, etc. can be set and adjusted according to the needs of the actual application scenario, and no specific limitation is made here.

[0092] Exemplarily, the target period can be the next 24 hours, with each 15 - minute period divided into a time slot, and the total number of time slots T included in the target period is 96. The purpose of this embodiment is to determine the output and operating status of each unit in each small time slot within the next 24 hours, and obtain the unit commitment information of the power system in the next 24 hours.

[0093] The power load prediction information of the power system during the future target period can be obtained from the power system or other devices. The specific method for predicting the power load of the power system in a future time slot can adopt the power load prediction methods in the prior art, which will not be elaborated here.

[0094] In this step, system data of the power system can be obtained from the server where the power system is located, the dispatching device of the power system, or other devices, specifically including unit configuration, grid data, unit cost, etc.

[0095] The unit configuration includes the number of units in the power system, the maximum and minimum allowable output powers of the units, the maximum upward and downward ramping rates of the units, the minimum continuous on - line time and minimum continuous off - line time of the units, the requirements for positive and negative reserve capacities of the system in the target period, the requirements for upward and downward spinning reserves in the target period, etc.

[0096] The unit cost includes: unit operation cost, start - up cost, etc.

[0097] The grid data includes: grid line topology, thermal stability limit of the line, generator output power transfer distribution factor of any unit i at a node to line l, generator output power transfer distribution factor of the bus load k at a node to line l, etc.

[0098] In this step, the electronic device can also obtain the unit combination information configured in the power system during the historical period from the dispatching device of the power system. Among them, each set of unit combination information corresponds to a historical integer feasible solution of the unit combination model of the power system. Although the historical integer feasible solution may not necessarily be a feasible solution of the unit combination model in the target period, an integer feasible solution of the unit combination model in the target period can be relatively easily obtained based on the historical integer feasible solution.

[0099] Step S202: Construct a unit combination model of the power system according to the power load prediction information and system data in the target period. The integer feasible solution of the unit combination model represents the available unit combinations of the power system in the target period.

[0100] In this embodiment, based on the power load prediction information and system data in the target period, a unit combination model is constructed to determine the unit combination information of the power system in the target period. Each integer feasible solution of this unit combination model represents a set of available unit combination information of the power system in the target period. By solving the optimal solution / better solution of this unit combination model, the optimal / better unit combination of the power system in the target period can be determined.

[0101] The unit combination model of the power system can adopt the construction method of common unit combination optimization models, such as the aforementioned SCUC model, which will not be elaborated here.

[0102] Step S203: Determine the initial integer feasible solution of the unit combination model according to the unit combination information configured in the historical period.

[0103] Since the integer feasible solutions of history may not be the feasible solutions of the unit commitment model within the target time period, in this step, an integer feasible solution of the unit commitment model within the target time period is obtained based on the integer feasible solutions of history. The specific implementation manner for determining the initial integer feasible solution of the current unit commitment model will be described in detail in the subsequent embodiments.

[0104] In an alternative embodiment, an integer solution of the unit commitment model can also be predicted based on the power load prediction information of the target time period by using an existing deep learning / reinforcement learning model for predicting the unit commitment of the power system, and one of the integer solutions is used to determine the initial integer feasible solution of the current unit commitment model.

[0105] Step S204: Use the initial integer feasible solution as the initial solution of the integer programming solver, warm-start the integer programming solver, and use the integer programming solver to determine the optimized unit commitment feasible solution.

[0106] The unit commitment model constructed in the above step S202 is an integer programming problem, and the optimal solution can be obtained by using an integer programming solver. Based on the optimal solution, a better unit commitment of the power system can be determined.

[0107] In this embodiment, after obtaining an initial integer feasible solution of the current unit commitment model, using this initial integer feasible solution as the initial solution, and utilizing the warm-start function of the integer programming solver to perform multiple rounds of iterative optimization based on the initial solution to determine the optimized unit commitment feasible solution can ensure the feasibility of the obtained unit commitment feasible solution and can accelerate the solving rate of the unit commitment model.

[0108] Step S205: Determine the unit commitment information of the power system within the target time period according to the optimized unit commitment feasible solution.

[0109] Among them, the unit commitment information includes the on / off state information and the output plan of the units.

[0110] After obtaining the optimized unit commitment feasible solution, the unit commitment information of the power system within the target time period can be determined based on the optimized unit commitment feasible solution, and a set of high-quality unit commitments is determined.

[0111] In this embodiment, by using a small amount of easily obtainable historical unit commitment information to determine the historical integer feasible solution of unit commitment, an initial integer feasible solution of the current unit commitment model is determined based on the historical integer feasible solution. This initial integer feasible solution serves as the initial solution to warm-start the integer programming solver, and the integer programming solver is used to determine the optimized unit commitment feasible solution, which can ensure the feasibility of the optimized unit commitment feasible solution in the target time period. According to the optimized unit commitment feasible solution, the unit commitment information of the power system in the target time period is determined, which can improve the efficiency of determining the unit commitment information of the power system in the target time period and improve the quality of the obtained unit commitment information, thereby enhancing the performance of the power system.

[0112] Furthermore, the unit commitment information of the power system in the target time period determined by the electronic device can be used to configure / optimize the unit commitment of the power system in the target time period, or to generate production plan information of the power system, etc.

[0113] In an alternative embodiment, after determining the unit commitment information of the power system in the target time period in step S205 above, according to the unit commitment information of the power system in the target time period, production plan information of the power system is generated and the production plan information is output. Combining with the method for determining the unit commitment of the power system provided in the above embodiment to determine the unit commitment information of the power system in the target time period can improve the optimization efficiency and quality of unit commitment. Further generating production plan information of the power system based on the determined unit commitment information of the power system in the target time period can improve the generation efficiency of the production plan and the effectiveness of the plan, thereby enhancing the performance of the power system.

[0114] In an alternative embodiment, the electronic device that determines the unit commitment information of the power system in the target time period by the determination method in the above embodiment may also have the function of scheduling / configuring / optimizing the power system. After determining the unit commitment information of the power system in the target time period in step S205 above, the electronic device can configure or optimize the unit commitment of the power system in the target time period according to the unit commitment information of the power system in the target time period, thereby optimizing the unit commitment of the power system in real time and enhancing the performance of the power system.

[0115] In an alternative embodiment, for the electronic device that determines the unit commitment information of the power system within the target period by the determination method in the above embodiment, if it does not have the scheduling / configuration / optimization function of the power system, after determining the unit commitment information of the power system within the target period in step S205 above, it may send the unit commitment information of the power system within the target period to the scheduling server of the power system. The scheduling server has the scheduling / configuration / optimization function of the power system, and based on the received unit commitment information of the power system within the target period, configures or optimizes the unit commitment of the power system within the target period, thereby optimizing the unit commitment of the power system in real time and improving the performance of the power system.

[0116] Figure 3 It is a flowchart of a method for determining the unit commitment of a power system provided by another exemplary embodiment of the present application. Based on any of the above method embodiments, the method for determining the unit commitment of the power system is described in detail in this embodiment. As Figure 3 shown, the specific steps of the method are as follows:

[0117] Step S301: Obtain the power load prediction information and system data of the power system within the future target period, and the unit commitment information configured by the power system within the historical period.

[0118] This step is the same as the implementation manner of step S201 above. For specific reference, see the relevant description of step S201 above, and details are not repeated here.

[0119] Step S302: Construct a unit commitment model of the power system according to the power load prediction information and system data within the target period. The integer feasible solution of the unit commitment model represents the available unit commitment of the power system within the target period.

[0120] This step is the same as the implementation manner of step S202 above. For specific reference, see the relevant description of step S201 above, and details are not repeated here.

[0121] In an alternative implementation manner, the initial integer feasible solution of the unit commitment model is determined according to the unit commitment information configured within the historical period in step S203 above through steps S303 - S306.

[0122] Step S303: Use the integer feasible solution corresponding to the unit commitment information configured within the historical period as the target solution.

[0123] Among them, each set of unit commitment information configured within the historical period may include the operating status and output of the units in the power system during that historical period. Among them, the operating status of the unit may be the on - state or off - state, represented by 1 or 0 respectively. The integer feasible solution of the unit commitment model includes a variable α representing the operating status of any unit i in the historical period t i，t, and the variable p representing the output of any unit i in the historical period t i，t . Among them, the variable α of the operating state of unit i in the historical period t i，t is an integer variable, taking values of 0 or 1.

[0124] According to a set of unit commitment information, a corresponding integer feasible solution can be determined, and according to an integer feasible solution of the unit commitment model, a corresponding set of unit commitment information can be determined. The integer feasible solution only contains the partial solution corresponding to the integer variable, that is, it contains the value of α in the solution. i，t value.

[0125] In this embodiment, multiple sets of unit commitment information within a relatively long historical period can be obtained, and the corresponding multiple integer feasible solutions can be obtained as the target solutions, so as to increase the possibility that the target solutions contain the integer feasible solutions of the current unit commitment model.

[0126] In another alternative embodiment, an integer solution of the unit commitment model can also be predicted based on the power load prediction information of the target period by using an existing deep learning / reinforcement learning model for predicting the unit commitment of the power system, and one of the integer solutions is used as the target solution. Further, based on the target solution, the initial integer feasible solution of the current unit commitment model is determined through steps S304 - S306.

[0127] Step S304: Check whether there is an integer feasible solution of the unit commitment model in the target solution.

[0128] In this embodiment, check whether there is an integer feasible solution of the current unit commitment model in the target solution.

[0129] If it exists, execute step S305, and directly select one of the integer feasible solutions of the current unit commitment model as the initial integer feasible solution.

[0130] If it does not exist, execute step S306, and based on a target solution (historical integer feasible solution), use the feasibility pump method to find an integer feasible solution of the current unit commitment model as the initial integer feasible solution.

[0131] Step S305: If there is an integer feasible solution of the unit commitment model in the target solution, then according to the objective function of the unit commitment model, select one of the integer feasible solutions in the target solution as the initial integer feasible solution of the unit commitment model.

[0132] If there is an integer feasible solution of the current unit commitment model in the target solution, one of the integer feasible solutions of the current unit commitment model can be randomly selected as the initial integer feasible solution, and the initial integer feasible solution of the current unit commitment model can be quickly found.

[0133] Step S306: If there is no integer feasible solution for the unit commitment model in the target solution, then according to the constraint conditions of the unit commitment model, select one of the target solutions as the starting solution. Based on the selected starting solution, use the feasibility pump method to generate an initial integer feasible solution for the unit commitment model.

[0134] If there is no integer feasible solution for the current unit commitment model in the target solution, then a target solution with a relatively smaller degree of violation of the constraints of the current unit commitment model can be selected as the starting solution. Based on this starting solution, use the feasibility pump method to determine an integer feasible solution for the current unit commitment model as the initial integer feasible solution of the unit commitment model.

[0135] Optionally, when selecting one of the target solutions as the starting solution according to the constraint conditions of the current unit commitment model, the target solution with a smaller number of violated constraint conditions can be used as the starting solution according to the number of constraint conditions violated by the target solution. Exemplarily, the target solution with the smallest (or the second smallest) number of violated constraint conditions can be used as the starting solution, so that a relatively better starting solution can be found, which can accelerate the speed of obtaining the initial integer feasible solution of the current unit commitment model.

[0136] Optionally, when selecting one of the target solutions as the starting solution according to the constraint conditions of the current unit commitment model, slack variables can be added to the constraint conditions of the current unit commitment model, a penalty term containing the slack variables can be added to the objective function of the current unit commitment model, substitute the target solution into the objective function containing the penalty term to determine the objective function value corresponding to the target solution, and use the target solution with a relatively smaller corresponding objective function value as the starting solution. Exemplarily, the target solution with the smallest (or the second smallest) corresponding objective function value can be used as the starting solution, so that a relatively better starting solution can be found, which can accelerate the speed of obtaining the initial integer feasible solution of the current unit commitment model.

[0137] Among them, the specific implementation method of adding slack variables to the constraint conditions of the current unit commitment model and adding a penalty term containing the slack variables to the objective function of the current unit commitment model can be implemented by using the method of constraint relaxation of adding a penalty term to the objective function in the prior art, which will not be elaborated here.

[0138] In addition, generating an initial integer feasible solution for the unit commitment model by using the feasibility pump method based on the selected starting solution will be described in the subsequent embodiments.

[0139] Step S307: Use the initial integer feasible solution as the initial solution of the integer programming solver, hot start the integer programming solver, and use the integer programming solver to determine the optimized unit commitment feasible solution.

[0140] After finding the initial integer feasible solution of the current unit commitment model, utilize the warm start function of the integer programming solver. Take the initial integer feasible solution as the initial solution of the integer programming solver, warm start the integer programming solver, and use the integer programming solver to perform iterative optimization repeatedly to find the optimal solution of the current unit commitment model, which serves as the optimized unit commitment feasible solution.

[0141] Step S308: Determine the unit commitment information of the power system during the target period based on the optimized unit commitment feasible solution.

[0142] This step is implemented in the same way as the above step S205. For specific details, refer to the relevant description of the above step S205 and will not be elaborated here.

[0143] In this embodiment, by using a small amount of easily obtainable historical unit commitment information to determine the historical integer feasible solution of the unit commitment, based on the historical integer feasible solution to determine an initial integer feasible solution of the current unit commitment model, taking this initial integer feasible solution as the initial solution, warm starting the integer programming solver, and using the integer programming solver to determine the optimized unit commitment feasible solution, it can ensure the feasibility of the optimized unit commitment feasible solution during the target period. Determine the unit commitment information of the power system during the target period based on the optimized unit commitment feasible solution, which can improve the efficiency of determining the unit commitment information of the power system during the target period and improve the quality of the obtained unit commitment information, thereby enhancing the performance of the power system.

[0144] In an alternative embodiment, in the above step S306, to generate the initial integer feasible solution of the unit commitment model based on the selected starting solution, it can also be implemented in the following manner:

[0145] Construct a linear programming model for the unit commitment of the power system according to the selected starting solution and the integer variables in the unit commitment model. The objective function of the linear programming model is the cumulative sum of the absolute values of the differences between the integer variables and the corresponding variable values in the starting solution. The constraint conditions of the linear programming model include the unit constraint conditions, system constraint conditions, and network security constraint conditions that the unit commitment of the power system needs to satisfy. Among them, the value range of the integer variables in the constraint conditions of the linear programming model is relaxed, so the solution obtained by solving the linear programming model is called the relaxed solution. Use the linear programming solver to solve the linear programming model to obtain the relaxed solution, which satisfies all the constraint conditions that the unit commitment of the power system needs to satisfy. However, due to the large number of network security constraint conditions included in the constructed linear programming model, the efficiency of solving the linear programming model is relatively low.

[0146] Considering the characteristics of the power system, in the unit commitment model of a large-scale (such as provincial or regional) power system, there are a large number of 0-1 variables (i.e., α) representing the unit start-stop states i，t) and there are a large number of complex network security constraints in the model, making it difficult to solve the integer feasible solution of the unit commitment model.

[0147] In another alternative embodiment, a method of solving the relaxation problem and checking the security constraints iteratively is embedded in the traditional feasibility pump method to accelerate the solution speed of the linear programming model (linear programming problem) in each iteration.

[0148] Specifically, referring to Figure 4 , in the above step S306, based on the selected starting solution, the feasibility pump method is used to generate an initial integer feasible solution of the unit commitment model, which can be specifically implemented by the following steps:

[0149] Step S401: According to the selected starting solution and the integer variables in the unit commitment model, a linear programming model of the power system unit commitment is constructed. The objective function of the linear programming model is the sum of the absolute values of the differences between the integer variables and the corresponding variable values in the starting solution. The constraint conditions of the linear programming model include the unit constraint conditions and system constraint conditions that the power system unit commitment needs to satisfy. The value range of the integer variables in the constraint conditions of the linear programming model is relaxed. The integer variables correspond one by one to the units in the power system, and each integer variable represents whether the corresponding unit is on or off during the target time period.

[0150] Among them, the integer variables in the current unit commitment model include the 0-1 variables (i.e., α i，t ) that characterize the operating state of the unit during the target time period. The value range of the integer variables in the constraint conditions of the linear programming model is relaxed to [0, 1].

[0151] According to the selected starting solution and the integer variables in the unit commitment model, the objective function of the linear programming model of the power system unit commitment can be expressed as: Among them, α i，t is the variable representing the operating state of unit i during target time period t, and α i，t′ is the variable value corresponding to the operating state of unit i during historical time period t'.

[0152] Optionally, use (α 1，t , α 2，t ,..., α N，t ) to represent the first vector composed of the integer variables in the current unit commitment model, and use (α 1，t′ , α 2，t′ ,..., α N，t′ ) to represent the second vector corresponding to the starting solution. The objective function of the linear programming model of the power system unit commitment can also be expressed as the L1 norm of the difference vector between the first vector and the second vector. Among them, the difference vector between the first vector and the second vector can be expressed as: (α 1，t -α1，t′ , α 2，t -α 2，t′ ,..., α N，t -α N，t′ )。

[0153] Step S402: Solve the linear programming model to obtain a relaxed solution.

[0154] This linear programming model is a linear programming model, and an existing linear programming solver can be used to solve the linear programming model, and the obtained solution is called a relaxed solution.

[0155] Step S403: Verify whether the relaxed solution satisfies the network security constraint conditions.

[0156] After obtaining the relaxed solution, substitute the relaxed solution into the network security constraint conditions of the current unit commitment model to verify whether the relaxed solution satisfies all the network security constraint conditions, that is, to verify whether there are network security constraint conditions that the relaxed solution does not satisfy (or violates, or exceeds the limit).

[0157] If the relaxed solution satisfies all the network security constraint conditions, that is, there are no network security constraint conditions that the relaxed solution does not satisfy (or violates, or exceeds the limit), then execute Step S405.

[0158] If there are network security constraint conditions that the relaxed solution does not satisfy (or violates, or exceeds the limit), then execute Step S405.

[0159] Step S404: If there are network security constraint conditions that the relaxed solution does not satisfy, add the network security constraint conditions that the relaxed solution does not satisfy to the constraint conditions of the linear programming model, and relax the value range of the integer variables in the added network security constraint conditions.

[0160] If there is at least one network security constraint condition that the relaxed solution does not satisfy (or violates, or exceeds the limit), add these network security constraint conditions that the relaxed solution does not satisfy to the constraint conditions of the linear programming model, and relax the value range of the integer variables in the added network security constraint conditions to [0, 1] to obtain a linear programming model with added network security constraint conditions.

[0161] Iteratively execute Steps S402 - S404 to solve the new linear programming model obtained after adding the network security constraint conditions, obtain a new relaxed solution, and verify whether the new relaxed solution satisfies all the network security constraint conditions until the obtained relaxed solution satisfies all the network security constraint conditions, that is, there are no network security constraint conditions that the relaxed solution does not satisfy.

[0162] In this way, through the above steps S402 - S404, the linear programming model can be solved, and the solution result can be verified according to the network security constraint conditions required for the unit commitment of the power system, and a relaxed solution that meets the network security constraint conditions can be obtained. This implementation method of embedding the method of solving the relaxation problem and checking the security constraints iteratively in the feasibility pump can speed up the speed of solving the linear programming model in each iteration.

[0163] Step S405: Check whether the relaxed solution is an integer solution.

[0164] After solving the linear programming model and obtaining a relaxed solution that satisfies all constraint conditions, check whether the relaxed solution is an integer solution.

[0165] If the obtained relaxed solution is an integer solution, then it can be determined that this relaxed solution is an integer feasible solution of the current unit commitment model, and step S407 is executed.

[0166] If the obtained relaxed solution is not an integer solution, that is, the relaxed solution is a fractional solution, then this relaxed solution is not an integer feasible solution of the current unit commitment model, and step S406 is continued to be executed.

[0167] Step S406: If the relaxed solution is a fractional solution, then determine the nearest integer solution of the relaxed solution, and update the objective function of the linear programming model to the cumulative sum of the absolute values of the differences between the integer variables and the corresponding variable values in the nearest integer solution.

[0168] If the obtained relaxed solution is not an integer solution, that is, the relaxed solution is a fractional solution, then this relaxed solution is not an integer feasible solution of the current unit commitment model. In order to find an integer feasible solution of the current unit commitment model, the nearest integer solution of this fractional solution can be determined, and the objective function of the linear programming model can be updated by replacing the starting solution with this nearest integer solution.

[0169] Exemplarily, assume that the variable value representing the operating state of unit i in the target time period t in the nearest integer solution of the fractional solution is α′ i，t , then the objective function of the linear programming model updated based on this nearest integer solution can be expressed as:

[0170] After updating the objective function, iteratively execute steps S402 - S405, iteratively solve the updated linear programming model, and verify the solution result according to the network security constraint conditions to obtain a relaxed solution that meets all network security constraint conditions..., until the verification result in S405 is that the relaxed solution is an integer solution, then stop the iteration and execute step S407.

[0171] Step S407: If the relaxed solution is an integer solution, stop the iteration, and use the relaxed solution obtained in the last iteration as the initial integer feasible solution of the unit commitment model.

[0172] If the obtained relaxed solution is an integer solution, it can be determined that the relaxed solution is an integer feasible solution of the current unit commitment model, and the relaxed solution obtained in the last iteration is used as the initial integer feasible solution of the unit commitment model.

[0173] In the method of this embodiment, considering the characteristics of the power system, in the unit commitment model of a large-scale (such as provincial or regional-level) power system, there are a large number of 0-1 variables (i.e., α i，t ) representing the start-stop states of units, and there are a large number of complex network security constraints in the model, making it difficult to solve the integer feasible solution of the unit commitment model. By embedding a method for solving the relaxed problem and iteratively checking the security constraints in the traditional feasibility pump method, the solution speed of the linear programming model (linear programming problem) in each iteration is accelerated, thereby improving the speed of obtaining the initial integer feasible solution of the current unit commitment model.

[0174] In another alternative embodiment, before the above step S402, the network security constraint conditions can also be added to the linear programming model, and the value range of the integer variables in the network security constraint conditions is relaxed to obtain a linear programming model including security constraints. The linear programming model including security constraints is solved by a linear programming solver to obtain a relaxed solution that satisfies the network security constraint conditions. Considering that the efficiency of directly solving the linear programming model with a large number of network security constraints is low, in this embodiment, it is preferably to use the method of steps S402 - S404 to obtain a relaxed solution that satisfies the network security constraint conditions.

[0175] The constraint conditions of the generally constructed unit commitment model usually include the system constraint conditions, unit constraint conditions, and network security constraint conditions of the power system, and the solution that satisfies all constraint conditions is obtained by solving the unit commitment model.

[0176] Considering the characteristics of the power system, in the unit commitment model of a large-scale (such as provincial or regional-level) power system, there are a large number of 0-1 variables (i.e., α i，t ) representing the start-stop states of units, and there are a large number of complex network security constraints in the model, making it difficult to solve the integer feasible solution of the unit commitment model. In an alternative embodiment, the constraint conditions of the unit commitment model constructed in the above step S202 may include the unit constraint conditions and system constraint conditions required for the unit commitment of the power system, but do not include the network security constraint conditions.

[0177] See Figure 5 , in the above steps S204 and S307, when using the initial integer feasible solution as the initial solution of the integer programming solver and hot-starting the integer programming solver to determine the optimized unit commitment feasible solution by the integer programming solver, it can be specifically implemented by the following steps:

[0178] Step S501: Use the initial integer feasible solution as the initial solution.

[0179] Step S502: Warm-start the integer programming solver, and use the integer programming solver to determine the optimal solution of the unit commitment model based on the initial solution.

[0180] In this embodiment, by using the warm-start function of the integer programming solver and based on a high-quality initial solution, a higher-quality optimal solution is obtained through multiple iterations of optimization. Moreover, since the initially constructed unit commitment model does not include network security constraint conditions, the speed of each iteration of the solution can be accelerated.

[0181] Step S503: Check whether the current optimal solution satisfies the network security constraint conditions not included in the unit commitment model.

[0182] After obtaining the optimal solution, substitute the optimal solution into the network security constraint conditions not included in the current unit commitment model to check whether the optimal solution satisfies the network security constraint conditions not included, that is, check whether there are any network security constraint conditions that the optimal solution does not satisfy (or violates, or exceeds the limit) in the network security constraint conditions not included in the current unit commitment model.

[0183] If the relaxed solution satisfies all the network security constraint conditions, that is, there are no network security constraint conditions that the relaxed solution does not satisfy (or violates, or exceeds the limit), then execute Step S505.

[0184] If there are network security constraint conditions that the relaxed solution does not satisfy (or violates, or exceeds the limit), then execute Step S504.

[0185] Step S504: If the current optimal solution violates at least one network security constraint condition, then add the network security constraint conditions violated by the current optimal solution to the constraint conditions of the unit commitment model.

[0186] If there are at least one network security constraint conditions that the current optimal solution does not satisfy (or violates, or exceeds the limit), then add these network security constraint conditions that the current optimal solution does not satisfy to the constraint conditions of the unit commitment model to ensure that the optimal solutions obtained in subsequent solutions can satisfy these network security constraint conditions.

[0187] Furthermore, use the current optimal solution as the new initial solution, and iterate the above Steps S502 - S504. Warm-start the integer programming solver, use the integer programming solver to determine the new optimal solution of the current unit commitment model, and check whether the new optimal solution satisfies the network security constraint conditions not included in the current unit commitment model... Stop the iteration until the verification result of S503 shows that the obtained optimal solution satisfies all network security constraint conditions, and determine the finally obtained optimal solution as the optimized unit commitment feasible solution.

[0188] Step S505: If the current optimized solution satisfies all network security constraint conditions, determine the finally obtained optimized solution as the feasible solution of the optimized unit commitment.

[0189] In this embodiment, considering the characteristics of the power system, in the unit commitment model of a large-scale (such as provincial or regional-level) power system, there are a large number of 0-1 variables (i.e., α i，t ) representing the start-stop states of units, and there are a large number of complex network security constraints in the model. It is difficult to solve the integer feasible solution of the unit commitment model. The constraint conditions of the initially constructed unit commitment model do not include network security constraint conditions. After obtaining the optimized solution, check whether it satisfies the network security constraint conditions, and gradually add a small number of network security constraint conditions to the constraint conditions of the unit commitment model, which can reduce the constraint conditions of the unit commitment model, thereby accelerating the speed of solving the unit commitment model in each iteration.

[0190] Optionally, before using the initial integer feasible solution as the initial solution of the integer programming solver, hot-starting the integer programming solver, and using the integer programming solver to determine the feasible solution of the optimized unit commitment, it further includes:

[0191] Determine the network security constraint conditions that take effect on the initial integer feasible solution, and add the effective network security constraint conditions to the constraint conditions of the unit commitment model. By adding the network security constraint conditions that take effect on the initial integer feasible solution to the constraint conditions of the unit commitment model, the quality of the obtained optimized solution can be improved, and thus an integer feasible solution of a high-quality unit commitment model that satisfies all constraint conditions can be quickly found.

[0192] Figure 6 It is a schematic structural diagram of a device for determining the unit commitment of a power system provided by an exemplary embodiment of the present application. The device provided in this embodiment is applied to execute the method for determining the unit commitment of a power system. As Figure 6 shown, the device 60 for determining the unit commitment of a power system includes: a power system related data acquisition module 61, a unit commitment model construction module 62, a feasible solution determination module 63, a unit commitment feasible solution optimization module 64, and a unit commitment information determination module 65.

[0193] Among them, the power system related data acquisition module 61 is used to acquire the power load prediction information and system data of the power system in the future target time period, as well as the unit commitment information configured by the power system in the historical time period.

[0194] The unit commitment model construction module 62 is used to construct a unit commitment model of the power system according to the power load prediction information and system data in the target time period. The integer feasible solution of the unit commitment model represents the available unit commitment of the power system in the target time period.

[0195] The feasible solution determination module 63 is configured to determine an initial integer feasible solution of the unit commitment model according to the unit commitment information configured within a historical period.

[0196] The unit commitment feasible solution optimization module 64 is configured to use the initial integer feasible solution as the initial solution of the integer programming solver, hot-start the integer programming solver, and use the integer programming solver to determine an optimized unit commitment feasible solution.

[0197] The unit commitment information determination module 65 is configured to determine the unit commitment information of the power system within a target period according to the optimized unit commitment feasible solution.

[0198] In an alternative embodiment, when implementing the determination of the initial integer feasible solution of the unit commitment model according to the unit commitment information configured within a historical period, the feasible solution determination module 63 is further configured to:

[0199] Use the integer feasible solution corresponding to the unit commitment information configured within the historical period as the target solution, and check whether there is an integer feasible solution of the unit commitment model in the target solution; if there is no integer feasible solution of the unit commitment model in the target solution, then according to the constraint conditions of the unit commitment model, select one of the target solutions as the starting solution, and based on the selected starting solution, use the feasibility pump method to generate an initial integer feasible solution of the unit commitment model.

[0200] In an alternative embodiment, when implementing the determination of the initial integer feasible solution of the unit commitment model according to the unit commitment information configured within a historical period, the feasible solution determination module 63 is further configured to:

[0201] If there is an integer feasible solution of the unit commitment model in the target solution, then according to the objective function of the unit commitment model, select an integer feasible solution of the unit commitment model in the target solution as the initial integer feasible solution of the unit commitment model.

[0202] In an alternative embodiment, when implementing the selection of one of the target solutions as the starting solution according to the constraint conditions of the unit commitment model, the feasible solution determination module 63 is further configured to:

[0203] According to the number of constraint conditions violated by the target solution, use the target solution with the least number of violated constraint conditions as the starting solution; or, add slack variables to the constraint conditions of the unit commitment model, add a penalty term including the slack variables to the objective function of the unit commitment model, substitute the target solution into the objective function including the penalty term to determine the objective function value corresponding to the target solution, and use the target solution with the smallest corresponding objective function value as the starting solution.

[0204] In an alternative embodiment, when implementing the generation of an initial integer feasible solution of the unit commitment model using the feasibility pump method based on the selected starting solution, the feasible solution determination module 63 is further configured to:

[0205] According to the selected initial solution and the integer variables in the unit commitment model, a linear programming model for the unit commitment of the power system is constructed. The objective function of the linear programming model is the cumulative sum of the absolute values of the differences between the integer variables and the corresponding variable values in the initial solution. The constraint conditions of the linear programming model include the unit constraint conditions and system constraint conditions that the unit commitment of the power system needs to satisfy. The value range of the integer variables in the constraint conditions of the linear programming model is relaxed. The integer variables correspond one by one to the units in the power system, and each integer variable represents whether the corresponding unit is on in the target time period.

[0206] Solve the linear programming model, and verify the solution according to the network security constraint conditions that the unit commitment of the power system needs to satisfy, to obtain a relaxed solution that satisfies the network security constraint conditions.

[0207] If the relaxed solution is a fractional solution, determine the nearest integer solution of the relaxed solution, and update the objective function of the linear programming model to the cumulative sum of the absolute values of the differences between the integer variables and the corresponding variable values in the nearest integer solution.

[0208] Iteratively solve the updated linear programming model, and verify the solution according to the network security constraint conditions, to obtain a relaxed solution that satisfies the network security constraint conditions, and stop the iteration until the obtained relaxed solution is an integer solution. Take the relaxed solution obtained in the last iteration as the initial integer feasible solution of the unit commitment model.

[0209] In an alternative embodiment, when implementing the solution of the linear programming model and verifying the solution according to the network security constraint conditions that the unit commitment of the power system needs to satisfy to obtain a relaxed solution that satisfies the network security constraint conditions, the feasible solution determination module 63 is further configured to:

[0210] Solve the linear programming model to obtain a relaxed solution; check whether the relaxed solution satisfies the network security constraint conditions; if there are network security constraint conditions that the relaxed solution does not satisfy, add the network security constraint conditions that the relaxed solution does not satisfy to the constraint conditions of the linear programming model, and relax the value range of the integer variables in the added network security constraint conditions. Repeatedly iteratively solve the linear programming model with the added network security constraint conditions to obtain a new relaxed solution, and check whether the new relaxed solution satisfies all network security constraint conditions until the obtained relaxed solution satisfies all network security constraint conditions.

[0211] In an alternative embodiment, the constraint conditions of the unit commitment model include the unit constraint conditions and system constraint conditions that the unit commitment of the power system needs to satisfy, and do not include network security constraint conditions.

[0212] When implementing using the initial integer feasible solution as the initial solution of the integer programming solver to hot-start the integer programming solver and using the integer programming solver to determine the optimized unit commitment feasible solution, the unit commitment feasible solution optimization module 64 is further configured to:

[0213] Use the initial integer feasible solution as the initial solution to hot-start the integer programming solver, and use the integer programming solver to determine the optimized solution of the unit commitment model based on the initial solution; check whether the current optimized solution satisfies the network security constraint conditions not included in the unit commitment model. If the current optimized solution violates at least one network security constraint condition, add the network security constraint conditions violated by the current optimized solution to the constraint conditions of the unit commitment model; use the current optimized solution as the new initial solution, iteratively hot-start the integer programming solver to determine the new optimized solution of the current unit commitment model, and check whether the new optimized solution satisfies the network security constraint conditions not included in the current unit commitment model until the obtained optimized solution satisfies all network security constraint conditions, and determine the finally obtained optimized solution as the optimized unit commitment feasible solution.

[0214] In an alternative embodiment, before using the initial integer feasible solution as the initial solution of the integer programming solver to hot-start the integer programming solver and using the integer programming solver to determine the optimized unit commitment feasible solution, the unit commitment feasible solution optimization module 64 is further configured to:

[0215] Determine the network security constraint conditions that act on the initial integer feasible solution, and add the acting network security constraint conditions to the constraint conditions of the unit commitment model.

[0216] In an alternative embodiment, the power system unit commitment determination device may further include: a power system production plan generation module, configured to:

[0217] After determining the unit commitment information of the power system during the target period according to the optimized unit commitment feasible solution, generate the production plan information of the power system according to the unit commitment information of the power system during the target period; output the production plan information.

[0218] In an alternative embodiment, the power system unit commitment determination device may further include: a power system scheduling module, configured to:

[0219] Configure or optimize the unit commitment of the power system during the target period according to the unit commitment information of the power system during the target period;

[0220] Or,

[0221] Send the unit commitment information of the power system during the target period to the scheduling server of the power system, so that the scheduling server configures or optimizes the unit commitment of the power system during the target period based on the received unit commitment information of the power system during the target period.

[0222] The device provided in this embodiment can be specifically used to execute the method for generating a downstream task model based on prompts provided in any of the above embodiments. The specific functions and achievable technical effects are not elaborated here.

[0223] Figure 7 It is a schematic structural diagram of an electronic device provided in an exemplary embodiment of the present application. As Figure 7 shown, the electronic device 70 includes: a processor 701, and a memory 702 communicatively connected to the processor 701. The memory 702 stores computer-executable instructions.

[0224] Among them, the processor executes the computer-executable instructions stored in the memory to implement the solutions provided in any of the above method embodiments. The specific functions and achievable technical effects are not elaborated here.

[0225] The embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the solutions provided in any of the above method embodiments. The specific functions and achievable technical effects are not elaborated here.

[0226] The embodiment of the present application also provides a computer program product. The computer program product includes: a computer program. The computer program is stored in a readable storage medium. At least one processor of the electronic device can read the computer program from the readable storage medium. Executing the computer program by at least one processor causes the electronic device to execute the solutions provided in any of the above method embodiments. The specific functions and achievable technical effects are not elaborated here.

[0227] In addition, in some of the processes described in the above embodiments and the accompanying drawings, there are multiple operations that appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. They are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types. The meaning of "multiple" is more than two, unless otherwise specifically and clearly defined.

[0228] Other embodiments of the present application will be readily apparent to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. The present application is intended to cover any variations, uses, or adaptations of the present application, which follow the general principles of the present application and include known common general knowledge or conventional technical means in the technical field not disclosed in the present application. The specification and examples are only to be considered as exemplary, and the true scope and spirit of the present application are pointed out by the following claims.

[0229] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

Claims

1. A method for determining unit commitment of a power system, characterized in that, it includes: Obtain the power load forecast information and system data of the power system within the target time period in the future, and the unit commitment information configured by the power system within the historical time period; Construct a unit commitment model of the power system according to the power load forecast information and system data within the target time period, and the integer feasible solution of the unit commitment model represents the available unit commitment of the power system within the target time period; Determine the initial integer feasible solution of the unit commitment model according to the unit commitment information configured within the historical time period; Use the initial integer feasible solution as the initial solution of the integer programming solver, hot-start the integer programming solver, and use the integer programming solver to determine the optimized unit commitment feasible solution; Determine the unit commitment information of the power system within the target time period according to the optimized unit commitment feasible solution, and the unit commitment information includes the on / off state information and output plan of the unit; The step of determining the initial integer feasible solution of the unit commitment model according to the unit commitment information configured within the historical time period includes: Use the integer feasible solution corresponding to the unit commitment information configured within the historical time period as the target solution, and check whether there is an integer feasible solution of the unit commitment model in the target solution; If there is no integer feasible solution of the unit commitment model in the target solution, then according to the constraint conditions of the unit commitment model, select one of the target solutions as the starting solution, and based on the selected starting solution, use the feasibility pump method to generate the initial integer feasible solution of the unit commitment model.

2. The method according to claim 1, characterized in that, it further includes: If there is an integer feasible solution of the unit commitment model in the target solution, then according to the objective function of the unit commitment model, select an integer feasible solution of the unit commitment model in the target solution as the initial integer feasible solution of the unit commitment model.

3. The method according to claim 1, characterized in that, The step of selecting one of the target solutions as the starting solution according to the constraint conditions of the unit commitment model includes: According to the number of constraint conditions violated by the target solution, select the target solution with the least number of violated constraint conditions as the starting solution; Or, Add slack variables to the constraint conditions of the unit commitment model, add a penalty term containing the slack variables to the objective function of the unit commitment model, substitute the target solution into the objective function containing the penalty term to determine the objective function value corresponding to the target solution, and select the target solution with the smallest corresponding objective function value as the starting solution.

4. The method according to claim 1, characterized in that, The step of generating the initial integer feasible solution of the unit commitment model based on the selected starting solution using the feasibility pump method includes: Construct a linear programming model for the unit commitment of the power system according to the selected initial solution and the integer variables in the unit commitment model. The objective function of the linear programming model is the sum of the absolute values of the differences between the integer variables and the corresponding variable values in the initial solution. The constraint conditions of the linear programming model include the unit constraint conditions and system constraint conditions that the unit commitment of the power system needs to satisfy. The value range of the integer variables in the constraint conditions of the linear programming model is relaxed. The integer variables correspond one by one to the units in the power system, and each integer variable represents whether the corresponding unit is turned on during the target period; Solve the linear programming model, and verify the solution result according to the network security constraint conditions that the unit commitment of the power system needs to satisfy, and obtain a relaxed solution that satisfies the network security constraint conditions; If the relaxed solution is a fractional solution, determine the nearest integer solution of the relaxed solution, and update the objective function of the linear programming model to the sum of the absolute values of the differences between the integer variables and the corresponding variable values in the nearest integer solution; Iteratively solve the updated linear programming model, and verify the solution result according to the network security constraint conditions, and obtain a relaxed solution that satisfies the network security constraint conditions, and stop the iteration until the obtained relaxed solution is an integer solution, and use the relaxed solution obtained in the last iteration as the initial integer feasible solution of the unit commitment model.

5. The method according to claim 4, wherein, the solving the linear programming model, and verifying the solution result according to the network security constraint conditions that the unit commitment of the power system needs to satisfy, and obtaining a relaxed solution that satisfies the network security constraint conditions includes: Solve the linear programming model to obtain a relaxed solution; Verify whether the relaxed solution satisfies the network security constraint conditions; If there are network security constraint conditions that the relaxed solution does not satisfy, add the network security constraint conditions that the relaxed solution does not satisfy to the constraint conditions of the linear programming model, and relax the value range of the integer variables in the added network security constraint conditions, and repeatedly iteratively solve the linear programming model with the added network security constraint conditions to obtain a new relaxed solution, and verify whether the new relaxed solution satisfies all network security constraint conditions until the obtained relaxed solution satisfies all network security constraint conditions.

6. The method according to claim 1, wherein, the constraint conditions of the unit commitment model include the unit constraint conditions and system constraint conditions that the unit commitment of the power system needs to satisfy, and do not include network security constraint conditions; the using the initial integer feasible solution as the initial solution of the integer programming solver, hot starting the integer programming solver, and using the integer programming solver to determine the optimized unit commitment feasible solution includes: Use the initial integer feasible solution as the initial solution, hot start the integer programming solver, and use the integer programming solver to determine the optimized solution of the unit commitment model based on the initial solution; Verify whether the current optimal solution satisfies the cybersecurity constraint conditions not included in the unit commitment model. If the current optimal solution violates at least one cybersecurity constraint condition, add the violated cybersecurity constraint conditions to the constraint conditions of the unit commitment model; Take the current optimal solution as the new initial solution, iterate to hot-start the integer programming solver to determine the new optimal solution of the current unit commitment model, and verify whether the new optimal solution satisfies the cybersecurity constraint conditions not included in the current unit commitment model until the obtained optimal solution satisfies all cybersecurity constraint conditions, and determine the finally obtained optimal solution as the optimized feasible solution of the unit commitment.

7. The method according to claim 6, wherein, before taking the initial integer feasible solution as the initial solution of the integer programming solver, hot-starting the integer programming solver, and using the integer programming solver to determine the optimized feasible solution of the unit commitment, further includes: Determine the cybersecurity constraint conditions that act on the initial integer feasible solution, and add the acting cybersecurity constraint conditions to the constraint conditions of the unit commitment model.

8. The method according to claim 1, wherein, after determining the unit commitment information of the power system during the target period according to the optimized feasible solution of the unit commitment, further includes: Generate the production plan information of the power system according to the unit commitment information of the power system during the target period; Output the production plan information.

9. The method according to claim 1, wherein, after determining the unit commitment information of the power system during the target period according to the optimized feasible solution of the unit commitment, further includes: Configure or optimize the unit commitment of the power system during the target period according to the unit commitment information of the power system during the target period; or, Send the unit commitment information of the power system during the target period to the dispatching server of the power system, so that the dispatching server configures or optimizes the unit commitment of the power system during the target period based on the received unit commitment information of the power system during the target period.

10. A device for determining the unit commitment of a power system, wherein, includes: A power system related data acquisition module, configured to acquire the power load prediction information and system data of the power system during a future target period, and the unit commitment information configured by the power system during a historical period; A unit commitment model construction module, configured to construct a unit commitment model of the power system according to the power load prediction information and system data during the target period, and the integer feasible solution of the unit commitment model represents the available unit commitment of the power system during the target period; A feasible solution determination module, configured to determine the initial integer feasible solution of the unit commitment model according to the unit commitment information configured during the historical period; The unit commitment feasible solution optimization module is used to use the initial integer feasible solution as the initial solution of the integer programming solver, hot-start the integer programming solver, and use the integer programming solver to determine the optimized unit commitment feasible solution; The unit commitment information determination module is used to determine the unit commitment information of the power system during the target period according to the optimized unit commitment feasible solution, where the unit commitment information includes on / off state information and output plan; The feasible solution determination module is specifically used to use the integer feasible solution corresponding to the unit commitment information configured during the historical period as the target solution, and check whether there is an integer feasible solution of the unit commitment model in the target solution; If there is no integer feasible solution of the unit commitment model in the target solution, then according to the constraint conditions of the unit commitment model, select one of the target solutions as the starting solution, and based on the selected starting solution, use the feasibility pump method to generate the initial integer feasible solution of the unit commitment model.

11. An electronic device, characterized in that, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer execution instructions; the processor executes the computer execution instructions stored in the memory to implement the method according to any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, the computer-readable storage medium stores computer execution instructions, and the computer execution instructions are used to implement the method according to any one of claims 1-9 when executed by a processor.

13. A computer program product, characterized in that, comprising a computer program, and the computer program is used to implement the method according to any one of claims 1-9 when executed by a processor.

Images