Unit maintenance plan optimization method and device considering elastic window

Abstract

This invention provides a method and apparatus for optimizing unit maintenance planning considering flexible windows, belonging to the field of power system maintenance technology. The method includes: receiving maintenance information for units to be maintained; determining the fixed maintenance window period for each unit and setting flexible window periods on both sides of the fixed window period; generating a window period deviation penalty term based on the unit's rated capacity and the degree to which the maintenance schedule deviates from the fixed window period; setting a compressible actual maintenance time variable that satisfies the minimum maintenance time constraint, and generating a maintenance time compression penalty term based on the compression amount; constructing a joint optimization model under unit combination constraints and maintenance constraints, setting a comprehensive objective function that includes grid operating costs and the aforementioned penalty term, and solving for the start and end times and maintenance duration of each unit's maintenance. The technical solution of this invention achieves flexible optimization of maintenance plans, improving planning feasibility and operational economy.

Description

Technical Field

[0001] This invention belongs to the field of power system maintenance technology, specifically relating to a method and apparatus for optimizing unit maintenance plans that considers flexible windows. Background Technology

[0002] Power generating units are a crucial component of the power grid. During long-term operation, these units are susceptible to performance degradation or potential malfunctions due to factors such as equipment aging and fluctuations in operating conditions. To ensure the safe and stable operation of equipment and maintain the designed performance of the units, power grid operation management typically requires the development of reasonable maintenance plans before unit overhauls. In recent years, the requirements for lean power grid operation management have been continuously increasing. Maintenance plans not only need to meet power grid safety constraints but also need to coordinate maintenance time under conditions of multiple units and multiple cycles, reducing the impact of repeated maintenance and maintenance conflicts on system operation.

[0003] Typically, power generation companies submit their unit maintenance windows. The grid dispatching department, considering factors such as grid safety and stability requirements, system reserve levels, load characteristics, and maintenance schedules for each unit, formulates annual maintenance tasks and optimizes the maintenance plan. However, in actual plan formulation and execution, situations often arise such as concentrated maintenance tasks, overlapping windows, and tight system operation constraints. This makes it difficult to complete some maintenance tasks within the predetermined fixed windows, requiring repeated coordination and adjustments between the grid and power generation companies to adjust maintenance schedules, increasing planning costs and execution uncertainty.

[0004] Furthermore, maintenance plans are often influenced by a combination of factors, including maintenance resource occupancy, maintenance duration, unit operating mode, and system safety constraints. When there are conflicts between maintenance duration, available maintenance resources, and system operating boundaries, traditional methods relying on experience or local adjustments are prone to problems such as insufficient plan feasibility, frequent adjustments, and impaired overall economic efficiency, making it difficult to meet the development needs of intelligent and refined power grid dispatch management.

[0005] Therefore, there is an urgent need for a method to optimize unit maintenance plans in order to improve the feasibility of maintenance plans and the flexibility of scheduling. Summary of the Invention

[0006] This invention provides a method for optimizing unit maintenance plans considering flexible windows. The method constructs a joint optimization model that includes setting fixed and flexible maintenance windows, generating window deviation penalties, setting compressible actual maintenance duration variables, and generating maintenance duration compression penalties. Under unit combination constraints and maintenance constraints, a comprehensive objective function is set that includes grid operation costs and the aforementioned penalties. The solution outputs the start and end times of maintenance for each unit and the corresponding maintenance duration. This addresses the problems of concentrated maintenance tasks and window conflict leading to plans that fail to meet grid operation constraints and require frequent coordination and adjustments under fixed maintenance windows and durations.

[0007] A first aspect of the present invention provides a method for optimizing unit maintenance plans considering flexible windows, the method comprising: In response to the input maintenance information of the unit to be maintained, a fixed maintenance window period corresponding to each unit to be maintained is determined, and flexible window periods are set on both sides of the fixed maintenance window period. Based on the rated capacity of the unit to be maintained and the degree of deviation of the maintenance schedule from the fixed maintenance window period, a window period deviation penalty factor is determined, and a window period deviation penalty item is generated. An actual maintenance time variable is set based on the normal maintenance time and minimum maintenance time of the unit to be maintained. The actual maintenance time variable is not less than the minimum maintenance time and not greater than the normal maintenance time. A maintenance time compression penalty factor is determined based on the compression amount of the actual maintenance time variable relative to the normal maintenance time, and a maintenance time compression penalty term is generated. A joint optimization model is constructed, which includes a set of decision variables related to unit operating status, unit output, unit maintenance schedule, and maintenance time compression. A comprehensive objective function is set, which includes at least the grid operating cost, the window period deviation penalty term, and the maintenance time compression penalty term. Unit combination constraints and maintenance constraints are added to the joint optimization model. The joint optimization model is solved, and the maintenance start and end times and corresponding maintenance times of each unit are output as the maintenance plan optimization results.

[0008] By adopting the above scheme, the unit maintenance plan optimization method of the present invention, which considers flexible windows, realizes the adjustable arrangement of maintenance tasks on the time axis by setting flexible windows on both sides of the fixed maintenance window period, reducing the scheduling conflicts caused by overlapping or concentrated maintenance of fixed windows; by constructing a window period deviation penalty term based on the rated capacity and deviation degree of the unit, it realizes the quantitative constraint and priority guidance of maintenance deviation, reducing the repeated manual coordination of window periods during the planning process; by introducing a compressible actual maintenance time variable that meets the minimum maintenance time constraint, it realizes the adaptive adjustment of maintenance duration under limited conditions, alleviating the infeasibility problem caused by the coupling of maintenance resources and system operation boundary; furthermore, by incorporating the grid operation cost, window period deviation penalty term, and maintenance time compression penalty term into the comprehensive objective function and solving them jointly under unit combination constraints and maintenance constraints, it realizes the integrated optimization output of maintenance plan and operation mode, significantly improving the executability and feasibility of maintenance plan under safety constraints.

[0009] In some embodiments of the present invention, the flexible window period is obtained by extending the fixed maintenance window period by a preset duration before the start time and after the end time, respectively, and the preset duration is set according to the system operating conditions.

[0010] In some embodiments of the present invention, the window period deviation penalty factor is obtained by combining the unit rated capacity normalization coefficient, the benchmark penalty factor, and the window period deviation factor, wherein the unit rated capacity normalization coefficient is determined by the unit rated capacity and the maximum rated capacity of the unit to be maintained.

[0011] In some embodiments of the present invention, the benchmark penalty factor is set based on the current system load level and dispatchable margin, the window period deviation factor is determined based on the number of deviation periods of the maintenance schedule from the fixed maintenance window period, and the mathematical form of the window period deviation factor adopts a linear function or an exponential function.

[0012] In some embodiments of the present invention, the window period deviation factor is an exponential function with the number of deviation periods as the independent variable, the exponential function including an adjustment coefficient, the adjustment coefficient representing the rate at which the penalty increases with the degree of deviation.

[0013] In some embodiments of the present invention, the normal maintenance duration is determined based on the unit maintenance type, maintenance workload, maintenance resources, and maintenance procedure arrangement, while the minimum maintenance duration is determined based on the shortest completion time of key procedures, equipment safety requirements, and personnel workload.

[0014] In some embodiments of the present invention, the maintenance time compression penalty factor is obtained by combining the normal maintenance time normalization coefficient, the benchmark penalty factor, and the maintenance time compression coefficient. The normal maintenance time normalization coefficient is determined by the normal maintenance time and the maximum normal maintenance time of the unit to be maintained. The benchmark penalty factor is set according to the unit maintenance type, maintenance resource occupancy rate, and system dispatch margin. The maintenance time compression coefficient is a monotonically increasing function of the maintenance time compression ratio.

[0015] In some embodiments of the present invention, the maintenance time compression coefficient adopts an exponential function; when performing a piecewise linear approximation of the exponential function, the compression ratio interval is divided into several continuous sub-intervals, an interpolation weight variable is introduced, and a second type of special ordered set SOS2 constraint is applied to the interpolation weight variable.

[0016] In some embodiments of the present invention, the maintenance constraints include at least maintenance continuity constraints, maintenance mutual exclusion constraints, maintenance duration constraints, and daily maintenance quantity limit constraints. The maintenance continuity constraints introduce maintenance start state variables and maintenance end state variables. The maintenance start state variable takes a first preset value during the maintenance start period, and the maintenance end state variable takes a first preset value during the maintenance end period.

[0017] Compared with existing technologies, the advantages of this invention are as follows: by extending a preset duration on both sides of a fixed maintenance window to form a flexible window period, the maintenance time can be adjusted within the limited window, reducing scheduling conflicts caused by overlapping windows and concentrated maintenance tasks; by designing the window period deviation penalty factor as consisting of a normalized coefficient of the unit's rated capacity, a benchmark penalty factor, and a deviation factor, and by linking the deviation factor with the number of deviation periods, a graded constraint and differentiated guidance on the degree of deviation can be achieved using linear or exponential forms, reducing repeated coordination and adjustments under the rigid constraints of a fixed window; and by setting a normal maintenance duration, a minimum maintenance duration, and introducing a compressible actual maintenance duration variable, the maintenance duration can be flexibly adjusted within the safety lower limit constraint, alleviating maintenance difficulties. The infeasibility problem caused by the coupling between resource consumption and system operation boundaries is addressed. This is achieved by constructing a maintenance time compression penalty factor composed of a normalized coefficient for normal maintenance time, a baseline penalty factor, and a compression coefficient. The compression coefficient is set to a monotonically increasing function of the compression ratio. Further, exponential penalties are employed, combined with piecewise linear approximation and SOS2 constraints, to controllably suppress and computationally characterize excessive compression behavior. Furthermore, by introducing maintenance start and end state variables into the maintenance constraints to characterize the continuity and boundary periods of the maintenance process, a continuous and executable expression of maintenance arrangements under compressible time constraints is achieved. Finally, the quantitative constraint capabilities for maintenance deviations and time compression are strengthened, reducing the frequency of plan adjustments and enhancing the controllability and consistency of maintenance arrangements.

[0018] A second aspect of the present invention provides a unit maintenance plan optimization system that takes into account flexible windows, comprising: The maintenance information acquisition module is used to acquire maintenance information of the units to be maintained. The window period generation module is used to determine the fixed maintenance window period corresponding to each unit to be maintained based on the maintenance information of the unit to be maintained, and to set flexible window periods on both sides of the fixed maintenance window period. The penalty item generation module is used to determine a window period deviation penalty factor based on the rated capacity of the unit to be maintained and the degree of deviation of the maintenance schedule from the fixed maintenance window period, and to generate a window period deviation penalty item; the penalty item generation module is also used to set an actual maintenance time variable based on the normal maintenance time and the minimum maintenance time of the unit to be maintained, wherein the actual maintenance time variable is not less than the minimum maintenance time and not greater than the normal maintenance time, and to determine a maintenance time compression penalty factor based on the compression amount of the actual maintenance time variable relative to the normal maintenance time, and to generate a maintenance time compression penalty item; The optimization solution module is used to construct a joint optimization model, which includes a set of decision variables related to unit operating status, unit output, unit maintenance schedule, and maintenance time compression. A comprehensive objective function is set, which includes at least the grid operating cost, the window period deviation penalty term, and the maintenance time compression penalty term. Unit combination constraints and maintenance constraints are added to the joint optimization model. The joint optimization model is solved, and the start and end times of maintenance for each unit and the corresponding maintenance duration are output as the maintenance plan optimization results. A third aspect of the present invention provides a unit maintenance plan optimization device that considers flexible windows, characterized in that the device includes a computer device, the computer device includes a processor and a memory, the processor stores computer instructions, and when the computer instructions are executed, the device implements the unit maintenance plan optimization method that considers flexible windows.

[0019] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the text, or may be learned by practice of the invention. The objects and other advantages of the invention will become apparent from the description and the accompanying drawings.

[0020] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0021] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0022] In the attached diagram: Figure 1 This is a flowchart illustrating a method for optimizing unit maintenance plans that considers flexible windows, provided as an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram illustrating the window period penalty in a unit maintenance plan optimization method considering flexible windows, provided as an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the unit maintenance time for a unit maintenance plan optimization method that considers flexible windows, provided as an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of a unit maintenance plan optimization system that takes into account the flexible window, provided as an embodiment of the present invention.

[0026] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0027] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0028] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0029] Figure 1 This is a flowchart illustrating a method for optimizing unit maintenance plans that considers flexible windows, provided by an embodiment of the present invention.

[0030] Example 1, as Figure 1 As shown, the present invention provides a method for optimizing unit maintenance plans considering flexible windows, the method comprising the following steps: S1. In response to the input maintenance information of the unit to be maintained, determine the fixed maintenance window period corresponding to each unit to be maintained, and set flexible window periods on both sides of the fixed maintenance window period. In this embodiment, step S1 specifically includes: Based on the operating years, maintenance procedures, and technical requirements of each generator unit, a fixed maintenance window is set for each unit. For example, for a 400MW unit, the fixed maintenance window is from the 120th to the 135th day of the year.

[0031] Based on this, and according to system operating conditions, adjacent time periods before and after the fixed window period are set as flexible window periods, allowing units to undergo maintenance earlier or later within this range. For example, extending the fixed window period by 15 days on each side creates a flexible window period of 45 days.

[0032] The flexible window period is used to provide time flexibility for overall maintenance feasibility, so that the maintenance time can be flexibly adjusted during periods of concentrated maintenance of a large number of units or when system resources are tight.

[0033] Determining the fixed maintenance window period for each unit includes the following steps: Step S11: The fixed maintenance window period for each unit is determined based on the number of units requiring maintenance, maintenance duration, maintenance resource occupancy, and system operating cycle. Specifically, a preliminary maintenance schedule can be developed using historical maintenance data or expert experience to ensure that the number of units undergoing maintenance within the same time period does not exceed the system's capacity.

[0034] Step S12: Based on the fixed window period, in order to improve the overall feasibility and flexibility of the maintenance plan, a flexible window period is set.

[0035] like Figure 2 As shown, the fixed maintenance window (rigid window period) of a certain unit to be inspected is used. Based on the fixed maintenance window period, flexible maintenance windows are set both before and after it. (See diagram) , , , These represent the different values ​​of the deviation penalty factor (which increases with the degree of deviation) when the maintenance time falls within different flexible window intervals outside the fixed window.

[0036] Fixed maintenance windows define the time interval for scheduled priority maintenance of generating units. Flexible windows provide an adjustable range for maintenance timing when all maintenance tasks cannot be completed within the fixed window, or when maintenance conflicts exist or system constraint margins are insufficient. Flexible windows are located on either side of the fixed window, with a preset length that can be dynamically adjusted based on system operating conditions (such as load levels, reserve capacity, and system constraint margins). For example, when system load is low or dispatchable reserves are sufficient, the length of the flexible window can be appropriately increased; when system operation is strained, the range of the flexible window can be shortened.

[0037] S2. Based on the rated capacity of the unit to be maintained and the degree of deviation of the maintenance schedule from the fixed maintenance window period, determine the window period deviation penalty factor and generate the window period deviation penalty item. In step S2, a window period deviation penalty factor is set to measure the system operation risk caused by the unit's maintenance time deviating from the fixed maintenance window period. The determination of the window period deviation penalty factor for each unit includes the following steps: Step S21: The window period deviation penalty factor is jointly determined by the rated capacity of the unit and the degree to which its maintenance time deviates from the fixed window period. Units with larger rated capacities have a more significant impact on system operation and therefore should be subject to higher penalties when deviating from the fixed window period; while units with smaller capacities can flexibly adjust their maintenance time within a certain range.

[0038] Step 22: Specifically, the window period deviation penalty factor is shown in equation (1): (1) In the formula: i It is a unit index; P i It is a generator set i Rated capacity; P max It is the maximum rated capacity among all units under maintenance; Φ MTW It is the benchmark penalty factor corresponding to the unit with the maximum rated capacity; γ MTW It is the window period deviation factor.

[0039] Among them, the benchmark penalty factor Φ MTW The decision is made based on a comprehensive assessment of the system's current load level, dispatchability margin, and the importance of unit capacity. This applies when the system load is high or the reserve margin is low. Φ MTW A larger value is chosen to increase the constraint strength of the fixed window period; when the system has ample operational flexibility, its value can be reduced accordingly to enhance the flexibility of the plan. Window period deviation factor γ MTW This reflects the degree of deviation of the unit maintenance plan from the fixed window period, and can be calculated based on the number of deviation periods. t i The mathematical form of this function can be linear or exponential. Its magnitude is directly proportional to or exponentially related to the number of days the unit maintenance time deviates from the fixed window period. For example, in a preferred embodiment, it can be determined according to equation (2): (2) In the formula: k This is an adjustment coefficient used to control the rate at which the penalty increases with the degree of deviation.

[0040] Through the aforementioned penalty factor design, the model can prioritize encouraging units to conduct maintenance within fixed windows, while allowing for moderate offset adjustments within flexible windows when system constraints are tight or maintenance conflicts are frequent. This setting ensures that the model possesses both robustness and safety, as well as flexibility, significantly improving the feasibility of overall maintenance plans in complex power grid scenarios.

[0041] S3. Set an actual maintenance time variable based on the normal maintenance time and minimum maintenance time of the unit to be maintained. The actual maintenance time variable satisfies that it is not less than the minimum maintenance time and not greater than the normal maintenance time. Determine the maintenance time compression penalty factor based on the compression amount of the actual maintenance time variable relative to the normal maintenance time, and generate a maintenance time compression penalty term. In step S3, the normal maintenance time and minimum maintenance time are set for each unit, such as... Figure 3 As shown. Determining the maintenance time for each unit includes the following steps: Step S31: The normal maintenance duration is determined based on the unit's maintenance type, workload, required resources, and procedures. It reflects the standard time required for the unit to complete maintenance tasks under normal conditions. Specifically, it can be adjusted based on historical maintenance data, equipment operating conditions, and maintenance standards, taking into account the average maintenance period of the unit and adjusting for seasonal load changes or resource constraints.

[0042] Step S32, the minimum maintenance duration, is the lower limit for ensuring maintenance quality and safety. It is usually determined based on the shortest completion time of critical processes, equipment safety requirements, and personnel workload. This duration ensures that the basic safety and quality requirements of unit maintenance can still be guaranteed even if the maintenance plan is adjusted or shortened.

[0043] Furthermore, to balance the feasibility and quality of unit maintenance plans, a maintenance time compression penalty factor is introduced. This factor is used to penalize excessive compression of maintenance time during the optimization process, guiding units to prioritize maintenance according to normal maintenance time. At the same time, in situations where system operating resources are limited or multiple units are undergoing concentrated maintenance, units are allowed to moderately compress maintenance time to ensure the overall scheduling feasibility of the system.

[0044] The determination of the maintenance time reduction penalty factor for each unit includes the following steps: Step S33: The maintenance time compression penalty factor is determined jointly by the normal maintenance time of each unit and the compression ratio. For units with shorter normal maintenance times, their maintenance tasks are usually more compact, with denser process connections, and time compression has a more significant impact on maintenance quality and safety; therefore, under the same time compression conditions, units with shorter schedules should be subject to a higher penalty. Conversely, units with longer normal maintenance times have relatively more leeway in terms of schedule and resource allocation, and can appropriately compress maintenance time within a certain range to improve the flexibility of the overall plan.

[0045] Step 34. Specifically, the maintenance time compression penalty factor is shown in equation (3): (3) In the formula: T i MTND It is a generator set i Normal maintenance time; T max It is the longest maintenance time among all maintenance units; Φ MTCD It is the baseline penalty factor corresponding to the unit with the longest maintenance time; γ i MTCD It is the maintenance time compression coefficient, which is a monotonically increasing function of the maintenance time compression ratio.

[0046] Among them, the benchmark penalty factor Φ MTCD The determination is based on a comprehensive consideration of the unit's maintenance type, maintenance resource occupancy rate, and system dispatchability margin. When a certain type of unit has a high maintenance resource occupancy rate or a low system reserve margin... Φ MTCD A larger value is chosen to strengthen the constraint on reducing maintenance time; when the system has ample operational flexibility or sufficient unit maintenance resources, the value can be appropriately reduced. Φ MTCD The value of γ is chosen to improve the flexibility of the maintenance plan. Maintenance time compression factor γ i MTCD This reflects the reduction ratio of unit maintenance time to normal maintenance time, and its magnitude increases monotonically as the reduction ratio increases. i MTCD Linear functions, exponential functions, or piecewise functions can be used to flexibly describe the rate at which the penalty increases with the degree of compression.

[0047] In a preferred embodiment, the maintenance time compression factor γ can be increased. MTCD Taken in exponential form, as shown in equation (4): (4) In equation (4): t iMTFD It is a generator set i The actual number of maintenance compression periods; T i MTCD It is a generator set i Normal maintenance time requirements; a and k This is an adjustable parameter used to control the magnitude and rate of penalty amplification. To balance solution efficiency, it is preferable to approximate the above exponential function using a piecewise linear function, and to use a breakpoint-based piecewise linear approximation during optimization. This allows for implementation within the mixed-integer linear programming framework using special ordered set (SOS2) constraints or interpolation weights.

[0048] S4. Construct a joint optimization model, which includes a set of decision variables related to unit operating status, unit output, unit maintenance schedule, and maintenance time compression; set a comprehensive objective function, which includes at least the grid operating cost, the window period deviation penalty term, and the maintenance time compression penalty term; add unit combination constraints and maintenance constraints to the joint optimization model; solve the joint optimization model and output the maintenance start and end times and corresponding maintenance times of each unit as the maintenance plan optimization results.

[0049] In step S4, the objective function of the joint optimization model for generator maintenance plan is to minimize grid operating cost, window period deviation penalty, and maintenance time compression penalty, as shown in equation (5): (5) In the formula: t It is a time index variable. NT This represents the number of time periods within a scheduling cycle; i It is a unit index. NG Indicates the number of thermal power units in the system; C op It is the cost of power grid operation; P win It is a penalty for deviating from the maintenance window period; P dur It is a penalty for the unit's compression maintenance time; ω 1, ω 2, ω 3 is the weighting coefficient.

[0050] Step S41, the power grid operating cost is shown in the following formula: (6) (7) (8) (9) In the formula:I it This is a Boolean variable representing the unit. i In the t The start / stop status of the segment. I it = 0 indicates that the unit is in a stopped state. I it = 1 indicates that the unit is in the powered-on state; π c This refers to the unit price of coal. It is a generator set Fuel consumption is generally considered to be a quadratic function, as shown in equation (7); It is a generator set In the Output power during a given time period; , , These are the coefficients of the unit cost function; SU i , SD i They represent the generating units. i The startup and shutdown costs are shown in equations (8) and (9); This is a Boolean variable representing the unit. In the Startup status of the time period Indicates the unit exist It is always in a shutdown state. It is always powered on; A Boolean variable, representing the unit. In the The shutdown status during a certain period of time. Indicates the unit exist It is always powered on. The machine is always in a stopped state; and They are the generator sets The cost of starting and stopping a single time.

[0051] Step S42, the penalty for deviation from the maintenance window period is as shown in equation (10): (10) In the formula: y it This is a Boolean variable representing the unit. i During the period t The unit is under maintenance. i During the period t If it is under maintenance, the value is 1; otherwise, the value is 0. λ idMTW It is a generator set i During the period t The maintenance window period deviation penalty factor is 0 in the fixed window period and not 0 in the flexible window period.

[0052] Step S43 The maintenance time compression penalty is shown in equation (11): (11) In the formula: t i MTCD It is a generator set i The actual number of maintenance compression periods; λ i MTCD It is a generator set i The penalty factor for compressed maintenance time.

[0053] Furthermore, in step 4, the unit combination constraints include unit start-up and shutdown constraints, unit minimum start-up and shutdown time constraints, active power balance constraints, upper and lower limits of conventional unit output constraints, conventional unit ramping constraints, and power grid flow safety constraints.

[0054] Step S44, Unit start-up and shutdown constraints: (12) (13) (14) Start-stop status Startup status Shutdown status The three variables need to satisfy equations (12)-(14).

[0055] Step S45, Minimum start-up and shutdown time constraints for the unit: (15) (16) In the formula: T i on It is a generator set i Minimum boot time, T i off It is a generator set i Minimum shutdown time. This constraint is a forward-looking approach. The start-stop state variables of the unit must satisfy the constraints for a continuous number of time periods after the unit starts or stops, as shown in equations (15) and (16).

[0056] Step S46, Active power balance constraint: (17) In the formula: j It is a wind farm index variable. NW Indicates the number of wind farms in the system. P W, jt Indicates wind farm j The predicted value; b It is a power grid node index variable. NB Indicates the number of nodes in the system. P D, bt Represents a node b Load forecast.

[0057] Step S47, Upper and lower limits of conventional unit output constraints: (18) In the formula: P Gi, min and P Gi, max They are conventional units i The minimum and maximum output.

[0058] Step S48, Conventional Unit Rate-Climbing Constraints: (19) (20) In the formula: RU i and RD i These are the upward and downward ramp constraints for the unit during operation. RSU i For the ramp-up constraints during the unit's startup process. RSD i This refers to the ramp-up constraint for the unit during shutdown.

[0059] Step S49, Power Flow Security Constraints: (twenty one) In the formula: l It is a power grid line index variable. PL l,max It is the upper limit of line power flow. G It is the line power flow transfer matrix.

[0060] Furthermore, in step 5, the maintenance constraints include allowable maintenance constraints, area power supply constraints, simultaneous maintenance constraints, maintenance continuity constraints, maintenance mutual exclusion constraints, maintenance duration constraints, and maintenance duration constraints.

[0061] Step S50, Allowing maintenance constraints: (twenty two) In the formula: I it It is a generator set i In the t The start / stop status of the segment. y it It is a generator set i In the t The maintenance status of the section.

[0062] Step S51, Regional Power Supply Constraints: If a region requires that there be no power outages during holidays or special meetings, then the equipment in that region cannot be maintained during these designated periods, as shown in Equation (22).

[0063] (twenty three) In the formula: Ω r It is a region r A collection of maintenance plans corresponding to the equipment; Ω r per For the region r The power supply protection zone.

[0064] Step S52, Simultaneously inspect constraints: To avoid repeated power outages, some maintenance equipment must meet the constraint of simultaneous maintenance, as shown in equation (23).

[0065] (twenty four) In the formula: Ω i It is a set of simultaneous maintenance plans, where units within the same simultaneous maintenance constraint need to be maintained simultaneously.

[0066] Step S53, Maintenance continuity constraint: (25) (26) (27) (28) (29) (30) (31) (32) (33) In the formula: Formula (25) is the maintenance continuity constraint when the unit maintenance time is fixed. y it and yi(t-1) It is the first i The maintenance status of each unit on two consecutive days is a Boolean variable, where 0 indicates that it is not under maintenance and 1 indicates that it is under maintenance. T This represents the total number of periods for overall maintenance and optimization. T i It is the first i Maintenance time requirements for each generating unit; Indicates the first i The last period of the maintenance plan for each unit. The value is determined by t Confirmed. When the unit maintenance time is compressible, the unit... i The last day of maintenance is indicated as Then the maintenance continuity constraint becomes (26), because the unit i The actual maintenance time is unknown before the solution is obtained. The value cannot be determined solely by t OK. To solve this problem, two variables are introduced. and They represent the generating units. i The start and end states of the maintenance are Boolean variables. α id The value is 1 at the start of maintenance, and 0 otherwise. β id The value is 1 when the maintenance is completed, otherwise it is 0. , and The relationships are shown in equations (27)-(29); equations (30) and (31) ensure that the unit i Each unit has only one start and one end state for maintenance, thus ensuring the continuity of unit maintenance. Ω n For the set of units that do not have maintenance tasks, equations (32) and (33) avoid the occurrence of non-zero solutions for the maintenance start and end state variables. In summary, the unit maintenance continuity constraints are equations (27)-(33).

[0067] Step S54: Inspect mutual exclusion constraints: The simultaneous power outage of certain maintenance equipment can cause other equipment to overload, so mutual exclusion constraints need to be met, as shown in equation (34).

[0068] (34) In the formula: Ω j A collection of units that cannot be repaired simultaneously; units under maintenance. s 1 and s 2 belongs to the maintenance mutual exclusion set Ω jA simultaneous power outage of certain maintenance equipment can cause overload on other equipment, therefore mutual exclusion constraints must be met.

[0069] Step S55, Maintenance time constraint: (35) (36) (37) (38) (39) In the formula: Formula (35) is the maintenance time constraint when the unit maintenance time is fixed. When the unit maintenance time can be compressed, Formula (35) cannot accurately describe the maintenance time constraint. T i MTRD For the unit i Minimum maintenance time; T i MTND For the unit i Normal maintenance time; t i MTCD For the unit i The optimized actual compressed maintenance time. The actual maintenance time for each unit shall not be less than the minimum maintenance time, as shown in equation (36); Unit i The optimized and actually compressed maintenance time is shown in equation (37), and is expressed by constraints as equations (38) and (39). In summary, the maintenance time constraints are equations (36), (38), and (39).

[0070] Step S56, Daily maintenance quantity limit constraint: Considering grid security and grid carrying capacity, it is necessary to limit the number of generating units that need to be maintained daily in order to ensure the normal and stable operation of the grid, as shown in equation (40): (40) In the formula: N t LMT Indicates time period t The daily maintenance limit for the number of units.

[0071] By adopting the above scheme, the unit maintenance plan optimization method of the present invention, which considers flexible windows, realizes the adjustable arrangement of maintenance tasks on the time axis by setting flexible windows on both sides of the fixed maintenance window period, reducing the scheduling conflicts caused by overlapping or concentrated maintenance of fixed windows; by constructing a window period deviation penalty term based on the rated capacity and deviation degree of the unit, it realizes the quantitative constraint and priority guidance of maintenance deviation, reducing the repeated manual coordination of window periods during the planning process; by introducing a compressible actual maintenance time variable that meets the minimum maintenance time constraint, it realizes the adaptive adjustment of maintenance duration under limited conditions, alleviating the infeasibility problem caused by the coupling of maintenance resources and system operation boundary; furthermore, by incorporating the grid operation cost, window period deviation penalty term, and maintenance time compression penalty term into the comprehensive objective function and solving them jointly under unit combination constraints and maintenance constraints, it realizes the integrated optimization output of maintenance plan and operation mode, significantly improving the executability and feasibility of maintenance plan under safety constraints.

[0072] In some embodiments of the present invention, the flexible window period is obtained by extending the fixed maintenance window period by a preset duration before the start time and after the end time, respectively, and the preset duration is set according to the system operating conditions.

[0073] In this embodiment, the IEEE-118 bus system is used as the test object, with a planning period of one year, a daily time granularity of one hour, and 54 units as the research objects. Each unit has a fixed maintenance window within the planning period. A flexible window and a maintenance duration compression mechanism are introduced, and both types of penalty terms are incorporated into the joint optimization objective to obtain the maintenance schedule and day-ahead scheduling plan. The fixed maintenance window for each unit is determined based on the number of units to be maintained, maintenance duration, maintenance resource occupancy, and system operating cycle. A preliminary maintenance schedule can be developed using historical maintenance data or expert experience to ensure that the number of units to be maintained within the same time period does not exceed the system's capacity.

[0074] In this embodiment, a flexible window period is set, located on either side of the fixed window period, with a preset length or dynamically adjusted according to system operating conditions. The length of the flexible window period is widened when system load is low or available spare capacity is sufficient, and shortened when system operation is strained. For example, the fixed window period is extended by 15 days on each side, forming a total flexible window period of 45 days. The flexible window period provides time flexibility for overall maintenance feasibility, allowing for adjustments to maintenance timing during concentrated unit maintenance or when system resources are strained.

[0075] In some embodiments of the present invention, the window period deviation penalty factor is obtained by combining the unit rated capacity normalization coefficient, the benchmark penalty factor, and the window period deviation factor, wherein the unit rated capacity normalization coefficient is determined by the unit rated capacity and the maximum rated capacity of the unit to be maintained.

[0076] In this embodiment, the window period deviation penalty factor is used to measure the system operation risk caused by the unit's maintenance time deviating from the fixed maintenance window period. The penalty factor is jointly determined by the unit's rated capacity and the degree to which its maintenance time deviates from the fixed window period. Units with larger rated capacity are subject to higher penalties when they deviate from the fixed window period, while units with smaller capacity are allowed to adjust flexibly within a certain range.

[0077] In some embodiments of the present invention, the benchmark penalty factor is set based on the current system load level and dispatch margin, the window period deviation factor is determined based on the number of deviation periods of the maintenance schedule from the fixed maintenance window period, and the mathematical form of the window period deviation factor adopts a linear function or an exponential function.

[0078] In this embodiment, the window period deviation penalty factor is as shown in Equation (1) above.

[0079] In some embodiments of the present invention, the window period deviation factor is an exponential function with the number of deviation periods as the independent variable, the exponential function including an adjustment coefficient, the adjustment coefficient representing the rate at which the penalty increases with the degree of deviation.

[0080] In this embodiment, the window period deviation factor is used to quantify the degree of deviation of the maintenance time from the fixed window period. The degree of deviation can be represented by the number of deviation periods / number of deviation days, and the mathematical form can be a linear function or an exponential function. In the preferred embodiment, an exponential form is used, as shown in equation (2) above.

[0081] The above settings guide maintenance within fixed time windows, while allowing for appropriate adjustments using flexible time windows when system constraints are complex or maintenance conflicts are frequent.

[0082] In some embodiments of the present invention, the normal maintenance duration is determined based on the unit maintenance type, maintenance workload, maintenance resources, and maintenance procedure arrangement, while the minimum maintenance duration is determined based on the shortest completion time of key procedures, equipment safety requirements, and personnel workload.

[0083] In this embodiment, normal maintenance time and minimum maintenance time are set for each unit: 1) The normal maintenance duration is determined based on the type of unit maintenance, maintenance workload, required maintenance resources and maintenance procedure arrangement. The average maintenance period can be calculated by combining historical maintenance records, equipment operation status assessment results and maintenance specifications, and can be fine-tuned according to seasonal load changes or resource constraints.

[0084] 2) The minimum maintenance time is the lower limit of maintenance time to ensure maintenance quality and safety. It is determined based on the shortest completion time of key processes, equipment safety requirements, and personnel workload.

[0085] In a preferred embodiment, the method for setting the normal maintenance time and minimum maintenance time of the unit includes the following steps: Determining the Normal Maintenance Duration: The normal maintenance duration characterizes the standard time required for the unit to complete all maintenance work under normal conditions. In this embodiment, the normal maintenance duration is determined comprehensively based on the unit's maintenance category, workload, required resources, and work order arrangement. Specifically, it can be based on historical maintenance records, equipment operating status assessment results, and the average maintenance duration of relevant maintenance specifications, and appropriately adjusted according to seasonal load changes, available resources for the maintenance team, and construction conditions to obtain a normal maintenance duration that meets the actual needs of the project.

[0086] Determining the minimum maintenance duration: The minimum maintenance duration is the lowest feasible timeframe that the unit must meet to complete maintenance work, ensuring maintenance quality and safety requirements. In this embodiment, the minimum maintenance duration is determined based on the shortest completion time of critical processes, equipment safety operation requirements, and personnel workload. This duration ensures that even when maintenance plans need to be compressed, parallel processes need to be optimized, or resource allocation is tight, the unit can still complete the necessary work within the minimum timeframe without compromising maintenance quality.

[0087] The combined use of normal maintenance duration and minimum maintenance duration: Through the above settings, the coordination of two types of schedules can be achieved during the overall maintenance plan optimization process. Normal maintenance duration guides maintenance arrangements to maintain project rationality and resource balance as much as possible, while minimum maintenance duration provides boundary constraints for maintenance compression. This allows the system to obtain feasible solutions by appropriately compressing the maintenance duration of some units, even in scenarios with many maintenance conflicts, resource constraints, or critical time windows. For example, a unit's normal maintenance duration is 12 days, and its minimum maintenance duration is 10 days.

[0088] In some embodiments of the present invention, the maintenance time compression penalty factor is obtained by combining the normal maintenance time normalization coefficient, the benchmark penalty factor, and the maintenance time compression coefficient. The normal maintenance time normalization coefficient is determined by the normal maintenance time and the maximum normal maintenance time of the unit to be maintained. The benchmark penalty factor is set according to the unit maintenance type, maintenance resource occupancy rate, and system dispatch margin. The maintenance time compression coefficient is a monotonically increasing function of the maintenance time compression ratio.

[0089] In a preferred embodiment, the determination of the maintenance time compression penalty factor includes the following steps: 1) Determining the penalty intensity based on normal maintenance time and compression ratio: In this embodiment, the maintenance time compression penalty factor is determined by the unit's normal maintenance time and its compression ratio. For units with shorter normal maintenance times, their maintenance tasks are usually compact with dense work connections, and the impact of maintenance time compression on maintenance quality and operational safety is more significant. Therefore, under the same compression ratio or number of compression periods, a higher penalty intensity is applied to such units. Conversely, units with longer normal maintenance times have more room for schedule planning and resource allocation, allowing for moderate compression of maintenance time within a certain range to improve the flexibility of the overall maintenance plan.

[0090] 2) Determine the mathematical expression of the maintenance time penalty factor: In this embodiment, the maintenance time compression penalty factor can be determined according to the above-mentioned formula (3).

[0091] In some embodiments of the present invention, the maintenance time compression coefficient adopts an exponential function; when performing a piecewise linear approximation of the exponential function, the compression ratio interval is divided into several continuous sub-intervals, an interpolation weight variable is introduced, and a second type of special ordered set SOS2 constraint is applied to the interpolation weight variable.

[0092] The maintenance time compression coefficient γ i MTCD Using exponential form, its mathematical expression is shown in equation (4) above.

[0093] In a preferred embodiment, in order to achieve efficient solution of the exponential function under the framework of mixed integer linear programming, it is approximated by a piecewise linear function.

[0094] Specifically, firstly, based on the range of values ​​for the maintenance time reduction ratio, the reduction ratio interval [0, ... r max The function is divided into several continuous sub-intervals, and the corresponding exponential function value is calculated at the endpoints of each segment, serving as the node value of the piecewise linear function. The number of segments can be determined based on the accuracy requirements and computational scale, preferably 2–4 segments, to balance model accuracy and solution efficiency.

[0095] In the optimization model, interpolation weight variables are introduced and subjected to Special Ordered Sets of Two Types (SOS2) constraints, ensuring that at any given time, only two adjacent piecewise nodes have non-zero weight variables, thus achieving a linear interpolation approximation of the exponential penalty function. This approach accurately characterizes the nonlinear growth relationship between the degree of maintenance time compression and the intensity of the penalty without introducing nonlinear constraints. Using this piecewise linear approximation and SOS2 constraints, the maintenance time compression penalty term can be seamlessly embedded into a maintenance plan optimization model based on mixed-integer linear programming. This ensures solution stability and computational efficiency while fully reflecting the constraint effect of maintenance time compression on maintenance quality and system safety. This embodiment uses a piecewise linear function approximation of the exponential function, achieving an efficient solution for the exponential penalty within the framework of mixed-integer linear programming. The interpolation weight variables are constrained by Special Ordered Sets of Two Types (SOS2) to achieve linear interpolation approximation, balancing model accuracy and solution efficiency.

[0096] In some embodiments of the present invention, the maintenance constraints include at least maintenance continuity constraints, maintenance mutual exclusion constraints, maintenance duration constraints, and daily maintenance quantity limit constraints. The maintenance continuity constraints introduce maintenance start state variables and maintenance end state variables. The maintenance start state variable takes a first preset value during the maintenance start period, and the maintenance end state variable takes a first preset value during the maintenance end period.

[0097] After defining the maintenance window period, the window period deviation penalty factor, and the maintenance time compression penalty factor, a joint optimization model for the coordinated optimization of generator maintenance plan and unit combination is constructed, and the optimal maintenance plan scheme is obtained by solving the model. The joint optimization model aims to minimize the overall system operating cost. Its objective function considers the grid operating cost, the maintenance window period deviation penalty, and the maintenance time compression penalty simultaneously, as shown in Equation (5) above. The grid operating cost is used to reflect the fuel consumption and start-up and shutdown costs of the generator set under different start-up and shutdown states; the window period deviation penalty is used to constrain the unit maintenance plan to fall within the fixed maintenance window period as much as possible; and the maintenance time compression penalty is used to suppress the excessive compression of the unit maintenance period. By setting weight coefficients, the above three types of objectives are comprehensively balanced.

[0098] In this embodiment, the grid operating cost consists of fuel cost and unit start-up and shutdown cost. The fuel cost is calculated based on the unit output level and the corresponding fuel consumption function, while the start-up and shutdown cost is characterized by unit start-up state variables and shutdown state variables, as shown in equations (6) to (9) above.

[0099] The maintenance window deviation penalty is implemented by weighted summation of the unit's maintenance status in each time period, as shown in Equation (10). When the unit maintenance occurs within a fixed window period, the penalty term is zero; when the maintenance occurs within a flexible window period, a penalty is applied according to a pre-set window deviation penalty factor.

[0100] S5-4. Maintenance time compression penalty: The actual number of compressed maintenance periods of the unit is modeled as shown in Equation (11), which is used to reflect the impact of maintenance time compression on maintenance quality and system safety.

[0101] To ensure that the optimization results meet the requirements of safe operation of the power grid, the model also introduces unit combination operation constraints, including unit start-stop logic constraints, minimum start-stop time constraints, active power balance constraints, upper and lower limits of conventional unit output constraints, ramping constraints, and power grid flow safety constraints, as shown in equations (12) to (21) above.

[0102] Based on this, multiple constraints related to the maintenance plan are further introduced, including allowable maintenance constraints, regional power supply constraints, simultaneous maintenance constraints, maintenance continuity constraints, maintenance mutual exclusion constraints, maintenance duration constraints, and daily maintenance quantity limits, as shown in equations (22) to (40) above. Among them, the maintenance continuity constraint ensures that the unit maintenance process is continuous in time and occurs only once by introducing maintenance start state variables and maintenance end state variables; the maintenance duration constraint ensures that the actual maintenance duration of the unit is not less than the minimum maintenance duration requirement while allowing for the compression of the maintenance period.

[0103] By jointly modeling the above objective function and constraints, unit operation scheduling decisions and maintenance plans can be optimized in a unified manner, and the solution can be obtained by commercial optimization solvers. Thus, under the premise of meeting the system's safe operation and maintenance engineering constraints, the optimal maintenance plan scheme that takes into account the operating economy, maintenance feasibility and scheduling flexibility can be obtained.

[0104] In this embodiment, the objective function of the joint optimization model is to minimize the power grid operating cost, the maintenance window deviation penalty, and the maintenance time compression penalty; the objective function includes weight coefficients ω1, ω2, and ω3 to comprehensively balance the three types of objectives.

[0105] Among them, the power grid operation cost consists of fuel cost and start-up and shutdown cost. Fuel consumption can be expressed as a quadratic function, and start-up and shutdown cost is characterized by start-up state variables and shutdown state variables. The maintenance window deviation penalty is based on the weighted summation of maintenance state variables, and the maintenance time compression penalty is modeled based on the actual number of compression periods.

[0106] In this embodiment, the joint optimization model introduces unit combination constraints and maintenance constraints.

[0107] Unit combination constraints include: unit start-up and shutdown constraints, unit minimum start-up and shutdown time constraints, active power balance constraints, upper and lower limits of conventional unit output constraints, conventional unit ramping constraints, and power grid flow safety constraints.

[0108] Maintenance constraints include: permissible maintenance constraints, regional power supply constraints, simultaneous maintenance constraints, maintenance continuity constraints, maintenance mutual exclusion constraints, maintenance duration constraints, and daily maintenance quantity limits.

[0109] Based on the objective function and constraints described above, the joint optimization model is solved, and the start and end times of maintenance for each unit and the corresponding maintenance duration are output as the optimization results of the maintenance plan.

[0110] Compared with existing technologies, the advantages of this invention lie in setting a flexible window period outside the fixed maintenance window period and introducing a window period deviation penalty factor. This allows generating units to prioritize maintenance during the optimization process within the fixed window period. When there are maintenance resource conflicts or operational constraints in the system, a feasible solution can be obtained by moderately deviating from the fixed window period. Simultaneously, the model sets the normal maintenance duration and minimum maintenance duration for the generating units and introduces a maintenance duration compression penalty factor. This allows for moderate compression of maintenance duration while ensuring maintenance quality and safety, thereby maintaining the flexibility and coordination of the dispatching scheme in complex power grid environments. Through the above design, this invention can effectively improve the feasibility and robustness of maintenance plans under multiple constraints and multi-unit coupling conditions, providing support for the safe and economical operation of the power grid.

[0111] Figure 4 This is a schematic diagram of a unit maintenance plan optimization system that takes into account the flexible window provided in an embodiment of the present invention.

[0112] Example 2, as Figure 4 As shown, the present invention also provides a unit maintenance plan optimization system that considers flexible windows, including: a maintenance information acquisition module S11, a window period generation module S12, a penalty term generation module S13, and an optimization solution module S14.

[0113] The maintenance information acquisition module is used to acquire maintenance information of the units to be maintained. The window period generation module is used to determine the fixed maintenance window period corresponding to each unit to be maintained based on the maintenance information of the unit to be maintained, and to set flexible window periods on both sides of the fixed maintenance window period. The penalty item generation module is used to determine a window period deviation penalty factor based on the rated capacity of the unit to be maintained and the degree of deviation of the maintenance schedule from the fixed maintenance window period, and to generate a window period deviation penalty item; the penalty item generation module is also used to set an actual maintenance time variable based on the normal maintenance time and the minimum maintenance time of the unit to be maintained, wherein the actual maintenance time variable is not less than the minimum maintenance time and not greater than the normal maintenance time, and to determine a maintenance time compression penalty factor based on the compression amount of the actual maintenance time variable relative to the normal maintenance time, and to generate a maintenance time compression penalty item; The optimization solution module is used to construct a joint optimization model, which includes a set of decision variables related to unit operating status, unit output, unit maintenance schedule, and maintenance time compression. A comprehensive objective function is defined, which includes at least the grid operating cost, the window period deviation penalty term, and the maintenance time compression penalty term. Unit combination constraints and maintenance constraints are added to the joint optimization model. The joint optimization model is solved, and the start and end times of maintenance for each unit and the corresponding maintenance duration are output as the maintenance plan optimization results.

[0114] Example 3: The present invention also provides a unit maintenance plan optimization device that considers flexible windows. The device includes a computer device, which includes a processor and a memory. The processor stores computer instructions. When the computer instructions are executed, the device implements the unit maintenance plan optimization method that considers flexible windows.

[0115] Example 4, as Figure 5 As shown, the present invention also provides an electronic device 100 for implementing a method for optimizing unit maintenance plans that takes into account flexible windows.

[0116] The electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on at least one processor 102, and at least one communication bus 104.

[0117] The memory 101 can be used to store the computer program 103. The processor 102 implements the steps of the unit maintenance plan optimization method considering the flexible window described in the first aspect of the present invention by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101.

[0118] The memory 101 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0119] At least one processor 102 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 102 may be a microprocessor or any conventional processor. Processor 102 is the control center of electronic device 100, connecting various parts of electronic device 100 via various interfaces and lines.

[0120] The memory 101 in the electronic device 100 stores multiple instructions to implement a unit maintenance plan optimization method that takes into account flexible windows, and the processor 102 can execute multiple instructions to achieve the following: In response to the input maintenance information of the unit to be maintained, a fixed maintenance window period corresponding to each unit to be maintained is determined, and flexible window periods are set on both sides of the fixed maintenance window period. Based on the rated capacity of the unit to be maintained and the degree of deviation of the maintenance schedule from the fixed maintenance window period, a window period deviation penalty factor is determined, and a window period deviation penalty item is generated. An actual maintenance time variable is set based on the normal maintenance time and minimum maintenance time of the unit to be maintained. The actual maintenance time variable is not less than the minimum maintenance time and not greater than the normal maintenance time. A maintenance time compression penalty factor is determined based on the compression amount of the actual maintenance time variable relative to the normal maintenance time, and a maintenance time compression penalty term is generated. A joint optimization model is constructed, which includes a set of decision variables related to unit operating status, unit output, unit maintenance schedule, and maintenance time compression. A comprehensive objective function is set, which includes at least the grid operating cost, the window period deviation penalty term, and the maintenance time compression penalty term. Unit combination constraints and maintenance constraints are added to the joint optimization model. The joint optimization model is solved, and the maintenance start and end times and corresponding maintenance times of each unit are output as the maintenance plan optimization results.

[0121] Example 5: If the modules / units integrated in the electronic device 100 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, and read-only memory (ROM).

[0122] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0123] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0124] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0125] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0126] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0127] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for optimizing unit maintenance plans considering flexible windows, characterized in that, The method includes: In response to the input maintenance information of the unit to be maintained, a fixed maintenance window period corresponding to each unit to be maintained is determined, and flexible window periods are set on both sides of the fixed maintenance window period. Based on the rated capacity of the unit to be maintained and the degree of deviation of the maintenance schedule to be optimized from the fixed maintenance window period, the window period deviation penalty factor is determined and the window period deviation penalty term is generated. An actual maintenance time variable is set based on the normal maintenance time and minimum maintenance time of the unit to be maintained. The actual maintenance time variable is not less than the minimum maintenance time and not greater than the normal maintenance time. A maintenance time compression penalty factor is determined based on the compression amount of the actual maintenance time variable relative to the normal maintenance time, and a maintenance time compression penalty term is generated. A joint optimization model is constructed, which includes a set of decision variables related to unit operating status, unit output, unit maintenance schedule, and maintenance time compression. A comprehensive objective function of the joint optimization model is generated based on grid operating costs, the window period deviation penalty term, and the maintenance time compression penalty term. Unit combination constraints and maintenance constraints are added to the joint optimization model. The joint optimization model is solved, and the maintenance start and end times and corresponding maintenance times of each unit are output as the maintenance plan optimization results.

2. The method for optimizing unit maintenance plans considering flexible windows according to claim 1, characterized in that, The flexible window period is obtained by extending the fixed maintenance window period by a preset duration before the start time and after the end time, respectively. The preset duration is set according to the system operating conditions.

3. The method for optimizing unit maintenance plans considering flexible windows according to claim 1, characterized in that, The window period deviation penalty factor is obtained by combining the unit rated capacity normalization coefficient, the benchmark penalty factor, and the window period deviation factor. The unit rated capacity normalization coefficient is determined by the unit rated capacity and the maximum rated capacity of the unit to be maintained.

4. The method for optimizing unit maintenance plans considering flexible windows according to claim 3, characterized in that, The benchmark penalty factor is set based on the current system load level and dispatch margin. The window period deviation factor is determined based on the number of deviation periods of the maintenance schedule from the fixed maintenance window period. The mathematical form of the window period deviation factor adopts a linear function or an exponential function.

5. The method for optimizing unit maintenance plans considering flexible windows according to claim 4, characterized in that, The window period deviation factor adopts an exponential function with the number of deviation periods as the independent variable. The exponential function includes an adjustment coefficient, which represents the rate at which the penalty increases with the degree of deviation.

6. The method for optimizing unit maintenance plans considering flexible windows according to claim 1, characterized in that, The normal maintenance duration is determined based on the unit maintenance type, maintenance workload, maintenance resources, and maintenance procedure arrangement, while the minimum maintenance duration is determined based on the shortest completion time of key procedures, equipment safety requirements, and personnel workload.

7. The method for optimizing unit maintenance plans considering flexible windows according to claim 1, characterized in that, The maintenance time compression penalty factor is obtained by combining the normal maintenance time normalization coefficient, the benchmark penalty factor, and the maintenance time compression coefficient. The normal maintenance time normalization coefficient is determined by the normal maintenance time and the maximum normal maintenance time of the unit to be maintained. The benchmark penalty factor is set according to the unit maintenance type, maintenance resource occupancy rate, and system dispatch margin. The maintenance time compression coefficient is a monotonically increasing function of the maintenance time compression ratio.

8. The method for optimizing unit maintenance plans considering flexible windows according to claim 7, characterized in that, The maintenance time compression coefficient adopts an exponential function; when performing a piecewise linear approximation of the exponential function, the compression ratio interval is divided into several continuous sub-intervals, an interpolation weight variable is introduced, and a second type of special ordered set SOS2 constraint is applied to the interpolation weight variable.

9. The method for optimizing unit maintenance plans considering flexible windows according to claim 1, characterized in that, The maintenance constraints include at least maintenance continuity constraints, maintenance mutual exclusion constraints, maintenance duration constraints, and daily maintenance quantity limits. The maintenance continuity constraints introduce maintenance start state variables and maintenance end state variables. The maintenance start state variable takes a first preset value during the maintenance start period, and the maintenance end state variable takes a first preset value during the maintenance end period.

10. A unit maintenance plan optimization device considering flexible windows, characterized in that, The apparatus includes a computer device, which includes a processor and a memory. The processor stores computer instructions, and when the computer instructions are executed, the apparatus implements the unit maintenance plan optimization method considering flexible windows as described in any one of claims 1 to 9.

Images