Optimization method, device, storage medium and computing equipment for substation maintenance decision

Through the substation maintenance optimization model based on Markov decision-making process, the problem of insufficient response to potential state information in the state maintenance decision-making of traditional power system is solved, and a more flexible and efficient maintenance plan is achieved, which improves the operational reliability of the power system. .

CN114493041BActive Publication Date: 2024-11-05STATE GRID JIANGSU ELECTRIC POWER CO LTD RESEARCH INSTITUTE +3
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210156485.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-02-21
Publication Date
2024-11-05
Estimated Expiration
2042-02-21

AI Technical Summary

Technical Problem

The traditional power system status maintenance decision model lacks a response strategy for potential status information during the research cycle, resulting in inadequate maintenance plans and ineffective in optimizing the economic and safety of the overall operation of the power grid.

Method used

A substation maintenance decision optimization model based on Markov's decision-making process is adopted, and an incentive mechanism is established through grid state transfer and maintenance decision-making, and the maintenance plan is optimized to minimize power outages, maximize user satisfaction and optimize equipment resource allocation.

Benefits of technology

It has achieved more accurate and effective maintenance decisions, improved the asset management and operation reliability of power system equipment, and met the economic and safety needs of the overall operation of the power grid.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114493041B_ABST
    Figure CN114493041B_ABST
Patent Text Reader

Abstract

The present invention discloses an optimization method, device, storage medium and computing equipment for substation maintenance decision. The method constructs a decision optimization objective function based on power grid safety constraints, maintenance relationship constraints, maintenance resource constraints and maintenance time constraints, including minimizing maintenance outage power, maximizing user power outage satisfaction and optimizing maintenance resource configuration of various types of equipment; establishing a substation maintenance decision optimization model based on Markov decision process; finally, based on a given Markov decision process, by maximizing the behavior value function and maximizing the state value function, the substation maintenance decision optimization model based on the Markov decision process is solved to obtain the optimal substation maintenance decision arrangement. The present invention provides substation maintenance decision optimization support for power system managers, which is of great significance to ensuring the economy and rationality of substation maintenance work, lean management of power grids and ensuring stable operation of power systems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an optimization method, device, storable medium and computing equipment for substation maintenance decision-making, belonging to the technical field of power maintenance. Background Art

[0002] With the continuous development of society and the economy, the content and methods of power engineering substation maintenance have changed. With the maturity of equipment condition monitoring, diagnosis, and prediction technologies, the traditional planned maintenance and outage maintenance has gradually evolved to maintenance based on equipment condition. Decisions are made based on both individual equipment and system risks (failure risk and maintenance risk), thus making condition-based maintenance strategies no longer solely based on individual equipment behavior. Substation equipment maintenance decisions must consider the impact of maintenance decisions on the economic and safety of the overall power grid operation. Traditional power system condition-based maintenance decision-making models use actual state information at the beginning of a study cycle and predicted equipment state changes during the study cycle to optimize decisions and obtain a fixed maintenance plan. However, these models lack strategies for responding to potential new state information during the study cycle. Summary of the Invention

[0003] The purpose of the present invention is to provide a method, device, storage medium and computing equipment for optimizing substation maintenance decision-making, with the constraints of minimizing the amount of maintenance power outages, maximizing user power outage satisfaction and optimizing the allocation of maintenance resources for various types of equipment, to solve the substation maintenance decision optimization model based on the Markov decision process, obtain the optimal substation maintenance decision arrangement, and provide substation maintenance decision optimization support for power system managers.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] In one aspect, the present invention provides a method for optimizing substation maintenance decision-making, comprising:

[0006] A substation maintenance decision optimization model based on the Markov decision process is established, with the grid state as the state of the Markov decision process, the substation maintenance decision as the action of the Markov decision process, the grid state transition as the strategy of the Markov decision process, and the grid operation state improvement value obtained after the substation maintenance decision arrangement is taken as the reward of the Markov decision process.

[0007] The substation maintenance decision optimization model based on the Markov decision process is solved to obtain the optimal substation maintenance decision.

[0008] Furthermore, the establishment of a substation maintenance decision optimization model based on a Markov decision process includes:

[0009] M={S,A,P s,a ,R},

[0010] Among them, S represents the grid state set, s∈S, s represents a certain grid state, A represents the substation maintenance decision set, a∈A, a represents a certain substation maintenance decision, P s,a represents the transition model, which expresses the probability of transitioning from grid state s to another grid state when substation maintenance decision a is taken. R is the improvement value of grid operation state obtained after substation maintenance decision is taken. The improvement value of grid operation state is quantitatively evaluated by establishing a corresponding indicator evaluation system.

[0011] The substation maintenance decision arrangement includes maintenance objects, maintenance start time and maintenance end time.

[0012] Furthermore, solving the substation maintenance decision optimization model based on the Markov decision process must meet the constraints of minimizing maintenance outage power, maximizing user outage satisfaction, and optimizing the allocation of maintenance resources for various types of equipment.

[0013] Further,

[0014] The minimum maintenance outage power consumption constraint is as follows:

[0015]

[0016] Among them, F ens Indicates the minimum power consumption for maintenance outage, i=1,...,N C N is the user number. C is the total number of users, Q i Represents the total power outage energy of user i after the power outage plan is implemented;

[0017] The constraint for maximizing user power outage satisfaction is as follows:

[0018]

[0019]

[0020] Among them, F sat represents the maximum value of user satisfaction with power outage, S i represents the satisfaction of user i after the power outage, C CoCi is the power outage time series of user i, Var(C CoCi ) is the variance of power outage time of user i, is the number of power outages for user i within the power outage plan cycle;

[0021] The constraints for optimizing the maintenance resource allocation for each type of equipment are as follows:

[0022]

[0023] Among them, Foptcon Indicates the minimum difference in maintenance time for each type of equipment. L, T, and K represent the types of line, transformer, and switchgear, respectively. Var(E L,T,K,j ) is the power outage time variance of type j equipment.

[0024] Furthermore, the following constraints must be met to calculate the minimum maintenance outage power consumption, maximize user outage satisfaction, and optimize the maintenance resource allocation for each type of equipment:

[0025] A. Grid security constraints, including node voltage constraints and line power flow constraints;

[0026] B. Maintenance relationship constraints, including simultaneous maintenance constraints and mutually exclusive maintenance constraints;

[0027] C. Maintenance resource constraints are expressed as:

[0028]

[0029] Among them, x i (t) is the state of device i at time t, r i,k The demand for material resource k for maintenance of equipment i, r k (t) is the available amount of personnel and material resources at time t, N M is the number of devices;

[0030] D. Maintenance time constraint is expressed as:

[0031] b i ≤x i (t)≤e i ;

[0032] Among them, b i is the earliest time allowed for maintenance of equipment i, e i The latest time allowed for maintenance of equipment i.

[0033] Furthermore, solving the substation maintenance decision optimization model based on the Markov decision process includes:

[0034] Calculate the state value function and action value function based on the rewards obtained after taking the substation maintenance decision;

[0035] With the constraints of minimizing the amount of maintenance outage power, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment, the maximum value of the state value function and the action value function is solved to obtain the optimal substation maintenance decision.

[0036] Further,

[0037] The expected future return of the grid operation state s at time t is used as the state value function;

[0038] The expected future reward of selecting substation maintenance decision a based on the grid operation status s at time t is used as the action value function;

[0039] Among them, the future return that can be obtained after choosing the substation maintenance decision is calculated as:

[0040]

[0041] Among them, U is the future reward that can be obtained after choosing the substation maintenance decision, γ t represents the discount rate at time t, γ t ∈[0,1],R(s t ) represents the improvement value of the grid operation status at time t.

[0042] A second aspect of the present invention provides an optimization device for substation maintenance decision-making, characterized by comprising:

[0043] A module is constructed to establish a Markov decision process-based substation maintenance decision optimization model using the power grid state as the state, the substation maintenance decision as the action, the power grid state transition as the strategy, and the improvement in the power grid operation state obtained after the substation maintenance decision is arranged as the reward of the Markov decision process.

[0044] The decision module is used to solve the substation maintenance decision optimization model based on the Markov decision process to obtain the optimal substation maintenance decision.

[0045] Furthermore, the building block is specifically used to:

[0046] The substation maintenance decision optimization model based on Markov decision process is established as follows:

[0047] M={S,A,P s,a ,R},

[0048] Among them, S represents the grid state set, s∈S, s represents a certain grid state, A represents the substation maintenance decision set, a∈A, a represents a certain substation maintenance decision, P s,a represents the transition model, which expresses the probability of transitioning from grid state s to another grid state when substation maintenance decision a is taken. R is the improvement value of grid operation state obtained after substation maintenance decision is taken. The improvement value of grid operation state is quantitatively evaluated by establishing a corresponding indicator evaluation system.

[0049] The substation maintenance decision arrangement includes maintenance objects, maintenance start time and maintenance end time.

[0050] Furthermore, the building block is also used to:

[0051] The constraints for minimizing maintenance outage power consumption, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment are as follows:

[0052] The constraint for minimizing maintenance outage power consumption is:

[0053]

[0054] Among them, F ens Indicates the minimum power consumption for maintenance, i=1,…,N C N is the user number. C is the total number of users, Q i Represents the total power outage energy of user i after the power outage plan is implemented;

[0055] The constraint to maximize user power outage satisfaction is:

[0056]

[0057]

[0058] Among them, F sat represents the maximum value of user satisfaction with power outage, S i represents the satisfaction of user i after the power outage, C CoCi is the power outage time series of user i, Var(C CoCi ) is the variance of power outage time of user i, is the number of power outages for user i within the power outage plan cycle;

[0059] The optimal resource allocation constraints for each type of equipment maintenance are:

[0060]

[0061] Among them, F optcon Indicates the minimum difference in maintenance time for each type of equipment. L, T, and K represent the types of line, transformer, and switchgear, respectively. Var(E L,T,K,j ) is the power outage time variance of type j equipment.

[0062] Furthermore, the decision module is specifically used to:

[0063] The expected future rewards obtained from the grid operation state s at time t are used as the state value function, and the expected future rewards obtained from selecting the substation maintenance decision a based on the grid operation state s at time t are used as the action value function. Based on the rewards obtained after taking the substation maintenance decision, the state value function and the action value function are calculated.

[0064] With the constraints of minimizing the amount of maintenance outages, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment, the maximum value of the state value function and the action value function is solved to obtain the optimal substation maintenance decision;

[0065] Among them, the future return that can be obtained after choosing the substation maintenance decision is calculated as:

[0066]

[0067] Among them, U is the future reward that can be obtained after choosing the substation maintenance decision, γ t represents the discount rate at time t, γ t ∈[0,1],R(s t ) represents the improvement value of the grid operation status at time t.

[0068] A third aspect of the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods described above.

[0069] A fourth aspect of the present invention provides a computing device, comprising:

[0070] One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any of the methods described above.

[0071] The beneficial effects of the present invention are:

[0072] The technology behind maintenance decision optimization is maturing, becoming a hot topic of research both domestically and internationally. By establishing a mathematical representation of grid maintenance risk and fault risk, and aiming to minimize total risk, a maintenance plan for substation equipment is developed. Based on stochastic optimization theory, a two-stage stochastic optimization model was constructed. The first stage examined maintenance decisions made before the uncertainty (equipment status) within the cycle was realized, while the second stage examined operational scheduling issues after the uncertainty was realized. Both achieved excellent decision optimization results, closely linking decisions to the equipment status change process, making maintenance decisions more accurate and effective, and further improving the equipment asset management and operational reliability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flow chart of the optimization method for substation maintenance decision-making provided by the present invention. DETAILED DESCRIPTION

[0074] The present invention will be further described below. The following examples are only used to more clearly illustrate the technical solution of the present invention and are not intended to limit the scope of protection of the present invention.

[0075] The present invention provides a method for optimizing substation maintenance decision-making, comprising: using the power grid state as the state of a Markov decision process, using the power substation maintenance decision as the action of the Markov decision process, using the power grid state transition as the strategy of the Markov decision process, using the improvement value of the power grid operation state obtained after the substation maintenance decision arrangement is taken as the reward of the Markov decision process, and establishing a substation maintenance decision optimization model based on the Markov decision process;

[0076] The substation maintenance decision optimization model based on the Markov decision process is solved to obtain the optimal substation maintenance decision.

[0077] A specific embodiment of the present invention provides an optimization method for substation maintenance decision making, such as Figure 1 As shown, the following steps are included:

[0078] Step 1: Obtain substation maintenance-related constraints, including grid security constraints, maintenance relationship constraints, maintenance resource constraints, and maintenance time constraints, as well as maintenance outage information.

[0079] The specific implementation process is:

[0080] Obtain grid security constraints, including node voltage constraints and line power flow constraints;

[0081] Obtain maintenance relationship constraints, including simultaneous maintenance constraints and mutually exclusive maintenance constraints;

[0082] Obtain maintenance resource constraints, including personnel, materials, etc.

[0083] The maintenance resource constraints are expressed as:

[0084]

[0085] Among them, x i (t) is the state of device i at time t, r i,k The demand for resource k for maintenance of equipment i, r k (t) is the available amount of resources such as personnel and materials at time t, N M The number of devices.

[0086] The equipment maintenance period must be limited to the operation and maintenance time interval. The maintenance time constraint is expressed as:

[0087] b i ≤x i (t)≤ei ;

[0088] Among them, b i is the earliest time allowed for maintenance of equipment i, x i (t) is the state of device i at time t, e i is the latest time allowed for maintenance of equipment i. This constraint means that the time the equipment is in maintenance state is controlled within [b i ,e i ] during this period.

[0089] The maintenance power outage information obtained includes the maintenance power outage power, user power outage time series and equipment power outage time series.

[0090] Step 2: Construct a decision optimization objective function based on various constraints, including minimizing the amount of power outages due to maintenance, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment.

[0091] The specific implementation process is:

[0092] First, define the minimum maintenance outage power consumption:

[0093]

[0094] Among them, F ens Represents the minimum maintenance outage power consumption, i=1,…,N C N is the user number. C is the total number of users, Q i Represents the total power outage energy consumed by user i after the power outage plan is executed.

[0095] Next, define the maximum user outage satisfaction:

[0096]

[0097] Among them, F sat Represents the maximum user satisfaction with power outage, i=1,…,N C N is the user number. C is the total number of users, S i Represents user i’s satisfaction after the power outage is executed.

[0098] User satisfaction S i The calculation method is:

[0099]

[0100] Among them, C CoCi is the power outage time series of user i, obtained based on the maintenance power outage information in step 1; Var(C CoCi) is the power outage time variance of user i, which is used to represent the difference in power outage time of user i; is the number of power outages for user i within the power outage plan cycle.

[0101] Finally, the optimal allocation of maintenance resources for each type of equipment is defined, and the maintenance time differences of each type of equipment are used to describe whether the maintenance schedule is reasonable.

[0102]

[0103] Among them, F optcon Represents minimizing the difference in maintenance time of various types of equipment. L, T, and K represent the types of equipment such as lines, transformers, and switches, respectively. Var(E L,T,K,j ) is the power outage time variance of type j equipment.

[0104] Step 3: Using the decision optimization objective function from step 2 as a constraint, establish a Markov decision process-based optimization model for the substation maintenance decision objective function. This Markov decision process uses the grid state as the state, the substation maintenance decision as the action, the grid state transition as the strategy, and the grid operation state improvement value obtained after the substation maintenance decision is made as the reward. The model represents a cyclical process in which substation maintenance decisions are used to change equipment state to improve grid operation state, which interacts with the grid operation environment.

[0105] The specific implementation process is:

[0106] The Markov decision process is used to represent the optimization model of the objective function of substation maintenance decision.

[0107] M={S,A,P s,a ,R},

[0108] Among them, S represents the set of power grid states, s∈S, s represents a specific power grid state, A represents the set of substation maintenance decision arrangements, a∈A, a represents a specific substation maintenance decision arrangement, P s,a represents the transition model, which indicates the probability of transitioning from the grid state s to another state when the substation maintenance decision arrangement a is adopted. R is the improvement value of the grid operation state obtained after the substation maintenance decision arrangement is adopted.

[0109] It should be noted that the improvement in the grid operation status obtained after the substation maintenance decision-making arrangement is made is quantitatively evaluated by establishing a corresponding indicator evaluation system.

[0110] Among them, the substation maintenance decision-making arrangement includes the maintenance object and the start and end time of the maintenance.

[0111]

[0112] Among them, a i Indicates a maintenance decision arrangement, including x i Maintenance object, b i Maintenance start time, d i Maintenance end time.

[0113] Step 4: Based on a given Markov decision process, by maximizing the behavioral value function and the state value function, with the constraints of minimizing the amount of maintenance outages, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment, solve the optimization model of the substation maintenance decision objective function based on the Markov decision process;

[0114] The specific implementation process is:

[0115] First, define the rewards obtained after taking the substation maintenance decision arrangement.

[0116] Define U(s0,s1,s2…) as the cumulative improvement in the operating status of all power grids after executing a set of maintenance plans. Since the sum of all power grid operating status improvements will lead to unbiasedness in an infinite time series, the concept of γ discount rate is introduced.

[0117]

[0118] Among them, γ∈[0,1], γ t represents the discount rate at time t, R(s t ) represents the improvement value of the power grid operation status after the substation maintenance decision arrangement is taken at time t; γ t Personalized adjustments are made based on specific scenarios. When it is close to 0, the agent cares more about short-term returns; when it is closer to 1, long-term returns become more important.

[0119] The above formula can also be written as:

[0120]

[0121] Among them, R max The maximum improvement in the grid operation status obtained after taking the substation maintenance decision arrangement.

[0122] Then, based on the reward U, the state value function is defined as the expectation of the future reward that the grid operation state s can obtain at time t:

[0123] v(s)=E[U t |S t =s].

[0124] Secondly, based on the reward U, the action value function is defined as the expected future reward that can be obtained by selecting the substation maintenance decision arrangement a based on the grid operation status s at time t:

[0125] q π =E π [U t |S t =s,A t =a];

[0126] S t is the grid state set at time t, A t is the set of substation maintenance decision arrangements at time t.

[0127] Finally, with the constraints of minimizing the amount of power outages due to maintenance, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment in step 2, the maximum values ​​of the state value function and the action value function are solved, which is the optimal combination for the power grid operation and maintenance decision arrangement.

[0128]

[0129] It should be understood that this solution is a common knowledge technique in the technical field and will not be described again.

[0130] Another embodiment of the present invention further provides an optimization device for substation maintenance decision-making, comprising:

[0131] A module is constructed to establish a Markov decision process-based substation maintenance decision optimization model using the power grid state as the state, the substation maintenance decision as the action, the power grid state transition as the strategy, and the improvement in the power grid operation state obtained after the substation maintenance decision is arranged as the reward of the Markov decision process.

[0132] The decision module is used to solve the substation maintenance decision optimization model based on the Markov decision process to obtain the optimal substation maintenance decision.

[0133] In this embodiment, the construction module is specifically used to:

[0134] The substation maintenance decision optimization model based on Markov decision process is established as follows:

[0135] M={S,A,P s,a ,R},

[0136] Among them, S represents the set of power grid states, s∈S, s represents a specific power grid state, A represents the set of substation maintenance decision arrangements, a∈A, a represents a specific substation maintenance decision arrangement, P s,arepresents the transition model, which expresses the probability of transitioning from grid state s to another grid state when substation maintenance decision arrangement a is adopted. R is the improvement value of the grid operation state obtained after the substation maintenance decision arrangement is adopted. The improvement value of the grid operation state is quantitatively evaluated by establishing a corresponding indicator evaluation system. Among them, the substation maintenance decision arrangement includes the maintenance object, maintenance start time and maintenance end time.

[0137] In this embodiment, the building block is further used to:

[0138] The constraints for minimizing maintenance outage power consumption, maximizing user satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment are as follows:

[0139] The constraints for minimizing maintenance outage power consumption are constructed as follows:

[0140]

[0141] Among them, F ens Indicates the minimum power consumption for maintenance, i=1,…,N C N is the user number. C is the total number of users, Q i Represents the total power outage energy of user i after the power outage plan is implemented;

[0142] The constraints for maximizing user power outage satisfaction are constructed as follows:

[0143]

[0144]

[0145] Among them, F sat represents the maximum value of user satisfaction with power outage, S i represents the satisfaction of user i after the power outage, C CoCi is the power outage time series of user i, Var(C CoCi ) is the variance of power outage time of user i, is the number of power outages for user i within the power outage plan cycle;

[0146] The constraints for optimizing the maintenance resource allocation for each type of equipment are as follows:

[0147]

[0148] Among them, F optcon Indicates the minimum difference in maintenance time for each type of equipment. L, T, and K represent the types of line, transformer, and switchgear, respectively. Var(E L,T,K,j ) is the power outage time variance of type j equipment.

[0149] In this embodiment, the decision module is specifically used to:

[0150] The expected future return that can be obtained from the grid operation state s at time t is used as the state value function, and the expected future return that can be obtained from selecting the substation maintenance decision arrangement a based on the grid operation state s at time t is used as the action value function. The state value function and action value function are calculated based on the returns obtained after the substation maintenance decision arrangement is taken;

[0151] The maximum value of the state value function and action value function is solved with the constraints of minimizing the power outage power consumption, maximizing the user's power outage satisfaction, and optimizing the allocation of maintenance resources for various types of equipment, to obtain the optimal substation maintenance decision arrangement;

[0152] Among them, the future return that can be obtained after selecting the substation maintenance decision arrangement is calculated as:

[0153]

[0154] Among them, U is the future return that can be obtained after choosing the substation maintenance decision arrangement, γ t represents the discount rate at time t, γ t ∈[0,1],R(s t ) represents the improvement value of the grid operation status at time t.

[0155] A third aspect of the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods described above.

[0156] A fourth aspect of the present invention provides a computing device, comprising:

[0157] One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any of the methods described above.

[0158] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0159] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0160] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0161] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0162] 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, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. An optimization method for substation maintenance decision-making, It is characterized in that include: Taking the power grid state as the state of the Markov decision process, taking the substation maintenance decision as the action of the Markov decision process, taking the power grid state transition as the strategy of the Markov decision process, taking the improvement value of the power grid operation state obtained after taking the substation maintenance decision arrangement as the reward of the Markov decision process, a substation maintenance decision optimization model based on the Markov decision process is established; Solving the substation maintenance decision optimization model based on the Markov decision process to obtain the optimal substation maintenance decision; The solution of the substation maintenance decision optimization model based on the Markov decision process must meet the constraints of minimizing the maintenance outage power, maximizing the user's power outage satisfaction, and optimizing the maintenance resource allocation of various types of equipment; The minimum maintenance outage power consumption constraint is as follows: Among them, F ens Indicates the minimum power outage power for maintenance, i=1,…,N C N is the user number. C is the total number of users, Q i Represents the total power outage amount of user i after the power outage plan is executed; The constraint for maximizing user power outage satisfaction is as follows: Among them, F sat represents the maximum value of user satisfaction with power outage, S i represents the satisfaction of user i after the power outage, C CoCi is the power outage time series of user i, Var(C CoCi ) is the power outage time variance of user i, is the number of power outages for user i within the power outage plan cycle; The optimization constraints for resource allocation of maintenance resources for various types of equipment are as follows: Among them, F optcon It represents the minimum difference of maintenance time of each type of equipment. L, T, and K represent the types of line, transformer, and switchgear respectively. Var(E L,T,K,j ) is the power outage time variance of type j equipment; The following constraints must also be met: A. Grid security constraints, including node voltage constraints and line flow constraints; B. Maintenance relationship constraints, including simultaneous maintenance constraints and mutually exclusive maintenance constraints; C. Maintenance resource constraints are expressed as: Among them, x i (t) is the state of device i at time t, r i,k The demand for material resource k for maintenance of equipment i, r k (t) is the available amount of personnel and material resources at time t, N M is the number of devices; D. Maintenance time constraint is expressed as: b i ≤x i (t)≤e i ; Among them, b i is the earliest time allowed for maintenance of equipment i, e i The latest time allowed for maintenance of equipment i.

2. According to claim 1, a method for optimizing substation maintenance decision making, It is characterized in that The establishment of a substation maintenance decision optimization model based on a Markov decision process includes: M={S,A,P s,a ,R} Among them, S represents the set of power grid states, s∈S, s represents a certain power grid state, A represents the set of substation maintenance decisions, a∈A, a represents a certain substation maintenance decision, P s,a represents the transfer model, which indicates the probability of transferring from the power grid state s to another power grid state after taking the substation maintenance decision a. R is the improvement value of the power grid operation state obtained after taking the substation maintenance decision. The improvement value of the power grid operation state is obtained by establishing a corresponding indicator evaluation system for quantitative evaluation. The substation maintenance decision arrangement includes maintenance objects, maintenance start time and maintenance end time.

3. According to claim 2, a method for optimizing substation maintenance decision-making, It is characterized in that Solving the substation maintenance decision optimization model based on the Markov decision process includes: Based on the rewards obtained after taking the substation maintenance decision, calculate the state value function and action value function; With the constraints of minimizing the amount of power outage for maintenance, maximizing the user's power outage satisfaction and optimizing the allocation of maintenance resources for various types of equipment, the maximum values ​​of the state value function and the action value function are solved to obtain the optimal substation maintenance decision.

4. The optimization method for substation maintenance decision according to claim 3, It is characterized in that The expected future return that the grid operation state s can obtain at time t is taken as the state value function; The expected future return of selecting substation maintenance decision a based on the grid operation status s at time t is used as the action value function; Among them, the future return that can be obtained after choosing the substation maintenance decision is calculated as: Among them, U is the future return that can be obtained after choosing the substation maintenance decision, γ t represents the discount rate at time t, γ t ∈[0,1],R(s t ) represents the improvement value of the power grid operation status at time t.

5. An optimization device for substation maintenance decision-making, It is characterized in that A method for optimizing substation maintenance decision-making according to any one of claims 1 to 4, the device comprising: A construction module is used to establish a substation maintenance decision optimization model based on the Markov decision process, using the power grid state as the state of the Markov decision process, the substation maintenance decision as the action of the Markov decision process, the power grid state transition as the strategy of the Markov decision process, and the power grid operation state improvement value obtained after the substation maintenance decision arrangement is taken as the reward of the Markov decision process; the model must meet the following constraints: The constraint for minimizing the power outage for maintenance is: Among them, F ens Indicates the minimum power outage power for maintenance, i=1,...,N C N is the user number. C is the total number of users, Q i Represents the total power outage amount of user i after the power outage plan is executed; The constraint to maximize user power outage satisfaction is: Among them, F sat represents the maximum value of user satisfaction with power outage, S i represents the satisfaction of user i after the power outage, C CoCi is the power outage time series of user i, Var(C CoCi ) is the power outage time variance of user i, is the number of power outages for user i within the power outage plan cycle; The optimal resource allocation constraints for each type of equipment maintenance are: Among them, F optcon It represents the minimum difference of maintenance time of each type of equipment. L, T, and K represent the types of line, transformer, and switchgear respectively. Var(E L,T,K,j ) is the power outage time variance of type j equipment; as well as, A. Grid security constraints, including node voltage constraints and line flow constraints; B. Maintenance relationship constraints, including simultaneous maintenance constraints and mutually exclusive maintenance constraints; C. Maintenance resource constraints are expressed as: Among them, x i (t) is the state of device i at time t, r i,k The demand for material resource k for maintenance of equipment i, r k (t) is the available amount of personnel and material resources at time t, N M is the number of devices; D. Maintenance time constraint is expressed as: b i ≤x i (t)≤e i ; Among them, b i is the earliest time allowed for maintenance of equipment i, e i The latest time allowed for maintenance of equipment i; The decision module is used to solve the substation maintenance decision optimization model based on the Markov decision process to obtain the optimal substation maintenance decision.

6. The optimization device for substation maintenance decision according to claim 5, It is characterized in that The building blocks are specifically used for: The substation maintenance decision optimization model based on Markov decision process is established as follows: M={S,A,P s,a ,R} Among them, S represents the set of power grid states, s∈S, s represents a certain power grid state, A represents the set of substation maintenance decisions, a∈A, a represents a certain substation maintenance decision, P s,a represents the transfer model, which indicates the probability of transferring from the power grid state s to another power grid state after taking the substation maintenance decision a. R is the improvement value of the power grid operation state obtained after taking the substation maintenance decision. The improvement value of the power grid operation state is obtained by establishing a corresponding indicator evaluation system for quantitative evaluation. The substation maintenance decision arrangement includes maintenance objects, maintenance start time and maintenance end time.

7. The optimization device for substation maintenance decision according to claim 6, It is characterized in that The decision module is specifically used to: The expected future return that can be obtained from the grid operation state s at time t is taken as the state value function, and the expected future return that can be obtained from selecting the substation maintenance decision a based on the grid operation state s at time t is taken as the action value function. Based on the return obtained after taking the substation maintenance decision, the state value function and the action value function are calculated; With the constraints of minimizing the power outage for maintenance, maximizing the user's satisfaction with power outages, and optimizing the allocation of maintenance resources for various types of equipment, the maximum value of the state value function and the action value function is solved to obtain the optimal substation maintenance decision; Among them, the future return that can be obtained after choosing the substation maintenance decision is calculated as: Among them, U is the future return that can be obtained after choosing the substation maintenance decision, γ t represents the discount rate at time t, γ t ∈[0,1],R(s t ) represents the improvement value of the power grid operation status at time t.

8. A computer-readable storage medium storing one or more programs, Features: The one or more programs include instructions which, when executed by a computing device, cause the computing device to perform any one of the methods according to claims 1 to 4.

9. A computing device, Features: include, One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for executing any one of the methods according to claims 1 to 4.