Method and system for constructing pre-disaster protection strategy of flexible power grid

Abstract

The invention relates to a tough power grid pre-disaster protection strategy construction method and system, and the method comprises the steps: obtaining parameters needed for formulating a protection strategy model, and constructing a risk neutral model and a risk avoidance model; solving the risk neutral model and the converted risk evasion model by using an algorithm to obtain a risk neutral protection strategy and a risk evasion protection strategy; and determining an optimal budget through parameter sensitivity analysis, and integrating a risk neutral protection strategy and a risk evasion protection strategy to obtain an optimal pre-disaster protection strategy adapted to the actual situation. Compared with the prior art, the method enables the final strategy to more comprehensively cope with challenges brought by natural disasters, guarantees the applicability of the strategy in all possible scenes, and achieves the effective balance between the protection cost and the toughness improvement.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method and system for constructing a pre-disaster protection strategy for a resilient power grid. Background Art

[0002] The frequency of extreme natural disasters caused by global climate change is increasing year by year. Effective pre-disaster defense strategies are crucial to improving the resilience of power systems to natural disasters. However, the current research on pre-disaster protection strategies for power grids still faces several unresolved issues.

[0003] First, the operation of other critical infrastructure such as gas transmission, water supply, and transportation systems depends on the electricity provided by the power system. In natural disaster events, damage to the power system can easily cause other infrastructure systems to fail to operate, amplifying the economic and social impacts of power outages. Therefore, the formulation of pre-disaster protection strategies for power systems should not only focus on the resilience of the system itself, but also consider the resilience of other infrastructure systems that rely on the power system. The formulation of protection strategies aimed at enhancing the resilience of the comprehensive system is more in line with actual needs. However, existing research has paid little attention to this.

[0004] Second, since the protection strategy is implemented before the disaster event occurs, it is impossible to accurately predict the disaster scenario and the damage state of the power system. Existing research focuses on potential disaster scenarios, with the goal of maximizing the resilience of the system in the worst-case scenario or the expected value of resilience in all possible scenarios. However, when formulating the protection strategy, two aspects must be fully considered: one is to ensure the applicability of the strategy in all possible scenarios, and the other is to avoid the system resilience value being too low or the performance being severely degraded under extreme conditions. Further research is still needed to fully address these two challenges.

[0005] Third, the degree of damage to the power system caused by disasters of different levels varies significantly, and the budget for pre-disaster protection measures is limited. For a disaster of a certain level, how to find the best balance between improving system resilience and controlling pre-disaster protection costs is an urgent problem to be solved. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a method and system for constructing a resilient power grid pre-disaster protection strategy. When constructing a power grid pre-disaster protection strategy, the strategy is customized based on the comprehensive system, while considering the system resilience and risk indicators, and giving an optimal budget, so that the final strategy can more comprehensively respond to the challenges brought by natural disasters, ensuring the applicability of the strategy in all possible scenarios, and achieving an effective balance between protection costs and resilience improvement.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for constructing a pre-disaster protection strategy for a resilient power grid, the method comprising:

[0009] Obtain the parameters required to formulate the protection strategy model and build the risk-neutral model and risk-averse model;

[0010] Using algorithms to solve the risk-neutral model and the transformed risk-aversion model, we can obtain the risk-neutral protection strategy and the risk-aversion protection strategy;

[0011] The optimal budget is determined through parameter sensitivity analysis, and the risk-neutral protection strategy and risk-averse protection strategy are combined to obtain the best pre-disaster protection strategy that adapts to the actual situation.

[0012] Furthermore, the parameters required for formulating the protection strategy model include:

[0013] Deterministic parameters, including network flow parameters of the power system and other infrastructure, location of system components, protection budget, unit cost of different protection measures, and typhoon intensity;

[0014] Uncertainty parameters, including a representative set of disaster scenarios in the power network obtained through scenario generation and K-Means clustering method.

[0015] Furthermore, the scenario generation includes:

[0016] Define the level of typhoon disaster based on deterministic parameters;

[0017] Enter the number of scenarios;

[0018] For each scenario, a typhoon path with probability of occurrence is generated based on historical data;

[0019] Calculate the failure probability of each edge in the affected area;

[0020] Generate edge damage conditions for the power system. For edges in the affected area, if the failure probability is greater than the damage threshold, the edge is damaged, otherwise, the edge is not damaged.

[0021] Furthermore, the risk-neutral model is a two-stage stochastic programming model, which includes a pre-disaster decision-making stage and a post-disaster operation decision-making stage. The pre-disaster decision-making stage determines the risk-neutral protection strategy under a limited budget, and the post-disaster operation decision-making stage determines the post-disaster system resilience improvement strategy.

[0022] The objective function of the risk-neutral model is:

[0023] max E(RE)=∑ ω∈Ψ p ω ×RE ω ,

[0024] Where ψ is the set of disaster scenarios of transmission line failure in the power network, ω is the index of the set of disaster scenarios of transmission line failure in the power network, and p ω is the probability of occurrence of scenario ω, RE ω is the resilience of the comprehensive system under scenario ω, and E(RE) is the expected value of the resilience of the comprehensive system.

[0025] Furthermore, the constraints of the risk-neutral model include policy budget constraints, protection measure effectiveness constraints, infrastructure service flow constraints and infrastructure association relationship constraints.

[0026] Furthermore, the policy budget constraint is:

[0027]

[0028] Among them, L E is the set of power network edges in the affected area, l∈L k is the edge l within the influence area, is the total number of nodes in the power network, is the node n in the power network, is the reinforcement cost of edge l in the power network, The cost of installing a distributed generation unit for node n, Indicates whether edge l is reinforced, Indicates whether node n is equipped with a backup power generation unit, and B is the protection strategy budget;

[0029] The protection measures are constrained to be effective as follows:

[0030]

[0031]

[0032] Where E represents the power system, indicates whether edge l is damaged after disaster scenario ω, is the service demand provided by node n under the disaster scenario ω, is the service demand of node n;

[0033] The infrastructure service traffic constraints are:

[0034]

[0035] Where K is the total number of infrastructure types affected by the disaster, k∈K is any infrastructure affected by the disaster, N is the total number of nodes of the infrastructure affected by the disaster, and denote nodes belonging to infrastructures g, t, and d, respectively, and f lk,ω For the disaster scenario ω, the service flow through l is is the service production capacity of node n under disaster scenario ω, It means that after the disaster scenario ω occurs, component j∈{n∈N k ,l∈L k}Is it running normally? represents the service demand of node n under disaster scenario ω, represents the service demand of node n, f l k For edge l∈L k Flow capacity;

[0036] The infrastructure association constraints are:

[0037]

[0038] in, is the service flow through l under the disaster scenario ω, θ E,k It is a set of dependencies between other infrastructure system nodes and power system nodes in the integrated system.

[0039] Furthermore, the risk aversion model is used to maximize the expected resilience value of the integrated system and minimize the downside risk index value, and the specific form includes:

[0040] Based on the risk model related to the downside value at risk, the objective function is:

[0041]

[0042] in, is the threshold of system resilience, ω is the index of the disaster scenario set of transmission line failure in the power network, p ω is the probability of occurrence of scenario ω, RE ω is the resilience of the comprehensive system under scenario ω, Δ ω It indicates whether the system resilience value is lower than the set threshold when the disaster scenario ω occurs, and DownRisk indicates the value of the relevant risk.

[0043] Furthermore, the algorithm includes an ε-constraint method and a step-by-step hedging algorithm.

[0044] Furthermore, the parameter sensitivity analysis includes disaster intensity analysis and protection measures cost analysis.

[0045] A system for constructing a resilient power grid pre-disaster protection strategy, comprising:

[0046] The parameter acquisition module is used to obtain the parameters required for formulating the protection strategy model;

[0047] Model building module, used to build risk-neutral models and risk-averse models;

[0048] The model solving module uses algorithms to solve the risk-neutral model and the transformed risk-averse model to obtain the risk-neutral protection strategy and the risk-averse protection strategy;

[0049] The strategy formulation module determines the optimal budget through parameter sensitivity analysis, and integrates risk-neutral protection strategy and risk-averse protection strategy to obtain the best pre-disaster protection strategy that adapts to the actual situation.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. This paper comprehensively considers the limited budget of defense strategies, the dependence of other infrastructure systems on electricity, and the uncertainty of disaster scenarios, and proposes a method for formulating pre-disaster protection strategies for power systems based on resilience and risk-driven. By optimizing the pre-disaster protection strategy, the resilience of the integrated system to natural disasters is improved, and a power grid protection strategy with risk neutrality and risk aversion characteristics can be formulated within a limited budget;

[0052] 2. Taking into account the dependence of other infrastructure on the power system, this paper proposes a two-stage stochastic model with the goal of enhancing the resilience of the integrated system consisting of the power system and other infrastructure that rely on electricity. This model can formulate a risk-neutral protection strategy for the power system, which has greater practical significance than strategies that only focus on improving the resilience of the power system and can more comprehensively respond to the challenges brought by natural disasters;

[0053] 3. In the present invention, in order to avoid the system resilience value being lower than the acceptable minimum level in some possible disaster scenarios, risk indicators are used to evaluate such extreme risks, and the constructed risk-neutral model is expanded into a multi-objective planning model that comprehensively considers resilience improvement and risk avoidance, and a risk avoidance protection strategy for the power system is formulated to reduce the risk brought by the uncertainty of the system resilience value in uncertain disaster scenarios, thereby ensuring the applicability of the strategy in all possible scenarios and avoiding the system resilience value being too low or the performance being seriously degraded under extreme conditions;

[0054] 4. The present invention conducts parameter sensitivity analysis on the pre-disaster protection strategy budget, comprehensively considers the disaster intensity and the cost of protection measures, determines the optimal budget of the pre-disaster protection strategy, and combines the risk-neutral protection strategy and the risk-averse protection strategy to formulate the best pre-disaster protection strategy that adapts to the actual situation. While improving the system resilience, it also takes into account the control of protection costs, achieves an effective balance between protection costs and resilience improvement, and saves costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flow chart for implementing the method of the present invention;

[0056] Figure 2 This is a result comparison diagram of the risk neutral model with different resilience improvement targets in an embodiment of the present invention;

[0057] Figure 3 A Pareto optimal curve diagram of a strategy obtained by applying a risk management model in an embodiment of the present invention;

[0058] Figure 4 This is a result diagram obtained when performing a sensitivity analysis on disaster intensity in an embodiment of the present invention;

[0059] Figure 5 This is a result diagram obtained when a sensitivity analysis is performed on the cost of protective measures in an embodiment of the present invention;

[0060] Figure 6 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0062] Example 1

[0063] This embodiment proposes a method for constructing a pre-disaster protection strategy for a resilient power grid. Figure 1 As shown, including:

[0064] Step S1, obtaining the parameters required for formulating the protection strategy model;

[0065] Step S2, constructing a risk-neutral model and a risk-averse model;

[0066] Step S3, using an algorithm to solve the risk neutral model and the transformed risk aversion model to obtain a risk neutral protection strategy and a risk aversion protection strategy;

[0067] Step S4, determine the optimal budget through parameter sensitivity analysis, and combine the risk-neutral protection strategy and the risk-averse protection strategy to obtain the optimal pre-disaster protection strategy that adapts to the actual situation.

[0068] Furthermore, obtaining the parameters required for formulating the protection strategy model includes obtaining deterministic parameters and uncertain parameters, wherein the deterministic parameters include network flow parameters of the power system and other infrastructure, system component locations, protection budgets, unit costs of different protection measures, and typhoon intensity. The uncertain parameters are the set of damaged edges in the power network.

[0069] Assuming that decision makers know the possible typhoon paths and the failure probability of transmission lines, a representative set of disaster scenarios is obtained through scenario generation and scenario reduction using the K-Means clustering method.

[0070] Furthermore, the scenario generation includes:

[0071] Define the level of typhoon disaster based on deterministic parameters;

[0072] Input scenario number |Ω|;

[0073] For each scenario ω=1,2,…,|Ω|, generate a typhoon path with occurrence probability based on historical data; calculate the failure probability of each edge in the affected area

[0074] Generate edge damage of the power system. For the edge l∈L in the affected area k , if the failure probability is greater than the damage threshold Edge damage Otherwise, the edge is not damaged

[0075] Furthermore, the risk neutral model is a two-stage stochastic programming model, and the two stages include a pre-disaster decision-making stage and a post-disaster operation decision-making stage.

[0076] Attached Figure 2 In the figure, EVR represents the expected value of system resilience, which shows the comparison of the expected values ​​of resilience obtained by choosing to improve the resilience of the power grid alone and improving the resilience of the comprehensive system under different budgets. It can be seen that compared with the goal of improving the resilience of the power grid alone (with objective 1), the expected value of system resilience of the goal of improving the resilience of the comprehensive system (with objective 2) is higher.

[0077] Under different budgets, the strategy obtained by improving the resilience of the overall system is more realistic than the strategy of only improving the resilience of the power grid.

[0078] The model aims to maximize the expected resilience value of the comprehensive system in possible disaster scenarios. The objective function of the pre-disaster decision-making stage is as follows:

[0079] max E(RE)=∑ ω∈Ψ p ω ×RE ω ,

[0080] Where ψ is the set of disaster scenarios of transmission line failure in the power network, ω is the index of the set of disaster scenarios of transmission line failure in the power network, and p ωis the probability of occurrence of scenario ω, RE ω is the resilience of the comprehensive system under scenario ω, and E(RE) is the expected value of the resilience of the comprehensive system (EVR).

[0081] The constraints in the model include policy budget constraints, protection measures effectiveness constraints, infrastructure service flow constraints, infrastructure association constraints, etc.

[0082] The strategic budget constraint ensures that the total cost of reinforcing the transmission lines and installing distributed generation units does not exceed the budget, which is in the following form:

[0083]

[0084] Among them, L E is the set of power network edges in the affected area, l∈L k is the edge l within the influence area, is the total number of nodes in the power network, is the node n in the power network, is the reinforcement cost of edge l in the power network, The cost of installing a distributed generation unit for node n, Indicates whether edge l is reinforced, Indicates whether node n is equipped with a backup power generation unit, and B is the protection strategy budget;

[0085] The protection measure effect constraint represents the connection between the protection decision in the first stage and the operation status and flow decision of the edge (transmission line) in the power system in the second stage, and is in the following form:

[0086]

[0087]

[0088] in, indicates whether edge l is damaged after disaster scenario ω, is the service demand provided by node n under the disaster scenario ω, is the service demand of node n;

[0089] The infrastructure service flow constraints are derived from the network flow model, which constrain the flow of various facilities (nodes and edges) in the infrastructure network in the following form:

[0090]

[0091]

[0092] Where K is the total number of infrastructure types affected by the disaster, k∈K is any infrastructure affected by the disaster, N is the total number of nodes of the infrastructure affected by the disaster, and denote nodes belonging to infrastructures g, t, and d, respectively, and f l k,ω For the disaster scenario ω, the service flow through l is is the service production capacity of node n under disaster scenario ω, It means that after the disaster scenario ω occurs, component j∈{n∈N k ,l∈L k}Is it running normally? represents the service demand of node n under disaster scenario ω, represents the service demand of node n, f l k For edge l∈L k Flow capacity;

[0093] Infrastructure association constraints represent the association between the operating status of nodes in the power grid and other infrastructure networks and the dependencies between systems, in the following form:

[0094]

[0095] in, is the service flow through l under the disaster scenario ω, θ E,k It is a set of dependencies between other infrastructure system nodes and power system nodes in the integrated system.

[0096] Furthermore, it is characterized in that the risk avoidance model aims to formulate risk management and protection strategies for the power system to avoid the risk of too low expected value of comprehensive system resilience in some disaster scenarios.

[0097] like Figure 3 As shown in the figure, when considering risk management, decision makers need to consider the actual risk aversion requirements and formulate effective risk aversion strategies by balancing EVR and downside risk. An excessive increase in the system's expected value will also lead to an increase in downside risk.

[0098] In the model, the downside risk value is introduced to measure the relevant risk, which is in the following form:

[0099]

[0100] in, is the threshold of system resilience, ω is the index of the disaster scenario set of transmission line failure in the power network, p ω is the probability of occurrence of scenario ω, RE ω is the resilience of the comprehensive system under scenario ω, Δ ωIt indicates whether the system resilience value is lower than the set threshold when the disaster scenario ω occurs, and DownRisk indicates the value of the relevant risk;

[0101] Furthermore, based on the risk neutral model, a multi-objective model for solving the pre-disaster protection strategy of power grid risk management is constructed. The model objectives are as follows:

[0102]

[0103] Furthermore, the algorithms used to solve the model include the ε-constraint method and the step-by-step hedging algorithm.

[0104] Furthermore, parameter sensitivity analysis includes disaster intensity analysis and protection measures cost analysis.

[0105] When determining the protection budget, it is necessary to comprehensively consider the disaster intensity (Wind speed) and the cost of protection measures to determine the optimal budget for the pre-disaster protection strategy. Figure 4 With attached Figure 5 As shown in the figure, for typhoons of different intensities, there is an optimal budget value that can balance the EVR of the comprehensive system and the cost of the protection strategy. This is because as the budget increases, the incremental marginal benefit of EVR decreases. Even if more budget is invested, there will be no significant improvement in the expected value of system resilience.

[0106] According to the actual situation, the cost of pre-disaster protection measures is set in combination with relevant research. The protection cost of some specific lines is proportional to their length. The budget obtained in this embodiment may be different from the actual problem, but combined with the set analysis process, a suitable strategy can still be obtained.

[0107] Example 2

[0108] This embodiment proposes a resilient power grid pre-disaster protection strategy construction system, such as Figure 6 As shown, including:

[0109] Parameter acquisition module 1, used to obtain the parameters required for formulating the protection strategy model;

[0110] Model building module 2, used to build risk neutral model 21 and risk aversion model 22;

[0111] Model solving module 3 uses algorithms to solve the risk neutral model and the transformed risk aversion model to obtain the risk neutral protection strategy and the risk aversion protection strategy;

[0112] Strategy formulation module 4 determines the optimal budget through parameter sensitivity analysis, integrates risk-neutral protection strategy and risk-averse protection strategy, and formulates the best pre-disaster protection strategy that adapts to the actual situation.

[0113] The specific details of the above modules can be understood by referring to the relevant descriptions and effects in Example 1.

[0114] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for constructing a pre-disaster protection strategy for a resilient power grid, characterized in that: The method comprises: Obtain the parameters required to formulate the protection strategy model and build the risk-neutral model and risk-averse model; Using algorithms to solve the risk-neutral model and the transformed risk-aversion model, we can obtain the risk-neutral protection strategy and the risk-aversion protection strategy; The optimal budget is determined through parameter sensitivity analysis, and the risk-neutral protection strategy and risk-averse protection strategy are combined to obtain the best pre-disaster protection strategy that adapts to the actual situation.

2. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 1, characterized in that: The parameters required for formulating the protection strategy model include: Deterministic parameters, including network flow parameters of the power system and other infrastructure, location of system components, protection budget, unit cost of different protection measures, and typhoon intensity; Uncertainty parameters, including a representative set of disaster scenarios in the power network obtained through scenario generation and K-Means clustering method.

3. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 2, characterized in that: The scenario generation includes: Define the level of typhoon disaster based on deterministic parameters; Enter the number of scenarios; For each scenario, a typhoon path with probability of occurrence is generated based on historical data; Calculate the failure probability of each edge in the affected area; Generate edge damage conditions for the power system. For edges in the affected area, if the failure probability is greater than the damage threshold, the edge is damaged, otherwise, the edge is not damaged.

4. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 1, characterized in that: The risk-neutral model is a two-stage stochastic programming model, which includes a pre-disaster decision-making stage and a post-disaster operation decision-making stage. The pre-disaster decision-making stage determines the risk-neutral protection strategy under a limited budget, and the post-disaster operation decision-making stage determines the post-disaster system resilience improvement strategy. The objective function of the risk-neutral model is: max E(RE)=∑ ω∈Ψ p ω ×RE ω , Where ψ is the set of disaster scenarios of transmission line failure in the power network, ω is the index of the set of disaster scenarios of transmission line failure in the power network, and p ω is the probability of occurrence of scenario ω, RE ω is the resilience of the comprehensive system under scenario ω, and E(RE) is the expected value of the resilience of the comprehensive system.

5. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 3, characterized in that: The constraints of the risk-neutral model include policy budget constraints, protection measure effectiveness constraints, infrastructure service flow constraints, and infrastructure association relationship constraints.

6. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 5, characterized in that: The policy budget constraint is: Among them, L E is the set of power network edges in the affected area, l∈L k is the edge l within the influence area, is the total number of nodes in the power network, is the node n in the power network, is the reinforcement cost of edge l in the power network, The cost of installing a distributed generation unit for node n, Indicates whether edge l is reinforced, Indicates whether node n is equipped with a backup power generation unit, and B is the protection strategy budget; The protection measures are constrained to be effective as follows: Where E represents the power system, Indicates whether edge l is damaged after disaster scenario ω, is the service demand provided by node n under the disaster scenario ω, is the service demand of node n; The infrastructure service traffic constraints are: Where K is the total number of infrastructure types affected by the disaster, k∈K is any infrastructure affected by the disaster, N is the total number of nodes of the infrastructure affected by the disaster, and denote nodes belonging to infrastructures g, t, and d, respectively, and f l k,ω For the disaster scenario ω, the service flow through l is is the service production capacity of node n under disaster scenario ω, It means that after the disaster scenario ω occurs, component j∈{n∈N k ,l∈L k }Is it running normally? represents the service demand of node n under disaster scenario ω, represents the service demand of node n, f l k For edge l∈L k Flow capacity; The infrastructure association constraints are: in, is the service flow through l under the disaster scenario ω, θ E,k It is a set of dependencies between other infrastructure system nodes and power system nodes in the integrated system.

7. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 1, characterized in that: The risk aversion model is used to maximize the expected resilience value of the comprehensive system and minimize the downside risk index value. The specific form includes: Based on the risk model related to the downside value at risk, the objective function is: in, is the threshold of system resilience, ω is the index of the disaster scenario set of transmission line failure in the power network, p ω is the probability of occurrence of scenario ω, RE ω is the resilience of the comprehensive system under scenario ω, Δ ω It indicates whether the system resilience value is lower than the set threshold when the disaster scenario ω occurs, and DownRisk indicates the value of the relevant risk.

8. The method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 1, characterized in that: The algorithms include the ε-constraint method and the step-by-step hedging algorithm.

9. A method for constructing a pre-disaster protection strategy for a resilient power grid according to claim 1, characterized in that: Parameter sensitivity analysis includes disaster intensity analysis and protection measures cost analysis.

10. A system for constructing a resilient power grid pre-disaster protection strategy, characterized in that: include: The parameter acquisition module is used to obtain the parameters required for formulating the protection strategy model; Model building module, used to build risk-neutral models and risk-averse models; The model solving module uses algorithms to solve the risk-neutral model and the transformed risk-averse model to obtain the risk-neutral protection strategy and the risk-averse protection strategy; The strategy formulation module determines the optimal budget through parameter sensitivity analysis, and integrates risk-neutral protection strategy and risk-averse protection strategy to obtain the best pre-disaster protection strategy that adapts to the actual situation.

Images