Extreme weather considering renewable energy main grid element reinforcement method and system

Abstract

The application discloses a method and system for strengthening main grid elements considering renewable energy under extreme weather, and mainly relates to the technical field of renewable energy research. The method comprises the following steps: collecting an expected value of system loss load caused by extreme weather, and establishing a first objective function with the minimum expected value; modeling an operation mode of a main grid; establishing a linear programming model for calculating a worst probability distribution of uncertain variables; establishing a three-layer mixed integer linear programming model according to the first objective function and the modeling of the operation mode of the main grid; and solving the three-layer mixed integer linear programming model and the linear programming model of the worst probability distribution and outputting a result. The application has the beneficial effects of solving the problem of difficult information acquisition, reducing the difficulty of information provided by a power grid operator, and improving the efficiency of power transmission grid restoration.

Description

Technical Field

[0001] This invention relates to the field of renewable energy research technology, specifically a method and system for strengthening main grid components that consider renewable energy under extreme weather conditions. Background Technology

[0002] Since the beginning of the new century, with the continuous improvement of industrialization in human society, humans have emitted large amounts of greenhouse gases into nature every year, leading to global warming and increased instability in the climate system. Global climate change is not only reflected in rising temperatures; extreme weather events are also showing a trend of increasing frequency, widespread occurrence, intensity, and concurrence. A report from the World Meteorological Organization indicates that in the past 50 years, the number of disasters has increased fivefold, and the losses caused by disasters have increased sevenfold. In September 2022, Typhoon Muifa brought strong winds and heavy rainfall to the coastal and central-northern parts of Zhejiang Province, affecting the power supply of more than 500,000 households. On September 18, 2022, Typhoon Nanmadol made landfall in Japan, sweeping through the Kyushu region, forcing more than 300,000 households to lose power. Building a power system based on new energy sources is the future development trend. However, the output of new energy sources such as wind and solar power is highly volatile and random, making them more susceptible to weather factors. Low-probability extreme events can cause greater damage to the power system. Therefore, improving the power system's ability to cope with low-probability, high-loss extreme events is of great significance for ensuring national energy security.

[0003] Traditional reliability metrics target high-probability, low-loss routine events and are inadequate for handling "NK" failures caused by extreme events, thus posing new requirements for power system security. Resilience, defined as "the system's ability to prepare for and prevent disasters, withstand and absorb damage during disasters, respond and adapt, and quickly return to a pre-set normal state after a disaster," was originally used to describe ecosystems and has since been introduced into engineering. Domestic and international scholars have conducted extensive research on the resilience of power systems. For predictable extreme events, proactive prevention can significantly reduce losses. Research on power system extreme event prevention focuses on two main aspects: firstly, optimizing the power system layout by considering low-probability extreme events during power system planning, improving the system's flexibility in responding to extreme events through optimized site selection, line expansion, and topology optimization; secondly, upgrading existing power systems by identifying weaknesses and strengthening key components to enhance robustness against extreme events, thereby reducing losses. The former requires significant investment and a long construction period, making it unsuitable for the transformation of existing power systems, while the latter only requires strengthening key system components, with less investment and a shorter construction period. Existing component reinforcement research mainly focuses on robust optimization and stochastic programming. The former seeks the worst-case scenario caused by extreme events, optimizes for the worst-case scenario, and calculates the optimal reinforcement strategy. The latter generates all possible scenarios caused by extreme events, optimizes all scenarios, and obtains component reinforcement strategies.

[0004] While robust optimization and stochastic programming are widely used in component reinforcement, they also have many drawbacks. Robust optimization requires accurate information such as the number of power system outages, which is extremely difficult to predict accurately for a power system in advance, and the decisions made by robust optimization are too conservative. Stochastic programming, while not as conservative as robust optimization, still requires obtaining the probability distribution function in advance, which is also very difficult. Furthermore, because stochastic programming considers more scenarios, it may increase the dimensionality of the problem and make the solution space more complex. Therefore, how to solve the problem of difficult information acquisition and improve the conservatism of the decisions made is a current challenge for component reinforcement. In addition, most existing component reinforcement methods are aimed at traditional power systems dominated by thermal power units, rarely considering the large-scale integration of renewable energy sources such as wind and solar power. The randomness and volatility of renewable energy in the power system increase the uncertainty of the system, and more factors need to be considered when formulating reinforcement strategies. Therefore, researching component reinforcement methods for power systems with high renewable energy penetration has significant theoretical and practical value. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for strengthening main grid components that consider renewable energy under extreme weather conditions. It solves the problem of difficulty in obtaining information, reduces the difficulty for grid operators to provide information, and improves the efficiency of transmission grid resilience.

[0006] To achieve the above objectives, the present invention employs the following technical solution:

[0007] A method for strengthening grid components considering renewable energy under extreme weather conditions includes the following steps:

[0008] S1: Collect the expected value of system load loss caused by extreme weather, and establish the first objective function with the minimum expected value;

[0009] S2: Model the main network's operating mode;

[0010] S3: Establish a linear programming model for calculating the worst-case probability distribution of uncertain variables;

[0011] S4: Establish a three-layer mixed integer linear programming model based on the first objective function and the main network's operating mode;

[0012] S5: Solve the above three-level mixed integer linear programming model and the worst probability distribution linear programming model and output the results.

[0013] Preferably, step S1 specifically involves: analyzing the uncertainty of renewable energy output and component damage under extreme weather conditions, taking into account the importance of the load at each node, and then collecting the minimum expected value of the system load loss caused by extreme weather as the first objective function, wherein the first objective function is:

[0014]

[0015] Among them, E l For the set of lines, E g Let W1 be the set of generators, φ be the load importance weight, γ be the fuzzy probability distribution space constructed based on historical data of the transmission network, and Z be the probability distribution function of the scenario. gen The component enhancement variables are 0 / 1 variables, v and v gen The attack variables are 0 / 1 variables, w and w gen These are refactored variables, where s represents the scenario and p... s Represents the probability of scenario s occurring. This represents the load shedding amount of node j in scenario s.

[0016] 5. Preferably, the modeling of the main grid's operation mode specifically involves: using the DCflow model to model the main grid's operation and using 0 / 1 variables to describe the state of the main grid components. This modeling is a transmission network operation model, and the specific form of the transmission network operation model is as follows:

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032] Where K is the set of iterative scenarios and V is the set of nodes. This represents the power flowing through line l in scenario s during the k-th iteration. Let g be the output of generator g in scenario s during the k-th iteration. Let g be the output of the photovoltaic system in the scenario during the k-th iteration. Let g be the output of the wind turbine in scenario s during the k-th iteration. This indicates the capacity of line l. Let P be the state of line l in scenario s during the k-th iteration. L,j Let j be the load of node j. Let g be the state of generator g in scenario s during the k-th iteration. and Let g be the minimum and maximum output of generator g. For the maximum output of photovoltaic g, This represents the maximum output of the fan g. Let b represent the phase angle of node i connected by line l in scenario s during the k-th iteration. l For the admittance of line l, M is a very large constant, Q l and Q gnn, All are auxiliary variables.

[0033] 6. Preferably, the establishment of the linear programming model for calculating the worst-case probability distribution of the uncertainty variables specifically involves: obtaining the worst-case scenario under different component reinforcement strategies and the number of component failures; and establishing a linear programming model for calculating the worst-case probability distribution of the uncertainty variables for this scenario. The uncertainty variables specifically include: renewable energy output and the number of component failures. The linear programming model for the probability distribution is expressed as follows:

[0034] ∑ s∈y p s =1 (17)

[0035]

[0036] in, ρ represents the probability of scenario s occurring in the past, calculated based on historical data, and represents the maximum deviation between the probability of scenario s occurring and the historical probability.

[0037] Preferably, the establishment of the three-layer mixed-integer linear programming model specifically involves: establishing a three-layer mixed-integer linear programming model based on the defense-attack-defense model. The three-layer mixed-integer linear programming model includes a first-layer model, used to optimize the main network operator for a set of attack scenarios under a specific probability distribution, formulate the optimal component reinforcement strategy, and minimize the losses caused by the attack; a second-layer model, used to formulate an attack plan against the component reinforcement strategy formulated by the attacker in the first-layer model, bypassing the component reinforcement strategy and causing the maximum loss; and a third-layer model, used to reduce load loss for the main network operator through network reconstruction and generator rescheduling when the component reinforcement and attack plan are determined. The second-layer model and the third-layer model are used to obtain the worst-case scenario under a specific component reinforcement strategy and the number of damaged components.

[0038] Preferably, the specific steps of solving and outputting the results of the three-level mixed integer linear programming model and the worst probability distribution linear programming model are as follows: the three-level mixed integer linear programming model is solved using a two-level column and constraint generation algorithm, and the probability distribution linear programming model is solved directly using a commercial solver. The two models are solved iteratively to obtain the optimal component strengthening scheme.

[0039] Preferably, the two-level column and constraint generation algorithm specifically involves decomposing the established three-level mixed integer linear programming model into an upper-level problem, a lower-level main problem, and lower-level sub-problems. The specific form of the upper-level problem is as follows:

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047] Where a is the objective function value of the upper-level problem, and N s The upper limit of the number of components to be reinforced. and These are known attack variables and are constants;

[0048] The specific form of the lower-level main problem is as follows:

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055]

[0056]

[0057] Where β is the objective function of the lower-level main problem. This represents the maximum number of damaged components in scenario s. This represents the load of node j in scenario s during the 0th iteration. and For known component reinforcement variables, and Reconstruct variables for the known network;

[0058] The specific form of the lower-level sub-problem is as follows:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] Where c is the objective function of the lower-level subproblem. and The auxiliary variable is known to be a constant;

[0066] The method of directly solving the probability distribution model using a commercial solver specifically involves solving the objective function of the linear programming model of the probability distribution, which is expressed as follows:

[0067]

[0068] Equation (40) is the objective function of the linear programming model of the probability distribution.

[0069] A grid reinforcement system considering renewable energy under extreme weather conditions includes a data acquisition module, a model building module, and an analysis and processing module. The data acquisition module is used to: collect the expected value of system load loss caused by extreme weather, and establish a first objective function with the minimum expected value; the model building module is used to: model the grid operation mode, establish a linear programming model for calculating the worst probability distribution of uncertain variables, and establish a three-level mixed integer linear programming model based on the first objective function and the grid operation mode; the analysis and processing module is used to: solve the three-level mixed integer linear programming model and the worst probability distribution linear programming model and output the results.

[0070] Preferably, the data acquisition module, model building module, and analysis and processing module are connected via data.

[0071] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0072] 1. A robust optimization model based on the defense-attack-defense model is proposed. This model considers the uncertainties of renewable energy and the number of damaged components, identifies the worst-case scenario for each scenario, and optimizes the reinforcement of transmission network components based on the probability distribution of each scenario using stochastic programming. This improves the conservatism of decision-making. Furthermore, the established model only requires historical data from the transmission network to construct the probability distribution space of uncertain variables such as renewable energy output and the number of damaged components, reducing the required information. The established robust optimization model combines the advantages of robust optimization and stochastic programming, preventing overly conservative decisions while not requiring accurate probability distribution information, thus solving the problem of difficult information acquisition.

[0073] 2. A probability distribution model is proposed based on historical data of the power transmission network to obtain the approximate probability distribution function of the uncertain variables. The probability distribution space is constructed with this probability distribution function as the center, and the worst probability distribution function is calculated, which reduces the difficulty for power grid operators to provide information.

[0074] 3. Using a two-level column and constraint generation algorithm to solve the established sub-bar optimization model, the global optimal solution of the model can be obtained within an acceptable time. The optimal component strengthening scheme can be obtained with limited investment, providing an efficient tool for improving the resilience of the power transmission network. Attached Figure Description

[0075] Figure 1 This is a flowchart of the method of the present invention.

[0076] Figure 2 This is a schematic diagram of the constraints included in the bibliometric optimization model of the present invention.

[0077] Figure 3 This is a schematic diagram of the decomposed model composition of the present invention.

[0078] Figure 4 This is a flowchart of the model solution process of this invention. Detailed Implementation

[0079] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined in this application.

[0080] like Figure 1 As shown, the present invention relates to a method for strengthening main grid components considering renewable energy under extreme weather conditions, which is achieved through the following steps:

[0081] S1: Considering the uncertainty of renewable energy output and the number of component failures, and taking into account the importance of the load at each node, the first objective function is to minimize the expected value of the system load loss caused by extreme weather.

[0082] S2: Use the DCflow model to model the operation of the main network, and use 0 / 1 variables to describe the state of the main network components;

[0083] S3: Based on the defense-attack-defense model, a three-layer mixed-integer linear programming model is established. In the first layer, the main network operator optimizes for a set of attack scenarios under a specific probability distribution, formulating the optimal component reinforcement strategy to minimize the losses caused by the attack. In the second layer, the attacker devises an attack plan against the component reinforcement strategy formulated in the first layer to bypass the strategy and cause the maximum loss. In the third layer, given the component reinforcement and attack plan, the main network operator reduces load loss through network reconfiguration and generator rescheduling. The second and third layers can obtain the worst-case scenario under a specific component reinforcement strategy and the number of damaged components.

[0084] S4: For the worst-case scenario with different component reinforcement strategies and the number of component failures, establish a linear programming model to calculate the worst probability distribution of uncertain variables such as renewable energy output and the number of component failures;

[0085] S5: The three-level mixed integer linear programming model is solved using a two-level column and constraint generation algorithm. The linear programming model with probability distribution is solved directly using a commercial solver. The two models are solved iteratively to obtain the optimal component strengthening scheme.

[0086] This invention proposes a sub-bar optimization model based on a defense-attack-defense model, considering the uncertainties in renewable energy output and the number of component failures. The objective function is to minimize the expected system load loss. The constraints included in the model are as follows: Figure 2 As shown, the constraints include power balance constraints, line power constraints, load shedding constraints, renewable energy and conventional generator output constraints, phase angle difference constraints at both ends of the line, component operating state constraints, and probability distribution constraints. The established sub-Bruker bar optimization model is solved using a column and constraint generation algorithm and a commercial solver. The constraints are shown below:

[0087] First objective function:

[0088]

[0089] Power transmission network operation model:

[0090] The component enhancement method proposed in this invention is aimed at the power transmission network. Since the impact of reactive power in the power grid does not need to be considered during planning, the DCflow model can be used to model the operation of the power transmission network. The specific form of the model is as follows:

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105]

[0106] Equation (2) is the node power balance constraint, Equation (3) is the line capacity constraint, the power flowing through the line cannot be greater than the line capacity, Equation (4) is the load shedding constraint, the load shedding cannot be greater than the original load of the node, Equation (5) is the thermal power generator output constraint, Equation (6) is the photovoltaic output constraint, Equation (7) is the wind turbine output constraint, Equation (8) is the phase angle difference constraint between the nodes at both ends of the line, Equations (9)-(12) represent the relationship between the power flowing through the line and the phase angle at both ends of the line, and Equations (13)-(16) are the element state constraints.

[0107] Probability distribution model:

[0108] ∑ s∈y p s =1 (17)

[0109]

[0110] Equation (17) indicates that the sum of the probabilities of all scenarios is 1, and Equation (18) indicates the relationship between the probability of scenario s occurring and the historical probability.

[0111] Solve the model:

[0112] like Figure 3 As shown, the decomposed Bruker optimization model established in this invention is a multi-level mixed integer linear programming model, which cannot be directly solved by existing commercial solvers. Therefore, a column and constraint generation algorithm is used to decompose the established model into an upper-level problem (UP), a lower-level master problem (LMP), a lower-level subproblem (LSP), and a probability distribution problem (PP) for iterative solution. The specific forms of each problem are as follows:

[0113] Upper-level issues:

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121] Equation (19) is the objective function of the upper-level problem, Equation (20) limits the number of components to be reinforced, Equation (21) calculates the value of the objective function, and Equations (22)-(25) are the state constraints of the components;

[0122] Lower-level main question:

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129]

[0130]

[0131] Equation (26) is the objective function of the lower-level main problem, Equation (27) limits the number of damaged components, Equation (28) calculates the objective function value, Equation (29) calculates the load loss of each node, and Equations (30)-(33) determine the state of each component;

[0132] Lower-level sub-problems:

[0133]

[0134]

[0135]

[0136]

[0137]

[0138]

[0139] Equation (34) is the objective function of the lower-level subproblem, Equation (35) calculates the objective function value of the lower-level subproblem, and constraints (36)-(39) determine the operating state of the component;

[0140] Objective function for probability distribution problems:

[0141]

[0142] The specific solution process is as follows: Figure 4 As shown:

[0143] 1) Set the upper bound UB = +∞, LB = -∞, and the number of upper-level iterations k = 1, and give any attack scheme;

[0144] 2) Solve the upper-level problem to obtain the component strengthening scheme. And the objective function a, update LB = max{LB, a}, and Pass this information to the next lower-level problem and update k = k + 1;

[0145] 3) Set LUB = +∞, LLB = -∞, and set the number of iterations in the next lower level to o = 1;

[0146] 4) Solve the lower-level master problem to obtain an attack plan. And the objective function value β, update LUB = min{LUB,β}, and and Passed on to the next level sub-problem;

[0147] 5) Solve the lower-level subproblems to obtain the reconstruction scheme. And the objective function value c, update LLB={LLB,c}, o=o+1;

[0148] 6) If LUB ≠ LLB, then Pass the problem to the next lower-level main problem, return to step 4, and if LUB = LLB, calculate the expected load value E for each scenario. p Update UB = min{UB, E} p};

[0149] 7) If UB≠LB, pass the objective function value β of the lower-level master problem for each scenario to the probability distribution problem, and solve for the worst-case probability distribution. Will and Pass the problem to the upper level, return to step 2, and if UB = LB, output the result of the strengthening element. and

Claims

1. A method of strengthening a main grid element considering renewable energy in extreme weather, characterized in that, Includes the following steps: S1: Collect the expected value of system load loss caused by extreme weather, and establish the first objective function with the minimum expected value; S2: Model the main network's operating mode; S3: Establish a linear programming model for calculating the worst-case probability distribution of uncertain variables; S4: Establish a three-layer mixed integer linear programming model based on the first objective function and the main network's operating mode; S5: Solve the above three-level mixed integer linear programming model and the worst probability distribution linear programming model and output the results; Step S1 specifically involves: analyzing the uncertainty of renewable energy output and component damage under extreme weather conditions, taking into account the importance of the load at each node, and then using the minimum expected value of the system load loss caused by extreme weather as the first objective function. The first objective function is: （1） in, For the collection of routes, For generator sets, As the weight of load importance, This represents a fuzzy probability distribution space constructed based on historical data from the power transmission network. The probability distribution function representing the scene. and The component enhancement variable is a 0 / 1 variable. and The attack variable is a 0 / 1 variable. and It is about refactoring variables. Representative scenarios, Representative scenarios The probability of occurrence Representing a scene Middle node The amount of load shedding; The establishment of the linear programming model for calculating the worst-case probability distribution of uncertain variables specifically involves: obtaining the worst-case scenario under different component reinforcement strategies and the number of component failures; and establishing a linear programming model for calculating the worst-case probability distribution of uncertain variables for this scenario. The uncertain variables specifically include: renewable energy output and the number of component failures. The linear programming model for the probability distribution is expressed as follows: （17） （18） in, Scenario calculated based on historical data The probability of history happening, Representing a scene The maximum deviation between the probability of occurrence and the historical probability; The establishment of the three-layer mixed-integer linear programming model specifically involves: based on the defense-attack-defense model, establishing a three-layer mixed-integer linear programming model, which includes a first-layer model used to optimize the main network operator for a set of attack scenarios under a specific probability distribution, formulate the optimal component reinforcement strategy, and minimize the losses caused by the attack; a second-layer model used to formulate an attack plan against the component reinforcement strategy formulated by the attacker in the first-layer model, bypassing the component reinforcement strategy and causing the maximum loss; and a third-layer model used to reduce load loss for the main network operator through network reconstruction and generator rescheduling when the component reinforcement and attack plan are determined. The second-layer model and the third-layer model are used to obtain the worst-case scenario under a specific component reinforcement strategy and the number of damaged components.

2. The method for strengthening main grid components considering renewable energy under extreme weather conditions according to claim 1, characterized in that, The specific steps for modeling the main grid's operation mode are as follows: The DCflow model is used to model the main grid's operation, and 0 / 1 variables are used to describe the state of the main grid components. This modeling is a transmission network operation model, and the specific form of the transmission network operation model is as follows: （2） （3） （4） （5） （6） （7） （8） （9） （10） （11） （12） （13） （14） （15） （16） in, To iterate over the scene set, For a set of nodes, Representing the Scenario during the next iteration Central route The power flowing through, For the first Scenario during the next iteration medium generator of efforts, For the photovoltaic scenario in the k-th iteration of efforts, For the first Scenario during the next iteration Stroke of efforts, Indicates the line capacity, For the first Scenario during the next iteration Central route state, For nodes The load, For the first Scenario during the next iteration medium generator state, and For generator The minimum and maximum output, For photovoltaic Maximum output For wind turbine Maximum output Indicates the first Scenario during the next iteration Central route Connected nodes phase angle, For the line Admittance, It is a very large constant. and All are auxiliary variables.

3. The method of claim 1, wherein the method is implemented in a power system including a plurality of power sources, a plurality of loads, and a plurality of power transmission lines, and the method is implemented in a power system including a plurality of power sources, a plurality of loads, and a plurality of power transmission lines. The solution and output of the three-level mixed integer linear programming model and the worst probability distribution linear programming model are as follows: the three-level mixed integer linear programming model is solved using a two-level column and constraint generation algorithm, and the probability distribution linear programming model is solved directly using a commercial solver. The two models are solved iteratively to obtain the optimal component strengthening scheme.

4. The method of claim 3, wherein the method is characterized by: The two-level column and constraint generation algorithm specifically involves decomposing the established three-level mixed integer linear programming model into an upper-level problem, a lower-level main problem, and lower-level sub-problems.

5. A system for strengthening of main grid elements considering renewable energy in extreme weather conditions for implementing a method for strengthening of main grid elements considering renewable energy in extreme weather conditions according to any one of claims 1 to 4, characterized in that, The system includes a data acquisition module, a model building module, and an analysis and processing module. The data acquisition module is used to: collect the expected value of system load loss caused by extreme weather, and establish a first objective function with the minimum expected value; the model building module is used to: model the operation mode of the main network, establish a linear programming model for calculating the worst probability distribution of uncertain variables, and establish a three-level mixed integer linear programming model based on the first objective function and the operation mode of the main network; the analysis and processing module is used to: solve the three-level mixed integer linear programming model and the worst probability distribution linear programming model and output the results.

6. The system for strengthening the main grid elements considering renewable energy in extreme weather conditions as claimed in claim 5 wherein, The data acquisition module, model building module, and analysis and processing module are connected via data.

Images