Distributed robust recovery method and terminal for electricity-gas integrated energy system

By establishing and reconstructing the power system recovery model, combining the natural gas system model, and using the analytical target cascade method to solve the problem of effectiveness and reliability in the recovery process of the electric-gas comprehensive energy system after failure, improving the recovery of the system Robustness in capability and decision-making.

CN119010140APending Publication Date: 2024-11-22山西省能源互联网研究院 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411088786.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing integrated electrical-gas energy systems lack effectiveness and reliability in the recovery process after failure, especially in the case of large wind power penetration, which makes it difficult to ensure the rapid and stable recovery of the system.

Method used

The upper power system recovery model is established based on pumped storage and wind power uncertainty, and the model is reconstructed through a distributed robust model, combined with the gas storage and pipe storage models of the natural gas system, and the analysis target cascade method is used to solve it. Obtain optimal recovery decisions for the power system.

Benefits of technology

It improves the effectiveness and reliability of system failure recovery, avoids overly conservative recovery decisions, ensures the privacy of the power network and natural gas system, and considers the impact of the natural gas network on the recovery of large-capacity power systems.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010140A_ABST
    Figure CN119010140A_ABST
Patent Text Reader

Abstract

The invention discloses an electricity-gas integrated energy system distributed robust recovery method and a terminal, an established upper-layer power system recovery model considers pumped storage and wind power uncertainty, the upper-layer power system recovery model is reconstructed based on a distributed robust model, and the reconstructed upper-layer power system recovery model is obtained. The established lower-layer natural gas system recovery model considers gas storage and management storage, the reconstructed upper-layer power system recovery model and the lower-layer natural gas system recovery model are solved by using an analysis target cascade method to obtain an optimal recovery decision of the power system, and the influence of a natural gas network on the recovery process of the high-capacity power system is considered; meanwhile, the concurrent influence of the pumped storage hydropower station and the wind power plant is considered, an over-conservative recovery decision is avoided, the balance point of the proposed double-layer optimization problem can be accurately calculated, and the privacy of the power network and the natural gas system is ensured, so that the effectiveness and the reliability of recovery after the system fails are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of integrated energy system operation, and in particular to a distributed robust recovery method and terminal for an electric-gas integrated energy system. Background Art

[0002] Rapidly restoring the power system after a power outage is one of the challenges faced by transmission system operators, which is particularly important in the case of wind power penetration, as wind farms have great potential in the rapid restoration of transmission systems. In recent years, the application of renewable energy in the power system restoration process has been widely studied, such as the use of a two-stage stochastic optimization method to evaluate the impact of wind farms in the power system restoration process, a risk-based approach for bulk power system restoration with high wind farm penetration, and a two-stage robust optimization algorithm that takes into account the bulk power system restoration process and power island synchronization under high wind farm penetration conditions. However, these methods cannot ensure the effectiveness and reliability of system restoration. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a distributed robust recovery method and terminal for an electric-gas integrated energy system, which can improve the effectiveness and reliability of recovery after a system failure.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] A distributed robust recovery method for an electric-gas integrated energy system comprises the following steps:

[0006] Establish an upper-level power system restoration model based on the uncertainty of pumped storage and wind power;

[0007] Reconstructing the upper power system restoration model based on the distributed robust model to obtain a reconstructed upper power system restoration model;

[0008] Establish a lower-level natural gas system recovery model based on gas storage and pipeline storage;

[0009] The reconstructed upper-layer power system recovery model and the lower-layer natural gas system recovery model are solved using an analytical target cascade method to obtain an optimal power system recovery decision.

[0010] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0011] A distributed robust recovery terminal for an electric-gas integrated energy system comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0012] Establish an upper-level power system restoration model based on the uncertainty of pumped storage and wind power;

[0013] Reconstructing the upper power system restoration model based on the distributed robust model to obtain a reconstructed upper power system restoration model;

[0014] Establish a lower-level natural gas system recovery model based on gas storage and pipeline storage;

[0015] The reconstructed upper-layer power system recovery model and the lower-layer natural gas system recovery model are solved using an analytical target cascade method to obtain an optimal power system recovery decision.

[0016] The beneficial effects of the present invention are as follows: the established upper-level power system recovery model takes into account the uncertainties of pumped storage and wind power, and the upper-level power system recovery model is reconstructed based on a distributed robust model to obtain a reconstructed upper-level power system recovery model, and the established lower-level natural gas system recovery model takes into account gas storage and pipeline storage, and the reconstructed upper-level power system recovery model and the lower-level natural gas system recovery model are solved using an analytical target cascade method to obtain an optimal recovery decision for the power system, taking into account the impact of the natural gas network on the recovery process of the large-capacity power system, and taking into account the concurrent impact of pumped-storage hydropower stations and wind farms, and reconstructing the upper-level power system recovery model based on a distributed robust model to avoid overly conservative recovery decisions, and using an analytical target cascade method to accurately calculate the equilibrium point of the proposed two-layer optimization problem, and ensure the privacy of the power network and the natural gas system, thereby improving the effectiveness and reliability of system recovery after a failure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of the steps of a distributed robust recovery method for an electric-gas integrated energy system according to an embodiment of the present invention;

[0018] Figure 2 This is a schematic diagram of the structure of a distributed robust recovery terminal for an electric-gas integrated energy system according to an embodiment of the present invention;

[0019] Figure 3 This is a solution flow chart of a distributed robust recovery method for an electric-gas integrated energy system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0021] Please refer to Figure 1 , a distributed robust recovery method for an electric-gas integrated energy system, comprising the steps of:

[0022] Establish an upper-level power system restoration model based on the uncertainty of pumped storage and wind power;

[0023] Reconstructing the upper power system restoration model based on the distributed robust model to obtain a reconstructed upper power system restoration model;

[0024] Establish a lower-level natural gas system recovery model based on gas storage and pipeline storage;

[0025] The reconstructed upper-layer power system recovery model and the lower-layer natural gas system recovery model are solved using an analytical target cascade method to obtain an optimal power system recovery decision.

[0026] From the above description, it can be seen that the beneficial effects of the present invention are: the upper-level power system recovery model established takes into account the uncertainties of pumped storage and wind power, and the upper-level power system recovery model is reconstructed based on the distributed robust model to obtain the reconstructed upper-level power system recovery model, and the lower-level natural gas system recovery model established takes into account gas storage and pipeline storage, and the analytical target cascade method is used to solve the reconstructed upper-level power system recovery model and the lower-level natural gas system recovery model to obtain the optimal recovery decision of the power system, taking into account the impact of the natural gas network on the recovery process of the large-capacity power system, and considering the concurrent impact of pumped-storage hydropower stations and wind farms, and reconstructing the upper-level power system recovery model based on the distributed robust model to avoid overly conservative recovery decisions, and using the analytical target cascade method to accurately calculate the equilibrium point of the proposed two-layer optimization problem, and ensure the privacy of the power network and the natural gas system, thereby improving the effectiveness and reliability of system recovery after failure.

[0027] Furthermore, the establishment of an upper-level power system restoration model based on pumped storage and wind power uncertainty includes:

[0028] Establishing the first objective function with the goal of maximizing the post-fault recovery capability of the power system;

[0029] Establish initial condition constraints, startup constraints for non-black start generators, generator constraints, pumped storage constraints, power system flow and power balance constraints, cold start load constraints, and dynamic reserve constraints;

[0030] An upper-level power system recovery model is generated based on the first objective function, the initial condition constraints, the starting constraints of the non-black start generator sets, the generator set constraints, the pumped storage constraints, the power flow and power balance constraints of the power system, the cold start load constraints and the dynamic reserve constraints.

[0031] From the above description, it can be seen that the first objective function is established with the goal of maximizing the post-fault recovery capability of the power system, and initial condition constraints, start-up constraints of non-black start generator sets, generator set constraints, pumped storage constraints, power system flow and power balance constraints, cold start load constraints and dynamic standby constraints are established to ensure the safety of power system operation during the recovery process.

[0032] Furthermore, the upper power system restoration model is reconstructed based on the distributed robust model to obtain the reconstructed upper power system restoration model, which includes:

[0033] The first objective function of the upper-layer power system restoration model is reconstructed based on a distributed robust model to obtain a reconstructed first objective function.

[0034] From the above description, it can be seen that by reconstructing the first objective function of the upper power system restoration model based on the distributed robust model, the uncertain behavior of the wind farm can be simulated through a distributed robust framework, which makes the restoration decision robust while avoiding the restoration decision being too conservative.

[0035] Furthermore, the reconstructed first objective function is:

[0036]

[0037] In the formula, Indicates that the probability distribution range of wind power is within the distributed robust model Θ, represents the expectation of wind power probability distribution, represents the upper limit of the active power of the traditional generator set g, represents the starting power of the traditional generator set g, represents the power generation state of the traditional generator set g at the recovery time t, represents the upper limit of the output power of the wind turbine w, u w,t represents the state of wind turbine w at recovery time t, α d represents the load priority factor, Indicates the maximum recoverable load, P d,t Represents the total load recovered within the recovery time t, Ω GFU represents the set of gas units, v g represents the first penalty coefficient, represents the fuel demand of the first gas unit, represents the known fuel demand of the second gas generator set, w g represents the second penalty coefficient, represents the Haldmann product.

[0038] From the above description, it can be seen that the objective function takes into account the start-stop unit combination of the generator set in the fault recovery state and the degree of change of the recovery load over time. At the same time, the ATC (Analysis Target Cascade) algorithm is used to consider the recovery capacity of the natural gas system to the power system through the gas units, so as to comprehensively consider the influence of various recovery factors and improve the reliability of the recovery decision.

[0039] Furthermore, the establishment of a lower-layer natural gas system recovery model based on gas storage and pipeline storage includes:

[0040] Establishing a second objective function with the goal of minimizing the net production cost of the natural gas system;

[0041] Establish the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system, and the safety constraints of the natural gas system;

[0042] A lower-layer natural gas system recovery model is generated according to the second objective function, the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system, and the safety constraints of the natural gas system.

[0043] From the above description, it can be seen that the establishment of the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system and the safety constraints of the natural gas system ensures the stable operation of the natural gas system during the recovery process.

[0044] Furthermore, the second objective function established with the goal of minimizing the net production cost of the natural gas system is:

[0045]

[0046] In the formula, C r GW represents the gas well production cost, Indicates the gas production of the gas well, Ω GFU represents the collection of gas units, represents the gas extraction cost of the gas unit, represents the gas intake of the gas generator set at time t, v g represents the first penalty coefficient, represents the fuel demand of the second gas unit, represents the known fuel demand of the first gas unit, w g represents the second penalty coefficient, represents the Haldmann product.

[0047] From the above description, it can be seen that since the sale of natural gas by the natural gas system to the power system through gas-fired units is regarded as the profit of the natural gas system, a second objective function is established with the goal of minimizing the net production cost of the natural gas system. Such an objective function can more strictly consider the production cost of the natural gas system, and at the same time make the decision of the natural gas system in supplying gas to the power system more optimistic, so as to provide more recovery capacity to the power system through gas-fired units.

[0048] Furthermore, the use of the analytical target cascade method to solve the reconstructed upper power system restoration model and the lower natural gas system restoration model to obtain the optimal restoration decision of the power system includes:

[0049] Performing inner loop iteration on the reconstructed upper power system recovery model and the lower natural gas system recovery model until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition, thereby obtaining an inner loop iteration result;

[0050] Determine whether the inner loop iteration result meets the second convergence condition. If so, output the optimal power system recovery decision. If not, update the penalty coefficients in the reconstructed upper power system recovery model and the lower natural gas system recovery model. After updating, return to execute the inner loop iteration step of the reconstructed upper power system recovery model and the lower natural gas system recovery model.

[0051] From the above description, it can be seen that since the interaction between the gas system operator and the power system operator is completed through the boundary information exchange between the power system and the gas system, the solution is obtained through the internal and external loop method, while ensuring the privacy of the power network and the gas system, the optimal recovery decision is obtained.

[0052] Further, the inner loop iteration is performed on the reconstructed upper power system recovery model and the lower natural gas system recovery model until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition, and the inner loop iteration result is obtained, including:

[0053] Solving the reconstructed upper-layer power system restoration model to obtain a first solution result;

[0054] Obtaining a known fuel demand of the first gas generator set from the first solution result, and inputting the known fuel demand of the first gas generator set into the lower-layer natural gas system recovery model, and then solving the lower-layer natural gas system recovery model to obtain a second solution result;

[0055] Obtaining a known fuel demand of the second gas generator set from the second solution result, and inputting the known fuel demand of the second gas generator set into the reconstructed upper power system recovery model, returning to execute the step of solving the reconstructed upper power system recovery model until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition;

[0056] An inner loop iteration result is generated according to the known fuel requirement of the first gas engine group and the known fuel requirement of the second gas engine group in the last iteration.

[0057] From the above description, it can be seen that the fuel demand of the first gas unit and the fuel demand of the second gas unit are obtained through interactive solution between the upper power system recovery model and the lower natural gas system recovery model to realize inner loop iteration, thereby improving the effectiveness of recovery after system failure.

[0058] Further, the determining whether the inner loop iteration result meets the second convergence condition includes:

[0059] Determine whether the absolute value of the difference between the known fuel demand of the first gas unit and the known fuel demand of the second gas unit in the last iteration is less than a first preset value; if so, determine that the inner loop iteration result meets the second convergence condition; if not, determine that the inner loop iteration result does not meet the second convergence condition.

[0060] From the above description, it can be seen that when the absolute value of the difference between the known fuel demand of the first gas unit and the known fuel demand of the second gas unit in the last iteration is less than the first preset value, it is considered that the inner loop iteration result reaches the second convergence condition, and the outer loop iteration can be terminated, indicating that the current recovery decision is optimal, thereby achieving the most effective system recovery after failure.

[0061] Please refer to Figure 2 Another embodiment of the present invention provides a distributed robust recovery terminal for an electricity-gas integrated energy system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step in the above-mentioned distributed robust recovery method for an electricity-gas integrated energy system is implemented.

[0062] The above-mentioned distributed robust recovery method and terminal of the electric-gas integrated energy system of the present invention can be applied to the power system restored after a fault, and the following is explained through specific implementation methods:

[0063] Please refer to Figure 1 and Figure 3 , Embodiment 1 of the present invention is:

[0064] A distributed robust recovery method for an electric-gas integrated energy system comprises the following steps:

[0065] S1. Establish an upper-level power system restoration model based on the uncertainty of pumped storage and wind power, including S11-S13:

[0066] S11. Establishing a first objective function with the goal of maximizing the post-fault recovery capability of the power system, specifically:

[0067]

[0068] In the formula, represents the upper limit of the active power of the traditional generator set g, represents the starting power of the traditional generator set g, It represents the power generation state of the traditional generator set g at the recovery time t, which is a binary variable. represents the upper limit of the output power of the wind turbine w, u w,t represents the state of wind turbine w at recovery time t, α d represents the load priority factor, Indicates the maximum recoverable load, P d,t Represents the total load recovered within the recovery time t, Ω GFU represents the set of gas units, v g represents the first penalty coefficient, represents the fuel demand of the first gas unit, represents the known fuel demand of the second gas generator set, w g represents the second penalty coefficient, represents the Haldmann product.

[0069] The first term of the first objective function is the total power generation of conventional units in the power system, the second term is the total power generation of wind turbines, the third term is the objective function of minimizing the unserviced load during the restoration of the power system, and the fourth and fifth terms represent the penalty terms of the structure of the iterative convergence of the ATC algorithm. When the iteration converges, these two terms are approximately 0. The establishment of the first objective function takes into account the combination of start-stop units of the generator sets in the fault recovery state and the degree of change of the recovery load over time. At the same time, the analytical target cascade (ATC) method is used to consider the recovery capacity of the natural gas system to the power system through the gas units, so as to comprehensively consider the influence of various recovery factors and improve the reliability of the optimized recovery decision.

[0070] S12. Establish initial condition constraints, startup constraints for non-black start generator sets, generator set constraints, pumped storage constraints, power system flow and power balance constraints, cold start load constraints, and dynamic reserve constraints.

[0071] Among them, before running the optimization problem, the power system operator must determine the initial state of the power system after the power outage. The main infrastructure of the power grid may suffer serious damage from natural disasters. Therefore, after the power outage, some network components may not be able to recover immediately. On the other hand, considering that some events such as substations, power plants or local facilities are attacked by cyber attacks, partial power outages may occur. In this case, the main part of the network is not affected and no major recovery measures are required. The present invention assumes that a complete power outage will not cause serious damage to the network equipment and the main infrastructure of the system. Both traditional generators and wind turbines can help restore the power grid, so the initial condition constraints are:

[0072]

[0073] In the formula, Indicates the startup state of the unit at the initial moment t = 0, Indicates the startup state of the unit at time t=1, Ω NBSU represents the set of non-black start units, u l,t=0 represents the state of the transmission line at the initial time t = 0, u n,t=0 represents the state of the transmission bus n at the initial time t = 0, Ω BSU The initial condition constraint indicates that at the initial time t = 0, all non-black start units are shut down and the transmission line / bus is not energized. When the black start units are turned on, the network recovery process starts from t = 1.

[0074] Restarting the unit and reconnecting the transmission line are necessary for the recovery process. It is necessary to start the non-black start unit whose bus is energized. Assuming that the transmission line l connects bus i and bus j, if both buses are cut off at the recovery time t, the transmission line l remains de-energized at time t. The starting constraint of the non-black start generator set is:

[0075]

[0076] In the formula, represents the startup state of the unit at time t, u l,t represents the state of the transmission line at time t, It represents the starting power of the non-black start generator set at time t, Indicates the start time of the generator to start outputting power. Indicates the duration of generator startup, u i,t Indicates the state of bus i, u j,t represents the state of bus j, represents the set of traditional starting units connected to bus j, represents the set of wind turbines connected to busbar i, ui(j),t represents the common state of buses i and j, represents the set of transmission lines connecting busbars i and j, u l,t+1 Indicates the state of the transmission line at time t+1.

[0077] The generator set constraints are:

[0078]

[0079] In the formula, Indicates the lower limit of the active power of the generator set, P g,t represents the active power of the generator set at time t, Indicates the lower limit of the reactive power of the generator set, Q g,t represents the reactive power of the generator set at time t, Indicates the upper limit of the reactive power of the generator set, P w,t represents the output power of the wind turbine at time t, Indicates the lower climbing limit, P g,t+1 represents the active power of the generator set at time t+1, Indicates the upper climbing limit, Q g,t+1 represents the reactive power of the generator set at time t+1, φ represents the electricity-gas conversion coefficient, η g Indicates the electricity-to-gas conversion efficiency.

[0080] The pumped storage constraints are:

[0081]

[0082] In the formula, H p,t represents the water storage capacity at time t, H p,t-1 represents the water storage capacity at time t-1, η c Indicates the charging efficiency. Indicates the charging power, represents the discharge power, η d Indicates the efficiency of discharge, H p represents the lower limit of water storage capacity, represents the upper limit of water storage capacity, Indicates the charging status. Indicates the lower limit of charging power, Indicates the upper limit of charging power. Indicates the state of discharge, Indicates the lower limit of discharge power, represents the upper limit of the discharge power, where the last state constraint is to ensure that the pumped storage power station does not charge and discharge at the same time.

[0083] The power flow and power balance constraints of the power system are:

[0084]

[0085] In the formula, represents the set of traditional starting units connected to bus i, represents the set of wind turbines connected to busbar i, represents the set of recoverable loads connected to bus i represents the set of pumped-storage hydroelectric units connected to busbar i, represents the set of transmission lines connecting busbars i and j, represents the active power flow, Q w,t represents the reactive power of wind turbine at time t, Q d,t represents the reactive power of the load that can be restored at time t, Represents reactive power flow, g ij Represents the susceptance of the transmission line, V i,t Represents the voltage amplitude of bus i, V j,t represents the voltage amplitude of bus j, θ i,t represents the phase angle of the voltage on transmission line i, b ij represents the conductance of the transmission line, θ j,t represents the phase angle of the voltage on transmission line j, represents the upper limit of active power flow, Represents the upper limit of reactive power flow, V i Indicates the lower limit of the voltage amplitude, Indicates the upper limit of the voltage amplitude, θ i Indicates the lower limit of the voltage phase angle, Indicates the upper limit of the voltage phase angle.

[0086] In load recovery, when a large load is started, a serious frequency drop will occur, which will lead to instability of the entire power system. The present invention defines a load recovery coefficient for each unit, that is, the maximum load that the generator can supply without causing a frequency drop. Therefore, the cold start load constraint is:

[0087]

[0088] In the formula, R g,t Indicates the dynamic standby of the unit, P d,t+1 represents the total load recovered within the recovery time t+1, λ g Indicates the cold start load factor.

[0089] Dynamic reserve constraints are used to ensure the stability of the system. In particular, when emergencies occur, the system frequency must be kept within an acceptable range. Dynamic reserve has three sources, including conventional generators, wind turbines, and underfrequency load reduction. In order to cope with the loss of large generators, the active power of conventional generators and wind turbines must not exceed the total dynamic reserve of the system. Therefore, the dynamic reserve constraint is:

[0090]

[0091] In the formula, R t Represents the total dynamic reserve of the system, R d,t Indicates dynamic load shedding standby, Ω D represents the system load set, Ω G Represents the set of traditional starting units in the system, Ω W represents the set of wind turbines in the system, R w,t represents dynamic wind power reserve, ΔP g Indicates the variable load starting power of the generator set.

[0092] S13. Generate an upper-level power system recovery model based on the first objective function, the initial condition constraints, the starting constraints of the non-black start generator sets, the generator set constraints, the pumped storage constraints, the power flow and power balance constraints of the power system, the cold start load constraints and the dynamic standby constraints.

[0093] S2. Reconstruct the upper-layer power system recovery model based on the distributed robust model to obtain a reconstructed upper-layer power system recovery model.

[0094] Specifically, the first objective function of the upper-layer power system restoration model is reconstructed based on the distributed robust model to obtain a reconstructed first objective function.

[0095] Distributed robust models can simulate the uncertain behavior of wind farms, making restoration decisions robust while avoiding overly conservative solutions. In distributed robust models, the objective function is not to minimize the expected value in the worst case, but to minimize the estimated distribution The expected value of the worst-case probability distribution in the fuzzy set Θ is considered in the distributed robust model, and all possible wind power distributions whose characteristics conform to the mean vector μ and the standard deviation matrix σ are considered.

[0096] Wherein, the distributed robust model θ is:

[0097]

[0098] In the formula, represents the set of all effective wind power probability distributions, represents the support set, whose purpose is to realize the uncertainty in the finite set, Γ1 represents the first uncertainty control parameter, Γ2 represents the second uncertainty control parameter, which is used to control the volume of the fuzzy set, which respectively satisfy: Γ1≥0, Γ2≥1.

[0099] The reconstructed first objective function is:

[0100]

[0101] In the formula, Indicates that the probability distribution range of wind power is within the distributed robust model Θ, represents the expectation of the wind power probability distribution.

[0102] Because wind power uncertainty only exists on the power system side, and the present invention does not consider the propagation of uncertainty on the power system side and the natural gas system side, only the power system side (ie, the upper power system recovery model) is reconstructed.

[0103] S3. Establish a lower-layer natural gas system recovery model based on gas storage and pipeline storage, specifically including S31-S33:

[0104] S31. Establishing a second objective function with the goal of minimizing the net production cost of the natural gas system, specifically:

[0105]

[0106] In the formula, C r GW represents the gas well production cost, Indicates the gas production of the gas well, Ω GFU represents the collection of gas units, represents the gas extraction cost of the gas unit, represents the gas intake of the gas generator set at time t, v g represents the first penalty coefficient, represents the fuel demand of the second gas unit, represents the known fuel demand of the first gas unit, w g Represents the second penalty coefficient.

[0107] The first term of the second objective function is the total cost of natural gas production, the second term represents the total cost of gas extraction for gas turbines, and the third and fourth terms represent the penalty terms of the iterative convergence structure of the ATC algorithm. When the iteration converges, these two terms are approximately 0. The purpose of this objective function is to minimize the net production cost of the natural gas system. The second term is a negative value because the sale of natural gas by the natural gas system to the power system through gas turbines is regarded as the profit of the natural gas system. This setting can more strictly consider the production cost of the natural gas system, and at the same time make the decision of the natural gas system in supplying gas to the power system more optimistic, so as to provide more recovery capacity to the power system through gas turbines.

[0108] S32. Establish the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system, and the safety constraints of the natural gas system.

[0109] The pressure and flow state equations of the natural gas system are:

[0110]

[0111] In the formula, represents the mass flow rate, represents the coefficient of the Weymouth equation, π n,t represents the pressure at node n, π m , t represents the pressure of node m, represents the mass flow rate entering the pipe connecting nodes n and m at time t, represents the mass flow rate out of the pipe connecting nodes n and m at time t, represents the gas well set of node n, represents the gas load set of node n, represents the set of gas units of node n, represents the set of gas tanks of node n, represents the set of pipes connecting nodes n and m, Indicates the civil gas load, Indicates the storage flow rate of the gas tank. Indicates the output flow of the gas tank.

[0112] The pipeline storage equation of the natural gas system is:

[0113]

[0114] Where, L n,m,t represents the storage at time t, represents the pipe storage coefficient, L n,m,t-1 represents the storage at time t-1, L n,m,0 Indicates the storage at the initial moment.

[0115] The gas storage equation of the natural gas system is:

[0116]

[0117] In the formula, Indicates the amount of gas stored in the gas tank at time t. Indicates the amount of gas stored in the gas tank at time t-1, represents the input gas storage efficiency coefficient, Indicates the input gas storage mass flow rate, Indicates the gas mass flow rate, Represents the gas transmission efficiency coefficient.

[0118] The safety constraints of the natural gas system are:

[0119]

[0120] In the formula, q w Indicates the lower limit of gas production of gas wells. Indicates the upper limit of gas production of gas wells, π n Indicates the lower limit of pressure, Indicates the upper pressure limit, Indicates the upper limit of the gas tank storage flow rate. Indicates the upper limit of the gas tank output flow rate. Indicates the lower limit of gas storage in the gas tank. Indicates the upper limit of gas storage in the gas tank. Indicates the amount of gas stored in the gas tank at the beginning of the day, It indicates the amount of gas stored in the gas tank at the end of the day, that is, the amount of gas stored at the beginning and end of the day should remain the same.

[0121] S33. Generate a lower-layer natural gas system recovery model according to the second objective function, the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system, and the safety constraints of the natural gas system.

[0122] S4, using the analysis target cascade method to solve the reconstructed upper power system recovery model and the lower natural gas system recovery model to obtain the optimal restoration decision of the power system, specifically including S41-S42:

[0123] Among them, the ATC algorithm adopted is an iterative message passing algorithm. In the ATC algorithm, each subsystem has local control variables and control variables shared with other subsystems. The subsystem exchanges the values ​​of shared variables with the coupled system and solves the optimization subproblems with corresponding constraints in parallel. Once the values ​​of shared variables converge, the optimal solution of the original optimization problem is obtained. The objective functions of the upper power system recovery model and the lower natural gas system recovery model both adopt the augmented Lagrangian form. The purpose of using this form is to establish the iterative convergence structure of the model so that the ATC algorithm can be used to solve it.

[0124] S41, performing inner loop iteration on the reconstructed upper power system recovery model and the lower natural gas system recovery model until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition, and obtaining an inner loop iteration result, specifically including S411-S414:

[0125] S411. Solve the reconstructed upper-layer power system restoration model to obtain a first solution result.

[0126] S412. Acquire a known fuel demand of the first gas generator set from the first solution result, input the known fuel demand of the first gas generator set into the lower-layer natural gas system recovery model, and then solve the lower-layer natural gas system recovery model to obtain a second solution result.

[0127] S413. Obtain the known fuel demand of the second gas generator set from the second solution result, and input the known fuel demand of the second gas generator set into the reconstructed upper power system recovery model, and then return to execute S411 until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach the first convergence condition.

[0128] and The actual physical meaning of is the fuel demand of the gas unit. The two are physically equivalent, but because the fuel demand value is used as the boundary information exchanged between the power system and the natural gas system in the ATC algorithm, such as Figure 3 As shown, its representation symbols in different models are different. Specifically, in the iterative process of the inner loop, the power system side receives the gas system side. And use it as a given target value to optimize the response value And the solved The gas system then sends the electricity from the As a given target value to optimize And the new The data are sent back to the power system, and this process continues until the two systems reach a balance, that is, the reconstructed upper power system recovery model and the lower natural gas system recovery model reach the first convergence condition.

[0129] The reconstructed upper power system recovery model and the lower natural gas system recovery model reach the first convergence condition including:

[0130] It is determined whether the objective function gaps between the two previous and subsequent iterations of the reconstructed upper power system recovery model and the lower natural gas system recovery model are less than or equal to a second preset value; if so, it is determined that the reconstructed upper power system recovery model and the lower natural gas system recovery model have reached a first convergence condition; otherwise, it is determined that the reconstructed upper power system recovery model and the lower natural gas system recovery model have not reached the first convergence condition.

[0131]

[0132] In the formula, represents the first objective function value of the k+1th iteration of the inner loop, represents the first objective function value of the kth iteration of the inner loop, ε2 represents the second preset value, represents the second objective function value of the k+1th iteration of the inner loop, Represents the second objective function value of the kth iteration of the inner loop.

[0133] Specifically, if the above equation is satisfied, it is determined that the reconstructed upper power system recovery model and lower natural gas system recovery model meet the first convergence condition. However, this only means that the objective functions of the power system and the natural gas system are stable, that is, the shared variables in the iteration process are and The absolute value of the gap has basically stabilized, and the inner loop iteration can no longer be shortened and The absolute value of the difference cannot explain and The absolute value of the difference is already smaller than the first preset value, so the above formula is only a sufficient condition for the overall ATC algorithm to converge, and can only terminate the inner loop.

[0134] S414: Generate an inner loop iteration result according to the known fuel demand of the first gas generator set and the known fuel demand of the second gas generator set in the last iteration.

[0135] S42. Determine whether the inner loop iteration result meets the second convergence condition. If so, output the optimal recovery decision for the power system. If not, update the penalty coefficients in the reconstructed upper power system recovery model and the lower natural gas system recovery model. After updating, return to execute S41.

[0136] Wherein, judging whether the inner loop iteration result meets the second convergence condition includes:

[0137] Determine whether the absolute value of the difference between the known fuel demand of the first gas unit and the known fuel demand of the second gas unit in the last iteration is less than a first preset value; if so, determine that the inner loop iteration result meets the second convergence condition; if not, determine that the inner loop iteration result does not meet the second convergence condition.

[0138] The updating of the penalty coefficients in the reconstructed upper power system recovery model and the lower natural gas system recovery model is specifically:

[0139]

[0140] In the formula, represents the first penalty coefficient of the updated i+1th iteration, represents the second penalty coefficient of the updated i+1th iteration, represents the first penalty coefficient of the i-th iteration of the outer loop, represents the second penalty coefficient of the i-th iteration of the outer loop, λ represents the penalty multiplier update coefficient, and its value range is greater than 1. represents the known fuel demand of the first gas unit at the i-th iteration, represents the known fuel demand of the second gas unit at the i-th iteration.

[0141] The present invention focuses on the recovery state of the power system, that is, the startup of the black start power plant and the energization of the critical load. After solving the optimal recovery strategy of the power system, the starting power required for the non-black start unit to restart, the power system enters the post-recovery state. In the post-recovery state, the most important part of the power system has been restored, and the power system can reliably provide the critical load. In order to evaluate the recovery efficiency of the power system and the natural gas system in this process, in an optional embodiment, it also includes:

[0142] S5. Calculate the recovery index, specifically:

[0143]

[0144] In the formula, RI represents the recovery index, RI = 100 represents a fully effective system, RI = 0 represents a system with complete performance loss, t irIndicates the end time of the system in the restored state, t r It indicates the time it takes for the system to enter the recovery state after a failure. R(t) indicates the elasticity index. pr Represents the loss of resilience of the power grid.

[0145] The restoration index is calculated using the energization of critical loads and the starting power of non-black start units during the restoration process of the bulk power system calculated by the proposed model as indicators, and evaluates the resilience of the power system in the entire restoration state.

[0146] Please refer to Figure 2 , Embodiment 2 of the present invention is:

[0147] A distributed robust recovery terminal for an electric-gas integrated energy system comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, each step of the distributed robust recovery method for an electric-gas integrated energy system in Embodiment 1 is implemented.

[0148] In summary, the present invention provides a distributed robust recovery method and terminal for an electric-gas integrated energy system. The upper-level power system recovery model established takes into account the uncertainty of pumped storage and wind power, and reconstructs the upper-level power system recovery model based on the distributed robust model to obtain the reconstructed upper-level power system recovery model. The established lower-level natural gas system recovery model takes into account gas storage and pipeline storage, and the reconstructed upper-level power system recovery model and the lower-level natural gas system recovery model are solved using the analytical target cascade method to obtain the optimal recovery decision for the power system, which takes into account the impact of the natural gas network on the recovery process of the large-capacity power system, and also takes into account the influence of the pumped storage network on the recovery process of the large-capacity power system. The concurrent impact of water-storage hydropower stations and wind farms is taken into account, and the upper-level power system recovery model is reconstructed based on the distributed robust model, which avoids overly conservative recovery decisions. The analytical target cascade method is used to accurately calculate the equilibrium point of the proposed two-level optimization problem, and the privacy of the power network and the natural gas system is guaranteed, thereby improving the effectiveness and reliability of system recovery after failure. In addition, the first objective function of the upper-level power system recovery model is reconstructed based on the distributed robust model, which can simulate the uncertain behavior of wind farms through a distributed robust framework, making the recovery decision robust while avoiding overly conservative recovery decisions.

[0149] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A distributed robust recovery method for an electric-gas integrated energy system, characterized in that: Includes steps: Establish an upper-level power system restoration model based on the uncertainty of pumped storage and wind power; Reconstructing the upper power system restoration model based on the distributed robust model to obtain a reconstructed upper power system restoration model; Establish a lower-level natural gas system recovery model based on gas storage and pipeline storage; The reconstructed upper-layer power system recovery model and the lower-layer natural gas system recovery model are solved using an analytical target cascade method to obtain an optimal power system recovery decision.

2. A distributed robust recovery method for an electric-gas integrated energy system according to claim 1, characterized in that: The upper power system restoration model established based on pumped storage and wind power uncertainty includes: Establishing the first objective function with the goal of maximizing the post-fault recovery capability of the power system; Establish initial condition constraints, startup constraints for non-black start generators, generator constraints, pumped storage constraints, power system flow and power balance constraints, cold start load constraints, and dynamic reserve constraints; An upper-level power system recovery model is generated based on the first objective function, the initial condition constraints, the starting constraints of the non-black start generator sets, the generator set constraints, the pumped storage constraints, the power flow and power balance constraints of the power system, the cold start load constraints and the dynamic reserve constraints.

3. A distributed robust recovery method for an electric-gas integrated energy system according to claim 2, characterized in that: The upper power system restoration model is reconstructed based on the distributed robust model to obtain the reconstructed upper power system restoration model, which includes: The first objective function of the upper-layer power system restoration model is reconstructed based on a distributed robust model to obtain a reconstructed first objective function.

4. A distributed robust recovery method for an electric-gas integrated energy system according to claim 3, characterized in that: The reconstructed first objective function is: In the formula, Indicates that the probability distribution range of wind power is within the distributed robust model Θ, represents the expectation of wind power probability distribution, represents the upper limit of the active power of the traditional generator set g, represents the starting power of the traditional generator set g, represents the power generation state of the traditional generator set g at the recovery time t, represents the upper limit of the output power of the wind turbine w, u w,t represents the state of wind turbine w at recovery time t, α d represents the load priority factor, Indicates the maximum recoverable load, P d,t Represents the total load recovered within the recovery time t, Ω GFU represents the set of gas units, v g represents the first penalty coefficient, represents the fuel demand of the first gas unit, represents the known fuel demand of the second gas generator set, w g represents the second penalty coefficient, and ο represents the Hadamard product.

5. A distributed robust recovery method for an electric-gas integrated energy system according to claim 1, characterized in that: The lower layer natural gas system recovery model established based on gas storage and pipeline storage includes: Establishing a second objective function with the goal of minimizing the net production cost of the natural gas system; Establish the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system, and the safety constraints of the natural gas system; A lower-layer natural gas system recovery model is generated according to the second objective function, the pressure and flow state equations of the natural gas system, the pipeline storage equation of the natural gas system, the gas storage equation of the natural gas system, and the safety constraints of the natural gas system.

6. A distributed robust recovery method for an electric-gas integrated energy system according to claim 5, characterized in that: The second objective function established with the goal of minimizing the net production cost of the natural gas system is: In the formula, C r GW represents the gas well production cost, Indicates the gas production of the gas well, Ω GFU represents the collection of gas units, represents the gas extraction cost of the gas unit, represents the gas intake of the gas generator set at time t, v g represents the first penalty coefficient, represents the fuel demand of the second gas generator set, represents the known fuel demand of the first gas unit, w g represents the second penalty coefficient, represents the Haldmann product.

7. A distributed robust recovery method for an electric-gas integrated energy system according to claim 1, characterized in that: The use of the analytical target cascade method to solve the reconstructed upper power system recovery model and the lower natural gas system recovery model to obtain the optimal power system recovery decision includes: Performing inner loop iteration on the reconstructed upper power system recovery model and the lower natural gas system recovery model until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition, thereby obtaining an inner loop iteration result; Determine whether the inner loop iteration result meets the second convergence condition. If so, output the optimal power system recovery decision. If not, update the penalty coefficients in the reconstructed upper power system recovery model and the lower natural gas system recovery model. After updating, return to execute the inner loop iteration step of the reconstructed upper power system recovery model and the lower natural gas system recovery model.

8. A distributed robust recovery method for an electric-gas integrated energy system according to claim 7, characterized in that: The inner loop iteration of the reconstructed upper power system recovery model and the lower natural gas system recovery model is performed until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition, and the inner loop iteration result is obtained, including: Solving the reconstructed upper-layer power system restoration model to obtain a first solution result; Obtaining a known fuel demand of the first gas generator set from the first solution result, and inputting the known fuel demand of the first gas generator set into the lower-layer natural gas system recovery model, and then solving the lower-layer natural gas system recovery model to obtain a second solution result; Obtaining a known fuel demand of the second gas generator set from the second solution result, and inputting the known fuel demand of the second gas generator set into the reconstructed upper power system recovery model, returning to execute the step of solving the reconstructed upper power system recovery model until the reconstructed upper power system recovery model and the lower natural gas system recovery model reach a first convergence condition; An inner loop iteration result is generated according to the known fuel requirement of the first gas engine group and the known fuel requirement of the second gas engine group in the last iteration.

9. A distributed robust recovery method for an electric-gas integrated energy system according to claim 8, characterized in that: The determining whether the inner loop iteration result meets the second convergence condition comprises: Determine whether the absolute value of the difference between the known fuel demand of the first gas unit and the known fuel demand of the second gas unit in the last iteration is less than a first preset value; if so, determine that the inner loop iteration result meets the second convergence condition; if not, determine that the inner loop iteration result does not meet the second convergence condition.

10. A distributed robust recovery terminal for an electric-gas integrated energy system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements each step of a distributed robust recovery method for an electric-gas integrated energy system as described in any one of claims 1 to 9.