Comprehensive energy system operation toughness calculation method for coping with influence of main shock and aftershock

By using polynomial chaotic expansion and influence incremental state enumeration method in an integrated energy system, the mapping relationship between the topology and load state of IES and the optimal load reduction is solved, and the toughness calculation problem in earthquakes, especially aftershocks, is solved. Fast and accurate resilience assessment and optimization are achieved, and the recovery capability and computing efficiency of the energy system are improved.

CN119004714APending Publication Date: 2024-11-22NORTHEAST DIANLI UNIVERSITY
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411010803.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The prior art is difficult to accurately calculate the resilience of the integrated energy system in earthquake disasters, especially ignoring the impact of aftershocks on the system, resulting in the inability to quickly recover and optimize energy supply, and the calculation efficiency is low.

Method used

Using a comprehensive energy system operation toughness calculation method based on arbitrary polynomial chaotic expansion (aPCE), combined with the influence incremental state enumeration method (IISE), the mapping relationship between IES topology, load state and optimal load reduction is constructed, and the polynomial is used to construct the mapping relationship between IES topology, load state and optimal load reduction. Computation replaces the complex optimization solution process and quickly calculates the impact of main aftershocks on IES.

Benefits of technology

The accuracy and rapidity of IES resilience calculations in high-impact and high-deterministic earthquake disasters are achieved, and the calculation time is shortened, real-time emergency response support is provided, and economic losses from energy disruptions and equipment damage are reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119004714A_ABST
    Figure CN119004714A_ABST
Patent Text Reader

Abstract

The invention discloses a comprehensive energy system operation toughness calculation method for coping with main and aftershock influences, and the offline solving of the optimal load reduction amount of an IES comprises the steps: calculating the univariate optimal orthogonal basis functions of node self-susceptance, a gas pipeline state variable, a power grid and a gas network load level, and carrying out the chaotic combination of the univariate optimal orthogonal basis functions, keeping a polynomial of which the highest order does not exceed PPCE as an optimal orthogonal basis function in the aPCE expression; on the basis of the optimal orthogonal basis function, V groups of random input variable samples Xsample and the optimal load reduction # imgabs0 # are substituted to calculate an approximation coefficient, and an aPCE expression of the optimal load reduction in the k-order fault state is obtained; the calculation of the operation toughness of the IES comprises the following steps: enumerating all k-order fault states occurring in the IES, and substituting a random input variable of each state into an aPCE expression of the optimal load reduction quantity to obtain the optimal load reduction quantity Lshed, l, s of the IES; and calculating the fault probability of the power transmission line and the gas pipeline in the IES according to the magnitude, and calculating the toughness index of the IES according to the fault probability and the optimal load reduction amount of the IES.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of resilience calculation of integrated energy systems, and in particular to a method for calculating the operational resilience of integrated energy systems that cope with the impact of main shocks and aftershocks. Background Art

[0002] Integrated Energy System (IES) can scientifically improve energy utilization and operational economy through the coordinated optimization of multiple types of energy, and is one of the typical ways of future energy supply. However, the deep coupling of heterogeneous energy flows also brings greater risks to the operational safety of IES. In recent years, the frequent occurrence of extreme weather and natural disasters has become more and more obvious. The system failures caused by them can easily spread across the energy chain through coupling elements, further expanding the scope and impact of failures. Therefore, in the context of the deepening degree of multi-energy coupling, it is of great significance to calculate the IES resilience affected by extreme disasters.

[0003] At present, resilience calculations are mostly aimed at surface disasters such as typhoons, ice disasters, and floods. Since they have little impact on buried gas pipelines, they usually only focus on the one-way impact of power system failures on natural gas systems. However, geological deformation and soil liquefaction caused by earthquakes will not only cause direct physical damage such as the collapse of power towers and the rupture of natural gas pipelines; at the same time, the rupture of natural gas pipelines will also lead to insufficient gas supply or shutdown of natural gas pipeline networks, directly affecting the normal operation of natural gas power plants and exacerbating the shortage of power supply. Therefore, the interactive impact of multi-energy system failures must also be considered in resilience calculations. However, existing literatures all simulate and analyze the fault status offline before the earthquake occurs, which cannot truly reflect the resilience of IES during the earthquake. In addition, the calculation process only considers the impact of the main shock with the largest energy in the earthquake sequence on IES, ignoring the impact of aftershocks on the system. Since earthquake disasters have a certain temporal sequence, multiple aftershocks with strong uncertainty may occur in succession after a strong earthquake, which may cause further damage or failure of IES, triggering a larger range of energy supply interruptions, and under the influence of aftershocks, load fluctuations may be more severe, and the impact on the system cannot be ignored. With the help of earthquake warning information and aftershock probability prediction data, the resilience level of IES is calculated online to provide real-time technical support for emergency response. The recovery plan is dynamically adjusted as aftershocks occur, key energy facilities and important energy channels are quickly restored, and the scope of energy supply interruption is minimized. Therefore, under the continued influence of high-impact and high-uncertainty earthquake disasters, it is particularly important to study the calculation of IES operation resilience.

[0004] Operational resilience calculation is used to measure the ability of IES to resist disaster impact and quickly recover to normal function in a short period of time under the influence of real-time operation status and external environment. Due to the wide impact range and strong uncertainty of earthquake disasters, the number of fault states is growing exponentially, which makes it impossible to quickly solve the optimal energy flow (OEF) under fault conditions, and thus cannot meet the timeliness requirements of IES operational resilience calculation under earthquake disasters. At present, the methods to improve the efficiency of resilience calculation are mainly divided into two aspects: reducing the number of states and improving the calculation speed of OEF. In terms of reducing the number of states, the main methods include system information entropy, K-means++ clustering algorithm, and scenario generation method based on slice sampling. Although the above methods can reduce the number of states, they may ignore some important states, especially the fault states with "small probability and large impact" under disasters. Similar to optimal power flow, OEF can also be used to optimize power generation scheduling or minimize load reduction under specific objective functions. In terms of improving calculation efficiency, the main methods include parametric linear programming theory, sequence programming method with adaptive convergence criteria, semi-invariant method, deep learning, etc. The use of a sequence programming method based on parametric linear programming theory and an adaptive convergence criterion requires multiple solutions to the OEF. The semi-invariant method is used to improve the efficiency of the OEF calculation considering the uncertainty of renewable energy output and load, but this method cannot consider the changes in system topology caused by random failures. Deep learning is used to form a mapping relationship between topology, source-load uncertainty and optimal load reduction to improve the efficiency of the OEF calculation, but the model has no clear expression and lacks interpretability.

[0005] Therefore, in response to the above-mentioned difficulties, it is particularly important to study and develop a method for calculating the operational resilience of an integrated energy system based on polynomial chaos expansion under the influence of main shocks and aftershocks. Summary of the invention

[0006] The present invention provides a method for calculating the operational resilience of an integrated energy system in response to the impact of main and aftershocks. The present invention considers the continuous impact of earthquake disasters on IES and proposes a framework for calculating the operational resilience of IES under the impact of main and aftershocks. The present invention constructs an IES optimal load reduction calculation method based on arbitrary polynomial chaos expansion (aPCE), uses aPCE expressions to construct a mapping relationship between IES topology, load state and optimal load reduction, and replaces the complex optimization solution process with polynomial calculation. The aPCE is combined with the Impact-Increments-based State Enumeration (IISE) method to quickly calculate the operational resilience index, quantify the impact of main and aftershocks on IES resilience, and can greatly improve the calculation efficiency of the resilience of the integrated energy system under the premise of ensuring the calculation accuracy, and can realize accurate and rapid calculation of the resilience of the integrated energy system under the impact of main and aftershocks. See the following description for details:

[0007] A method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks, the method comprising: offline solving IES optimal load reduction and IES operational resilience calculation,

[0008] The offline solution of IES optimal load reduction includes:

[0009] Calculate the single variable optimal orthogonal basis functions of node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels, and perform chaotic combination on the single variable optimal orthogonal basis functions, retaining the highest order not exceeding P PCE The polynomial is used as the optimal orthogonal basis function in the aPCE expression; based on the optimal orthogonal basis function, V groups of random input variable samples X are introduced. sample and optimal load reduction Calculate the approximation coefficient and obtain the aPCE expression of the optimal load reduction amount for the j-order fault state;

[0010] The IES operational resilience calculation includes:

[0011] Enumerate all j-order fault states that occur in IES, and substitute the random input variables of each state into the aPCE expression of the optimal load reduction to obtain the optimal load reduction L of IES shed,l,s ; Calculate the failure probability of power transmission lines and gas pipelines in IES according to the earthquake magnitude, and calculate the IES resilience index according to the failure probability and the IES optimal load reduction.

[0012] Among them, the V groups of random input variable samples X sample The acquisition is:

[0013] From the j-order topological state set V j-order topological states are extracted from the grid, and combined with V load levels to obtain V j-order fault states. The node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels in each state are extracted to obtain V groups of random input variable samples X representing the j-order fault state. sample .

[0014] The optimal load reduction amount is For V groups of random input variable samples X sample Each group of samples in performs the IES optimal load reduction calculation and obtains V corresponding IES optimal load reduction amounts.

[0015] The IES resilience index is calculated according to the failure probability and the IES optimal load reduction amount as follows: the failure probability and the IES optimal load reduction amount are substituted into the following formula to calculate and obtain the IES resilience index;

[0016]

[0017] Among them, N order is the maximum fault order; N level is the number of load levels considered; P l is the probability of the lth load level; is the set of j-order topological states enumerated; ΔP s is the modified probability of topological state s; ΔL shed,l,s is the influence increment of topological state s under the lth load level; n s is the number of fault states representing the topological state s; L shed,l,r is the j-order state set of topological state s The optimal load reduction amount of topological state r under the lth load level in N PCE is the total number of optimal orthogonal basis functions; c w,l,r is the wth approximation coefficient of the optimal load reduction of topological state r under the lth load level; φ w,l,r is the wth optimal orthogonal basis function of the optimal load reduction of topological state r under the lth load level.

[0018] Furthermore, the IES operational resilience calculation uses magnitude, intensity, and earthquake source, combined with the system operating status and topological structure to calculate the IES, and calculates the resilience index of the period based on the probability prediction of aftershocks in the future T period and the failures that have occurred in the current system, and updates it once every ΔT.

[0019] The calculation of IES is specifically as follows: according to the predicted aftershock sequence in the future T period, the failure probability of the power transmission line and the gas pipeline under the influence of each aftershock is calculated, and the probability of each component failing at least once in the future T period is solved, and then combined with the probability of aftershock occurrence, the probability of failure in the future T period is obtained. and It is updated as the predicted aftershock sequence and system conditions change.

[0020] Wherein, the optimal orthogonal basis function is:

[0021]

[0022] Where: p e is the order of the optimal orthogonal polynomial for the power grid load level; p g is the order of the optimal orthogonal polynomial for the gas grid load level; p i is the order of the optimal orthogonal basis function of the self-susceptance of the i-th node; p m is the order of the optimal orthogonal polynomial of the mth gas pipeline state variable, is the grid load level p e Order univariate optimal orthogonal polynomial; is the gas grid load level p g Order univariate optimal orthogonal polynomial; is the self-susceptance of the ith node p i Order univariate optimal orthogonal polynomial; is the p of the mth gas pipeline state variable m The optimal orthogonal polynomial of univariate p. e 、p g 、p i and p m satisfy:

[0023]

[0024] Among them, N bus is the number of grid nodes; N pipe is the number of pipelines; P PCE is the highest expansion order of aPCE expression; p w is the order of the w-th optimal orthogonal basis function.

[0025] The beneficial effects of the technical solution provided by the present invention are:

[0026] 1. The present invention considers the multiple impacts that the main shock and aftershocks with time sequence may have on IES resilience, can more comprehensively and accurately identify various potential risks faced by IES, and uses a rolling method to calculate IES operational resilience, which can provide reliable data support for grid operators to reasonably allocate emergency resources and formulate emergency repair strategies, shorten power outage time and recovery time; avoid system overload or a larger range of cascading failures, and reduce economic losses caused by energy interruptions and equipment damage caused by earthquakes; it can also provide important data support for long-term energy system planning and promote the construction of safer, more reliable and sustainable energy infrastructure;

[0027] 2. The optimal load reduction calculation method based on aPCE proposed in the present invention constructs a mapping relationship between IES topology, load state and optimal load reduction, which can effectively handle the impact of topology and load uncertainty on the optimal load reduction. It replaces the traditional OEF optimization solution with polynomial calculation, which can significantly reduce the calculation complexity. While taking into account the calculation accuracy, it greatly shortens the calculation time and realizes the rapid calculation of IES operation resilience in a large number of fault scenarios.

[0028] 3. The present invention combines IISE with aPCE to improve the efficiency of resilience calculation from two perspectives: reducing the number of fault states and increasing the speed of OEF calculation. It can realize accurate and rapid calculation of IES resilience under the continuous influence of high-impact and high-uncertainty earthquake disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Flowchart of aPCE expression for offline calculation of IES optimal load reduction;

[0030] Figure 2 Flowchart for running resilience calculations for IES;

[0031] Figure 3 This is the framework diagram of IES operation resilience calculation under the influence of the main aftershock;

[0032] Figure 4 This is a schematic diagram of the IES E24-G14 test system;

[0033] Figure 5 Schematic diagram of the IES E24-G14 test system under the influence of a main shock;

[0034] Figure 6 Comparison chart of calculation error and time consumption of IES E24-G14 test system under the influence of main shock;

[0035] Figure 7 A sequence diagram for predicting aftershocks;

[0036] Figure 8This is the operational resilience change curve of the IES E24-G14 test system under the influence of the main aftershock. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention are described in further detail below.

[0038] Firstly, a calculation framework for IES operational resilience under the influence of mainshocks and aftershocks is proposed. Then, to address the problem that the traditional optimal load reduction calculation method is difficult to meet the calculation efficiency requirements of IES operational resilience under the influence of earthquakes, an optimal load reduction calculation method for IES based on aPCE is constructed. The mapping relationship between IES topology, load state and optimal load reduction is constructed using aPCE expression, and polynomial calculation is used to replace the complex optimization solution process. Furthermore, aPCE is combined with IISE to quickly calculate the operational resilience calculation index, and the impact of mainshocks and aftershocks on IES resilience is quantified.

[0039] Example 1

[0040] A method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks, the method comprising: offline solution of IES optimal load reduction and IES operational resilience calculation.

[0041] Part 1: Offline solution for IES optimal load reduction

[0042] Extract V groups of j-order fault states, extract the node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels in each state to obtain V groups of random input variable samples X sample , and calculate the corresponding IES optimal load reduction for each group of samples

[0043] Calculate the optimal orthogonal basis functions of the node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels, and perform chaotic combination on them, retaining the highest order not exceeding P PCE The polynomial of is used as the optimal orthogonal basis function in the aPCE expression;

[0044] X sample and Substitute it into the formula to calculate the approximation coefficient, and then obtain the aPCE expression of the optimal load reduction amount in the j-order fault state;

[0045] Part II: IES Operational Resilience Calculation

[0046] Enumerate all possible j-order fault states in IES, and substitute the random input variables of each state into the aPCE expression of optimal load reduction to obtain the optimal load reduction amount L of IES shed,l,s ;

[0047] The failure probability of power transmission lines and gas pipelines in IES is calculated according to the earthquake magnitude, and the failure probability and the IES optimal load reduction are substituted into the calculation formula to calculate and output the IES resilience index.

[0048] In summary, the embodiments of the present invention avoid system overload or larger-scale cascading failures, reduce economic losses caused by energy outages and equipment damage due to earthquakes; and provide important data support for long-term energy system planning, promoting the construction of safer, more reliable and sustainable energy infrastructure.

[0049] Example 2

[0050] The embodiment of the present invention provides an IES operation resilience calculation method based on aPCE, which is divided into two parts: offline solution of the aPCE expression of the IES optimal load reduction and IES operation resilience calculation. Figure 1 and Figure 2 , the implementation mode of the present invention is further described in detail below, wherein 101-107 are detailed steps for offline solving the aPCE expression of the IES optimal load reduction; 108-115 are detailed steps for IES operation resilience calculation.

[0051] See also Figure 1 , 101-107 are the detailed steps for offline solving the aPCE expression of IES optimal load reduction as follows:

[0052] 101: Input IES subsystem data, historical load data, and maximum fault order N order , initialize the fault order j = 1;

[0053] 102: From the j-order topological state set V j-order topological states are extracted from the grid, and combined with V load levels to obtain V j-order fault states. The node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels in each state are extracted to obtain V groups of random input variable samples X representing the j-order fault state. sample ;

[0054] 103: Obtain the single variable optimal orthogonal basis functions of node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels;

[0055] 104: Perform chaotic combination of the single variable optimal orthogonal basis functions of node self-susceptance, gas pipeline state variables, power grid, and gas grid load level and only retain the highest order not exceeding P PCE The polynomial of is used as the optimal orthogonal basis function in the aPCE expression;

[0056] 105: Randomly input variable sample X to V groups sampleEach group of samples in performs the IES optimal load reduction calculation and obtains V corresponding IES optimal load reduction amounts.

[0057] 106: X sample and Substitute it into the formula to calculate the approximation coefficient, and then obtain the aPCE expression of the optimal load reduction amount in the j-order fault state;

[0058] 107: Check whether all fault orders (i.e. j = N) have been analyzed. order ), if yes, the calculation ends, otherwise, j=j+1, and execute step 102;

[0059] See also Figure 2 , 108-115 are the detailed steps for IES to run toughness calculation:

[0060] 108: Input IES subsystem data, load data, and maximum fault order N order , real-time data of earthquake disasters such as magnitude, intensity, and epicenter distance;

[0061] 109: Enumerate all possible j-order fault states in IES;

[0062] 110: Select the a-th j-th fault state, extract the node self-susceptance, gas pipeline state variables, and power grid and gas grid load levels of the state as random input variables;

[0063] 111: Substitute the random input variables into the aPCE expression of optimal load reduction to obtain the optimal load reduction amount L of IES shed,l,s ;

[0064] 112: Check whether the optimal load reduction amount is calculated for all j-order fault states. If yes, initialize a=1 and execute step 106. Otherwise, a=a+1 and execute step 113.

[0065] 113: Check whether all fault orders (i.e. j = N) have been analyzed. order ), if yes, execute step 107, otherwise, j=j+1, execute step 112;

[0066] 114: Calculate the failure probability of power transmission lines and gas pipelines in IES according to the magnitude of the earthquake;

[0067] 115: Substitute the IES optimal load reduction and component failure probability into equation (35) to calculate and output the IES resilience index.

[0068] In summary, the embodiment of the present invention can greatly improve the calculation efficiency of the resilience of the integrated energy system under the premise of ensuring the calculation accuracy through the above steps 101-115, and can realize accurate and rapid calculation of the resilience of the integrated energy system under the influence of main shocks and aftershocks.

[0069] Example 3

[0070] The present invention introduces the IES operation resilience calculation framework under the influence of main and aftershocks, see Figure 3 , the framework calculates the operational resilience of IES online in a rolling manner to respond in real time to the impact of earthquakes and their subsequent aftershocks on the resilience of IES. When the seismic detection instruments deep underground detect non-destructive longitudinal waves, the early warning system sends an earthquake alarm and quickly transmits real-time earthquake data to the computer for data processing. The framework uses relevant parameters such as magnitude, intensity, and earthquake source, combined with the system operation status and topological structure to calculate the IES, and then calculates the resilience index of the period based on the probability prediction of aftershocks in the future T period and the faults that have occurred in the current system, and updates it every ΔT. Over time, the calculation framework can provide continuous online resilience operation calculations, thereby assisting power grid operators to formulate more effective response measures and optimize resource allocation to mitigate the further impact of earthquakes and aftershocks on IES. The proposed operational resilience calculation framework is mainly divided into four parts: earthquake disaster impact, fault state generation, fault state analysis, and resilience index calculation.

[0071] 1) Impact of earthquake disasters. The failure probability of transmission lines under the influence of earthquakes is determined by the transmission towers, and its calculation formula is:

[0072]

[0073] Where: is the failure probability of the oth transmission tower of the dth transmission line; N tower is the number of towers on the dth transmission line.

[0074] Gas pipelines may cross multiple seismic intensity zones. Therefore, the pipeline is divided into several elementary pipelines with sufficiently small lengths. The failure probability of the gas pipeline is determined by the elementary pipelines, and its calculation formula is:

[0075]

[0076] Where: is the failure probability of the nth section of the mth gas pipeline; N unit is the number of element pipelines of the mth gas pipeline.

[0077] In real-time calculation, according to the predicted aftershock sequence in the future T period, the failure probability of power transmission lines and gas pipelines under the influence of each aftershock is calculated, and then the probability of each component failing at least once in the future T period is solved by using it. Combined with the probability of aftershock occurrence, the probability of failure in the future T period is obtained. and It is updated as the predicted aftershock sequence and system conditions change.

[0078] 2) Fault state selection. A fault state consists of topology state and load level. To reduce the number of fault states, the maximum fault order N is set. order , and obtain the j-order topological state set And combined with the load level to generate all possible j-order fault states.

[0079] 3) Fault state analysis: According to the current topological state and load conditions of the system, the OEF model with the minimum load reduction as the objective function is solved for all possible fault states in IES one by one, and recalculated as the fault state changes.

[0080] 4) Calculation of toughness index. In order to further improve the efficiency and accuracy of toughness calculation, the toughness index based on IISE is used to quantify the toughness level of IES:

[0081]

[0082] Where: N level is the number of load levels considered; P l is the probability of the first load level; N order is the maximum fault order enumerated; is the set of j-order topological states enumerated; ΔP s is the modified probability of topological state s; ΔL shed,l,s is the impact increment of topological state s under the ath fault state, that is, the lth grid load level.

[0083]

[0084] Where: n s is the number of fault states representing the topological state s; is a k-order subset of s.

[0085] The modified probability ΔP of the topological state s s From the faulty power transmission line and gas pipeline in this state, we can get:

[0086]

[0087] Where: is the failure probability of the dth transmission line; The probability of failure of the mth gas pipeline.

[0088] As the scale of the system continues to expand, even if the state is reduced, the number of fault states to be calculated is still very large. The traditional model-driven method is used to solve the optimal load reduction under a large number of fault states, which is difficult to meet the timeliness requirements of IES resilience calculation. Therefore, considering the uncertainty of system topology and load under the influence of earthquakes, an IES operational resilience calculation method based on aPCE is proposed. The aPCE expression is used to fit the complex nonlinear relationship between system topology, load state and optimal load reduction, and the IES optimal load reduction is quickly output to achieve rapid calculation of IES operational resilience.

[0089] Example 4

[0090] The scheme in Example 1 is further introduced below in combination with a specific calculation formula and Example 3, as described below for details:

[0091] 201: Basic principles of chaotic expansion of arbitrary polynomials;

[0092] Consider a Z-dimensional random input variable ξ=[ξ1,ξ2,…,ξ Z ], the aPCE expression of the output response Y of the function is:

[0093]

[0094] Where: N PCE is the total number of optimal orthogonal basis functions; c w is the w-th approximation coefficient in the aPCE expression; φ w (ξ) is the w-th optimal orthogonal basis function of the Z-dimensional random input variable, which is the chaotic combination of the single variable optimal orthogonal basis function of each random input variable:

[0095]

[0096] Where: (h) (ξ z ) represents the z The related h-order univariate orthogonal polynomial. The w-th optimal orthogonal basis function order p w The order p of the optimal orthogonal basis function of a single variable z The following relationship exists:

[0097] 0≤p w =p1+p2+…+p Z ≤P PCE (8)

[0098] Where: P PCE is the highest expansion order of the aPCE expression.

[0099] Therefore, the Z-dimensional random input variable is expanded to P PCE In the aPCE expression of order, the total number of optimal orthogonal basis functions is:

[0100]

[0101] aPCE is applicable to random input variables of any distribution. Therefore, the optimal orthogonal basis function of each random input variable needs to be constructed based on its sample data. Randomly select V groups of Z-dimensional random input variable samples According to formula (6), the output response of the function is calculated for each group of samples Variable ξ z The h-order univariate optimal orthogonal basis function is:

[0102]

[0103] Where: is the u-order coefficient of the h-order univariate optimal orthogonal basis function of the z-th random input variable.

[0104] After obtaining the optimal orthogonal basis function of the single variable of the Z-dimensional random input variable, the optimal orthogonal basis functions of each order of the aPCE expression are constructed by formula (7). The approximation coefficients of each order in the aPCE expression of formula (6) can be solved by the least squares method (LS):

[0105]

[0106] In the formula, is the vth group of samples of the input random variable, is the vth response value calculated by the original model Y=f(ξ).

[0107] 202: Optimal load reduction calculation method based on aPCE;

[0108] (1) Random variable selection

[0109] The uncertainty of transmission line and gas pipeline faults will lead to IES topology uncertainty, which in turn leads to uncertainty in the optimal load reduction of IES. At the same time, the load uncertainty of the power grid and gas grid will also affect the optimal load reduction of IES under the same topology state. Therefore, it is necessary to convert topology uncertainty and load uncertainty into random input variables.

[0110] a. Transmission line fault uncertainty. The node self-susceptance B is selected as the random input variable to characterize the transmission line fault uncertainty in IES:

[0111]

[0112] Where: B i is the self-admittance of node i; N bus is the number of grid nodes.

[0113] b. Uncertainty of gas pipeline failure. 0 and 1 are used to represent pipeline failure and operation status:

[0114]

[0115] Where: X 01,m represents the fault condition of the mth gas pipeline, X 01,m =1 indicates that the corresponding gas pipeline is in normal state. 01,m =0 indicates that the corresponding gas pipeline is in fault state; N pipe is the number of pipelines.

[0116] c. Load uncertainty. Using the grid load level L e and gas grid load level L g As a random input variable that characterizes the uncertainty of load levels.

[0117] Therefore, the random input variable that characterizes the IES fault state is:

[0118] X=[B,X 01 ,L e ,L g ] (14)

[0119] From the j-order topological state set V j-order topological states are extracted from the V load levels to obtain V j-order fault states. B and X in each state are extracted. 01 , L e and L g Then we can get V groups of random input variable samples X representing the j-order fault state. sample :

[0120]

[0121] (2) Construction of optimal orthogonal basis functions

[0122] aPCE uses the sample value statistical moments of random input variables that characterize the IES fault state to construct the single variable optimal orthogonal basis function, and then performs chaotic combination to obtain the optimal orthogonal basis function in the expression.

[0123] Next, we take the self-susceptance of the i-th node B i As an example, the calculation process of the optimal orthogonal basis function of a single variable is introduced. Assuming that the highest expansion order of the polynomial is P PCE , then the expansion order of the univariate optimal orthogonal basis function is h=0,1,…,P PCEAccording to formula (10), B i The h-order univariate optimal orthogonal basis function is:

[0124]

[0125] Where: For B i The u-order term coefficient of the h-order univariate optimal orthogonal basis function is the quantity to be determined; For B i The u power.

[0126] When h = 0, B i The 0th-order univariate optimal orthogonal basis function has only one term, and its coefficient The calculation expression is:

[0127]

[0128] Obviously, B i The 0th-order univariate optimal orthogonal basis function is:

[0129]

[0130] When h = 1, B i The first-order univariate optimal orthogonal basis function has 0th-order terms and 1st-order terms, and its coefficients and The calculation expression is:

[0131]

[0132] Where: B,i,0 and λ B,i,1 B i The 0th and 1st order statistical moments of the V group samples are:

[0133]

[0134] Calculated by equations (19) and (20) and After that, we can get B i The first-order univariate optimal orthogonal basis function is:

[0135]

[0136] When h=P PCE When B i P PCE The optimal orthogonal basis functions of univariate order range from 0 to (P PCE +1) order terms, the coefficient calculation expressions are:

[0137]

[0138] Where: For B i The V group of samples is 0 to (2P PCE -1) order statistical moment is:

[0139]

[0140] Through equations (22) and (23), (P PCE +1) coefficients. Therefore, B i P PCE The optimal orthogonal basis function of univariate order is:

[0141]

[0142] Similarly, taking the mth gas pipeline state variable X 01,m For variables from 0 to P PCE The optimal orthogonal basis function of univariate order is:

[0143]

[0144] The grid load level L e For variables from 0 to P PCE The optimal orthogonal basis function of univariate order is:

[0145]

[0146] Based on the gas grid load level L g For variables from 0 to P PCE The optimal orthogonal basis function of univariate order is:

[0147]

[0148] According to formula (7), the random input variables from 0 to P are PCE The optimal orthogonal basis functions of the single-variable order are combined chaotically, and the optimal orthogonal basis functions of each item in the aPCE expression of the IES optimal load reduction are obtained as follows:

[0149]

[0150] Where: p e is the order of the optimal orthogonal polynomial for the power grid load level; p g is the order of the optimal orthogonal polynomial for the gas grid load level; p i is the order of the optimal orthogonal basis function of the self-susceptance of the i-th node; p m is the order of the optimal orthogonal polynomial of the mth gas pipeline state variable, p e 、p g 、pi and p m satisfy:

[0151]

[0152] (3) Solving the approximation coefficient

[0153] The optimal load reduction is calculated once for each set of random input variable samples representing the IES fault state, and finally the optimal load reduction of IES under the influence of V earthquakes is obtained.

[0154]

[0155] According to formula (11), the approximation coefficients of each order in the aPCE expression of the optimal load reduction model are:

[0156]

[0157] Where: is the vth group of fault state vector samples; is the optimal load reduction of the vth IES.

[0158] The least squares solution of formula (31) is:

[0159]

[0160] Where: Φ is V groups of random input variable samples substituted into N PCE V×N optimal orthogonal basis functions PCE matrix

[0161]

[0162] After obtaining the approximation coefficients of each order, the optimal orthogonal basis function constructed according to formula (28) can be obtained as follows:

[0163]

[0164] Substituting formula (34) into formula (3), the IES operation resilience index based on aPCE is obtained as follows:

[0165]

[0166] Compared with formula (3), the IES optimal load reduction calculation model in formula (35) is replaced by the aPCE expression. When the system topology state and load level are known, the optimal load reduction can be output by substituting the random input variables into the aPCE expression without the need for OEF optimization solution, which reduces the calculation complexity and greatly improves the resilience calculation efficiency of IES.

[0167] Example 5

[0168] In the following, a method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks proposed in the embodiment of the present invention is combined with a specific example. In this example, the accuracy and effectiveness of the proposed method are verified by using the IES E24-G14 test system consisting of the IEEE RTS 79 power grid and the 14-node natural gas grid. There are 4 gas-fired power plants in the test system, accounting for 41.06% of the total installed capacity. Figure 4 Schematic diagram of the IES E24-G14 test system.

[0169] (1) Calculation of the optimal load reduction model based on aPCE

[0170] In order to take into account the complexity and accuracy of the model and ensure good fitting ability, the highest expansion order of the aPCE expression of the optimal load reduction of the IES E24-G14 test system is set to P PCE =3, the number of node self-susceptances representing the uncertainty of transmission line faults is 24, the number of gas pipeline state variables representing the uncertainty of gas pipeline faults is 12, and the number of grid load levels and gas grid load levels representing load fluctuations is 1 each. Therefore, the optimal load reduction aPCE expression of the IES E24-G14 test system has a total of 38 random input variables, and the optimal number of orthogonal basis functions is:

[0171]

[0172] Taking the first random input variable, i.e., the first node self-susceptance B1, in the aPCE expression of the first-order optimal load reduction of the IES E24-G14 test system as an example, the solution process of the single variable optimal orthogonal basis function is introduced. PCE =3, therefore, we first need to calculate the statistical moments of the sample values ​​from 0 to 5, and the results are shown in Table 1.

[0173] Table 1B1 Statistical moments of various orders

[0174]

[0175] The obtained 0 to 5 order statistical moments are used to solve the coefficients of the 0 to 3 order single variable optimal orthogonal basis functions of B1. The results are shown in Table 2, where h is the order of the single variable optimal orthogonal basis function, are the 0th, 1st, 2nd and 3rd order coefficients in the h-order univariate optimal orthogonal basis function respectively.

[0176] Table 2B1 optimal orthogonal basis function coefficients of each order of single variable

[0177]

[0178] According to equations (18), (21) and (24), we can obtain the optimal orthogonal basis functions of B1 from 0 to 3 orders. The solution process of other optimal orthogonal basis functions is similar to the above process and will not be repeated here. After obtaining the optimal orthogonal basis functions of each random input variable from 0 to 3 orders, the chaotic combination is used to obtain the 10660 optimal orthogonal basis functions φ in the aPCE expression. w (X). Then for each optimal orthogonal basis function φ w (X) Solve for the approximation coefficient c w , and the aPCE expression of the first-order optimal load reduction of the IES E24-G14 test system is obtained.

[0179] (2) Calculation of IES operational resilience under the influence of the main shock

[0180] Assume that the magnitude of the main shock is M=6 and the focal coordinates are (27.5,110). Figure 5 This is a schematic diagram of the IES E24-G14 test system under the influence of a main shock. The left picture shows the power subsystem, and the right picture shows the natural gas subsystem. The two systems overlap in space.

[0181] Figure 5 In the left figure, the earthquake source is taken as the center and the interval is 5 km, which is divided into 35 intervals. Four equal points are taken in the interval in turn. The failure probability is calculated according to the seismic intensity at the equal points and the average value is taken as the failure probability of the transmission tower in the interval; in the right figure, it is divided into three earthquake intensity areas of 6 degrees, 7 degrees, and 8 degrees. The failure probability is calculated based on the regional intensity as the failure probability of the element pipeline in the area. The failure probability of the element pipeline outside the three intensity areas is 0.

[0182] Assuming that the distance between towers is d = 500m and the length of the element pipeline is ΔL = 500m, the failure probability of the transmission line and the gas pipeline under the influence of the main earthquake can be calculated. Taking element 41, i.e. the 5th gas pipeline, as an example, the calculation process of the failure probability of the gas pipeline is introduced. The total length of the 5th gas pipeline is 47.12km. The gas pipeline is divided into 95 sections of element pipelines, and the midpoint coordinates of each section of the element pipeline are obtained to determine the intensity zone to which each section of the element pipeline belongs. It can be seen that 54 sections of the element pipeline are located in the intensity 6 zone and 41 sections of the element pipeline are located in the intensity 7 zone. The failure probability of the 5th gas pipeline is:

[0183]

[0184] Where: p I=6 、p I=7 They are the failure probabilities of the elementary pipelines in intensity zones 6 and 7. The failure probability of the transmission line is approximately calculated by judging the earthquake intensity zone to which each transmission tower belongs, which is similar to the above calculation process and will not be repeated here.

[0185] Monte Carlo simulation (MCS), state enumeration (SE), IISE, and the combination of IISE and aPCE (hereinafter referred to as “IISE-aPCE”) are used to solve the resilience index of the IES E24-G14 test system under the influence of the main shock. The calculation results of the MCS method are used as the benchmark, and the sampling times of the MCS method are 10. 7 The convergence result variance coefficient is 0.0032. The calculation results of the four methods are shown in Table 3 and Figure 6 shown.

[0186] Table 3 Calculation results of toughness index of IES E24-G14 test system under the influence of main shock

[0187]

[0188]

[0189] From Table 3 and Figure 6 It can be seen that when the IISE toughness index is calculated to the third order, the relative error of the IISE method is much smaller than that of the SE method. The calculation results of the IISE-aPCE method are basically consistent with those of the IISE method, and the relative error with the benchmark value is only 2.7842%. This is because the IISE method can transfer the impact of high-order fault states to related low-order fault states, increase the proportion of low-order fault impact in the index, and thus reduce the impact of ignoring high-order fault states on the calculation accuracy.

[0190] It can be further seen from Table 3 that the calculation time of the SE method and the IISE method is about 2315 times that of the IISE-aPCE method. The proposed method greatly shortens the calculation time. This is because the number of third-order fault states is huge. The SE method and the IISE method are based on the traditional OEF calculation method. The optimal load reduction amount must be solved for each fault state one by one, which obviously cannot meet the requirements of IES operational resilience calculation for computational efficiency. The IISE-aPCE method adopts the optimal load reduction calculation method based on aPCE. The mapping relationship between the system topology, load state and optimal load reduction amount established by the IISE-aPCE method can quickly analyze the impact of a large number of uncertain factors on the optimal load reduction amount. The random input variables representing the fault state to be calculated can be substituted into the aPCE expression to output the IES optimal load reduction amount, which greatly simplifies the calculation process and greatly improves the calculation efficiency of the IES operational resilience index while ensuring the calculation accuracy.

[0191] (3) IES operational resilience calculation under aftershocks

[0192] After the main shock occurs, the operation resilience index for the period is calculated based on the predicted aftershock sequence for the next 24 hours and the faults that have occurred in the system during the current period, and is updated every hour. Figure 7 shown.

[0193] Assuming that the probability of aftershocks on the first day after the main shock is 1, according to Omori's law, the probability of aftershocks on the second day decays to 1 / 2, and the probability of aftershocks on the third day decays to 1 / 3, and so on. The main shock caused component 22 in the system to fail. Under the influence of aftershocks, component 41 failed in the first period, component 21 failed in the third period, and no component failed in subsequent periods; due to post-disaster repairs, component 22 was repaired in the 74th period, component 21 was repaired in the 120th period, and component 41 was repaired in the 150th period. Each time the index is calculated, the failure probability of the component under the influence of aftershocks that may occur in the next day is calculated, the failure probability of the failed component is set to 1, and the failure probability of the non-failed component is calculated according to the magnitude of the aftershock.

[0194] Table 4 shows the calculation results of resilience index for some periods of time seven days after the main shock, based on the calculation results of the MCS method. The sampling times of the MCS method are 10. 7 The variance coefficient of the convergence result is 0.0030. The change curve of the operating toughness of the IES E24-G14 test system under the influence of the main shock and aftershock is as follows: Figure 8 shown.

[0195] Table 4 Calculation results of toughness index of IES E24-G14 test system under aftershock

[0196] Tab.4Calculated resilience indexes of IES E24-G14 test systeminmainshock in

[0197] aftershock

[0198]

[0199] From Table 4 and Figure 8 It can be seen that the calculation results of the proposed IISE-aPCE method are basically consistent with those of the IISE method, with the maximum relative error being only 1.2741%, and the calculation time is reduced by 3 orders of magnitude compared with the IISE method, which significantly improves the calculation efficiency and verifies the accuracy and effectiveness of the proposed method. Figure 8The resilience index in the first period decreased because the failure rate of each component was large under the influence of the main shock. However, only component 22 failed after the main shock. When calculating the resilience index in the first period, although the failure rate of component 22 became 1, the load reduction was not large and the magnitude of the aftershock was smaller than that of the main shock, which caused the component failure rate to decay. Therefore, the resilience index value decreased. The resilience index in the second period increased significantly because the aftershock in the first period caused the failure of component 41, which caused a large load reduction in the system, and thus had a greater impact on the index calculation results. The index in the third period decreased because the component failure rate decayed with the attenuation of the magnitude. The index in the fourth period continued to decrease because although the aftershock in the previous period caused the failure of component 21, the attenuation of the component failure rate had a greater impact on the index. The index remained basically unchanged from the fifth period to the 74th period because the subsequent aftershocks did not cause component failures, the system load reduction remained unchanged, and the index results showed a downward trend with the change of component failure rate. The index in the 75th, 121st and 151st periods decreased because the components returned to normal after the post-disaster repair and the system load resumed supply.

[0200] Those skilled in the art will appreciate that the accompanying drawing is only a schematic diagram of a preferred embodiment, and the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0201] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for calculating the operational resilience of an integrated energy system in response to the impact of main and aftershocks, characterized in that: The method comprises: solving the IES optimal load reduction amount and IES operation resilience calculation offline, The offline solution of IES optimal load reduction includes: Calculate the single variable optimal orthogonal basis functions of node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels, and perform chaotic combination on the single variable optimal orthogonal basis functions, retaining the highest order not exceeding P PCE The polynomial is used as the optimal orthogonal basis function in the aPCE expression; based on the optimal orthogonal basis function, V groups of random input variable samples X are introduced. sample and optimal load reduction Calculate the approximation coefficient and obtain the aPCE expression of the optimal load reduction amount for the j-order fault state; The IES operational resilience calculation includes: Enumerate all j-order fault states that occur in IES, and substitute the random input variables of each state into the aPCE expression of the optimal load reduction to obtain the optimal load reduction L of IES shed,l,s ; Calculate the failure probability of power transmission lines and gas pipelines in IES according to the earthquake magnitude, and calculate the IES resilience index according to the failure probability and the IES optimal load reduction.

2. The method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks according to claim 1 is characterized in that: The V set of random input variable samples X sample The acquisition is: From the j-order topological state set V j-order topological states are extracted from the grid, and combined with V load levels to obtain V j-order fault states. The node self-susceptance, gas pipeline state variables, power grid, and gas grid load levels in each state are extracted to obtain V groups of random input variable samples X representing the j-order fault state. sample .

3. The method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks according to claim 1 is characterized in that: The optimal load reduction amount For V groups of random input variable samples X sample Each group of samples in performs the IES optimal load reduction calculation and obtains V corresponding IES optimal load reduction amounts.

4. The method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks according to claim 1 is characterized in that: The IES resilience index is calculated according to the failure probability and the IES optimal load reduction amount as follows: Substitute the failure probability and the IES optimal load reduction amount into the following formula to calculate and obtain the IES resilience index; Among them, N order is the maximum fault order; N level is the number of load levels considered; P l is the probability of the lth load level; is the set of j-order topological states enumerated; ΔP s is the modified probability of topological state s; ΔL shed,l,s is the influence increment of topological state s under the lth load level; n s is the number of fault states representing the topological state s; L shed,l,r is the k-order state set of topological state s The optimal load reduction amount of topological state r under the lth load level in N PCE is the total number of optimal orthogonal basis functions; c w,l,r is the wth approximation coefficient of the optimal load reduction of topological state r under the lth load level; φ w,l,r is the wth optimal orthogonal basis function of the optimal load reduction of topological state r under the lth load level.

5. The method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks according to claim 1 is characterized in that: The IES operational resilience calculation uses magnitude, intensity, and earthquake source, combined with the system operating status and topological structure to calculate the IES, and calculates the resilience index of the period based on the probability prediction of aftershocks in the future T period and the failures that have occurred in the current system, and updates it every interval ΔT.

6. The method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks according to claim 5 is characterized in that: The calculation of IES is specifically as follows: according to the predicted aftershock sequence in the future T period, the failure probability of the power transmission line and the gas pipeline under the influence of each aftershock is calculated, and the probability of at least one failure of each component in the future T period is solved, and then combined with the probability of aftershock occurrence, the probability of failure in the future T period is obtained. and It is updated as the predicted aftershock sequence and system conditions change.

7. The method for calculating the operational resilience of an integrated energy system to cope with the impact of main and aftershocks according to claim 5 is characterized in that: The optimal orthogonal basis function is: Where: p e is the order of the optimal orthogonal polynomial for the power grid load level; p g is the order of the optimal orthogonal polynomial for the gas grid load level; p i is the order of the optimal orthogonal basis function of the self-susceptance of the i-th node; p m is the order of the optimal orthogonal polynomial of the mth gas pipeline state variable, is the grid load level p e Order univariate optimal orthogonal polynomial; is the gas grid load level p g Order univariate optimal orthogonal polynomial; is the self-susceptance of the ith node p i Order univariate optimal orthogonal polynomial; is the p of the mth gas pipeline state variable m The optimal orthogonal polynomial of univariate p. e 、p g 、p i and p m satisfy: Among them, N bus is the number of grid nodes; N pipe is the number of pipelines; P PCE is the highest expansion order of aPCE expression; p w is the order of the w-th optimal orthogonal basis function.

Citation Information

Cited By

  • Power grid toughness evaluation method and system based on deep learning under extreme disaster

    CN119941029A