WO2017016021A1 - 一种基于影响增量的状态枚举可靠性评估方法及其装置 - Google Patents

一种基于影响增量的状态枚举可靠性评估方法及其装置 Download PDF

Info

Publication number
WO2017016021A1
WO2017016021A1 PCT/CN2015/088389 CN2015088389W WO2017016021A1 WO 2017016021 A1 WO2017016021 A1 WO 2017016021A1 CN 2015088389 W CN2015088389 W CN 2015088389W WO 2017016021 A1 WO2017016021 A1 WO 2017016021A1
Authority
WO
WIPO (PCT)
Prior art keywords
power system
state
influence
sensitivity
increment
Prior art date
Application number
PCT/CN2015/088389
Other languages
English (en)
French (fr)
Inventor
贾宏杰
侯恺
穆云飞
余晓丹
Original Assignee
天津大学
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 天津大学 filed Critical 天津大学
Priority to US15/747,522 priority Critical patent/US20180375373A1/en
Publication of WO2017016021A1 publication Critical patent/WO2017016021A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/001—Methods to deal with contingencies, e.g. abnormalities, faults or failures
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
    • G01R19/25—Arrangements for measuring currents or voltages or for indicating presence or sign thereof using digital measurement techniques
    • G01R19/2513—Arrangements for monitoring electric power systems, e.g. power lines or loads; Logging
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/40—Testing power supplies
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S40/00—Systems for electrical power generation, transmission, distribution or end-user application management characterised by the use of communication or information technologies, or communication or information technology specific aspects supporting them
    • Y04S40/20—Information technology specific aspects, e.g. CAD, simulation, modelling, system security

Definitions

  • the present invention relates to the field of power system reliability assessment, and in particular, to a state enumeration reliability evaluation method based on impact increment and a device thereof.
  • the current common methods for power system reliability assessment are divided into state enumeration method and Monte Carlo simulation method.
  • the state enumeration method calculates the probability and impact of each power system state by enumerating all possible power system states, and then calculates the reliability index of the power system.
  • the number of power system states increases exponentially.
  • state enumeration can quickly and efficiently calculate reliability metrics, but for complex large power systems, it is difficult to enumerate all power system states. Therefore, for large power systems, high-order faults are often ignored to improve computational efficiency. However, this will result in a decrease in the accuracy of the resulting reliability index, especially for power systems with high probability of component failure.
  • the state enumeration method is more suitable for power systems with small scale, simple structure and low component failure probability.
  • the Monte Carlo simulation method is also called random sampling method.
  • the method obtains the state of the power system by sampling the state of each component in the power system, and then calculates the reliability index.
  • the Monte Carlo simulation method can be divided into sequential Monte Carlo method and non-sequential Monte Carlo method.
  • Monte Carlo simulation method is a statistical test method, which is relatively intuitive and easy to understand. Its characteristic is that the sampling times are not affected by the scale and complexity of the power system, and it is easy to handle the random variation characteristics of the load. However, its error is closely related to the number of simulations. In order to obtain a reliability index with higher accuracy, it is necessary to increase the number of simulations and prolong the calculation time. Therefore, the Monte Carlo simulation method is less efficient in dealing with a simple power system, and is more suitable for a power system with a larger scale, a higher component failure probability or multiple fault effects that cannot be ignored.
  • Analytic method and Monte Carlo simulation method each have their own advantages, and the applicable situations complement each other. Therefore, the hybrid method that combines the two is an ideal evaluation method.
  • the hybrid method is characterized by the use of an analytical method in the case of an analytical method, and the Monte Carlo method is applied beyond the scope of the analytical method. And in the application of Monte Carlo method, the information provided by the analytical method is used as much as possible to reduce the calculation time and improve the calculation precision.
  • the invention provides a state enumeration reliability evaluation method based on influence increment and a device thereof, and the invention improves the meter Calculating the accuracy and computational efficiency of the reliability index reduces the complexity of the calculation reliability index, as described below:
  • a state enumeration reliability evaluation method based on impact increments comprising the following steps:
  • the breadth-first search method is used to verify the reachability of all elements in the independent adjacency matrix corresponding to the selected power system state. If there is an unreachable element, the influence of the selected power system state is incremented to zero, and the power system state is reselected;
  • the influence of the power system state under all load levels is evaluated by the optimal power flow algorithm, and the influence expectation of the power system state under each load level is obtained, thereby obtaining the influence increment of the power system state;
  • the power system reliability index is obtained by affecting the increment.
  • the method before the step of verifying the reachability of all elements in the independent adjacency matrix corresponding to the selected power system state by the breadth-first search method, the method further includes:
  • the sensitivity of each device impedance to the power flow of each branch is obtained by the perturbation method, and the independence between the devices is determined according to the sensitivity;
  • a power system state is selected from the state set, and an independent adjacency matrix of the power system state is created by the independence between the devices.
  • the method further includes: inputting power system data, device reliability data, and preset parameters, and initializing the fault order.
  • the preset parameters include: a maximum fault search order and a device independence sensitivity threshold.
  • the step of determining the independence between the devices according to the sensitivity is specifically:
  • the sensitivity index of the impedance of one faulty device to the branch power flow distribution is greater than the device independence sensitivity threshold, and the sensitivity index of the impedance of the other faulty device to the branch power flow distribution is greater than the device independence.
  • the two faulty devices are not independent.
  • a state enumeration reliability evaluation device based on influence increment comprising:
  • the verification module is configured to verify the reachability of all elements in the independent adjacency matrix corresponding to the selected power system state by the breadth-first search method, and if there is an unreachable element, the influence of the selected power system state is incremented to zero. Select the power system status;
  • the first obtaining module is configured to evaluate the influence of the state of the power system under all load levels by using an optimal power flow algorithm if the unreachable element is not present, obtain the influence expectation of the state of the power system under each load level, and then obtain the state of the power system. Impact increment
  • the second obtaining module is configured to obtain the power system reliability indicator by affecting the increment when all the power system states in the state set have been analyzed and the maximum fault search order has been reached.
  • the device further comprises:
  • a third obtaining module configured to obtain, by using a perturbation method, sensitivity of each device impedance to each branch power flow
  • a module is created for selecting a power system state from a set of states, and creating an independent adjacency matrix of power system states by independence between devices.
  • the device further comprises:
  • Input and initialization modules for inputting power system data, device reliability data and preset parameters, and initializing the fault order.
  • the preset parameters include: a maximum fault search order and a device independence sensitivity threshold.
  • the determining module includes:
  • the technical solution provided by the present invention has the beneficial effects that the core of the present invention is to replace the influence of the enumerated power system state with the influence increment, and can effectively improve the weight of the low-order fault state in the reliability index;
  • the lower-order state calculates a more accurate reliability index; the invention proves that the calculation of the reliability index can ignore the influence increment of the high-order fault and greatly improve the calculation efficiency.
  • 1 is a flow chart of a state enumeration reliability evaluation method based on an influence increment
  • FIG. 2 is a schematic diagram of a state enumeration reliability evaluation device based on an influence increment
  • FIG. 3 is another schematic diagram of a state enumeration reliability evaluation device based on an influence increment
  • FIG. 4 is another schematic diagram of a state enumeration reliability evaluation device based on an influence increment
  • Figure 5 is a schematic diagram of a determination module
  • FIG. 6 is a topological structure diagram of an IEEE 118 node system
  • FIG. 7a is a schematic diagram showing a convergence curve of the obtained EENS index when the method, the traditional state enumeration method and the Monte Carlo method are applied to the IEEE 118 node system;
  • FIG. 7b is a schematic diagram showing a comparison of the convergence curves of the obtained PLC indicators when the method, the traditional state enumeration method and the Monte Carlo method are applied to the IEEE 118 node system;
  • FIG. 8a is a schematic diagram showing a comparison of the relative error convergence curves of the EENS index obtained by the method, the traditional state enumeration method and the Monte Carlo method applied to the IEEE 118 node system;
  • FIG. 8b is a schematic diagram showing a comparison of the relative error convergence curves of the obtained PLC indicators when the method, the traditional state enumeration method and the Monte Carlo method are applied to the IEEE 118 node system;
  • Figure 9a is a schematic diagram showing the comparison of the convergence curves of the obtained EENS indicators when the method, the traditional state enumeration method and the Monte Carlo method are applied to the PEGASE 1354 node system;
  • Figure 9b is a schematic diagram showing the comparison of the convergence curves of the obtained PLC indicators when the method, the traditional state enumeration method and the Monte Carlo method are applied to the PEGASE 1354 node system;
  • Figure 10a is a schematic diagram showing the comparison of the relative error convergence curves of the EENS index obtained by the method, the traditional state enumeration method and the Monte Carlo method applied to the PEGASE 1354 node system;
  • Fig. 10b is a schematic diagram showing the comparison of the relative error convergence curves of the obtained PLC indicators when the method, the traditional state enumeration method and the Monte Carlo method are applied to the PEGASE 1354 node system.
  • 1 inspection module
  • 2 first acquisition module
  • 3 a second acquisition module
  • 4 a third acquisition module
  • the state enumeration reliability evaluation method based on the impact increment includes the following steps:
  • step 104 Verifying the reachability of all elements in the independent adjacency matrix D s by the breadth-first search method. If there is an unreachable element, the power system state s corresponding to the faulty device may be divided into at least two independent subsets, and the steps are performed. 103; otherwise, perform step 105;
  • the corresponding reliability indicators can be obtained, as follows:
  • the obtained reliability index is the EENS index.
  • the obtained index is the PLC index.
  • step 106 Calculate the influence increment ⁇ I s of the power system state s; check whether all power system states in the k-th order state set ⁇ A k have been analyzed, and if yes, perform step 107; if not, execute step 103;
  • the method improves the accuracy and calculation efficiency of the calculation reliability index by the above steps 101-108, and reduces the complexity of the calculation reliability index.
  • the power system data includes: power system nodes, branches, generator sets parameters, load levels of each node, annual load change curves, etc.; equipment reliability data includes: unavailability of equipment such as lines, transformers, generator sets, etc.;
  • the parameters include: maximum fault search order N CTG and device independence sensitivity threshold ⁇ s .
  • A represents a set of power system equipment
  • the power system state s is a set of faulty devices used to indicate the state of the power system at the time of failure of these devices
  • Card(s) represents the order of failure of the state of the power system s.
  • the reachability is defined as: in the undirected connected graph determined by the independent adjacency matrix D s , if a certain node V 1 can be connected to another node V 2 through the edge in the graph, then V 1 is called Up to V 2 .
  • the faulty device in the power system state s may be divided into at least two mutually independent subsets, so that the impact increment ⁇ I s is 0, Performing the calculation, performing step 205; otherwise, if all the nodes are mutually reachable, step 207 is performed;
  • P l is the probability of load level l
  • n l is the total number of load levels.
  • n s is the total number of faulty devices in the power system state s; ⁇ s k is the set of k-order subsets of the power system state s; u is an element in ⁇ s k ; ⁇ I u is the load of u Loss increment.
  • ⁇ s k is defined as follows:
  • Is a subset symbol
  • s 1 is a subset of the power system state s
  • Card(s 1 ) represents the fault order of the power system state s 1 .
  • step 209 Check whether all power system states in the state set ⁇ A k have been analyzed, if yes, go to step 210, otherwise go to step 205;
  • R is the system reliability indicator
  • P i is the unavailability of device i
  • N is the total number of devices in the system.
  • the calculation method of the sensitivity S PZ of each device in step 202 is:
  • a device failure can be equivalent to a sudden rise in impedance of the device from a nominal value to infinity.
  • Equipment failures have a direct impact on the power flow distribution of the power system. Therefore, the sensitivity of the device impedance to the power flow of each branch of the power system can be used to describe the independence between the faulty device and the branches of the power system.
  • the sensitivity index is labeled as S PZ , and the sensitivity index can be calculated by using the perturbation method. The process of calculating the sensitivity is well known to those skilled in the art, and will not be described in detail in the embodiments of the present invention.
  • the calculation method of the independence device flag d ij between the faulty devices in step 203 is:
  • the independence flag between the faulty devices i and j is d ij .
  • ⁇ s is the preset parameter device independence sensitivity threshold
  • A represents the power system equipment set
  • S PZ (h, i) is the sensitivity index of the fault device i impedance to the branch h tidal current distribution
  • S PZ (h, j) is the sensitivity index of the impedance of the faulty device j to the flow distribution of the branch h.
  • the faulty device in the power system state s can be divided into at least two mutually independent subsets, and thus the influence increment ⁇ I s is 0,
  • the basic proof process is as follows:
  • the node is proved to have at least one set of faulty devices and other faulty devices in the state of the power system, and thus the faulty devices in the power system state s can be divided into at least two mutually independent subsets s 1 and s 2 .
  • the formula for calculating the reliability index of the power system in step 211 is as follows:
  • the reliability indicator can be expressed as
  • ⁇ is the set of all power system states that may occur in the power system; I(s) is the influence function of the power system state s; P(s) is the probability of occurrence of the power system state s.
  • the reliability index R is the product of the probability of failure of the device and its loss, plus the probability of the normal operation of the device and the product of the loss at that time.
  • the reliability index calculation can be formulated into a form based on the influence increment, in which all normal running probabilities are eliminated, and the fault impact is replaced by the influence increment.
  • the incremental influence ⁇ I s of the high-order fault state can be expressed in the form of a formula, and the equation can be further simplified to
  • R 2 I ⁇ +P 1 ⁇ I 1 +P 2 ⁇ I 2 +P 1 P 2 ⁇ I 12 ⁇ *MERGEFORMAT(12)
  • the power system contains n+1 devices.
  • the reliability index R n+1 of the new power system can be derived from the original power system indicator R n .
  • ⁇ n+1 ⁇ indicates the state of the power system with only the newly added equipment failure
  • P n+1 They are the availability and unavailability of newly added equipment
  • k' and k 1 represent the order of failure.
  • k' and k 1 should be calculated from the 0th order; For the k 1st order sub-set of the power system state s, the definition is as shown in the formula; u is An element in the middle. The equation can be further transformed into the following form:
  • k 2 represents the number of failure orders; It is a k 2 order sub-set of the power system state s, and its definition is as shown in the equation.
  • the influence increment ⁇ I s of any state s can be calculated according to the system reliability index. Different reliability indicators can be obtained depending on the state influence function I s used.
  • the method improves the accuracy and calculation efficiency of the calculation reliability index by the above steps 201-211, and reduces the complexity of the calculation reliability index.
  • a state enumeration reliability evaluation device based on influence increment see FIG. 2, the device includes:
  • the verification module 1 is configured to verify the reachability of all elements in the independent adjacency matrix corresponding to the selected power system state by the breadth-first search method, and if there is an unreachable element, the influence of the selected power system state is zero. Reselect the power system status;
  • the first obtaining module 2 is configured to: if there is no unreachable element, evaluate the influence of the power system state under all load levels by using an optimal power flow algorithm, obtain the influence expectation of the power system state under each load level, and obtain the power system state. Incremental impact;
  • the second obtaining module 3 is configured to obtain the power system reliability index by affecting the increment when all the power system states in the state set have been analyzed and the maximum fault search order has been reached.
  • the device further includes:
  • the third obtaining module 4 is configured to obtain, by using a perturbation method, sensitivity of each device impedance to each branch power flow;
  • Determining module 5 determining the independence between the devices according to the sensitivity
  • the creation module 6 is configured to select a power system state from the state set, and create an independent adjacency matrix of the power system state by the independence between the devices.
  • the device further includes:
  • the input and initialization module 7 is configured to input power system data, device reliability data and preset parameters, and initialize the fault order.
  • the preset parameters include: a maximum fault search order and a device independence sensitivity threshold.
  • the determining module 5 includes:
  • the determining sub-module 51 is configured to: if there is one branch, the sensitivity index of the impedance of the faulty device to the branch power flow distribution is greater than the device independence sensitivity threshold, and the impedance of the other faulty device is sensitive to the branch power flow distribution When the indicator is greater than the device independence sensitivity threshold, the two faulty devices are not independent.
  • modules and sub-modules can be implemented by a device having a computing function, such as a single-chip microcomputer or a PC.
  • a computing function such as a single-chip microcomputer or a PC.
  • the embodiment of the present invention does not limit the type and type of the device.
  • the device improves the accuracy and calculation efficiency of the calculation reliability index by the verification module 1, the first acquisition module 2, the second acquisition module 3, the third acquisition module 4, the determination module 5, the creation module 6, the input and initialization module 7, Reduce the complexity of calculating reliability indicators.
  • the implementation method and practical effects of the present invention will be described below with reference to an example.
  • This example is tested on an IEEE 118 node test system, and its network topology is shown in Figure 6.
  • the test system consists of 118 nodes, 54 generator sets, 186 branches, 54 generator nodes, and 64 load nodes.
  • the total installed capacity and load demand are 9966 MW and 4242 MW, respectively.
  • This example verifies the efficiency and accuracy of the method by comparing the method with the traditional state enumeration method and the Monte Carlo method.
  • the EENS and PLC indicators of the test system can be calculated, as shown in Table 1.
  • IISE comparative analysis of the method
  • MCS Monte Carlo method
  • SE state enumeration method
  • the search depth N CTG of the method is also set to 2; in the Monte Carlo method, the convergence criterion total number of samples N MCS is set to 10 6 . Due to the large sample size, the Monte Carlo method can produce sufficiently accurate results, so the results of its calculations can be used as a benchmark for evaluating the accuracy of other examples.
  • the evaluation results of the above three methods are shown in Table 1, Figure 7a, Figure 7b, Figure 8a and Figure 8b.
  • Table 1 shows the evaluation results of two reliability indicators (EENS and PLC). It can be seen that the Monte Carlo method and the method are very close, and their relative error is about 1% (the error of EENS is 0.8182% and the error of PLC is 1.3157%). The error of the two indicators obtained by the traditional enumeration method exceeds 6% (the error of EENS is 6.2357% and the error of PLC is 7.2863%), which is much higher than this method. At the same time, the CPU time consumed by this method is much smaller than the other two algorithms, indicating that this method is more efficient than the traditional evaluation method.
  • the Monte Carlo method and the method are very close, and their relative error is about 1% (the error of EENS is 0.8182% and the error of PLC is 1.3157%). The error of the two indicators obtained by the traditional enumeration method exceeds 6% (the error of EENS is 6.2357% and the error of PLC is 7.2863%), which is much higher than this method. At the same time, the CPU time consumed
  • Fig. 7a and Fig. 7b respectively show the convergence curves of EENS and PLC obtained by Monte Carlo method
  • Fig. 8a and Fig. 8b respectively show the relative error convergence curves of these two indicators.
  • the calculation results of this method and the state enumeration method are also given in these figures.
  • the calculation accuracy of this method is much higher than that of the state enumeration method, and the calculation time of this method is about 1/10 of the state enumeration method.
  • the Monte Carlo method requires about 10 4 seconds for the relative error to be stable within 1%, and the method can achieve the same accuracy in 100 seconds, which is about 1/100 of the Monte Carlo method.
  • the EENS and PLC indicators of the test system can be calculated, as shown in Table 2.
  • IISE comparative analysis of the method
  • MCS Monte Carlo method
  • SE state enumeration method
  • the search depth N CTG of the method is also set to 1; in the Monte Carlo method, the convergence criterion total number of samples N MCS is set to 10 5 . Due to the large sample size, the Monte Carlo method can produce sufficiently accurate results, so the results of its calculations can be used as a benchmark for evaluating the accuracy of other examples.
  • the evaluation results of the above three methods are shown in Table 2, Fig. 9a, Fig. 9b, Fig. 10a and Fig. 10b.
  • Table 2 shows the evaluation results of two reliability indicators (EENS and PLC). It can be seen that the Monte Carlo method and the method are very close, and their relative error is about 2% (the error of EENS is 1.4590% and the error of PLC is 2.1644%). The two indicators obtained by the traditional enumeration method have a very large error of about 98% (the error of EENS is 98.1514% and the error of PLC is 98.0834%), which is much higher than other methods. At the same time, because only the first-order fault is considered, the operation time of ISE and SE is similar, but it is much shorter than the MCS. It shows that this method is more efficient than the traditional evaluation method in this power system.
  • the Monte Carlo method and the method are very close, and their relative error is about 2% (the error of EENS is 1.4590% and the error of PLC is 2.1644%).
  • the two indicators obtained by the traditional enumeration method have a very large error of about 98% (the error of EENS is 98.1514%
  • Fig. 9a and Fig. 9b respectively show the convergence curves of EENS and PLC obtained by Monte Carlo method
  • Fig. 10a and Fig. 10b respectively show the relative error convergence curves of these two indicators.
  • the calculation results of this method and the state enumeration method are also given in these figures. It can be seen from Fig. 10a and Fig. 10b that the calculation accuracy of the method is much higher than that of the state enumeration method, and the calculation time of the method is similar to the state enumeration method. It can be seen from the relative error convergence curve that the Monte Carlo method requires about 3 ⁇ 10 4 seconds to stabilize the relative error within 2%, and the method can achieve the same accuracy in 1500 seconds, which is about 1/ of the Monte Carlo method. 20.

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

一种基于影响增量的状态枚举可靠性评估方法及其装置，方法包括：通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性，若存在不可达元素，则所选状态的影响增量为零，重新选择电力***状态；若不存在不可达元素，通过最优潮流算法评估所有负荷水平下的电力***状态的影响，获取电力***状态在各负荷水平下的影响期望，进而获取电力***状态的影响增量；当状态集中所有电力***状态已被分析，且已达到最大故障搜索阶数时，通过影响增量获取电力***可靠性指标。装置包括：检验模块（1）、第一获取模块（2）和第二获取模块（3），通过这些模块实现了可靠性指标的计算。上述方法和装置提高了计算精度和计算效率，降低了计算复杂度。

Description

一种基于影响增量的状态枚举可靠性评估方法及其装置 技术领域
本发明涉及电力***可靠性评估领域，尤其涉及一种基于影响增量的状态枚举可靠性评估方法及其装置。
背景技术
目前电力***可靠性评估的常用方法分为状态枚举法和蒙特卡洛模拟法。
状态枚举法是通过枚举出所有可能出现的电力***状态，计算各个电力***状态的发生概率和影响，进而计算得到电力***的可靠性指标。在实际应用中，随着元件数量的增加，电力***状态的个数呈指数增长。对于规模较小的电力***，状态枚举法能够快速高效地计算出可靠性指标，但对于复杂的大电力***，该方法很难枚举出所有的电力***状态。因此，对于电力大***，通常会忽略高阶故障以提高计算效率。然而这会造成所得可靠性指标精度的下降，尤其是对于元件失效概率较高的电力***。总之，由于其物理概念清晰，模型精度高的特点，状态枚举法较适用于规模小，结构简单，元件失效概率低的电力***。
蒙特卡罗模拟法又称为随机抽样法，该方法通过抽样电力***内各元件的状态，得到电力***状态，进而计算出可靠性指标。根据抽样原理的不同，蒙特卡罗模拟法又可分为序贯蒙特卡罗法和非序贯蒙特卡罗法。蒙特卡罗模拟法属于统计试验方法，较为直观，便于理解；其特点是采样次数不受电力***规模和复杂程度影响，便于处理负荷的随机变化特性。然而它的误差与模拟次数密切相关，为了获得具有较高精确度的可靠性指标，需要增加模拟次数，延长了计算时间。因此蒙特卡罗模拟法在处理结构简单的电力***时效率较低，而更适用于规模较大，具有较高元件失效概率或多重故障影响不容忽视的电力***。
解析法和蒙特卡罗模拟法各有自己的优势，且适用的情况相互补充，因此将二者有机结合起来的混合法是一种较为理想的评估方法。混合法的特点是在适合解析法的情况下使用解析法，在超出解析法适用范围的情况应用蒙特卡罗法。并且在蒙特卡罗法的应用中尽可能使用解析法所提供的信息，以减少运算时间，提高计算精度。
然而，现有方法均无法满足在线应用对计算效率和精度的要求，为实现电力***可靠性评估的实时应用，迫切需要一种效率更高、精度更好的评估方法。
发明内容
本发明提供了一种基于影响增量的状态枚举可靠性评估方法及其装置，本发明提高了计 算可靠性指标的精度和计算效率，降低了计算可靠性指标的复杂度，详见下文描述：
一种基于影响增量的状态枚举可靠性评估方法，所述方法包括以下步骤：
通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性，若存在不可达元素，则所选电力***状态的影响增量为零，重新选择电力***状态；
若不存在不可达元素，则通过最优潮流算法评估所有负荷水平下的电力***状态的影响，获取电力***状态在各负荷水平下的影响期望，进而获取电力***状态的影响增量；
当状态集中所有电力***状态已被分析，且已达到最大故障搜索阶数时，通过影响增量获取电力***可靠性指标。
其中，在通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性的步骤之前，所述方法还包括：
通过微扰法获取各设备阻抗对各支路潮流的灵敏度，根据灵敏度确定各设备间的独立性；
从状态集中选择一个电力***状态，通过设备间的独立性创建电力***状态的独立性邻接矩阵。
其中，所述方法还包括：输入电力***数据，设备可靠性数据和预置参数，并初始化故障阶数。
进一步地，所述预置参数包括：最大故障搜索阶数和设备独立性灵敏度阈值。
其中，所述根据灵敏度确定各设备间的独立性的步骤具体为：
若存在一条支路，使得一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值，且另一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值时，两故障设备之间不独立。
一种基于影响增量的状态枚举可靠性评估装置，所述装置包括：
检验模块，用于通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性，若存在不可达元素，则所选电力***状态的影响增量为零，重新选择电力***状态；
第一获取模块，用于若不存在不可达元素，通过最优潮流算法评估所有负荷水平下的电力***状态的影响，获取电力***状态在各负荷水平下的影响期望，进而获取电力***状态的影响增量；
第二获取模块，用于当状态集中所有电力***状态已被分析，且已达到最大故障搜索阶数时，通过影响增量获取电力***可靠性指标。
其中，所述装置还包括：
第三获取模块，用于通过微扰法获取各设备阻抗对各支路潮流的灵敏度；
确定模块，根据灵敏度确定各设备间的独立性；
创建模块，用于从状态集中选择一个电力***状态，通过设备间的独立性创建电力***状态的独立性邻接矩阵。
其中，所述装置还包括：
输入和初始化模块，用于输入电力***数据，设备可靠性数据和预置参数，并初始化故障阶数。
进一步地，所述预置参数包括：最大故障搜索阶数和设备独立性灵敏度阈值。
进一步地，所述确定模块包括：
确定子模块，用于若存在一条支路，使得一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值，且另一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值时，两故障设备之间不独立。
本发明提供的技术方案的有益效果是：本发明的核心在于将枚举出的电力***状态的影响替换为影响增量，能够有效提升低阶故障状态在可靠性指标中的权重；仅利用少数低阶状态计算出较为准确的可靠性指标；本发明证明了计算可靠性指标时可忽略高阶故障的影响增量，大大提升了计算效率。
附图说明
图1为基于影响增量的状态枚举可靠性评估方法的流程图；
图2为基于影响增量的状态枚举可靠性评估装置的示意图；
图3为基于影响增量的状态枚举可靠性评估装置的另一示意图；
图4为基于影响增量的状态枚举可靠性评估装置的另一示意图；
图5为确定模块的示意图；
图6为IEEE 118节点***拓扑结构图；
图7a为本方法、传统状态枚举法及蒙特卡洛法应用于IEEE 118节点***时，所得EENS指标收敛曲线对比示意图；
图7b为本方法、传统状态枚举法及蒙特卡洛法应用于IEEE 118节点***时，所得PLC指标收敛曲线对比示意图；
图8a为本方法、传统状态枚举法及蒙特卡洛法应用于IEEE 118节点***时，所得EENS指标相对误差收敛曲线对比示意图；
图8b为本方法、传统状态枚举法及蒙特卡洛法应用于IEEE 118节点***时，所得PLC指标相对误差收敛曲线对比示意图；
图9a为本方法、传统状态枚举法及蒙特卡洛法应用于PEGASE 1354节点***时，所得EENS指标收敛曲线对比示意图；
图9b为本方法、传统状态枚举法及蒙特卡洛法应用于PEGASE 1354节点***时，所得PLC指标收敛曲线对比示意图；
图10a为本方法、传统状态枚举法及蒙特卡洛法应用于PEGASE 1354节点***时，所得EENS指标相对误差收敛曲线对比示意图；
图10b为本方法、传统状态枚举法及蒙特卡洛法应用于PEGASE 1354节点***时，所得PLC指标相对误差收敛曲线对比示意图。
附图中，各标号所代表的部件列表如下：
1：检验模块；         2：第一获取模块；
3：第二获取模块；     4：第三获取模块；
5：确定模块；         6：创建模块；
7：初始化模块；       51：确定子模块。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面对本发明实施方式作进一步地详细描述。
实施例1
如图1所示，本发明实施例提供的基于影响增量的状态枚举可靠性评估方法包括以下几个步骤：
101：输入电力***数据，设备可靠性数据和预置参数，并初始化故障阶数k＝1；
102：利用微扰法计算各设备阻抗对各支路潮流的灵敏度SPZ，根据灵敏度SPZ确定各设备间的独立性；
103：从k阶状态集ΩA k中选择一个电力***状态s，通过设备间的独立性创建电力***状态s的独立性邻接矩阵Ds；
104：通过广度优先搜索法检验独立性邻接矩阵Ds中所有元素的可达性，若存在不可达元素，则电力***状态s对应故障设备可划分为至少两个互相独立的子集，执行步骤103；否则，执行步骤105；
105：通过OPF(最优潮流算法)评估所有负荷水平下的电力***状态s的影响Is,l，获取电力***状态在各负荷水平下的影响期望；
根据采用的影响函数不同，可得到对应的可靠性指标，具体如下：
(1)期望缺电电量EENS(expected energy not supplied，MWh/年)
当影响函数I为全年负荷损失量(MWh/年)时，所得的可靠性指标为EENS指标。
(2)负荷削减概率PLC(probability ofload curtailments)
当影响函数I为损失标志位时，所得指标为PLC指标。
106：计算电力***状态s的影响增量ΔIs；检验k阶状态集ΩA k中所有电力***状态是否已被分析，如果是，执行步骤107；如果否，执行步骤103；
107：若k＝NCTG(最大故障搜索阶数)，执行步骤108；否则令k＝k+1，执行步骤103；
108：计算电力***可靠性指标。
本方法通过上述步骤101-步骤108提高了计算可靠性指标的精度和计算效率，降低了计算可靠性指标的复杂度。
实施例2
下面结合具体的计算公式，对实施例1中的方案进行详细说明，详见下文描述：
201：输入电力***数据、设备可靠性数据和预置参数，并初始化故障阶数(即故障设备的个数)k＝1；
其中，电力***数据包括：电力***节点、支路、发电机组参数、各节点负荷水平、年负荷变化曲线等；设备可靠性数据包括：线路、变压器、发电机组等设备的不可用率；预置参数包括：最大故障搜索阶数NCTG和设备独立性灵敏度阈值δs。
202：计算电力***中各支路(包括线路及变压器)之间的灵敏度SPZ；
203：根据各支路间的灵敏度SPZ确定各支路间的独立性；
若设备i、j独立，则记dij＝0；否则记dij＝1。
204：创建k阶状态集ΩA k如下：
Figure PCTCN2015088389-appb-000001
式中，A表示电力***设备集合；电力***状态s是一个由故障设备构成的集合，用于表示这些设备故障时的电力***状态；Card(s)表示电力***状态s的故障阶数。
205：从k阶状态集ΩA k中选择一个电力***状态s，通过下式创建电力***状态s的独立性邻接矩阵Ds：
Ds＝[dij], i,j∈s    \*MERGEFORMAT(2)
206：通过广度优先搜索法检验独立性邻接矩阵Ds中所有节点的可达性；
其中，可达性的定义为：在由独立性邻接矩阵Ds确定的无向连通图中，若某一节点V1能够通过该图中的边连接到另一节点V2，则称V1可达V2。
若该无向连通图中存在任意两个节点是互相不可达的，则电力***状态s中故障设备必然可划分为至少两个互相独立的子集，因而其影响增量ΔIs为0，不必进行计算，执行步骤205；否则，若所有节点均相互可达，执行步骤207；
207：通过OPF评估所有负荷水平下的电力***状态s的影响Is,l，则该电力***状态s在各负荷水平下的影响期望为：
Figure PCTCN2015088389-appb-000002
式中，Pl为负荷水平l的概率；nl为负荷水平的总个数。
208：根据下式计算电力***状态s的影响增量ΔIs；
Figure PCTCN2015088389-appb-000003
式中，ns为电力***状态s下故障设备的总个数；Ωs k是电力***状态s的k阶子集的集合；u为Ωs k中的一个元素；ΔIu为u的负荷损失增量。
其中，Ωs k的定义如下：
Figure PCTCN2015088389-appb-000004
式中，
Figure PCTCN2015088389-appb-000005
是子集符号，s1是电力***状态s的一个子集；Card(s1)表示电力***状态s1的故障阶数。
209：检验状态集ΩA k中所有电力***状态是否已被分析，如果是，执行步骤210，否则执行步骤205；
210：若k等于最大故障搜索阶数NCTG，执行步骤211；否则令k＝k+1，执行步骤204；
211：通过下式计算电力***可靠性指标。
Figure PCTCN2015088389-appb-000006
式中，R表示***可靠性指标，Pi是设备i的不可用率；N为***中设备总数。
其中，步骤202中的各设备的灵敏度SPZ的计算方法为：
在电力***中，设备故障可等效为该设备阻抗由额定值突然上升至无穷大。而设备故障会对电力***的潮流分布产生直接影响。因此，可以利用设备阻抗对电力***各支路潮流的灵敏度描述故障设备与电力***各支路间的独立性。本发明将灵敏度指标记作SPZ，可采用微扰法计算该灵敏度指标，具体计算灵敏度的过程为本领域技术人员所公知，本发明实施例对此不做赘述。
其中，步骤203中各故障设备间独立性标志位dij的计算方法为：
故障设备i和j间的独立性标志位为dij。当以下条件成立时，认为两者不独立，即dij＝1；否则认为两者独立，即dij＝0。
存在h∈A,使得，
|SPZ(h,i)|＞δs且|SPZ(h,j)|＞δs
\*MERGEFORMAT(7)
式中，δs是预设参数设备独立性灵敏度阈值；A表示电力***设备集合；SPZ(h,i)是故障设备i的阻抗对支路h潮流分布的灵敏度指标；SPZ(h,j)是故障设备j的阻抗对支路h潮流分布的灵敏度指标。
其中，步骤206中的独立性邻接矩阵Ds中若存在不可达元素，则电力***状态s中故障设备可划分为至少两个互相独立的子集，因而其影响增量ΔIs为0，其基本证明过程如下：
设条件一为：在电力***中，假设一个高阶电力***状态s(当s的故障阶数大于或等于2时，认为它是一个高阶状态)对应的独立性邻接矩阵Ds中存在不可达节点，则证明该电力***状态中至少存在一组故障设备与其他故障设备均相互独立，因而电力***状态s中的故障设备可划分为至少两个互相独立的子集s1和s2。
则可通过数学归纳法可证明，条件一成立时，ΔIs＝0。
首先，对于一个***状态s＝{i1，i2}，其中i1、i2为两个故障设备，若条件一成立，则ΔIs＝Is–I{i1}-I{i2}＝0。因此ns＝2时，ΔIs＝0成立。假设ΔIS＝0对于k阶以下故障状态(2<ns<k)成立，则任一对于(k+1)阶状态s，若条件一成立，同样可证明ΔIs＝0成立。根据数学归纳法可知，该结论对于所有高阶故障状态s均成立。
步骤211中电力***可靠性指标的计算方法公式证明如下：
在电力***中，可靠性指标可表示为
Figure PCTCN2015088389-appb-000007
式中，Ω为电力***中可能出现的所有电力***状态的集合；I(s)为电力***状态s的影响函数；P(s)为电力***状态s的发生概率。
设电力***共有n个设备，Pi和
Figure PCTCN2015088389-appb-000008
分别为设备i发生故障和正常运行的概率；Is为电力***状态s所造成的影响；Iφ为电力***正常运行时的影响，则
Figure PCTCN2015088389-appb-000009
对于某个由两个设备构成的电力***，可靠性指标R为设备故障概率与其造成损失的乘积，再加上设备正常工作的概率和此时损失的乘积。
Figure PCTCN2015088389-appb-000010
将式代入并化简得
Figure PCTCN2015088389-appb-000011
通过公式推导，可将可靠性指标计算公式化为一种基于影响增量的形式，该形式下所有的正常运行概率均被消除，且故障影响被替换为影响增量。其中，高阶故障状态的增量影响ΔIs可表示为式的形式，则式可进一步简化为
R2＝Iφ+P1ΔI1+P2ΔI2+P1P2ΔI12    \*MERGEFORMAT(12)
由上式可以看出，多项式的项数并没有改变，但将所有设备正常运行的概率均消去。由于低阶故障包含了较多的正常元件和较少的故障元件，因此消去正常运行概率可以提高低阶状态的权重。此外，式中电力***各状态的影响已被影响增量所代替。由于高阶故障的影响较大，但影响增量相对较小，因而式中高阶故障所占的权重减小。
将式拓展到含有N个设备的电力***，即可得式。其基本证明过程如下：
由式可知N＝2时式成立。由数学归纳法可知，若假设N＝n时式成立，则若N＝n+1时该式也成立，则证明完成。
设原电力***包含n个设备，现新加入一个设备，则电力***包含n+1个设备。该新电力***的可靠性指标Rn+1可由原电力***指标Rn推导得出。
Figure PCTCN2015088389-appb-000012
其中，{n+1}表示仅有新加设备故障的电力***状态；Pn+1、
Figure PCTCN2015088389-appb-000013
分别为新加设备的可用率和不可用率；k’、k1表示故障阶数，为考虑仅有新加设备故障的电力***状态，k’及k1应从0阶开始计算；
Figure PCTCN2015088389-appb-000014
为电力***状态s的k1阶子集合，其定义如式所示；u为
Figure PCTCN2015088389-appb-000015
中的一个元素。该等式可进一步化为如下形式：
Figure PCTCN2015088389-appb-000016
其中，k2表示故障阶数；
Figure PCTCN2015088389-appb-000017
为电力***状态s的k2阶子集合，其定义如式所示。
因此对于目标n+1阶***，式成立。根据数学归纳法可知，式对于任意阶***均成立。
根据式可计算任意状态s的影响增量ΔIs，根据可计算***可靠性指标。根据所采用的状态影响函数Is的不同，可得到不同可靠性指标。
本方法通过上述步骤201-步骤211提高了计算可靠性指标的精度和计算效率，降低了计算可靠性指标的复杂度。
实施例3
一种基于影响增量的状态枚举可靠性评估装置，参见图2，该装置包括：
检验模块1，用于通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性，若存在不可达元素，则所选电力***状态的影响增量为零，重新选择电力***状态；；
第一获取模块2，用于若不存在不可达元素，通过最优潮流算法评估所有负荷水平下的电力***状态的影响，获取电力***状态在各负荷水平下的影响期望，进而获取电力***状态的影响增量；
第二获取模块3，用于当状态集中所有电力***状态已被分析，且已达到最大故障搜索阶数时，通过影响增量获取电力***可靠性指标。
其中，参见图3，该装置还包括：
第三获取模块4，用于通过微扰法获取各设备阻抗对各支路潮流的灵敏度；
确定模块5，根据灵敏度确定各设备间的独立性；
创建模块6，用于从状态集中选择一个电力***状态，通过设备间的独立性创建电力***状态的独立性邻接矩阵。
其中，参见图4，该装置还包括：
输入和初始化模块7，用于输入电力***数据，设备可靠性数据和预置参数，并初始化故障阶数。
进一步地，预置参数包括：最大故障搜索阶数和设备独立性灵敏度阈值。
进一步地，参见图5，确定模块5包括：
确定子模块51，用于若存在一条支路，使得一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值，且另一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值时，两故障设备之间不独立。
具体实现时，上述模块、子模块均可以通过单片机、PC机等具有运算功能的器件实现，本发明实施例对器件的型号、类型不做限制。
本装置通过检验模块1、第一获取模块2、第二获取模块3、第三获取模块4、确定模块5、创建模块6、输入和初始化模块7提高了计算可靠性指标的精度和计算效率，降低了计算可靠性指标的复杂度。
实施例4
下面结合一个实例来介绍本发明的实施方法和实际效果。本实例在IEEE 118节点测试***上进行测试，其网络拓扑示意图如图6所示。该测试***包括118个节点、54台发电机组、186条支路、54个发电机节点、64个负荷节点，发电总装机和负荷需求分别为9966MW和4242MW。本实例通过将本方法与传统状态枚举法和蒙特卡洛法进行对比，验证本方法的高效性和准确性。
输入***数据，设置最大故障搜索阶数NCTG＝2和设备独立性灵敏度阈值δs＝0.02。初始化故障阶数k＝1。依次对每条支路增加阻抗0.01p.u.，并进行潮流计算，记录每条支路阻抗增加前后***各支路的潮流变化量，其与0.01的比值即为支路阻抗对各支路潮流的的灵敏度SPZ。其余的操作步骤参见实施例1和2，本发明实施例对此不做赘述。
根据以上步骤可计算测试***的EENS及PLC指标，如表1所示。此外，为对比分析本方法(IISE)的效果，将其与传统蒙特卡洛法(MCS)与状态枚举法(SE)进行对比。本方法的搜索深度NCTG同样设置为2；在蒙特卡罗法中，将收敛判据总采样数NMCS设置为106。由于样本量巨大，蒙特卡罗法可以得出足够精确的结果，因此可将其算例结果作为评估其他算例精度的基准。以上三种方法的评估结果如表1、图7a、图7b、图8a和图8b所示。
表1 三种评估方法结果(IEEE-118)
Figure PCTCN2015088389-appb-000018
表1展示了两种可靠性指标(EENS和PLC)的评估结果。可以看出采用蒙特卡罗法和本方法得出的指标非常接近，它们的相对误差在1％左右(通过表1可以看出EENS的误差为0.8182％，PLC的误差为1.3157％)。而传统枚举法得出的两指标误差均超过6％(通过表1可以看出EENS的误差为6.2357％，PLC的误差为7.2863％)，远高于本方法。同时采用本方法计算所消耗的CPU时间也远小于其他两种算法，表明本方法比传统的评估方法效率更高。
图7a、图7b分别给出了蒙特卡罗法所得EENS和PLC的收敛曲线，而图8a和图8b分别给出了这两个指标的相对误差收敛曲线。本方法与状态枚举法的计算结果也在该些图中给出。从图8a和图8b中可以看出，本方法计算精度远高于状态枚举法，同时本方法计算用时 约为状态枚举法的1/10。通过相对误差收敛曲线可以看出，蒙特卡洛法相对误差稳定在1％以内需要大约104秒，而本方法可在100秒内达到同样精度，用时约为蒙特卡洛法的1/100。
因此可以得出结论，本方法比其他两种传统的可靠性评估方法具有更高的精度和计算效率。
实施例5
下面结合另一个实例来介绍本发明在实际***中的实施方法和实际效果。本实例在PEGASE实际电力***上进行测试，该电力***是一个欧洲输电网***。该电力***包括1354个节点、260台发电机组、1991条支路、260个发电机节点、1094个负荷节点，发电总装机和负荷需求分别为128739MW和73060MW。本实例通过将本方法与传统状态枚举法和蒙特卡洛法进行对比，验证本方法的实用价值。
输入电力***数据，设置最大故障搜索阶数NCTG＝1和设备独立性灵敏度阈值δs＝0.02。初始化故障阶数k＝1。依次对每条支路增加阻抗0.01p.u.，并进行潮流计算，记录每条支路阻抗增加前后***各支路的潮流变化量，其与0.01的比值即为支路阻抗对各支路潮流的的灵敏度SPZ。其余的操作步骤参见实施例1和2，本发明实施例对此不做赘述。
根据以上步骤可计算测试***的EENS及PLC指标，如表2所示。此外，为对比分析本方法(IISE)的效果，将其与传统蒙特卡洛法(MCS)与状态枚举法(SE)进行对比。本方法的搜索深度NCTG同样设置为1；在蒙特卡罗法中，将收敛判据总采样数NMCS设置为105。由于样本量巨大，蒙特卡罗法可以得出足够精确的结果，因此可将其算例结果作为评估其他算例精度的基准。以上三种方法的评估结果如表2、图9a、图9b、图10a和图10b所示。
表2 三种评估方法结果(IEEE-118)
Figure PCTCN2015088389-appb-000019
表2展示了两种可靠性指标(EENS和PLC)的评估结果。可以看出采用蒙特卡罗法和本方法得出的指标非常接近，它们的相对误差在2％左右(通过表2可以看出EENS的误差为1.4590％，PLC的误差为2.1644％)。而传统枚举法得出的两指标误差极大，约为98％(通过表2可以看出EENS的误差为98.1514％，PLC的误差为98.0834％)，远高于其他方法。同时由于只考虑了一阶故障，IISE和SE的运算时间相近，但比MCS用时短很多。表明在该电力***中本方法比传统的评估方法效率更高。
图9a、图9b分别给出了蒙特卡罗法所得EENS和PLC的收敛曲线，而图10a和图10b 分别给出了这两个指标的相对误差收敛曲线。本方法与状态枚举法的计算结果也在该些图中给出。从图10a和图10b中可以看出，本方法计算精度远高于状态枚举法，同时本方法计算用时与状态枚举法近似。通过相对误差收敛曲线可以看出，蒙特卡洛法相对误差稳定在2％以内需要大约3×104秒，而本方法可在1500秒内达到同样精度，用时约为蒙特卡洛法的1/20。
因此可以得出结论，本方法比其他两种传统的可靠性评估方法具有更高的精度和计算效率。
本领域技术人员可以理解附图只是一个优选实施例的示意图，上述本发明实施例序号仅仅为了描述，不代表实施例的优劣。
以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。

Claims (10)

  1. 一种基于影响增量的状态枚举可靠性评估方法，其特征在于，所述方法包括以下步骤：
    通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性，若存在不可达元素，则所选电力***状态的影响增量为零，重新选择电力***状态；
    若不存在不可达元素，则通过最优潮流算法评估所有负荷水平下的电力***状态的影响，获取电力***状态在各负荷水平下的影响期望，进而获取电力***状态的影响增量；
    当状态集中所有电力***状态已被分析，且已达到最大故障搜索阶数时，通过影响增量获取电力***可靠性指标。
  2. 根据权利要求1所述的一种基于影响增量的状态枚举可靠性评估方法，其特征在于，在通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性的步骤之前，所述方法还包括：
    通过微扰法获取各设备阻抗对各支路潮流的灵敏度，根据灵敏度确定各设备间的独立性；
    从状态集中选择一个电力***状态，通过设备间的独立性创建电力***状态的独立性邻接矩阵。
  3. 根据权利要求2所述的一种基于影响增量的状态枚举可靠性评估方法，其特征在于，所述方法还包括：输入电力***数据，设备可靠性数据和预置参数，并初始化故障阶数。
  4. 根据权利要求3所述的一种基于影响增量的状态枚举可靠性评估方法，其特征在于，所述预置参数包括：最大故障搜索阶数和设备独立性灵敏度阈值。
  5. 根据权利要求4所述的一种基于影响增量的状态枚举可靠性评估方法，其特征在于，所述根据灵敏度确定各设备间的独立性的步骤具体为：
    若存在一条支路，使得一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值，且另一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值时，两故障设备之间不独立。
  6. 一种基于影响增量的状态枚举可靠性评估装置，其特征在于，所述装置包括：
    检验模块，用于通过广度优先搜索法检验所选电力***状态对应的独立性邻接矩阵中所有元素的可达性，若存在不可达元素，则所选电力***状态的影响增量为零，重新选择电力 ***状态；
    第一获取模块，用于若不存在不可达元素，通过最优潮流算法评估所有负荷水平下的电力***状态的影响，获取电力***状态在各负荷水平下的影响期望，进而获取电力***状态的影响增量；
    第二获取模块，用于当状态集中所有电力***状态已被分析，且已达到最大故障搜索阶数时，通过影响增量获取电力***可靠性指标。
  7. 根据权利要求6所述的一种基于影响增量的状态枚举可靠性评估装置，其特征在于，所述装置还包括：
    第三获取模块，用于通过微扰法获取各设备阻抗对各支路潮流的灵敏度；
    确定模块，根据灵敏度确定各设备间的独立性；
    创建模块，用于从状态集中选择一个电力***状态，通过设备间的独立性创建电力***状态的独立性邻接矩阵。
  8. 根据权利要求7所述的一种基于影响增量的状态枚举可靠性评估装置，其特征在于，所述装置还包括：
    输入和初始化模块，用于输入电力***数据，设备可靠性数据和预置参数，并初始化故障阶数。
  9. 根据权利要求8所述的一种基于影响增量的状态枚举可靠性评估装置，其特征在于，所述预置参数包括：最大故障搜索阶数和设备独立性灵敏度阈值。
  10. 根据权利要求9所述的一种基于影响增量的状态枚举可靠性评估装置，其特征在于，所述确定模块包括：
    确定子模块，用于若存在一条支路，使得一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值，且另一故障设备的阻抗对支路潮流分布的灵敏度指标大于所述设备独立性灵敏度阈值时，两故障设备之间不独立。
PCT/CN2015/088389 2015-07-28 2015-08-28 一种基于影响增量的状态枚举可靠性评估方法及其装置 WO2017016021A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US15/747,522 US20180375373A1 (en) 2015-07-28 2015-08-28 Impact increments-based state enumeration reliability assessment approach and device thereof

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510456039.9A CN105071381B (zh) 2015-07-28 2015-07-28 一种基于影响增量的状态枚举可靠性评估方法及其装置
CN201510456039.9 2015-07-28

Publications (1)

Publication Number Publication Date
WO2017016021A1 true WO2017016021A1 (zh) 2017-02-02

Family

ID=54500694

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/088389 WO2017016021A1 (zh) 2015-07-28 2015-08-28 一种基于影响增量的状态枚举可靠性评估方法及其装置

Country Status (3)

Country Link
US (1) US20180375373A1 (zh)
CN (1) CN105071381B (zh)
WO (1) WO2017016021A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111950894A (zh) * 2020-08-11 2020-11-17 杜金其 一种基于改进蒙特卡洛法的电力***可靠性评估方法

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105680442B (zh) * 2016-03-07 2018-06-15 重庆大学 考虑潮流和灵敏度一致性等值的期望缺供电量评估方法
CN107508279B (zh) * 2017-08-08 2020-02-18 国网山东省电力公司荣成市供电公司 一种电力网络的稳定性仿真方法
CN110298766B (zh) * 2019-07-02 2021-11-02 广东电网有限责任公司 适于继电保护整定计算的负荷筛选方法、装置及设备
CN111244944B (zh) * 2020-01-23 2023-04-18 天津大学 一种基于影响增量的交直流配电网可靠性评估方法
CN111313411B (zh) * 2020-03-11 2023-08-04 国网天津市电力公司 基于重要抽样影响增量的电力***可靠性评估方法及装置
CN111651911B (zh) * 2020-04-17 2022-09-02 北京理工大学 一种集总元件阻抗灵敏度快速计算方法及优化方法
CN111673291B (zh) * 2020-06-03 2021-11-05 广东省智能制造研究所 一种激光切割机精度保持性的评估方法
CN112039211B (zh) * 2020-09-07 2022-03-01 国网四川省电力公司电力科学研究院 智能变电站二次安措可靠性与复杂度的优化方法
CN112288326B (zh) * 2020-11-23 2023-04-18 天津大学 一种适用于输电***韧性评估的故障场景集削减方法
CN112668177B (zh) * 2020-12-25 2022-12-16 天津大学 一种对多重故障的配电***进行可靠性评估方法
CN113177717B (zh) * 2021-05-06 2022-08-26 天津大学 一种基于影响增量灵敏度的输电***韧性快速评估方法
CN113609721B (zh) * 2021-07-13 2023-11-07 天津大学 多类型极端灾害电气互联***韧性计算方法及装置
CN113887989B (zh) * 2021-10-15 2024-01-16 中国南方电网有限责任公司超高压输电公司柳州局 电力***可靠性评估方法、装置、计算机设备和存储介质
CN116720324A (zh) * 2023-05-15 2023-09-08 中铁第四勘察设计院集团有限公司 基于预测模型的牵引变电所关键设备故障预警方法及***
CN116579181B (zh) * 2023-06-02 2023-11-24 天津大学 一种基于激活约束的电力***可靠性快速评估方法及装置

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103324744A (zh) * 2013-07-03 2013-09-25 国家电网公司 基于配电环网自动成图的拓扑搜索方法
CN103632313A (zh) * 2013-12-05 2014-03-12 国家电网公司 一种基于pmu数据的电网动态可观方法
CN103902814A (zh) * 2014-03-10 2014-07-02 中国南方电网有限责任公司 基于动态分区的电力***运行状态检测方法
JP2014138488A (ja) * 2013-01-17 2014-07-28 Espec Corp パワーサイクル試験装置
CN104156883A (zh) * 2014-08-01 2014-11-19 重庆大学 基于分块枚举法的风电场集电***可靠性评估方法
CN104518488A (zh) * 2014-12-30 2015-04-15 广西大学 用于配电网可靠性分析的负荷点故障区域类型划分方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9300134B2 (en) * 2012-06-26 2016-03-29 Eleon Energy, Inc. Methods and systems for power restoration planning
CN103914788B (zh) * 2014-03-04 2017-08-08 广东电网公司电力科学研究院 电网多环节***的可靠性评估方法
CN104376504B (zh) * 2014-11-06 2017-10-27 国家电网公司 一种基于解析法的配电***概率可靠性评估方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014138488A (ja) * 2013-01-17 2014-07-28 Espec Corp パワーサイクル試験装置
CN103324744A (zh) * 2013-07-03 2013-09-25 国家电网公司 基于配电环网自动成图的拓扑搜索方法
CN103632313A (zh) * 2013-12-05 2014-03-12 国家电网公司 一种基于pmu数据的电网动态可观方法
CN103902814A (zh) * 2014-03-10 2014-07-02 中国南方电网有限责任公司 基于动态分区的电力***运行状态检测方法
CN104156883A (zh) * 2014-08-01 2014-11-19 重庆大学 基于分块枚举法的风电场集电***可靠性评估方法
CN104518488A (zh) * 2014-12-30 2015-04-15 广西大学 用于配电网可靠性分析的负荷点故障区域类型划分方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111950894A (zh) * 2020-08-11 2020-11-17 杜金其 一种基于改进蒙特卡洛法的电力***可靠性评估方法

Also Published As

Publication number Publication date
CN105071381B (zh) 2017-04-12
CN105071381A (zh) 2015-11-18
US20180375373A1 (en) 2018-12-27

Similar Documents

Publication Publication Date Title
WO2017016021A1 (zh) 一种基于影响增量的状态枚举可靠性评估方法及其装置
Liu et al. Recognition and vulnerability analysis of key nodes in power grid based on complex network centrality
CN110417011A (zh) 一种基于互信息与迭代随机森林的在线动态安全评估方法
CN106021097A (zh) 基于测试特征的软件可靠性指标区间估计方法
CN109657913B (zh) 一种考虑分布式电源的输配电网联合风险评估方法
CN111881124A (zh) 一种基于改进算法的状态估计的数据处理方法及***
CN107305651B (zh) 一种输电***可靠性评估方法和***
CN114660461A (zh) 电池健康状态的评估方法、装置、电子设备和存储介质
Binelo et al. Mathematical modeling and parameter estimation of battery lifetime using a combined electrical model and a genetic algorithm
US11169188B2 (en) Low-observability matrix completion
CN107204616B (zh) 基于自适应稀疏伪谱法的电力***随机状态估计方法
Barzegkar-Ntovom et al. Methodology for evaluating equivalent models for the dynamic analysis of power systems
Guillen et al. Data‐driven short‐circuit detection and location in microgrids using micro‐synchrophasors
Correa-Henao et al. Representation of electric power systems by complex networks with applications to risk vulnerability assessment
KR101988670B1 (ko) 배전계통의 상태추정 장치
CN102738794B (zh) 基于赛德尔式递推贝叶斯估计的电网拓扑错误辨识方法
CN103957114A (zh) 一种基于变异系数的网络抗毁性评估方法
Vasudevan et al. Improved state estimation by optimal placement of measurement devices in distribution system with ders
CN111191867B (zh) 电力***复杂网络的可靠性评估方法
CN115964951A (zh) 电力***运行可靠性评估方法、装置及电子设备
Kabiri et al. Enhancing power system state estimation by incorporating equality constraints of voltage dependent loads and zero injections
CN114386510A (zh) 一种电力***量测错误辨识方法及***
CN109861214B (zh) 判断区域电网暂态功角稳定薄弱线路的方法、***
Fu et al. Bayesian estimation based parameter estimation for composite load
CN105930613A (zh) 一种分布式发电***等值建模方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15899373

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15899373

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 01.08.2018)

122 Ep: pct application non-entry in european phase

Ref document number: 15899373

Country of ref document: EP

Kind code of ref document: A1