CN102915515A - Cascade outage recognition and risk evaluation method of power grid - Google Patents

Abstract

The invention relates to the technical field of safety protection of a power system, and in particular relates to a cascade outage recognition and risk evaluation method of a power grid. The cascade outage recognition and risk evaluation method comprises the following steps of: obtaining power grid historical fault information, determining a power gird anticipated accident list, determining a cascade fault propagation and evolution mechanism, determining possibility, ponderance and a risk measurement value of a power grid cascade fault, and establishing a cascade fault recognition and risk evaluation model. According to an actual physical propagation process and a mathematical property of the power grid cascade fault, credibility measurement and total load loss of accident occurrence possibility and ponderance are established by a credibility theory, and are solved by adopting a fuzzy simulation method and load flow calculation, and the contradiction between simulation precision and sampling frequency in the traditional method is solved. Simulation analysis on a WSCC 9 node testing system proves that the method has correctness and robustness, is little in dependence of a sample, and has important theoretical significance and practical significance in preventing large-area power failure accidents of a power grid.

Description

A kind of power grid cascading Fault Identification and methods of risk assessment

Technical field

The present invention relates to power system security guard technology field, be specifically related to a kind of power grid cascading Fault Identification and methods of risk assessment.

Background technology

Along with the day by day increase of net capacity and scale, electrical network becomes increasingly complex, and when improving electrical network validity and robustness, the security risk of electrical network has also increased thereupon.In recent years, a lot of large-area power-cuts that caused by cascade failure have occured in the whole world, as 1996 7, August US West the accident of having a power failure on a large scale occurs twice in succession; Large area blackout has all successively occured in North America and Canada, London, Sweden and Denmark, Italy in the second half year in 2003; In January, 2008 south China the large-area power-cuts event that causes of ice and snow calamity, all caused huge economic loss and serious social danger, even had influence on national security.Therefore, for formulating effective Prevention-Security system, the raising electrical network is taken precautions against natural calamities, anti-disaster ability, needs accurately to identify the cascade failure that electrical network may occur, and the potential risk that electrical network faces is assessed.

The power grid cascading fault is affected by many inducements, as: weather, environment, topological structure of electric etc. have complicated uncertain.Existing method is take Monte Carlo simulation as the basis, and with the uncertain attribute of classical probability measure or fuzzy mearue portrayal cascade failure, concrete grammar can be divided into stochastic appraisal method and THE FUZZY EVALUATING METHOD.Wherein, the stochastic appraisal method is very responsive to this class rare event of cascade failure, need carry out bulk sampling and calculate when guaranteeing simulation accuracy.Contradiction for simulation accuracy and simulation precision in the solution stochastic appraisal method, be called " Optimal locations for protection system enhancement:a simulation of cascading outages[J] in name " (Wang H Y, Thorp J S.IEEE Transaction on PowerDelivery, 2001,16 (4): propose variance in document 528-533) and reduce technology, be called " Generatingcapacity reliability evaluation based on Monte Carlo simulation and cross entropymethods[J] in name ". (Leite da Silva A M, Fernandez R A G, Singh C.IEEE Transactions on PowerSystems, 2010,25 (1): propose cross entropy in document 129-137) and reduce technology, attempt to remedy by change of scale the deficiency of traditional stochastic appraisal method, but choosing the assessment result impact of optimal scale function is very large.THE FUZZY EVALUATING METHOD utilizes expertise to find the solution, though can improve to a certain extent simulation efficiency, the method complicated operation, and lack tight mathematical theory support.

Summary of the invention

The object of the present invention is to provide a kind of power grid cascading Fault Identification and methods of risk assessment, solve present identification for the power grid cascading fault and methods of risk assessment precision and efficient not high, and the problem of complicated operation.

For solving above-mentioned technical matters, the present invention by the following technical solutions:

A kind of power grid cascading Fault Identification and methods of risk assessment may further comprise the steps:

Obtain electrical network historical failure information: the running status of electrical network is provided by the telemetry intelligence (TELINT) that provides in the electrical network SCADA system and remote signalling information, is obtained parameter information under the electrical network current state, obtain the inducement and the statistical information that affect power grid security;

Determine the electrical network contingency set: by historical statistics information or manually set out contingency set;

Determine cascade failure Propagation and evolution mechanism: determine the branch road group that next stage may break down by the branch road group who breaks down on last stage;

Determine possibility, seriousness and Risk Measurement value that the power grid cascading fault occurs: the possibility and the seriousness that adopt respectively credibility measure and total load loss portrayal accident to occur, with both Risk Measurements of product metric accident;

Set up cascade failure identification and risk evaluation model: the Monte-Carlo Simulation number of times is set, with the initial incident of contingency set as cascade failure, identify contingent cascade failure, calculate credibility measure, total load loss and the Risk Measurement of cascade failure;

Comprehensive all possible cascade failure calculates the potential risk that electrical network faces under the current state.

Further above-mentioned telemetry intelligence (TELINT) comprises electric parameters and the load trend of electrical network, and remote signalling information comprises various protections and switching value information.

Further above-mentioned telemetry intelligence (TELINT) and the remote signalling information running status of determining electrical network by providing in the electrical network SCADA system obtained parameter information under the electrical network current state, and its parameter information comprises node voltage, load and generator output.

The element that further above-mentioned forecast accident is concentrated is key element or the element that the most easily breaks down.

Compared with prior art, the invention has the beneficial effects as follows: according to actual physics communication process and the mathematical properties of power grid cascading fault, credibility measure and the total load loss of accident possibility occurrence and seriousness have been set up by Credibility Theory, and calculated by fuzzy simulation method and trend and to find the solution, solved simulation accuracy in the classic method and the contradiction between the frequency in sampling.By WSCC 9 node test systems are carried out simulation analysis, proved correctness, the robustness of the method, little to the sample dependence, the method has important theory significance and practical significance to electrical network prevention large area blackout.

Description of drawings

Fig. 1 is the schematic flow sheet of a kind of power grid cascading Fault Identification of the present invention and a kind of embodiment of methods of risk assessment.

Fig. 2 is that a kind of power grid cascading Fault Identification of the present invention and a kind of embodiment of methods of risk assessment are used for WSCC 9 node test system of systems electrical wiring schematic diagram.

Embodiment

In order to make purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, is not intended to limit the present invention.

Fig. 1 shows a kind of embodiment of a kind of power grid cascading Fault Identification of the present invention and methods of risk assessment: a kind of power grid cascading Fault Identification and methods of risk assessment may further comprise the steps:

Obtain electrical network historical failure information: the running status that electrical network is provided by the telemetry intelligence (TELINT) that provides in the electrical network SCADA system and remote signalling information, obtain parameter information under the electrical network current state, obtain the inducement and the statistical information that affect power grid security, as: " closing " or " closing " state etc. that is in of switch, this information can be determined the current topological sum running state information of electrical network, the power grid cascading fault occurs and the factor of propagation is a lot of and affect, and comprising: external environment (such as hail, thunderbolt etc.), manually-operated error, wire touch tree, aging of electrical element etc.According to the historical failure statistical information, determine the inducement that has the greatest impact and the component information that the most easily breaks down;

Determine the electrical network contingency set: by historical statistics information or manually set out contingency set, generally speaking, the key element in the electrical network is electric pressure is higher, trend is heavier branch road, interconnection etc.And the element that the most easily breaks down should be and be in for a long time rugged surroundings or near the element of operational limit, can be by the historical failure acquisition of information.The element that forecast accident is concentrated may be single element, also may be a plurality of elements;

Determine cascade failure Propagation and evolution mechanism: determine the branch road group that next stage may break down by the branch road group who breaks down on last stage; studies show that; the hidden failure of protective device is the main cause that causes the power grid cascading fault; usually caused by the unreasonable of hardware deficiency or setting valve; these hidden danger only just occur under the states such as electric network fault mostly, and its consequence is to cause protecting malfunction or tripping.Actual electric network one general configuration the protections of two covers, the probability of simultaneously trippings of two covers protections is very low, the present invention ignores its impact, only considers to cause protecting the hidden failure of malfunction;

Determine possibility, seriousness and Risk Measurement value that the power grid cascading fault occurs: the possibility and the seriousness that adopt respectively credibility measure total load loss portrayal accident to occur, with both Risk Measurements of product metric accident;

Set up cascade failure identification and risk evaluation model: the Monte-Carlo Simulation number of times is set, with the initial incident of contingency set as cascade failure, identify contingent cascade failure, calculate credibility measure, total load loss and the Risk Measurement of cascade failure;

Comprehensive all possible cascade failure calculates the potential risk that electrical network faces under the current state.

SCADA (Supervisory Control And Data Acquisition) system wherein, i.e. data acquisition and supervisor control.The SCADA system is take DCS and the power automation supervisory system of computing machine as the basis; Its application is very wide, can be applied to the data acquisition and the numerous areas such as supervision control and process control in the fields such as electric power, metallurgy, oil, chemical industry.

A kind of power grid cascading Fault Identification and methods of risk assessment according to the present invention, credibility measure is a kind of of fuzzy mearue, can be used for portraying the possibility that fuzzy event occurs.Credibility measure is that possibility is estimated the comprehensive measurement of estimating with necessity.Wherein, possibility is estimated the easy degree that the description event occurs, and necessity is estimated the impossibility of describing the complementary events generation.Compare with other fuzzy mearue, credibility measure has and the similar self-duality of probability measure, and namely when credibility measure was 1, event must occur; And when credibility measure was 0, event is inevitable not to be occured.It is suitable with the status of probability measure in theory of probability that the status of credibility measure in fuzzy theory is considered to.

In the cascade failure communication process, the branch road that every one-phase breaks down directly depends on fault branch on last stage.When cascade failure is transmitted to j during the stage, suppose total n _jPlanting branch road may break down, and can randomly draw the branch road that breaks down with certain " estimating " this moment by Monte Carlo method, then total

Plant fault sampling form.Therefore, under the effect of initial incident, electrical network may comprise a large amount of cascade failure sequences, and available Monte Carlo method is identified.

As more detailed explanation:

Determine in the present invention possibility, seriousness and the Risk Measurement value that the power grid cascading fault occurs, the base attribute of power grid cascading fault is described with " ambiguity ", based on Credibility Theory and the electric network swim theory of computation, with possibility and seriousness that credibility measure and total load loss amount portrayal accident occur, it is defined as follows:

1, credibility measure: when cascade failure is transmitted to j during the stage, at first determine the active-power P of each branch road _Ij, and find the solution P _IjCorresponding membership function μ _Ij(P _Ij).Then calculate M according to the fuzzy simulation method _Pos(A _j) and M _Pec(A _j), obtain thus the credibility measure M that j stage accident occurs _Cr(A _j).By M _Cr(A _j) can obtain the credibility measure M that cascade failure occurs _Cr(A).

2, total load loss: by the total load amount of each node of trend calculative determination, then subtract each other with initial load first, obtain the total load loss M of electrical network _Sev(A).

3, Risk Measurement: the Risk Measurement value M that is obtained accident by the product of credibility measure and total load loss _Risk(A).

Its concrete computing method are as follows:

1, the credibility measure M of power grid cascading fault A generation _Cr(A) can be expressed as:

$M_{\mathrm{cr}}\left(A\right)=\frac{1}{2}(M_{\mathrm{pos}}\left(A\right)+M_{\mathrm{nec}}\left(A\right))---\left(1\right)$

Wherein,

$M_{\mathrm{nec}}\left(A\right)=1-M_{\mathrm{pos}}\left(\stackrel{\‾}{A}\right)---\left(2\right)$

In the formula:

Supplementary set for A; Possibility is estimated M _Pos(A) the possible degree of expression A generation, necessity is estimated M _Pec(A) expression

Impossible degree.

Credibility measure M _CrHave and the similar self-duality of probability measure.Obviously, for cascade failure A, work as M _Cr(A) value is 1 o'clock, and A must occur; And work as M _Cr(A) value is 0 o'clock, and A is inevitable not to be occured.

The credibility measure M that cascade failure A occurs _Cr(A) be transmitted to the credibility measure M that the j stage breaks down with accident _Cr(A _j) relevant.The present invention intends adopting the fuzzy simulation method to find the solution M _Cr(A _j), step is as follows:

Step 1: from fuzzy vector P ₁, P ₂Domain in evenly produce P _{1, t}, P _{2, t}, so that μ _Ij(P _{1, t}) 〉=ε, μ _Ij(P _{2, t}) 〉=ε.Wherein t is frequency in sampling, and t=1,2, and, T, ε are abundant little positive number.

Step 2: order

Step 3: calculate M _Cr(A _j), then have

$M_{\mathrm{cr}}\left(A_j\right)=\frac{1}{2}\left(\underset{1\&le;t\&le;T}{\max }\right\{v_h|P_{1,t}\&le;{\stackrel{\&OverBar;}{P}}_1,{\mathrm{<figure-callout\; id="2"\; label="P"\; filenames="HDA00002383904900012.png"\; state="\{\{state\}\}">P</figure-callout>}}_{2,t}\&GreaterEqual;{\stackrel{\&OverBar;}{P}}_2\}+$ (4)

$\underset{1\&le;t\&le;T}{\min }\{1-v_t|P_{1,t}\&GreaterEqual;{\stackrel{\&OverBar;}{P}}_1,{\mathrm{<figure-callout\; id="2"\; label="I"\; filenames="HDA00002383904900012.png"\; state="\{\{state\}\}">I</figure-callout>}}_{2,t}\&le;{\stackrel{\&OverBar;}{P}}_2\})$

In the formula: i is the branch road sequence number; J is the cascade failure propagation stage; P _IjBe branch road L _IjActive power, its membership function is μ _Ij(P _Ij);

Be branch road L before j stage accident occurs _IjOn active power; Vector P ₁,

Be respectively the j stage because of the cut branch power P of hidden failure _IjWith

The vector that consists of; Vector P ₂,

With the branch power P that is respectively the j stage and is in normal operating condition _IjWith

The vector that consists of.In like manner can find the solution the credibility measure M that cascade failure occurs _Cr(A).

The present invention adopts following function representation membership function, is determined by base attribute and the sample data of actual physics phenomenon in the reality.

$\mathrm{\&mu;}\left(P\right)=\left\{\begin{array}{cc}0& P<P_{\mathrm{HF}}\\ \frac{2P-P_{\mathrm{HF}}}{2P_{\mathrm{set}}-P_{\mathrm{HF}}}& P_{\mathrm{HF}}\&le;P\&le;P_{\mathrm{set}}\\ 1& P>P_{\mathrm{set}}\end{array}\right.---\left(5\right)$

In the formula, P _HF, P _SetTrigger threshold value and the setting valve of hidden failure for protective device.The physical significance of formula (5) is: when branch road is gained merit P greater than protection setting valve P _SetThe time, this branch road is cut because of the overcurrent protection action; When P less than P _HFThe time, this branch road normally moves; When P is positioned at P _HFAnd P _SetBetween the time, this branch road may be cut because of hidden failure.

2, the seriousness of accident generation is estimated M _Sev(A)

Because load is one of topmost variable of electrical network, in the linksystem accident communication process, when line voltage or Frequency generated depart from, the operation of taking mostly with load relevant.Therefore, can load with electrical network total losses part the order of severity of portrayal accident generation.

3, the Risk Measurement M of accident generation _Risk(A)

Risk Measurement is the comprehensive measurement of accident possibility occurrence and seriousness, is embodied as:

M _risk(A)＝M _cr(A)M _sev(A) (6)

When setting up cascade failure identification and risk evaluation model among the present invention, the step of cascade failure identification and risk assessment is:

Step 1: the Monte-Carlo Simulation number of times is set, for the degree of dependence of explanation the present invention to sample, selects respectively several different simulation times to study.It should be noted that the test macro scale is larger, reach the required simulation times of certain precision just more.

Step 2: with the initial incident of contingency set as cascade failure, cascade failure is triggered by single failure usually, and the element fault that the present invention is concentrated by forecast accident is as the initial incident of cascade failure.

Step 3: identify contingent cascade failure:

1) the concentrated branch road generation primary fault of forecast accident, the protection action, fault branch is cut;

2) cause that electric network swim shifts behind certain branch trouble on last stage, when the active power that flows through other branch road surpasses a certain threshold value, trigger hidden failure with certain " estimating ", corresponding branch road is cut.

3) repetitive process 2.In the cascade failure communication process, if take effective defensive measure, cascade failure will stop; Otherwise when fault recurs, and harm just may cause the electric grid large area power cut accident when acquiring a certain degree.Therefore, the branch road group that breaks down of each stage is out-of-limit or the branch road of hidden failure occurs for trend.

In addition, the present invention intends adopting the mode of topology search to identify the branch road that stages breaks down.

Step 4: the credibility measure, total load loss and the Risk Measurement that calculate cascade failure

Calculate credibility measure, total load loss and the Risk Measurement that the research cascade failure occurs with fuzzy simulation and trend.

Step 5: comprehensive all possible cascade failure, calculate the potential risk that electrical network faces under the current state

Under the effect of contingency set, a lot of cascade failures may occur in electrical network, and average risk is estimated M _{Average_risk}For:

$M_{\mathrm{average}\_\mathrm{rsik}}=\frac{1}{N}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{M}_{\mathrm{risk}}\left(j\right)---\left(7\right)$

In the formula, M _Risk(j) be the Risk Measurement of j cascade failure.

The another kind of embodiment of a kind of power grid cascading Fault Identification and methods of risk assessment according to the present invention, telemetry intelligence (TELINT) comprises electric parameters and the load trend of electrical network, remote signalling information comprises various protections and switching value information.

The another kind of embodiment of a kind of power grid cascading Fault Identification and methods of risk assessment according to the present invention, the running status of electrical network is provided by the telemetry intelligence (TELINT) that provides in the electrical network SCADA system and remote signalling information, obtain parameter information under the electrical network current state, its parameter information comprises node voltage, load and generator output.

The element that the another kind of embodiment of a kind of power grid cascading Fault Identification and methods of risk assessment according to the present invention, forecast accident concentrate is key element or the element that the most easily breaks down.

The below carries out cascade failure identification and risk-assessment as an example of WSCC 9 node test systems example, the system electrical wiring as shown in Figure 2.

Choose branch road L ₁Be contingency set, adopt respectively sample calculation 10000,8000,6000,4000,2000 times, institute's extracting method can accurately identify 104 kinds of whole cascade failures among the present invention.Wherein, be 10000 o'clock in simulation times, 5 kinds of accident sequences that possibility occurrence is larger such as following table.

Ask for credibility measure, the seriousness of various linksystem accidents generations and estimate and Risk Measurement, ask for the potential risk that electrical network faces under the current running status by single Risk Measurement again.Below 2 tables be followed successively by 5 kinds of larger measure values that accident is corresponding of possibility occurrence, and the average risk of electrical network under various frequency in samplings estimated.As seen, simulation value and the actual value of the method are more or less the same, and maximum relative error only is 2.37%.

Via power grid cascading Fault Identification and methods of risk assessment based on Credibility Theory that the present invention proposes, little to the sample dependence, method is accurate and effective.The method is formulated the measure of preventing and fighting natural adversities important theoretical foundation is provided for identification electrical network weak link.

Although invention has been described with reference to a plurality of explanatory embodiment of the present invention here, but, should be appreciated that, those skilled in the art can design a lot of other modification and embodiments, and these are revised and embodiment will drop within the disclosed principle scope and spirit of the application.More particularly, in the scope of, accompanying drawing open in the application and claim, can carry out multiple modification and improvement to building block and/or the layout of subject combination layout.Except modification that building block and/or layout are carried out with improving, to those skilled in the art, other purposes also will be obvious.

Claims (4)

1. a power grid cascading Fault Identification and methods of risk assessment is characterized in that may further comprise the steps:

Obtain electrical network historical failure information: the running status of electrical network is provided by the telemetry intelligence (TELINT) that provides in the electrical network SCADA system and remote signalling information, is obtained parameter information under the electrical network current state, obtain the inducement and the statistical information that affect power grid security;

Determine the electrical network contingency set: by historical statistics information or manually set out contingency set;

Determine cascade failure Propagation and evolution mechanism: determine the branch road group that next stage may break down by the branch road group who breaks down on last stage;

Determine possibility, seriousness and Risk Measurement value that the power grid cascading fault occurs: the possibility and the seriousness that adopt respectively credibility measure and total load loss portrayal accident to occur, with both Risk Measurements of product metric accident;

Set up cascade failure identification and risk evaluation model: the Monte-Carlo Simulation number of times is set, with the initial incident of contingency set as cascade failure, identify contingent cascade failure, calculate credibility measure, total load loss and the Risk Measurement of cascade failure;

Comprehensive all possible cascade failure calculates the potential risk that electrical network faces under the current state.

2. a kind of power grid cascading Fault Identification according to claim 1 and methods of risk assessment is characterized in that: described telemetry intelligence (TELINT) comprises electric parameters and the load trend of electrical network, and remote signalling information comprises various protections and switching value information.

3. a kind of power grid cascading Fault Identification according to claim 1 and methods of risk assessment, it is characterized in that: described telemetry intelligence (TELINT) and the remote signalling information running status of determining electrical network by providing in the electrical network SCADA system, obtain parameter information under the electrical network current state, its parameter information comprises node voltage, load and generator output.

4. a kind of power grid cascading Fault Identification according to claim 1 and methods of risk assessment is characterized in that: the element that described forecast accident is concentrated is key element or the element that the most easily breaks down.

Images