CN103296677A - On-line large power grid recovery assistant decision-making system - Google Patents

Abstract

The invention relates to an on-line large power grid recovery assistant decision-making system. The system comprises two work modes, namely a real-time assistant mode and a research and analysis mode, a fracture surface operation analysis module, a calculation configuration module, a calculation function module and a calculation management module, wherein the fracture surface operation analysis module is used for fracture surface reading and fracture surface adjustment, the calculation configuration module is used for setting calculation parameters and selecting a black-start power supply and a started power supply, the calculation function module is used for system partitioning, load flow calculation, electromagnetic transient calculation, electromechanical transient calculation and path search and assessment, and the calculation management module is used for displaying system partitioning schemes, voltage waveforms, ranking of available paths and a black-start generating scheme. When the on-line large power grid recovery assistant decision-making system is used for developing an on-line electric power system recovery assistant decision-making supporting system, existing data resources in a dispatching center can be sufficiently utilized, the advantages of the method of pre-arranged scheme recovery and intelligent dispatching are combined, a model does not need to be rebuilt, the construction time and cost of the system are reduced, the real-time performance and the accuracy of suggestions to decision-making are improved, and a powerful support is provided for rapid and reliable recovery of a power grid after power outages.

Description

A kind of online big power system restoration aid decision-making system

Technical field

The present invention relates to field of power, be specifically related to a kind of online big power system restoration aid decision-making system.

Background technology

Since electric power system occurred, the massive blackout accident happened occasionally, and along with the expansion of electric power system scale, the frequency of its generation is more apparent frequently, influence is more abominable.On August 3rd, 1996 the Malaysian whole nation have a power failure on a large scale, zone, Brazilian capital on the 21st January in 2002 has a power failure on a large scale, August in 2003 14 Japan-US add have a power failure on a large scale, on May 25th, 2005 Russia have a power failure on a large scale, on September 26th, 2005 Hainan the whole province have a power failure on a large scale, on September 15th, 2011 Korea S have a power failure on a large scale and history on influence the maximum India in 30 days July in 2012 of number and have a power failure on a large scale, it is influenced to surpass 6.7 hundred million people.Massive blackout brings enormous economic loss to the mankind, has had a strong impact on people's life, and may cause social unrest to a certain extent.The electric power system accident theoretical research of having a power failure on a large scale shows: the generation of having a power failure on a large scale concerning any one electric power system always exists with small probability, and probability of happening and influence degree increase along with the expansion of system scale, and the accident of having a power failure on a large scale of electric power system is difficult to avoid.After electrical network was had a power failure on a large scale accident, the system of restoring electricity had very important meaning fast.

Current, the black common way that starts of electric power system both at home and abroad is to study and define out feasible black startup prediction scheme by experienced management and running expert in advance according to the concrete condition of system, or calculate feasible black startup prediction scheme by off-line system, help dispatcher's recovery system operation when system blackout.When being the prediction scheme formulation, its advantage passed through a large amount of fail safe verifications, and generally all can organize and deceive firing test, the related personnel can therefrom run up to valuable experience, the each side that participation system recovers, comprise the staff who dispatches mechanism, power plant and transformer station, comparatively familiar to prediction scheme, rulyly in can the system after the accident of having a power failure on a large scale takes place recovering carry out by scheduled operation.As long as reality satisfies the prediction scheme condition, carry out black start prediction scheme be fast, the reliable fashion of safety, reliable recovery system power supply.Its shortcoming is that off-line system modeling workload is big, expert's experience and off-line system model often in time the radical actual electric network change, and the accident in the recovery process and some uncontrollable factors (the electric power facility damage that causes as overhaul of the equipments stoppage in transit, equipment fault, natural calamity etc.) may make the partial content in the prediction scheme no longer feasible.In the face of new situation, the dispatcher need do the analysis of large amount of complex can determine new recovery scheme, causes interruption, delay even the failure of black start-up course easily.

During " 12 ", national grid will build up North China, East China, Central China (" three China ") extra-high voltage grid, form the ultra high voltage rack of " three vertical three horizontal looped networks ".The big electrical network of extra-high voltage alternating current-direct current series-parallel connection that electric pressure is the highest in the world, transmission capacity is maximum, technical merit is most advanced, operation characteristic is the most complicated forms.Simultaneously, along with the electrical network scale is increasing, management and running information becomes increasingly complex, and technical merit and the complexity of electrical network are more and more higher, the difficulty of management and running management is also increasing, and the power system restoration solution formulation after the massive blackout accident is even more complex and difficulty also.Dominating down of State Grid Corporation of China, the research and development of intelligent grid dispatching technique back-up system have obtained interim achievement, and apply in more than 20 the provincial scheduling of net mechanisms.System follows the design concept of " laterally integrated, vertical perforation " and " maintenance of source end, the whole network are shared ", has made up four big classes at integrated basic platform and has used, and integrated scheduling provides technical support.Its research and development and pilot are used and to have been promoted scheduling at different levels mechanism overall application level significantly, and dispatching for intelligent grid provides strong technical support, has brought huge economic and social benefit simultaneously.

Strengthened in the designing and developing of intelligent grid dispatching technique back-up system integrated, in linearize, become more meticulous, practicability, visual and intelligentized thought, unified technical standard and functional specification are followed in design and research and development.By working out electrical network universal model Naming conventions, electric network model standard is described, the electrical network figure is described technical specifications such as standard, for all softwares provide unified data-interface and model specification, this starts that model data research, real time data interaction schemes design, software is designed and developed etc. unified technical standard is provided for black, for solid foundation has been established in the software development of online big electrical network black start assistant decision making.

The online power system recovery auxiliary decision technology that possesses adaptive ability based on intelligent grid dispatching technique back-up system has quite high fail safe and reliability, and good adaptability and interactivity, the black startup of the big electrical network prediction scheme formulation that can be under the complex environment provides effectively auxiliary, alleviate dispatcher's work load, also can be at any time according to the system mode in the recovery process, recovery policy is carried out rapid adjustment, adapt to various unexpected variations, escort for system safety fast quick-recovery in back takes place the accident of having a power failure on a large scale.

Summary of the invention

At the deficiencies in the prior art, the purpose of this invention is to provide a kind of online big power system restoration aid decision-making system, use the present invention and develop online power system recovery auxiliary decision-making support system, can take full advantage of control centre's available data resource, combine the advantage of prediction scheme restoration methods and intelligent scheduling, need not modeling again, minimizing system building time and cost, real-time and the accuracy of decision recommendation have been improved, for the fast and reliable that electrical network is had a power failure on a large scale after the accident recovers to provide strong support, employing has vast market prospect based on intelligent grid dispatching technique back-up system specification data interface and man-machine displaying scheme.

The objective of the invention is to adopt following technical proposals to realize:

A kind of online big power system restoration aid decision-making system, its improvements are that described system comprises real-time auxiliary mode and research and analyse two kinds of mode of operations of pattern that described system comprises:

Section Operations Analyst module: be used for section and read and the section adjustment, comprise that recovering target section read module, real-time section read module, the basic section read module of research, research section adjusting module and section recovers module;

Calculating configuration module: be used for calculating parameter setting, black startup Power Selection and be activated Power Selection, comprise that parameter arranges module, black startup Power Selection module and is activated the Power Selection module;

Computing function module: be used for system partitioning, trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment, comprise system partitioning module, trend computing module, electro-magnetic transient computing module, electromechanical transient computing module and route searching and evaluation module;

Management of computing module: be used for system partitioning scheme, voltage waveform, available path rank and the black demonstration that starts the generation scheme, comprise the waveform display module.

Wherein, under real-time auxiliary mode, described section operational module is used for reading recovery target section and real-time section; Researching and analysing under the pattern, described section read module is used for reading the basic section of research and recovers the target section and study the section adjustment and the section recovery;

Read and recover the target section by recovering the realization of target section read module, comprising: from the section storehouse of system, read and specify section as the target section that recovers, load value and the system running state of optimizing each required load bus is provided for analytical calculation;

Read real-time section and realized by real-time section read module, comprising: reading system SCADA uses measurement and the state of all devices down, obtains system's operation present situation;

Read the basic section of research and realized by the basic section read module of research, comprising: from the section storehouse of system, read and specify section as the basic section of researching and analysing usefulness;

The adjustment of research section is realized by research section adjusting module, comprise: (comprising that equipment running status setting, switch and the set of disconnecting link state, unit output and load power size arrange) is set at the basic section of research, obtain simulating the system mode after the accident of having a power failure on a large scale takes place, i.e. black startup problem initial state, this initial state is called the research section;

Section recovers to recover module by section to be realized, comprising: will study basic section revert to last section adjust before the state of (the last study section adjusting module be called before).

Wherein, at real-time auxiliary mode with research and analyse under the pattern, described calculating configuration module all is used for calculating parameter setting, black startup Power Selection and is activated Power Selection;

The calculating parameter setting arranges module by parameter and realizes, described parameter comprises for describing the route searching rule and be used for the calculating parameter of path evaluation ordering, and computation schema is selected, the control parameter of storing path as a result;

The black Power Selection that starts is chosen the module realization by black the startup, and described black startup Power Selection is to specify the black startup power supply that adopts when being used for researching and analysing;

Be activated Power Selection and realize by being activated the Power Selection module, the described Power Selection that is activated is used to specify and researchs and analyses or be activated really when auxiliary in real time power supply.

Wherein, under real-time auxiliary mode, described computing function module is used for trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment; Researching and analysing under the pattern, described computing function module is used for system partitioning, trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment;

Trend is calculated by the trend computing module and is realized, comprises the direction of energy distribution situation in the designated model;

Electro-magnetic transient calculates by the electro-magnetic transient computing module and realizes, adopts the EMTP kit to carry out the electro-magnetic transient analysis and obtains the operating voltage check results;

Electromechanical transient calculates and is realized by the electromechanical transient computing module.

Wherein, route searching is realized by route searching and evaluation module that with assessment route searching is to select from the black power supply that starts to the path that is activated power supply combination property optimum;

Described route searching and evaluation module comprise that restoration path generation module and the black path that starts generate to guide module;

Generating the black restoration path that starts comprises the steps:

I, determine to be activated power supply: the described power supply that is activated adopts fired power generating unit;

II, the rule-based initial path searching algorithm of employing obtain from the black power supply that starts to the feasible electrical path that is activated power supply; Described initial path searching algorithm is followed following rule:

1) voltage transitions number of times restriction principle: the voltage transitions number of times in the black start-up course of control;

2) path restriction principle: the black restoration path length that starts of restriction;

3) factory station restriction principle;

4) capacity limits principle: required minimum load and recovers jumbo unit less than starting exerting oneself of unit when being activated unit starting;

5) important load priority principle;

6) minimum principle of time: the unit of selecting to be in hot stand-by duty;

7) avoid the loop principle: to the path that is activated unit, do not have loop from starting unit.

III, filter out the electrical path by the feasibility verification: the Interventions Requested of described feasibility verification comprise: the verification of generator self-excitation, low-frequency oscillation verification, power-frequency overvoltage verification, switching overvoltage verification and the verification of electric voltage frequency stability;

IV, definite from deceiving the startup power supply to the path that is activated power supply combination property optimum, be the optimized database restore path:, select from the black power supply that starts to the path that is activated power supply combination property optimum assessing by all black startup restoration paths of feasibility verification and sorting; Evaluation index comprise select voltage transitions number of times, path, start-up time, be activated unit capacity, the good and bad degree of technology verification and be activated the priority level of power supply, the good and bad index of technology verification wherein is that summation after generator self-excitation, power-frequency overvoltage multiple, switching overvoltage multiple, variation, the frequency shift (FS) normalization is obtained.

Wherein, described system partitioning is based on the black boot partition of electric power system of fuzzy cluster analysis, realizes that by the system partitioning module described system partitioning is realized by following steps:

(1) set up each node of reflection and each black raw data matrix that starts electrical distance between the power supply:

Defined node i and the black electrical distance that starts between the power supply j are:

$L_{\mathrm{ij}}=\underset{k=1}{\overset{l}{\mathrm{\Σ}}}{\mathrm{\ω}}_{\mathrm{ij}\left(k\right)}{r}_{\mathrm{ij}\left(k\right)}^{\′}$ ①;

Wherein,

r _{Ij (k)}For weighing k desired value of electrical link power, r _{Ij (k)}' be numerical value after its standardization, r _Max, r _MinBe respectively r _{Ij (k)}Maximum, minimum value corresponding to different i; ω _iBe the weights of i index; L represents the index number, is taken as 3; I=1,2 ..., m, j=1,2 ..., n; Node comprises and is activated power supply, transformer station and load bus;

As if being object of classification with the node of the m in the electrical network, black startup power supply number is n, then node i and the black electrical distance formation individual features index that starts between the power supply j, i.e. each element in the raw data matrix:

x _ij＝L _ij ②；

Then set up raw data matrix X=(x _Ij) _{M * n}I=1 wherein, 2 ..., m, j=1,2 ..., n

(2) each element in the described raw data matrix is carried out standardization: adopt the extreme difference standardized method that raw data matrix is carried out standardization, its expression formula is:

$y_{\mathrm{ik}}=\frac{x_{\mathrm{ik}}-\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003168181200031.png"\; state="\{\{state\}\}">min</figure-callout>}}\left\{x_{\mathrm{ik}}\right\}}{\underset{1\&le;i\&le;m}{\max }\left\{x_{\mathrm{ik}}\right\}-\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003168181200031.png"\; state="\{\{state\}\}">min</figure-callout>}}\left\{x_{\mathrm{ik}}\right\}},$ ③;

Wherein: k=1,2 ..., n;

(3) demarcate object to be sorted, set up fuzzy similarity matrix:

Set up fuzzy similarity matrix and be called demarcation again, refer to mark the statistic r that measurement is classified similarity degree between object _IjIf domain X={x ₁, x ₂..., x _m, x _i=(x _I1, x _I2..., x _In), x then _iWith x _jSimilarity degree be r _Ij=R (x _i, x _j), i, j=1,2 ..., m;

Adopt the absolute value index method to find the solution r _Ij, its expression formula is:

$r_{\mathrm{ij}}=\exp (-c\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}|y_{\mathrm{ik}}-y_{\mathrm{jk}}\left|\right)$ ④;

Wherein: c is positive number, and its value is 0≤r _Ij≤ 1;

(4) use Transitive Closure Method each object carried out cluster, obtain the system partitioning scheme of corresponding different threshold values:

Choose appropriate threshold λ ∈ [0,1], according to as shown in the formula the λ that 5. asks for fuzzy matrix R-cut matrix Obtain the partition scheme of corresponding different λ values, be:

$r_{\mathrm{ij}}^{\left(\mathrm{\&lambda;}\right)}=\left\{\begin{array}{c}1,r_{\mathrm{ij}}\&GreaterEqual;\mathrm{\&lambda;}\\ 0,r_{\mathrm{ij}}<\mathrm{\&lambda;}\end{array}\right.,$ ⑤;

For x _i, x _j∈ X, if

, then on threshold value λ level with object x _iWith object x _jBe classified as same class; For different threshold value λ, according to R _λClassify, resulting classification is exactly the equivalence classification on the λ level, i.e. the system partitioning scheme of corresponding different threshold values;

(6) carry out merging and the adjustment of system partitioning with reference to electrical network black startup power distribution and trend check results, obtain final black boot partition scheme.

Wherein, in black boot partition process in conjunction with following rule:

(1) the voltage transitions number of times in the black start-up course of control reduces that empty to fill the voltage that circuit causes out-of-limit;

(2) the restriction path reduces the capacitive load that circuit forms, and reduces the probability of generator generation self-excitation;

(3) factory's station quantity of process reduces recovery operation in the black start-up course of restriction, reduces the risk of scheme and shortens start-up time.

Compared with the prior art, the beneficial effect that reaches of the present invention is:

1, online big power system restoration aid decision-making system provided by the invention, all kinds of verifications that can be used for when provincial and above scale electrical network is black to start solution formulation are calculated, and also can be used for the accident of having a power failure on a large scale the scheme verification of back system recovery process and the outer optimal path generation of scheme etc. take place.

2, dispatcher when the real-time mode method has kept the traditional prediction scheme formula of employing system recovery scheme, the advantage that power plant and substation equipment operator comparatively are familiar with, with scientific methods prediction scheme is formulated and to be instructed and to assist, alleviate prediction scheme and formulated personnel's working strength, embody the rapidity of power system recovery aid decision, accuracy and reliability, simultaneously can be rapid at power network development, actual conditions and prediction scheme imagination situation are inconsistent, prediction scheme relates to shortcomings such as restoration path is infeasible, after the electrical network accident of having a power failure on a large scale takes place, take full advantage of already present remote signalling in the dispatch automated system, real time datas such as remote measurement and power network topology information, check the feasibility of predetermined scheme, and when predetermined scheme is unavailable, generate fast power system recovery optimal path and scheme, the auxiliary dispatching operations staff carries out system's recovery operation, farthest shorten interruption duration, the loss that reduces to have a power failure and bring improves the reliable ability of power supply.Adopt the TipWizard module, be convenient to the researcher and go up manual manipulation.The TipWizard module is included in route searching and the evaluation module, for route searching and assessment preparation calculating parameter, calls route searching and assessment in the final step of this guide then.

3, almost whole compute type that the research mode method is integrated General Electric Network is black to start that solution formulation may relate to, and increased the distinctive electrical network subregion of big power system restoration and calculated, also possesses scheme rough draft systematic function, the result that the user of service can be easily produces its analytic process is converted to the scheme rough draft, has saved greatly the Different Results integrating documents to together time overhead.Adopt the path to generate guide, can avoid setting omission because of some and bring the inaccurate situation of result to take place.The research section is adjusted abundant section storehouse in the function combined with intelligent dispatching of power netwoks technology support system, for the user of service provides sufficient research material.It is complete respectively to go on foot the result of calculation preservation under the research mode, and personnel easy to use verify the correctness of result of calculation.

4, system adopts the multinomial achievement in research of intelligent grid dispatching technique back-up system (SG-OSS), data circulation, system architecture and data-interface satisfy the requirement of intelligent grid dispatching technique back-up system relevant criterion comprehensively, can be applied to more than 20 provincial control centre quickly and easily.Take full advantage of the existing resource among the SG-OSS, reduce construction period and cost.

Description of drawings

Fig. 1 is the functional structure chart under the real-time auxiliary mode provided by the invention;

Fig. 2 is the functional structure chart of researching and analysing under the pattern provided by the invention;

Fig. 3 is that initial section provided by the invention reads surface chart;

Fig. 4 is that target section provided by the invention reads surface chart;

Fig. 5 is that power failure scope provided by the invention arranges surface chart;

Fig. 6 is black startup Power Selection surface chart provided by the invention;

Fig. 7 is that black start-up parameter provided by the invention arranges surface chart;

Fig. 8 is that black startup provided by the invention path generates guide affirmation surface chart;

Fig. 9 is that the power supply that is activated that black startup provided by the invention path generates in the guide is selected surface chart;

Figure 10 is electro-magnetic transient analysis result oscillogram provided by the invention;

Figure 11 is SG-OSS basic platform hierarchical chart provided by the invention;

Figure 12 is real-time auxiliary mode workflow diagram provided by the invention;

Figure 13 is the pattern workflow diagram of researching and analysing provided by the invention.

Embodiment

Below in conjunction with accompanying drawing the specific embodiment of the present invention is described in further detail.

The purpose of this invention is to provide a kind of online power system recovery aid decision-making system, this system can obtain electric power system topological connection relation and device parameter by dispatch automated system, and the recovery scheme during for big grid power blackout provides subregion to instruct and blackly starts the power source planning suggestion; Also can in time obtain the real-time grid state under the disabled situation of prediction scheme and recover target black startup the after the accident of having a power failure on a large scale, by the topological analysis to the initial launch state, reasonably deceive in the selecting system and start power supply, automatically search generates the black power supply that starts to the initial path that is activated between the power supply, to assessing by all paths of every verification and sorting, select the more excellent path of combination property, for dispatcher's reference, decided the black startup path of final employing by the dispatcher, and real-time tracking system recovering process, provide load restoration optimization order in conjunction with the load restoration target, assist the dispatcher to recover possible system's power supply as early as possible, the extent of damage lightens the load.Use method of the present invention and develop online power system recovery auxiliary decision-making support system, can take full advantage of control centre's available data resource, combine the advantage of prediction scheme restoration methods and intelligent scheduling, need not modeling again, minimizing system building time and cost, real-time and the accuracy of decision recommendation have been improved, for the fast and reliable that electrical network is had a power failure on a large scale after the accident recovers to provide strong support, employing has vast market prospect based on intelligent grid dispatching technique back-up system specification data interface and man-machine displaying scheme.

Below term is described:

1) the black startup: after whole system hinders stoppage in transit for some reason, do not rely on the help of other networks, by having the startup of self-startup ability unit in the system, drive the scope of the unit expansion system gradually recovery of no self-startup ability, finally realize the process of the recovery of whole system.

2) the black power supply that starts: refer to that whole system hinders collapse for some reason and stops transport when being in complete " deceiving " state, do not rely on external electrical network, self have the unit of self-startup ability.The black power supply that starts is chosen usually: Hydropower Unit, gas turbine, HVDC Light system, generation of electricity by new energy, little unit initiatively solution are classified isolated island etc. as.

3) restoration path: the sequencing that each node (power plant, transformer station etc.) recovers.

4) section: divided data section and model section.The data section refers to the operation conditions description set of all devices in a certain moment electrical network.Be gaining merit of passing through on all circuits, transformer, capacity reactance device, the unit in this moment electrical network, idle as the 2013-02-2619:45:02 information that section comprises second, the voltage of all buses, the folding condition of all circuit breakers and disconnecting link, and each equipment running status (normally, maintenance, stoppage in transit, fault etc.); The model section refers to parameter and all annexations of all devices in a certain moment electrical network.Section mentioned in this article is not if directly indicate, and then it has the implication of data section and model section simultaneously.

5) feasibility verification: the initial path that automatic search is generated carries out verification, the path that verification is not satisfied in deletion.Mainly comprise: self-excitation verification, switching overvoltage verification, electric voltage frequency skew etc.

6) self-excitation verification: in black initial start stage, system forms by starting power supply, step-up transformer and idle load long line, capacity current is bigger in the system, this easily produces the black power supply self-excitation problem that starts, terminal voltage and the capacitance current of generator constantly increase, cause generator overvoltage and overcurrent, make the stator end of generator seriously generate heat.Therefore, the self-excitation problem becomes a black major issue that starts power supply.Power supply does not produce the self-excitation condition:, wherein be the generator d-axis reactance, be the step-up transformer leakage reactance, be the equivalent capacitive reactance of nonloaded line.When this condition did not satisfy, selected black startup power supply did not satisfy and does not produce the self-excitation requirement, needs the black power supply that starts of alternative.The self-excitation condition is the black major issue that power supply is selected that starts, and also is one of important indicator of path evaluation.

7) switching overvoltage verification: consider deceive in the start-up course, each switch successively after the closure on the path, respective bus bars sky fill afterwards busbar voltage whether transient state are out-of-limit.Its implementation is when each bar bus closes a floodgate on the path, by it is carried out the electromagnetic transient simulation analysis, and obtains transient voltage, and compares and estimated accordingly with the permission transient voltage value of rules defined.

8) evaluation index: all feasible paths are carried out comprehensive performance evaluation and ordering, in order to pick out optimum, and the feasible path of suboptimum.Usually choose voltage transitions number of times, path, start-up time, be activated unit capacity, the good and bad degree of technology verification and be activated 6 indexs such as priority level of power supply, the good and bad index of technology verification wherein is that summation after the normalization of every technology check results is obtained.

9) main grid structure recovers: to the critical path that is activated between the power supply, progressively main force's generating set, important transformer station and the part transmission line in the start-up system comes progressively recovery system main grid structure according to the black startup power supply that generates.This part work comprises two parts: the one, set up the target rack, and the 2nd, establish concrete recovery order.

10) load restoration: when fired power generating unit oneself through starting and have certain generating capacity, and set up after the comparatively stable rack, begin the stage of recovering to load gradually.Load restoration stage main target is to recover load as much as possible as soon as possible, and constraints mainly comprises the thermally-stabilised of voltage, circuit, transformer, and system frequency.

11) network topology analysis: according to connectivity and the power supply state of real-time device state analysis grid equipment, and constitute the bus model to support senior application function and to show the situation that is connected in user interface.The network topology analysis is the prerequisite that starts route searching and path technique verification.Important requirement of electric power networks topological analysis is exactly real-time and rapidity, when changing in the position of circuit breaker, disconnecting link, can provide the topological analysis result of electrical network at once, and the topology coloring of electrical network changes at once.

Online big power system restoration aid decision-making system provided by the invention comprises real-time auxiliary mode and researchs and analyses two kinds of mode of operations of pattern, and described system comprises:

Section Operations Analyst module: be used for section and read and the section adjustment, comprise that recovering target section read module, real-time section read module, the basic section read module of research, research section adjusting module and section recovers module;

Calculating configuration module: be used for calculating parameter setting, black startup Power Selection and be activated Power Selection, comprise that parameter arranges module, black startup Power Selection module and is activated the Power Selection module;

Computing function module: be used for system partitioning, trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment, comprise system partitioning module, trend computing module, electro-magnetic transient computing module, electromechanical transient computing module and route searching and evaluation module;

Management of computing module: be used for system partitioning scheme, voltage waveform, available path rank and the black demonstration that starts the generation scheme, comprise the waveform display module.

Functional structure such as Fig. 1 under the real-time auxiliary mode, in case enter real-time mode, to periodically calculate current trend, after the recovery operation that dispatcher's appointment is wanted to carry out, will carry out the technology verification to this operation, if verification is by can being not starting point with preoperative charged node then, be that terminal point carries out route searching to be activated power supply, and verification, assessment and ordering are carried out in all paths that search out, provide optimal path for dispatcher's reference.

Research and analyse functional structure such as Fig. 2 under the pattern, its calculating configuration and computing function are consistent with real-time mode, the aspect that reads section reads real-time section function and becomes and read the basic section of research and the research section is adjusted function, increased the section restore funcitons, the result of calculation demonstration comprises that then voltage waveform displaying, available path rank, black startup scheme generate.System and user's interaction point is a lot of under the research mode, and each computing module is also relatively independent, and the flow chart as real-time auxiliary mode does not now provide schematic flow sheet such as accompanying drawing 2.It has represented for obtaining the needed fundamental operation step of a certain class result, and the general contents that comprises of some data or calculating.

One, section reads:

Under real-time auxiliary mode, described section operational module is used for reading recovery target section and real-time section; Researching and analysing under the pattern, described section read module is used for reading the basic section of research and recovers the target section and study the section adjustment and the section recovery;

Read and recover the target section by recovering the realization of target section read module, comprising: from the section storehouse of system, read and specify section as the target section that recovers, load value and the system running state of optimizing each required load bus is provided for analytical calculation; Its interface such as Fig. 4.

Read real-time section and realized by real-time section read module, comprising: reading system SCADA uses measurement and the state of all devices down, obtains system's operation present situation.

Read the basic section of research and realized by the basic section read module of research, comprising: from the section storehouse of system, read and specify section as the basic section of researching and analysing usefulness; Its interface such as Fig. 3.

The adjustment of research section is realized by research section adjusting module, comprise: carry out a series of setting (comprising that equipment running status setting, switch and the set of disconnecting link state, unit output and load power size arrange) at the basic section of research, obtain simulating the system mode after the accident of having a power failure on a large scale takes place, i.e. black startup problem initial state, this initial state is called the research section;

Section recovers to recover module by section to be realized, comprising: will study basic section revert to last section adjust before the state of (the last study section adjusting module be called before).

Two, section adjustment:

The adjustment of research section: carry out a series of setting at the basic section of research, the i.e. initial state of black startup problem of system mode after taking place with the accident of having a power failure on a large scale that obtains simulating, this initial state is also referred to as the research section.

Section recovers: will study the state before section reverts to last section adjustment.

Research section setting is to research and analyse the important subfunction that aid decision recovers in system under the pattern, mainly the power failure scope setting that realizes by the folding operation to circuit breaker and unit is sent injection rate that power and payload arrange adjust function and realize.The injection rate adjustment can realize that the power failure scope arranges interface such as Fig. 5 at the figure that stands of factory, and it is described as follows:

1, power failure scope arranges.The tree of power failure scope setting area is to generate according to regional attaching relation in the electric network model, and the equipment under the factory station is only showed circuit breaker and two kinds of equipment of disconnecting link.Initial condition is for selecting state entirely.When choosing a certain node, its child node also can be chosen, and after the state of choosing of its a certain child node was cancelled, the selected state of the child node of this node, this node and all parent nodes thereof all should be cancelled.Click counter select button after, the state of the choosing negate of all nodes is clicked disconnecting link that button then can be selected with all and the quality position branch of circuit breaker, the painted white that changes into of line topological that the factory station that all are chosen is connected is set.The function of searching input frame arranges and searches the input frame functional similarity in the interface with the black power supply that starts, search the equipment in the combobox in elected after, its corresponding node should be selected in dendrogram, and dendrogram launches.

2, the data section obtains.Click is obtained and all will be ejected section behind initial section or the select target section button interface is set, and difference is that the initial state that the section that ejects arranges the interface has difference.Initial section is loaded in the real-time data base on the one hand, will stay an end on the other hand, and the section in the storehouse can change or change with remote signalling remote measurement renewal with operation in real time.The target section also takes out, but is not loaded in the real-time storehouse, describes in detail to see that the data section obtains part.

3, clicking the routine processes arrange behind the button is to carry out on the initial section basis that obtains.

4, for specifying the factory station to restore electricity.Choose the factory that wants to remove from power supply interrupted district to stand in the station tabulation of power failure factory, click recovery button, then the remote signalling state of all switch tools in this factory station is set by the value in the initial section.

5, the tabulation of power failure factory station is corresponding with certain epiphase in the real-time storehouse, show in this interface for the territory that " whether has a power failure " in this table be 1 record, the every row that reads all comprises unique identification id, permission is made amendment to starting the priority level territory, if revise, this value will be updated to and concern in the table corresponding in the storehouse when click refreshes.

What 6, power failure information was had a guide look of every data demonstration is the comparison result of current system mode and initial section, carries out this comparison when clicking refresh button.

Three, calculate configuration:

At real-time auxiliary mode with research and analyse under the pattern, described calculating configuration module all is used for calculating parameter setting, black startup Power Selection and is activated Power Selection;

The calculating parameter setting arranges module by parameter and realizes, parameter arranges object that module can arrange and is included in being used for that last joint mentions and describes the calculating parameter route searching rule and that be used for path evaluation ordering usefulness, comprises that also computation schema is selected, control class parameter such as storing path as a result.Parameter arranges interface such as Fig. 7.

The black Power Selection that starts is chosen the module realization by black the startup, the black Power Selection that starts is namely specified the black startup power supply that adopts when researching and analysing, when real-time mode, need not artificial appointment, system will judge unit or the bus that puts into operation at first in the power supply interrupted district, locate the black startup power supply of native system when working as time black the startup with this.The black interface that starts Power Selection is shown in 6, and the content in its tabulation is the unit that unit, converter, interconnection and supposition that system possesses black start-up ability have black start-up ability.

Be activated Power Selection and realize by being activated the Power Selection module, be activated power supply and select to be used to specify and want to research and analyse or be activated power supply really when auxiliary in real time, its interface is selected similar with the black power supply that starts.

Four, computing function module:

Under real-time auxiliary mode, described computing function module is used for trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment; Researching and analysing under the pattern, described computing function module is used for system partitioning, trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment;

Trend is calculated by the trend computing module and is realized, comprises the direction of energy distribution situation in the designated model; Starting power supply to the black start-up course that is activated power supply from deceiving in real time, force device progressively joins in the electric island alive one by one, thereby calculates the dynamically generation of model of trend.The trend module can become corresponding computation model according to the black startup restoration path of user's appointment and the bus of wanting to study next life, and result of calculation is used for the power-frequency overvoltage verification; Under real-time mode, also can calculate the real-time trend of electric power system in recovery, be used for the control of system voltage frequency stability.

Electro-magnetic transient calculates by the electro-magnetic transient computing module and realizes, adopts the EMTP kit to carry out the electro-magnetic transient analysis and obtains the operating voltage check results; Exist a large amount of breaker operators to make circuit and transformer by not charged to electriferous state in the black start-up course, thereby this process may produce the switching overvoltage damage equipment, thereby need carry out the switching overvoltage verification.Native system adopts the EMTP kit to carry out electro-magnetic transient analysis and obtains the operating voltage check results.The basic function of EMTP program is to carry out electric system simulation to calculate, it is the prediction electric power system in certain disturbance (as switch on-off or fault) time dependent rule of interested variable afterwards that the typical case uses, and just is being suitable for the switching overvoltage check problem analysis of black start-up course.

The voltage response of system when the electro-magnetic transient analysis result can embody circuit or transformer and puts into operation, whether the maximum voltage in the verification response process surpasses limit value, also participate in the path evaluation ordering as an index, its result also can come out by the formal representation of curve display module with curve, as shown in figure 10.

Electromechanical transient calculates and is realized by the electromechanical transient computing module.Electromechanical transient is analyzed, and mainly studies the steady stability performance after electric power system is subjected to the transient stability after the big disturbance and is subjected to microvariations.Wherein transient stability analysis is that the research electric power system is subjected to such as short trouble, and excision circuit, generator, load, generator lose under the big perturbation action such as excitation or impact load, the ability of the dynamic behaviour of electric power system and the operation of maintenance synchronism stability.In black start-up course, be activated power supply auxiliary system can to start that power supply drive be the black the most key factor of recovery process initial stage that starts via black.The impact auxiliary system is equivalent to the impact induction-motor load, is suitable for adopting the electromechanical transient analysis.According to the electromechanical transient analysis principle, write the electromechanical transient routine analyzer, show itself and PSASP(Power System Analysis Software Package power system analysis synthesizer in the result of example network analysis) consistency as a result better.

Five, route searching and assessment:

Route searching is realized by route searching and evaluation module that with assessment route searching is to select from the black power supply that starts to the path that is activated power supply combination property optimum;

Described route searching and evaluation module comprise that restoration path generation module and the black path that starts generate to guide module;

Generating the black restoration path that starts comprises the steps:

I, determine to be activated power supply: the described power supply that is activated adopts fired power generating unit.

II, the rule-based initial path searching algorithm of employing obtain from the black power supply that starts to the feasible electrical path that is activated power supply; Described initial path searching algorithm is followed following rule:

1) voltage transitions number of times restriction principle: the voltage transitions number of times in the black start-up course of control;

2) path restriction principle: the black restoration path length that starts of restriction;

3) factory station restriction principle;

4) capacity limits principle: required minimum load and recovers jumbo unit less than starting exerting oneself of unit when being activated unit starting;

5) important load priority principle;

6) minimum principle of time: the unit of selecting to be in hot stand-by duty;

7) avoid the loop principle: to the path that is activated unit, do not have loop from starting unit.

III, filter out the electrical path by the feasibility verification: the Interventions Requested of described feasibility verification comprise: the verification of generator self-excitation, low-frequency oscillation verification, power-frequency overvoltage verification, switching overvoltage verification and the verification of electric voltage frequency stability;

IV, definite from deceiving the startup power supply to the path that is activated power supply combination property optimum, be the optimized database restore path:, select from the black power supply that starts to the path that is activated power supply combination property optimum assessing by all black startup restoration paths of feasibility verification and sorting; Evaluation index comprise select voltage transitions number of times, path, start-up time, be activated unit capacity, the good and bad degree of technology verification and be activated the priority level of power supply, the good and bad index of technology verification wherein is that summation after generator self-excitation, power-frequency overvoltage multiple, switching overvoltage multiple, variation, the frequency shift (FS) normalization is obtained.The technology contents that the good and bad degree of the index of considering during path evaluation and technology verification comprises is distinguished as shown in Table 1 and Table 2:

The index of considering during table 1 path evaluation

The technology contents that the good and bad degree of table 2 technology verification comprises

Need in several interfaces, arrange because generating the black required initial condition in path that starts, can't obtain legitimate result thereby the user may omit some steps in use.For making things convenient for the user of service to grasp the black path generating method that starts, introduced the guide pattern.The black path that starts generates guide and is actually and blackly starts power supply with arranging, arrange and be activated power supply, the route searching parameter integration be set to together procedure aggregate.Need with all display in certain sequence to arranging, before the interface of calling the route searching operation program (black start path generate guide confirm that the interface is as shown in Figure 8), all necessary data are all ready.In the figure, click is finished button and will be called route searching program, to finish specifying startup power supply and appointment to be activated the search in path between power supply.The black power supply that is activated that starts in the path generation guide is selected the interface as shown in Figure 9.

Six, system partitioning:

According to current scheduling mechanism formulate face the massive blackout accident time the power system restoration scheme, what take substantially is that power supply interrupted district is divided into a plurality of subregions by Utilities Electric Co.'s ownership, arrange all that in all subregion the black power supply that starts is to guarantee the self-startup ability of subregion, main unit and load in all subregion parallel recovery zone, all subregion is selected suitable simultaneous interconnecting on opportunity under the coordination of higher level's scheduling, to realize the interconnected of whole electrical network.The shortcoming of this division methods is that the combination of the black startup power supply in the interior load bus in zone and the zone may not be optimum, for example, P node in the a-quadrant starts via the black startup power supply SA in the a-quadrant may need 2 hours, need 1.5 hours and only start via the SB in the B zone, and the fast quick-recovery load in the back that has a power failure is significant.Carry out the subregion of system according to the black distribution situation that starts power supply, and considered starting characteristic and the system mode of unit, but consider not enough for fail safe and generating in system's recovery and load balancing restriction.Based on the recovery subsystem partitioning algorithm of complex network corporations structural theory, but it just selectedly before calculating is activated power supply, and will deceive and start power supply and become a node to being activated the node merging that comprises in the path between the power supply, has certain subjectivity.Based on the black boot partition strategy of ordered binary decision diagram, but this method mainly is to seek optimal solution by the network topology search, and is less to the consideration of other influencing factors.

The present invention has adopted the black boot partition method of a kind of electric power system based on fuzzy cluster analysis, and system partitioning realizes that by the system partitioning module described system partitioning is realized by following steps:

(1) set up each node of reflection and each black raw data matrix that starts electrical distance between the power supply:

For guaranteeing black carrying out smoothly of starting, when subregion, guarantee that each subregion has a black power supply that starts at least.Each black several middle transformer station of power supply process that start that each subregion is interior start other power plant or recovery to the power supply of load.Consider reliability and timeliness, can weigh each with following several indexs and be activated power plant, important transformer station, load bus and the black power that starts electrical link between the power supply: 1) with the distance of deceiving between the startup power supply; 2) voltage transitions number of times; 3) middle through the power station number.These three indexs all are cost type indexs, and the control target of subregion is that these three desired values are the smaller the better.

Defined node i and the black electrical distance that starts between the power supply j are:

$L_{\mathrm{ij}}=\underset{k=1}{\overset{l}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_{\mathrm{ij}\left(k\right)}{r}_{\mathrm{ij}\left(k\right)}^{\&prime;}$ ①;

Wherein, r _{Ij (k)}For weighing k desired value of electrical link power, r _{Ij (k)}' be numerical value after its standardization, r _Max, r _MinBe respectively r _{Ij (k)}Maximum, minimum value corresponding to different i; ω _iBe the weights of i index; L represents the index number, is taken as 3; I=1,2 ..., m, j=1,2 ..., n; Node comprises and is activated power supply, transformer station and load bus;

As if being object of classification with the node of the m in the electrical network, black startup power supply number is n, then node i and the black electrical distance formation individual features index that starts between the power supply j, i.e. each element in the raw data matrix:

x _ij＝L _ij ②；

Then set up raw data matrix X=(x _Ij) _{M * n}I=1 wherein, 2 ..., m, j=1,2 ..., n

(2) in practical problem, different data generally have different dimensions, the king-sized characteristic index of some order of magnitude may play too outstanding effect to classification in calculating process, and the effect of the very little characteristic index of some order of magnitude may be reduced even ignore.For different dimensions also can be compared, need do suitable conversion to data usually.According to the requirement of fuzzy matrix, need be with data compression on interval [0,1].Adopt more standardized method to mainly contain at present: standard deviation standardization, maximum standardization, extreme difference standardization, average standardization, central standardization, logarithm standardization etc.The present invention adopts the extreme difference standardized method that raw data matrix is carried out standardization, and its expression formula is:

$y_{\mathrm{ik}}=\frac{x_{\mathrm{ik}}-\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003168181200031.png"\; state="\{\{state\}\}">min</figure-callout>}}\left\{x_{\mathrm{ik}}\right\}}{\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="max"\; filenames="HDA00003168181200031.png"\; state="\{\{state\}\}">max</figure-callout>}}\left\{x_{\mathrm{ik}}\right\}-\underset{1\&le;i\&le;m}{\mathrm{<figure-callout\; id="1"\; label="min"\; filenames="HDA00003168181200031.png"\; state="\{\{state\}\}">min</figure-callout>}}\left\{x_{\mathrm{ik}}\right\}},$ ③;

Wherein: k=1,2 ..., n;

(3) demarcate object to be sorted, set up fuzzy similarity matrix:

Set up fuzzy similarity matrix and be called demarcation again, refer to mark the statistic r that measurement is classified similarity degree between object _IjIf domain X={x ₁, x ₂..., x _m, x _i=(x _I1, x _I2..., x _In), x then _iWith x _jSimilarity degree be r _Ij=R (x _i, x _j), i, j=1,2 ..., m; Its method for solving mainly contains: similarity factor method (comprising scalar product method, included angle cosine method, correlation coefficient process, index similarity factor method, minimax method, the minimum method of arithmetic average, the minimum method of geometric average), Furthest Neighbor (comprising absolute value counting backward technique, absolute value index method, direct range method) and subjective scoring method etc.

The present invention adopts the absolute value index method to find the solution r _Ij, its expression formula is:

$r_{\mathrm{ij}}=\exp (-c\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}|y_{\mathrm{ik}}-y_{\mathrm{ik}}\left|\right)$ ④;

Wherein: c is positive number, and its value is 0≤r _Ij≤ 1; Learn from else's experience and test numerical value c=1.

(4) use Transitive Closure Method each object carried out cluster, obtain the system partitioning scheme of corresponding different threshold values:

Choose appropriate threshold λ ∈ [0,1], according to as shown in the formula the λ that 5. asks for fuzzy matrix R-cut matrix

Obtain the partition scheme of corresponding different λ values, be:

$r_{\mathrm{ij}}^{\left(\mathrm{\&lambda;}\right)}=\left\{\begin{array}{c}1,r_{\mathrm{ij}}\&GreaterEqual;\mathrm{\&lambda;}\\ 0,r_{\mathrm{ij}}<\mathrm{\&lambda;}\end{array}\right.,$ ⑤;

For x _i, x _j∈ X, if

Then on threshold value λ level with object x _iWith object x _jBe classified as same class; For different threshold value λ, according to R _λClassify, resulting classification is exactly the equivalence classification on the λ level, i.e. the system partitioning scheme of corresponding different threshold values;

(6) carry out merging and the adjustment of system partitioning with reference to electrical network black startup power distribution and trend check results, obtain final black boot partition scheme.In black boot partition process in conjunction with following rule:

(1) the voltage transitions number of times in the black start-up course of control reduces that empty to fill the voltage that circuit causes out-of-limit;

(2) the restriction path reduces the capacitive load that circuit forms, and reduces the probability of generator generation self-excitation;

(3) factory's station quantity of process reduces recovery operation in the black start-up course of restriction, reduces the risk of scheme and shortens start-up time.

Seven, black startup scheme (rough draft) generates:

Generally speaking, deceiving the final goal that starts analysis under the research attitude is to form the black analysis report that starts of complete and effective electrical network, the black startup scheme generation module of system can directly be filed aforementioned analysis result arrangement and export with the form of report, form the black analysis report that starts of preliminary electrical network, exempted the user and from a plurality of destination files, extracted data to carry out the worry of confluence analysis.

Eight, use design:

The present invention also provides the application design of black start assistant decision making method.

Basic platform provides the interface of various services, for application and development provides data exchange mechanism, man-machine support, data support, public service module and system management function, and supporting business customization and adjustment, as shown in figure 11.

Big power system restoration aid decision-making system is developed based on the corresponding interface and specification requirement, mainly comprises database, program and man-machine interface.

Database

Online big power system restoration aid decision-making system comprises system database (various public equipment list), uses calculated data storehouse (parameter list, result of calculation table etc.), private data storehouse (only recovering relevant table with scale grid line).The database of online power system recovery aid decision is the bridge of contact procedure, picture, input/output information, plays a part to deposit initial data, preservation computational process record and computing information, output result of calculation.

The database of power system recovery aid decision is based on intelligent grid dispatching technique back-up system support platform, and database mainly comprises following three parts:

1. equipment list: be used for depositing each device parameter and interconnected relationship, the information that has comprised most needs in the basic platform common block list of intelligent grid dispatching technique back-up system support system, but lack the parameter that is activated the corresponding subsidiary engine of unit, newly-built auxiliary system table is to comprise relevant parameter.

---bus table (busbarsection)

---generator table (generatingunit)

---transformer table (powertransformer)

---circuit table (aclinesegment)

---load meter (energyconsumer)

---switch (breaker)

---disconnecting link table (disconnector)

---auxiliary system table (auxiliary_motor)

Wherein the auxiliary system table is as shown in table 3.

Table 3 auxiliary system table

Sequence number	English name	Chinese name	Data type
					1	id	Sign	long
2	un_id	The unit id that serves	long
				3	percent_ld	The dynamic load accounting	float
4	tj	Inertia time constant	float
				5	xs	Stator winding induction reactance	float
6	rr	The rotor winding resistance	float

7	xr	Rotor winding leakage reactance	float
					8	alfa	The constant force square	float
9	ro	The mechanical property index	float
				10	s0	The stable state slippage	float
11	t1	Stator open circuit transient state time constant	float
				12	percent_ld1	Constant-impedance load accounting	float

2. scheme table:

---partition scheme table: the black boot partition scheme result of storage.The partition scheme table is the expansion of platform device table, and " black boot partition " territory that increases an int type on original each equipment list (bus table, generating set, transformer table, circuit table, load meter, circuit breaker table, disconnecting link table) basis gets final product.

---route scheme table: store the scheme that obtains behind the rule-based route searching.Its list structure is as shown in table 4:

Table 4 route scheme table

Sequence number	English name	Chinese name	Data type
					1	id	Sign	long
2	area_no	Regional number	int
				3	route_no	Path number	int
4	position	The position of element in the path	int
				5	type_dev	Component type	int
6	id_dev	Element i d	long
				7	name_dev	The element title	string

3. calculating parameter table:

---calculating control parameter list: the weight of each index when being used for depositing black startup Power Selection process for route searching restriction, evaluates calculation, parameters such as result of calculation store path.

Table 5 calculates the control parameter list

Sequence number	English name	Chinese name	Data type
					1	id	Sign	long
2	mode	Pattern	int
					3	path_case	The scheme storing path	string
4	path_emtp	The transient state result path	string
					5	path_steady	The stable state result path	string
6	limit_transfer_v	The restriction of voltage transitions number of times	int
				7	limit_length	The path restriction	float
8	limit_st	By way of factory's station number restriction	int
				9	limit_starttime	Restriction start-up time	float
10	limit_vipload	The important load restriction	float
				11	weight_transfer_v	Voltage transitions number of times weight	float
12	weight_length	The path weight	float
				13	weight_tech	Technology verification weight	float
14	weight_time	Start-up time weight	float
				15	weight_unmva	Be activated the unit capacity weight	float
16	weight_highkv	The highest voltage level weight	float

---operation demonstration table: be used for depositing being about to carry out apparatus operating, carry out verification after reading for program.

Table 6 operation demonstration table

Sequence number	English name	Chinese name	Data type
					1	id	Sign	long
2	dev_id	Equipment id	long
				3	seq	Operating sequence	int
4	dev_name	Device name	string

4. evaluation form

---path evaluation is table as a result: be used for depositing from specifying the black power supply that starts to being activated each path evaluation ranking the power supply.As shown in table 7.

Table 7 path evaluation is table as a result

Sequence number	English name	Chinese name	Data type
					1	id	Sign	long
2	route_no	Path number	int
					3	score	The assessment score	float
4	rank	Rank	int

Initial restoration path assessment result table: the relative effectiveness order of storage feasible program.

Man-machine interface

Draw the exploitation of the aid decision calculating picture of rationing the power supply based on intelligent grid dispatching technique back-up system support system, picture browser and editing machine and man-machine interface realization man-machine interaction and the result's displaying of adopting basic platform to provide.

The present invention also provides a kind of online big power system restoration aid decision-making method, and described method selects different working modes to carry out corresponding steps, and described mode of operation comprises real-time auxiliary mode and researchs and analyses pattern.

1. the flow process of the described method under the real-time mode comprises the steps: as shown in figure 12

＜1〉confirm to enter real-time back work flow process: the back work flow process is generally by Event triggered in real time, after the accident of having a power failure on a large scale takes place, the system monitoring module judges that current system mode is the state of having a power failure on a large scale, then send the state event of having a power failure on a large scale, big power system restoration aid decision-making system preparation enters real-time back work flow process, treat formally to enter real-time back work flow process after the user confirms, cause mistake to enter the situation of real-time back work flow process to avoid the event error signal.

＜2〉read the distance event of having a power failure on a large scale and constantly nearest system's normal section takes place as the target section that recovers, this section is that other modules by intelligent grid dispatching technique back-up system generate, and unified being stored in the section storehouse.

＜3〉read and be activated unit;

＜4〉reading system state is judged power failure scope and the charged situation of equipment;

＜5〉calculate current trend;

＜6〉the operation demonstration table in the reading database judges whether the equipment that puts into operation;

＜7〉judge that whether the equipment put into operation be and be activated the transformer that unit links to each other: if carry out step＜8 〉, otherwise carry out step＜9 〉;

＜8〉electromechanical transient verification, and the machine that the is activated group id that will reach is set to 1;

＜9〉carry out the feasibility verification to being activated unit;

＜10〉judge whether by the feasibility verification: if carry out step＜11, otherwise carry out step＜14 〉;

＜11〉provide this path (this path refer to from black start power supply just be connected to the operation demonstration table when the equipment of verification in the middle of the equipment ordered set of process) feasible prompting;

＜12〉with the unit deletion of having verified in the operation demonstration table that puts into operation;

＜13〉continue to judge through step＜12〉after the operation demonstration table whether reached and be activated unit: if then carry out step＜20 〉, otherwise, return step＜4 〉;

＜14〉to be starting point with (this equipment refers in the operation demonstration table current just at the equipment of verification) the charged bus that will apparatus operating links to each other, be that terminal point carries out route searching to be activated unit;

＜15〉to meeting step＜14〉the path of requirement carry out the feasibility verification;

＜16〉path evaluation and ordering;

＜17〉judge whether to exist path by the feasibility verification: if then carry out step＜19 〉, otherwise carry out step＜18 〉;

＜18〉provide the path of not having by the feasibility verification, and provide the assessment rank order of preset path;

＜19〉provide path by the feasibility verification, and provide the highest path of appreciable levels;

＜20〉be activated power supply as if having returned to, and withdraw from real-time auxiliary mode, otherwise, repeating step＜10 〉-＜19 〉.

2. research and analyse under the pattern and said method comprising the steps of:

A, enter the pattern of researching and analysing.The pattern of researching and analysing is specified by the user and is entered, as the step 1 among Figure 13.

B, research section obtain.Obtain the basic research section from the section storehouse, and serve as that the back of making amendment, basis generates the user and wants the section state studied with it.The user can preserve its amended section in order to using later on.As step 2,3,4,5,6,7 among Figure 13.

C, black boot partition calculate.The research section of setting according to the user carries out system ambiguous cluster subregion calculating, reference when providing the subregion result for the formulation partition scheme.As step 8 among Figure 13 and 15.

D, route searching.User of service's designated analysis zone, and specify the black startup power supply in this zone and be activated behind the power supply and after some calculate control parameters is calculated the path of satisfying search condition and is showed by the path generation module.As step 9,10,11,12,13,14 among Figure 13.

E, path evaluation.All kinds of verifications are carried out in all paths that obtain behind the route searching, the verification structure is participated in path evaluation as index, obtain to describe the scoring of the good and bad degree in each path and the rank in each path.As step 23,24,25 among Figure 13.

F, labor and waveform are showed.Route searching finish obtain all paths after, can specify and want the path studied and the bus of concern, call trend calculating, electro-magnetic transient calculating and electromechanical transient respectively and calculate, adopt the waveform display module just the voltage waveform of paying close attention to the bus correlation computations can be displayed.As step 16,17,18,19,20,21,22 among Figure 13.

G, in generation pass and after finishing all verifications of a paths, specify to generate adopt the black startup scheme in this path, system will gather about the result of this path when carrying out all kinds of verification, generate black startup scheme draft.As the step 26 among Figure 13,27.

Online big power system restoration aid decision-making system provided by the invention and method can provide the suggestion of aspects such as system partitioning, optimal path for formulating the black startup scheme of large-scale interconnected electric power system (provincial and above scale electrical network), can generate the scheme rough draft fast, can assist the dispatcher to carry out the stable control of electric voltage frequency in the recovery process after the accident that causes electrical network to be deceived entirely takes place, and can when the recovery prediction scheme is unavailable, provide other feasible recovery schemes suggestions for the dispatcher.Researching and analysing under the pattern, can be according to the network configuration of large-scale interconnected electric power system and the system initial state of envisioning, obtain the required recovery partition scheme that recovers the zone by fuzzy cluster analysis, and can this mechanism be planned by the black power supply that starts, after treating the black startup power supply of each subregion and be activated power supply to determine, carry out the network topology analysis based on condition, carry out the technology verification after being met the path of condition, with the technology check results as an index, other indexs of considering when coupling system recovers are carried out overall merit with entropy power fuzzy evaluation method, and being restored is activated the optimal path of power supply.Under in real time auxiliary recovery pattern, the SCADA(Supervisory Control And Data Acquisition that obtains in real time in the on-line system is data acquisition and supervision control) data, EMTP(Electro-Magnetic Transient Program electro-magnetic transient routine analyzer in the coupling system) module and flexible trend computing module, preceding up-to-date normal section taking place as the load restoration target with accident, draws the Optimal Load recovery scheme in conjunction with electric voltage frequency control; When equipment related in the black startup prediction scheme causes it unavailable owing to objective factors such as maintenance, fault or natural calamities, be given in adoptable black startup optimal path under the current electrical network situation, assist dispatcher's rapid adjustment recovery scheme, to realize the recovery of whole electric power system.System is intelligent grid dispatching technique back-up system based on SG-OSS(Smart Grid Operating and Supporting System) exploitation, satisfy its every related specifications requirement.

Should be noted that at last: above embodiment is only in order to illustrate that technical scheme of the present invention is not intended to limit, although with reference to above-described embodiment the present invention is had been described in detail, those of ordinary skill in the field are to be understood that: still can make amendment or be equal to replacement the specific embodiment of the present invention, and do not break away from any modification of spirit and scope of the invention or be equal to replacement, it all should be encompassed in the middle of the claim scope of the present invention.

Claims (7)

1. an online big power system restoration aid decision-making system is characterized in that, described system comprises real-time auxiliary mode and research and analyse two kinds of mode of operations of pattern that described system comprises:

Section Operations Analyst module: be used for section and read and the section adjustment, comprise that recovering target section read module, real-time section read module, the basic section read module of research, research section adjusting module and section recovers module;

Calculating configuration module: be used for calculating parameter setting, black startup Power Selection and be activated Power Selection, comprise that parameter arranges module, black startup Power Selection module and is activated the Power Selection module;

Computing function module: be used for system partitioning, trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment, comprise system partitioning module, trend computing module, electro-magnetic transient computing module, electromechanical transient computing module and route searching and evaluation module;

Management of computing module: be used for system partitioning scheme, voltage waveform, available path rank and the black demonstration that starts the generation scheme, comprise the waveform display module.

2. online big power system restoration aid decision-making system as claimed in claim 1 is characterized in that, under real-time auxiliary mode, described section operational module is used for reading recovery target section and real-time section; Researching and analysing under the pattern, described section read module is used for reading the basic section of research and recovers the target section and study the section adjustment and the section recovery;

Read and recover the target section by recovering the realization of target section read module, comprising: from the section storehouse of system, read and specify section as the target section that recovers, load value and the system running state of optimizing each required load bus is provided for analytical calculation;

Read real-time section and realized by real-time section read module, comprising: reading system SCADA uses measurement and the state of all devices down, obtains system's operation present situation;

Read the basic section of research and realized by the basic section read module of research, comprising: from the section storehouse of system, read and specify section as the basic section of researching and analysing usefulness;

The adjustment of research section is realized by research section adjusting module, comprising: arranges at the basic section of research, obtains simulating the system mode after the accident of having a power failure on a large scale takes place, i.e. and black startup problem initial state, this initial state is called the research section;

Section recovers to recover module by section to be realized, comprising: will study the state before basic section reverts to last section adjustment.

3. online big power system restoration aid decision-making system as claimed in claim 1 is characterized in that, at real-time auxiliary mode with research and analyse under the pattern, described calculating configuration module all is used for calculating parameter setting, black startup Power Selection and is activated Power Selection;

The calculating parameter setting arranges module by parameter and realizes, described parameter comprises for describing the route searching rule and be used for the calculating parameter of path evaluation ordering, and computation schema is selected, the control parameter of storing path as a result;

The black Power Selection that starts is chosen the module realization by black the startup, and described black startup Power Selection is to specify the black startup power supply that adopts when being used for researching and analysing;

Be activated Power Selection and realize by being activated the Power Selection module, the described Power Selection that is activated is used to specify and researchs and analyses or be activated really when auxiliary in real time power supply.

4. online big power system restoration aid decision-making system as claimed in claim 1 is characterized in that, under real-time auxiliary mode, described computing function module is used for trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment; Researching and analysing under the pattern, described computing function module is used for system partitioning, trend calculating, electro-magnetic transient calculating, electromechanical transient calculating and route searching and assessment;

Trend is calculated by the trend computing module and is realized, comprises the direction of energy distribution situation in the designated model;

Electro-magnetic transient calculates by the electro-magnetic transient computing module and realizes, adopts the EMTP kit to carry out the electro-magnetic transient analysis and obtains the operating voltage check results;

Electromechanical transient calculates and is realized by the electromechanical transient computing module.

5. online big power system restoration aid decision-making system as claimed in claim 4 is characterized in that,

Route searching is realized by route searching and evaluation module that with assessment route searching is to select from the black power supply that starts to the path that is activated power supply combination property optimum;

Described route searching and evaluation module comprise that restoration path generation module and the black path that starts generate to guide module;

Generating the black restoration path that starts comprises the steps:

I, determine to be activated power supply: the described power supply that is activated adopts fired power generating unit;

II, the rule-based initial path searching algorithm of employing obtain from the black power supply that starts to the feasible electrical path that is activated power supply; Described initial path searching algorithm is followed following rule:

1) voltage transitions number of times restriction principle: the voltage transitions number of times in the black start-up course of control;

2) path restriction principle: the black restoration path length that starts of restriction;

3) factory station restriction principle;

4) capacity limits principle: required minimum load and recovers jumbo unit less than starting exerting oneself of unit when being activated unit starting;

5) important load priority principle;

6) minimum principle of time: the unit of selecting to be in hot stand-by duty;

7) avoid the loop principle: to the path that is activated unit, do not have loop from starting unit.

III, filter out the electrical path by the feasibility verification: the Interventions Requested of described feasibility verification comprise: the verification of generator self-excitation, low-frequency oscillation verification, power-frequency overvoltage verification, switching overvoltage verification and the verification of electric voltage frequency stability;

IV, definite from deceiving the startup power supply to the path that is activated power supply combination property optimum, be the optimized database restore path:, select from the black power supply that starts to the path that is activated power supply combination property optimum assessing by all black startup restoration paths of feasibility verification and sorting; Evaluation index comprise select voltage transitions number of times, path, start-up time, be activated unit capacity, the good and bad degree of technology verification and be activated the priority level of power supply, the good and bad index of technology verification wherein is that summation after generator self-excitation, power-frequency overvoltage multiple, switching overvoltage multiple, variation, the frequency shift (FS) normalization is obtained.

6. online big power system restoration aid decision-making system as claimed in claim 4 is characterized in that, described system partitioning is based on the black boot partition of electric power system of fuzzy cluster analysis, realizes that by the system partitioning module described system partitioning is realized by following steps:

(1) set up each node of reflection and each black raw data matrix that starts electrical distance between the power supply:

Defined node i and the black electrical distance that starts between the power supply j are:

$L_{\mathrm{ij}}=\underset{k=1}{\overset{l}{\mathrm{\&Sigma;}}}{\mathrm{\&omega;}}_{\mathrm{ij}\left(k\right)}{r}_{\mathrm{ij}\left(k\right)}^{\&prime;}$ ①；

Wherein, r _{Ij (k)}For weighing k desired value of electrical link power, r _{Ij (k)}' be numerical value after its standardization, r _Max, r _MinBe respectively r _{Ij (k)}Maximum, minimum value corresponding to different i; ω _iBe the weights of i index; L represents the index number, is taken as 3; I=1,2 ..., m, j=1,2 ..., n; Node comprises and is activated power supply, transformer station and load bus;

As if being object of classification with the node of the m in the electrical network, black startup power supply number is n, then node i and the black electrical distance formation individual features index that starts between the power supply j, i.e. each element in the raw data matrix:

x _ij＝L _ij ②；

Then set up raw data matrix X=(x _Ij) _{M * n}I=1 wherein, 2 ..., m, j=1,2 ..., n

(2) each element in the described raw data matrix is carried out standardization: adopt the extreme difference standardized method that raw data matrix is carried out standardization, its expression formula is:

$y_{\mathrm{ik}}=\frac{x_{\mathrm{ik}}-\underset{1\&le;i\&le;m}{\min }\left\{x_{\mathrm{ik}}\right\}}{\underset{1\&le;i\&le;m}{\max }\left\{x_{\mathrm{ik}}\right\}-\underset{1\&le;i\&le;m}{\min }\left\{x_{\mathrm{ik}}\right\}},$ ③；

Wherein: k=1,2 ..., n;

(3) demarcate object to be sorted, set up fuzzy similarity matrix:

Set up fuzzy similarity matrix and be called demarcation again, refer to mark the statistic r that measurement is classified similarity degree between object _IjIf domain X={x ₁, x ₂..., x _m, x _i=(x _I1, x _I2..., x _In), x then _iWith x _jSimilarity degree be r _Ij=R (x _i, x _j), i, j=1,2 ..., m;

Adopt the absolute value index method to find the solution r _Ij, its expression formula is:

$r_{\mathrm{ij}}=\exp (-c\underset{k=1}{\overset{n}{\mathrm{\&Sigma;}}}|y_{\mathrm{ik}}-y_{\mathrm{jk}}\left|\right)$ ④；

Wherein: c is positive number, and its value is 0≤r _Ij≤ 1;

(4) use Transitive Closure Method each object carried out cluster, obtain the system partitioning scheme of corresponding different threshold values:

Choose appropriate threshold λ ∈ [0,1], according to as shown in the formula the λ that 5. asks for fuzzy matrix R-cut matrix

Obtain the partition scheme of corresponding different λ values, be:

$r_{\mathrm{ij}}^{\left(\mathrm{\&lambda;}\right)}=\left\{\begin{array}{cc}1,& r_{\mathrm{ij}}\&GreaterEqual;\mathrm{\&lambda;}\\ 0,& r_{\mathrm{ij}}<\mathrm{\&lambda;}\end{array}\right.,$ ⑤；

For x _i, x _j∈ X, if

Then on threshold value λ level with object x _iWith object x _jBe classified as same class; For different threshold value λ, according to R _λClassify, resulting classification is exactly the equivalence classification on the λ level, i.e. the system partitioning scheme of corresponding different threshold values;

(6) carry out merging and the adjustment of system partitioning with reference to electrical network black startup power distribution and trend check results, obtain final black boot partition scheme.

7. online big power system restoration aid decision-making system as claimed in claim 6 is characterized in that, in black boot partition process in conjunction with following rule:

(1) the voltage transitions number of times in the black start-up course of control reduces that empty to fill the voltage that circuit causes out-of-limit;

(2) the restriction path reduces the capacitive load that circuit forms, and reduces the probability of generator generation self-excitation;

(3) factory's station quantity of process reduces recovery operation in the black start-up course of restriction, reduces the risk of scheme and shortens start-up time.

Images