CN103606921B - Based on the Transient Voltage Stability determination methods of real-time broad sense Thevenin's equivalence - Google Patents

Abstract

A kind of Transient Voltage Stability determination methods based on real-time broad sense Thevenin's equivalence being applied to electric power system, according in transient state process of electric power system each time the update the system admittance matrix Y ' that inscribes and voltage, the power rating of system, obtain Jacobian matrix J and the inverse matrix J thereof of corresponding moment lower node power equation ^-1.Obtain the relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity further, obtain the relational expression between equivalent Nodes load current variable quantity and load power variable quantity, then the Thevenin's equivalence impedance in corresponding moment is obtained, thus calculate Transient Voltage Stability judge index, for tracking and the early warning of Power System Off-line analysis and presence.The Thevenin's equivalence parameter that the present invention is current according to the state computation of discontinuity surface time single in transient state process of electric power system, thus whether judgement system there is transient voltage unstability, has intuitive and accurate feature.

Description

Based on the Transient Voltage Stability determination methods of real-time broad sense Thevenin's equivalence

Technical field

What the present invention relates to is a kind of data judgment method being applied to electric power system, specifically a kind of Transient Voltage Stability determination methods based on real-time broad sense Thevenin's equivalence.

Background technology

Transient Voltage Stability in Electric Power System refers in a short time, and electric power system is subject to as line fault, loses the ability that all buses of system after the large disturbances such as generator or load keep burning voltage.How accurately to judge that electric power system generation transient voltage unstability is still the difficult problem needing to solve, current China generally adopts following Practical Criteria: continue (being generally 1s) lower than limit value (being generally 0.75p.u.) when system voltage backbone point busbar voltage in transient process declines, just think that system or load voltage are unstable; Otherwise then system or load voltage are stablized.This practical approach is the expedient method of one adopted when not having accurate Transient Voltage Stability method of discrimination, lacks theory support, and cannot ensure the accuracy of result.

Thevenin's equivalence principle is a kind of reduced chemical reaction kinetics model conventional in linear circuit, be introduced in from the nineties in last century in the static voltage stability analysis of complicated electric power system, concrete thought is: the system outside the load bus that study is regarded as the ball bearing made using that an equivalent voltage source and an equivalent impedance are in series, then judge voltage whether steady stability by the size of the mould comparing load impedance and system equivalent impedance, namely so-calledly wait impedance criterion.In theory, this criterion is equally applicable to the judgement of Enhancement of Transient Voltage Stability, but prerequisite wants to calculate the Thevenin's equivalence parameter in each moment in transient process by real-time tracking.

Through finding there is multiple Thevenin's equivalence parameter tracking algorithm at present, be broadly divided into three classes according to square ratio juris thinking, and all there is respective deficiency to the retrieval of prior art: 1) based on the algorithm of local status monitoring or Load flow calculation.Such algorithm utilize operating point and neighborhood point multiple time discontinuity surface status data, calculate Thevenin's equivalence parameter by the method such as least square regression.Its prerequisite is that in hypothesis operating point neighborhood, Thevenin's equivalence parameter immobilizes, thus there is parameter drift problem (" the drifting problem research of Thevenin's equivalence parameter ", Proceedings of the CSEE, 2005, 25 (20): 1-5), although document " node Thevenin's equivalence parameter on-line tracing short-cut method " (Proceedings of the CSEE, 2006, 26 (10): 40-44) and Chinese patent (document number CN103258103, publication date 2013.08.21) improve for parameter drift problem, but still fail fundamentally to change the defect based on discontinuity surface time multiple.2) based on total differential algorithm.Document " based on total differential Thevenin's equivalence parameter tracking algorithm " (Proceedings of the CSEE, 2009,29 (13), 48-53) with " adapting to the electric power system Thevenin's equivalence parameter tracking algorithm of large disturbances " (Proceedings of the CSEE, 2012,32 (22), these class methods 126-132) proposed consider the change of Thevenin's equivalence parameter between different operating point, there is not parameter drift problem, but higher to the degree of dependence of initial value, initial value arranges accurately whether larger to subsequent calculations Influence on test result.3) based on the algorithm of time-domain-simulation.Document " the Thevenin's equivalence parameter tracking computational methods based on time-domain-simulation " (electrical engineering journal, 2010,30(34): 63-68) and " receiving-end system Transient Voltage Stability study mechanism " (thesis for the doctorate, the second month in a season realizes it, Beijing: China Electric Power Research Institute, 2008) such algorithm proposed, to each time step in time-domain-simulation process, short circuit current when open circuit voltage when employing waits value node to open a way and short circuit is divided by and obtains Thevenin's equivalence impedance.The problem of these class methods is directly the conclusion of linear circuit to be applied to nonlinear electric power system, have ignored the change of Thevenin's equivalence parameter under this two states, thus there is certain deviation.

Summary of the invention

The present invention is directed to prior art above shortcomings, a kind of Transient Voltage Stability determination methods based on real-time broad sense Thevenin's equivalence is provided, take into full account the nonlinear characteristic of electric power system, Thevenin's equivalence parameter that can be current according to the state computation of discontinuity surface time single in transient state process of electric power system, thus judgement Enhancement of Transient Voltage Stability, overcome in existing method and there is parameter drift based on discontinuity surface time multiple, ignore the change of Thevenin's equivalence parameter when load bus open circuit and short circuit and there is deviation, needing the deficiencies such as initialization.

The present invention is achieved by the following technical solutions.The present invention according in transient state process of electric power system each time the update the system admittance matrix Y ' that inscribes and voltage, the power rating of system, obtain Jacobian matrix J and the inverse matrix J thereof of corresponding moment lower node power equation ^-1.By inverse matrix J ^-1obtain the relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity further, obtain the relational expression between equivalent Nodes load current variable quantity and load power variable quantity by the voltage of system, power rating, then calculate the Thevenin's equivalence resistance value in corresponding moment according to two kinds of relational expressions.Then the Transient Voltage Stability judge index in corresponding moment is calculated by the load impedance of the Thevenin's equivalence impedance in each moment and equivalent Nodes, for tracking and the early warning of Power System Off-line analysis and presence.For tracking and the early warning of electric power system warning device.

Whole technical scheme can pass through software simulating; it is off-line simulation for electric power system that its application scenario mainly contains two kinds: one; under analyzing specific fault or disturbance, whether electric power system key node can there is transient voltage unstability, thus determines operational mode, formulates protection and safety control strategy provides beneficial reference for relevant departments.Two is realize application on site when condition maturity, be installed on main important transformer station, by calculating Transient Voltage Stability judge index in real time and monitoring its situation of change, early warning was sent in time before close to Voltage Instability, be conducive to Regulation department and correctly take corresponding control measure, ensure safe operation of power system.

The preparation method of described update the system admittance matrix Y ' comprises the following steps: first according to the network parameter of moment t a certain in transient state process of electric power system, generation system admittance matrix Y; Secondly, all the other load buses beyond reciprocity value node process, and obtain update the system admittance matrix Y '.

Described processing method is: constant power load model does not deal with; Constant current load is considered as constant power load model; Induction motor load is as the process of constant-impedance load; Then constant-impedance loads all in system are integrated in the corresponding element of system admittance matrix Y and obtain update the system admittance matrix Y '.

The described computing formula be integrated in system admittance matrix Y by constant-impedance load is Y ' _jj=Y _jj+ 1/Z _lj, wherein, subscript L represents load, and j is the numbering of the constant-impedance load bus except waiting value node; Y ' _jjand Y _jjbe respectively the element in system admittance matrix Y and update the system admittance matrix Y '; Z _ljfor the resistance value of constant-impedance load.

The described step obtaining relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity comprises:

I) voltage of nodes all in system is represented with Cartesian form, according to amplitude and the phase angle of update the system admittance matrix Y ', each node voltage of current time system, and the node injecting power of the current each node of system, calculate the Jacobian matrix J generating current state lower node power equation, and invert and obtain J ^-1.Wherein, Δ e, Δ f are respectively the bias vector of node voltage real part and imaginary part, and Δ P, Δ Q are the bias vector that node injects active power and reactive power, Δ U ²for the bias vector of generator node voltage amplitude square, negative sign represents that load power is contrary with node injecting power direction: $\left[\begin{array}{c}\mathrm{\Δf}\\ \mathrm{\Δe}\end{array}\right]=-J^{-1}\left[\begin{array}{c}\mathrm{\ΔP}\\ \mathrm{\ΔQ}\\ \mathrm{\Δ}{U}^2\end{array}\right].$

The node injecting power of the current each node of described system does not comprise the load be integrated in system admittance matrix.

II) obtain relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity as shown in the formula.Wherein, subscript L represents load, and i is the numbering waiting value node; for equivalent Nodes node voltage variable quantity, Δ e _i, Δ f _ibe respectively its real part and imaginary part; Δ P _li, Δ Q _libe respectively load active power variable quantity and the reactive power variable quantity of equivalent Nodes; for J ^-1middle corresponding element: $\left\{\begin{array}{c}\mathrm{\Δ}{e}_i=\frac{\∂\mathrm{\Δ}{e}_i}{\∂\mathrm{\Δ}{P}_{\mathrm{Li}}}\mathrm{\Δ}{P}_{\mathrm{Li}}+\frac{\∂\mathrm{\Δ}{e}_i}{\∂\mathrm{\Δ}{Q}_{\mathrm{Li}}}\mathrm{\Δ}{Q}_{\mathrm{Li}}\\ \mathrm{\Δ}{f}_i=\frac{\∂\mathrm{\Δ}{f}_i}{\∂\mathrm{\Δ}{P}_{\mathrm{Li}}}\mathrm{\Δ}{P}_{\mathrm{Li}}+\frac{\∂\mathrm{\Δ}{f}_i}{\∂\mathrm{\Δ}{Q}_{\mathrm{Li}}}\mathrm{\Δ}{Q}_{\mathrm{Li}}\\ \mathrm{\Δ}{\stackrel{\·}{U}}_i=\mathrm{\Δ}{e}_i+\mathrm{j\Δ}{f}_i\end{array}\right.;$ After overload process, except waiting all the other load buses except value node to be all constant power load model, its power variation perseverance is zero, and the variable quantity of generator node voltage amplitude square is also zero, therefore $\left[\begin{array}{c}\mathrm{\Δf}\\ \mathrm{\Δe}\end{array}\right]=-J^{-1}\left[\begin{array}{c}\mathrm{\ΔP}\\ \mathrm{\ΔQ}\\ \mathrm{\Δ}{U}^2\end{array}\right]$ Rightmost side matrix is only containing Δ P _liwith Δ Q _litwo elements.

Relational expression between described equivalent Nodes load current variable quantity and load power variable quantity as shown in the formula, wherein, " * " represents conjugate complex number, for the variable quantity of equivalent Nodes load current, represent the node voltage of equivalent Nodes, represent the load current of equivalent Nodes, the voltage current by equivalent Nodes and injecting power calculate: $\mathrm{\Δ}{\stackrel{\·}{I}}_{\mathrm{Li}}=(\mathrm{\Δ}{P}_{\mathrm{Li}}-\mathrm{j\Δ}{Q}_{\mathrm{Li}}-\mathrm{\Δ}{U}_i^{*}{\stackrel{\·}{I}}_{\mathrm{Li}})/(U_i^{*}+\mathrm{\Δ}{U}_i^{*}).$

The method of described computing system Thevenin's equivalence impedance is: order equivalent Nodes node voltage variable quantity and the relational expression of load power variable quantity, the relational expression between equivalent Nodes load current variable quantity and load power variable quantity are substituted into following formula, and get wherein, Q _li, P _libe respectively load active power and the reactive power in equivalent Nodes corresponding moment, Z _iThevthevenin's equivalence impedance for equivalent node i place:

By in linear circuit conclusion be generalized in nonlinear power system, equivalent Nodes node voltage to load current derivative as Thevenin's equivalence impedance Z _iThev, be exactly broad sense Dai Weinan principle.The prerequisite applying this principle is: must be about analytical function.For electric power system, when load power growing direction is determined, this condition meets, therefore can be by be taken as the ratio of equivalent Nodes current loads reactive power and active power.In order after, omit high-order in a small amount, the expression formula of load current becomes due to and j ^-1, all known, therefore with be Δ P _liexpression formula, after being divided by, Δ P _lijust in time cancellation, namely tries to achieve Z _iThev.

The method of described calculating Transient Voltage Stability judge index as shown in the formula, wherein, i is the numbering waiting value node, CRI ⁱfor the Transient Voltage Stability judge index of equivalent Nodes, Z _lifor the load impedance of equivalent Nodes, Z _iThevthevenin's equivalence impedance for equivalent Nodes: ${\mathrm{CRI}}^i=\frac{\left|Z_{\mathrm{Li}}\right|-\left|Z_{\mathrm{iThev}}\right|}{\left|Z_{\mathrm{Li}}\right|}.$

The load impedance Z of described equivalent Nodes _licalculate according to node voltage and load power.

In described judgement transient process, the mode of voltage stability is: work as CRI ⁱduring >0, Transient Voltage Stability; Work as CRI ⁱwhen=0, transient voltage neutrality; Work as CRI ⁱduring <0, transient voltage unstability.Because transient voltage unstability is an irreversible process, therefore once there is CRI ⁱthe situation of≤0, just thinks and transient voltage unstability has occurred.

When the present invention is based on single, the system mode of discontinuity surface calculates Thevenin's equivalence parameter, there is not parameter drift problem; By broad sense Thevenin's equivalence principle, solve and directly the conclusion of linear system is applied to nonlinear power system and the deviation that causes; In addition, there is not initial value On The Choice yet.

The present invention has the features such as real-time, directly perceived, accurate, can meet the demand that system off-line is analyzed completely.On the other hand, along with the day by day perfect of PMU synchronized measurement system and the continuous enhancing of Computer Analysis, storage capacity, the on-line analysis that the Transient Voltage Stability determination methods of real-time broad sense Thevenin's equivalence is applied to system also becomes possibility, has a good application prospect.

Accompanying drawing explanation

Fig. 1 is schematic flow sheet of the present invention.

The IEEE-39 node power system topology schematic diagram that Fig. 2 provides for embodiment 1.

Fig. 3 is the actual change situation schematic diagram of embodiment 1 10s interior nodes 8 voltage magnitude after being disturbed.

Fig. 4 is the result schematic diagram using the Thevenin's equivalence track algorithm based on time-domain-simulation of the prior art to judge embodiment 1 gained.

Fig. 5 is the result schematic diagram using method of the present invention to judge embodiment 1 gained.

Embodiment

Elaborate to embodiments of the invention below, the present embodiment is implemented under premised on technical solution of the present invention, give detailed execution mode and concrete operating process, but protection scope of the present invention is not limited to following embodiment.

Embodiment 1

As shown in Figure 1, the present embodiment method therefor comprises the following steps:

Steps A: according to the network parameter of moment t a certain in transient process, generation system admittance matrix Y;

Step B: all the other load buses beyond reciprocity value node process, obtains update the system admittance matrix Y ':

Constant power load model does not deal with; Constant current load is approximate as constant power load model (ratio is generally less, little on the impact of result accuracy); Induction motor load is as the process of constant-impedance load (in transient process, the impedance of induction motor load can change along with slippage, but slippage is fixing in particular moment, thus shows as constant-impedance part throttle characteristics); Then constant-impedance loads all in system are integrated in the corresponding element of system admittance matrix Y and obtain update the system admittance matrix Y '.

The computing formula be integrated into by constant-impedance load in system admittance matrix Y is Y ' _jj=Y _jj+ 1/Z _lj, wherein, subscript L represents load, and j is the numbering of the constant-impedance load bus except waiting value node; Y ' _jjand Y _jjbe respectively the element in system admittance matrix Y and update the system admittance matrix Y '; Z _ljfor the resistance value of constant-impedance load.

Step C: according to amplitude and the phase angle of update the system admittance matrix Y ', each node voltage of current time system, and the node injecting power of the current each node of system (load be integrated in system admittance matrix is no longer considered), try to achieve Jacobian matrix J and the inverse matrix J thereof of node power equation ^-1:

Represented with Cartesian form by the voltage of nodes all in system, Δ e, Δ f are respectively the bias vector of node voltage real part and imaginary part, and Δ P, Δ Q are the bias vector that node injects active power and reactive power, Δ U ²for the bias vector of generator node voltage amplitude square, negative sign represents that load power is contrary with node injecting power direction: $\left[\begin{array}{c}\mathrm{\&Delta;f}\\ \mathrm{\&Delta;e}\end{array}\right]=-J^{-1}\left[\begin{array}{c}\mathrm{\&Delta;P}\\ \mathrm{\&Delta;Q}\\ \mathrm{\&Delta;}{U}^2\end{array}\right].$

Step D: obtain the relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity, and the relational expression between equivalent Nodes load current variable quantity and load power variable quantity:

Relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity is as follows, and wherein, subscript L represents load, and i is the numbering waiting value node; for equivalent Nodes node voltage variable quantity, Δ e _i, Δ f _ibe respectively its real part and imaginary part; Δ P _li, Δ Q _libe respectively load active power variable quantity and the reactive power variable quantity of equivalent Nodes; for J ^-1middle corresponding element: $\left\{\begin{array}{c}\mathrm{\&Delta;}{e}_i=\frac{\&PartialD;\mathrm{\&Delta;}{e}_i}{\&PartialD;\mathrm{\&Delta;}{P}_{\mathrm{Li}}}\mathrm{\&Delta;}{P}_{\mathrm{Li}}+\frac{\&PartialD;\mathrm{\&Delta;}{e}_i}{\&PartialD;\mathrm{\&Delta;}{Q}_{\mathrm{Li}}}\mathrm{\&Delta;}{Q}_{\mathrm{Li}}\\ \mathrm{\&Delta;}{f}_i=\frac{\&PartialD;\mathrm{\&Delta;}{f}_i}{\&PartialD;\mathrm{\&Delta;}{P}_{\mathrm{Li}}}\mathrm{\&Delta;}{P}_{\mathrm{Li}}+\frac{\&PartialD;\mathrm{\&Delta;}{f}_i}{\&PartialD;\mathrm{\&Delta;}{Q}_{\mathrm{Li}}}\mathrm{\&Delta;}{Q}_{\mathrm{Li}}\\ \mathrm{\&Delta;}{\stackrel{\&CenterDot;}{U}}_i=\mathrm{\&Delta;}{e}_i+\mathrm{j\&Delta;}{f}_i\end{array}\right.;$ Relational expression between equivalent Nodes load current variable quantity and load power variable quantity is as follows, and " * " represents conjugate complex number, for the variable quantity of equivalent Nodes load current, represent the node voltage of equivalent Nodes, represent the load current of equivalent Nodes, the voltage current by equivalent Nodes and injecting power calculate: $\mathrm{\&Delta;}{\stackrel{\&CenterDot;}{I}}_{\mathrm{Li}}=(\mathrm{\&Delta;}{P}_{\mathrm{Li}}-\mathrm{j\&Delta;}{Q}_{\mathrm{Li}}-\mathrm{\&Delta;}{U}_i^{*}{\stackrel{\&CenterDot;}{I}}_{\mathrm{Li}})/(U_i^{*}+\mathrm{\&Delta;}{U}_i^{*}).$

Step e: two relational expressions drawn by step D are divided by, and make load current variable quantity trend towards zero, obtain system Thevenin's equivalence impedance Z _iThev:

Order the result of step D is substituted in following formula, and gets wherein, Q _li, P _libe respectively load active power and the reactive power in equivalent Nodes corresponding moment:

Step F: the Transient Voltage Stability judge index CRI calculating a certain moment t ⁱ, wherein, Z _liload impedance (calculating according to node voltage and load power) for equivalent Nodes:

Repeat each step of above-mentioned A to F, calculate the Transient Voltage Stability judge index in electric power system critical load node each moment in transient process, thus judge the voltage stability of system in transient process: work as CRI ⁱduring >0, Transient Voltage Stability; Work as CRI ⁱwhen=0, transient voltage neutrality; Work as CRI ⁱduring <0, transient voltage unstability.Because transient voltage unstability is an irreversible process, therefore once there is CRI ⁱthe situation of≤0, just thinks and transient voltage unstability has occurred.

The present embodiment is described for the off-line analysis of IEEE-39 node standard test system, the topological structure of system as shown in Figure 2, and do following setting: the load of node 3,4,7,8 is induction motor load, the load of node 31,39 is constant power load model, and all the other node loads are constant-impedance load; During 0.2s there is instantaneity three-phase shortcircuit in node 6 side of circuit 6-11, and the 0.22s moment fixes a breakdown; Using load bus 8 as waiting value node.Specifically comprise the steps:

Step 1) forms system admittance matrix.System admittance matrix is tried to achieve by network parameter.It should be noted that, before and after disturbance, the individual element of admittance matrix may change, and as in the present embodiment, system admittance matrix needs amendment twice altogether, respectively corresponding three-phase shortcircuit moment and failture evacuation moment;

Wherein, network parameter comprises topology of networks, the resistance of circuit, reactance, conductance, susceptance, and the no-load voltage ratio etc. of transformer branch.

Step 2) processing load.In the present embodiment, the active power that the impedance of each moment induction motor is absorbed by voltage and the induction motor of its place node and reactive power are tried to achieve, and are then integrated in admittance matrix; Constant-impedance load is directly integrated in admittance matrix; Constant power load model does not deal with;

Step 3) calculates Jacobian matrix J and inverse J thereof ^-1.The computing reference flow calculation program of Jacobian matrix J, inverts, and adopts special mathematical method and instrument;

Step 4) is according to J ^-1the voltage current with node 8 and load current, determine its node voltage variable quantity with load current variable quantity expression formula;

Step 5) calculates Thevenin's equivalence impedance Z _8Thevand load impedance Z _l8, thus obtain the Transient Voltage Stability judge index CRI of current time;

Step 6) is according to step 1) ~ step 5) try to achieve the Transient Voltage Stability judge index CRI in each moment in 10s before and after disturbance.In theory when there is the situation of CRI≤0, transient voltage unstability can be judged as.

Result of calculation is as seen in figures 3-5:

Wherein, Fig. 3 is the situation of change of system 10s interior nodes 8 voltage magnitude after being disturbed, can see that the voltage of egress 8, between 9s ~ 10s, Transient Instability occurs, if but using the Practical Criteria of " voltage is lower than the time remaining 1 second of 0.75PU " as standard, so after 10s, just will can be judged as transient voltage unstability, with reality gap to some extent;

Fig. 4 is system judges Enhancement of Transient Voltage Stability after being disturbed with the existing Thevenin's equivalence track algorithm based on time-domain-simulation result in 10s, result display CRI is greater than zero all the time, therefore None-identified system there occurs transient voltage unstability, and this is caused by Thevenin's equivalence calculation of parameter deviation;

Fig. 5 is system judges Enhancement of Transient Voltage Stability after being disturbed by the method for the invention result in 10s, and result is presented at the situation occurring CRI≤0 near t=9.5s, thus is judged as transient voltage unstability, consistent with the result of Fig. 3.Therefore, the method for the invention accurately, intuitively can judge Enhancement of Transient Voltage Stability.

The application scenarios of the present embodiment acquired results divides two kinds:

One, when off-line analysis, when can there is certain fault in order to accurately to judge system, whether load bus there is Voltage Instability, and when there occurs Voltage Instability, after this gives the judgement of fault severity level and fault, the formulation of safety control strategy provides reference frame;

When two, being applied to on-line analysis, a warning value (as 0.2) can be set, when CRI sends early warning lower than during this warning value, so that Regulation department takes effective control measure in time, this is more accurate compared to existing criterion, can earlier identify Voltage Instability risk, thus be conducive to the safe operation ensureing electrical network.

Claims (5)

1. the Transient Voltage Stability determination methods based on real-time broad sense Thevenin's equivalence, it is characterized in that, according in transient state process of electric power system each time the update the system admittance matrix Y ' that inscribes and voltage, the power rating of system, obtain Jacobian matrix J and the inverse matrix J thereof of corresponding moment lower node power equation ^-1; By inverse matrix J ^-1obtain the relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity further, obtain the relational expression between equivalent Nodes load current variable quantity and load power variable quantity by the voltage of system, power rating, then calculate the Thevenin's equivalence resistance value in corresponding moment according to two kinds of relational expressions; Then the Transient Voltage Stability judge index in corresponding moment is calculated by the load impedance of the Thevenin's equivalence impedance in each moment and equivalent Nodes, for tracking and the early warning of Power System Off-line analysis and presence;

The step of the described relational expression obtained between equivalent Nodes node voltage variable quantity and load power variable quantity comprises:

I) voltage of nodes all in system is represented with Cartesian form, according to amplitude and the phase angle of update the system admittance matrix Y ', each node voltage of current time system, and the node injecting power of the current each node of system, calculate the Jacobian matrix J of current state lower node power equation, and invert and obtain described inverse matrix J ^-1, wherein, Δ e, Δ f are respectively the bias vector of node voltage real part and imaginary part, and Δ P, Δ Q are the bias vector that node injects active power and reactive power, Δ U ²for the bias vector of generator node voltage amplitude square, negative sign represents that load power is contrary with node injecting power direction: $\left[\begin{array}{c}\mathrm{\&Delta;f}\\ \mathrm{\&Delta;e}\end{array}\right]=-J^{-1}\left[\begin{array}{c}\mathrm{\&Delta;P}\\ \mathrm{\&Delta;Q}\\ \mathrm{\&Delta;}{U}^2\end{array}\right];$

II) obtain relational expression between equivalent Nodes node voltage variable quantity and load power variable quantity as shown in the formula, wherein, subscript L represents load, and i is the numbering waiting value node; for equivalent Nodes node voltage variable quantity, Δ e _i, Δ f _ibe respectively its real part and imaginary part; Δ P _li, Δ Q _libe respectively load active power variable quantity and the reactive power variable quantity of equivalent Nodes; for J ^-1middle corresponding element: $\left\{\begin{array}{c}\mathrm{\&Delta;}{e}_i=\frac{\&PartialD;\mathrm{\&Delta;}{e}_i}{\&PartialD;\mathrm{\&Delta;}{P}_{\mathrm{Li}}}\mathrm{\&Delta;}{P}_{\mathrm{Li}}+\frac{\&PartialD;\mathrm{\&Delta;}{e}_i}{\&PartialD;\mathrm{\&Delta;}{Q}_{\mathrm{Li}}}\mathrm{\&Delta;}{Q}_{\mathrm{Li}}\\ \mathrm{\&Delta;}{f}_i=\frac{\&PartialD;\mathrm{\&Delta;}{f}_i}{\&PartialD;\mathrm{\&Delta;}{P}_{\mathrm{Li}}}\mathrm{\&Delta;}{P}_{\mathrm{Li}}+\frac{\&PartialD;\mathrm{\&Delta;}{f}_i}{\&PartialD;\mathrm{\&Delta;}{Q}_{\mathrm{Li}}}\mathrm{\&Delta;}{Q}_{\mathrm{Li}}\\ \mathrm{\&Delta;}{\stackrel{.}{U}}_i=\mathrm{\&Delta;}{e}_i+\mathrm{j\&Delta;}{f}_i\end{array}\right.;$

Relational expression between described equivalent Nodes load current variable quantity and load power variable quantity as shown in the formula, wherein, " * " represents conjugate complex number, for the variable quantity of equivalent Nodes load current, represent the node voltage of equivalent Nodes, represent the load current of equivalent Nodes, the voltage current by equivalent Nodes and injecting power calculate: $\mathrm{\&Delta;}{\stackrel{.}{I}}_{Li}=(\mathrm{\&Delta;}{P}_{\mathrm{Li}}-\mathrm{j\&Delta;}{Q}_{\mathrm{Li}}-\mathrm{\&Delta;}{U}_i^{*}{\stackrel{.}{I}}_{\mathrm{Li}})/(U_i^{*}+\mathrm{\&Delta;}{U}_i^{*});$

The method of described computing system Thevenin's equivalence impedance is: order equivalent Nodes node voltage variable quantity and the relational expression of load power variable quantity, the relational expression between equivalent Nodes load current variable quantity and load power variable quantity are substituted into following formula, and get wherein, Q _li, P _libe respectively load active power and the reactive power in equivalent Nodes corresponding moment, Z _iThevthevenin's equivalence impedance for equivalent node i place:

The method of described calculating Transient Voltage Stability judge index as shown in the formula, wherein, i is the numbering waiting value node, CRI ⁱfor the Transient Voltage Stability judge index of equivalent Nodes, Z _lifor the load impedance of equivalent Nodes, Z _iThevthevenin's equivalence impedance for equivalent Nodes: ${\mathrm{CRI}}^i=\frac{\left|Z_{\mathrm{Li}}\right|-\left|Z_{\mathrm{iThev}}\right|}{\left|Z_{\mathrm{Li}}\right|}.$

2. method according to claim 1, is characterized in that, the preparation method of described update the system admittance matrix Y ' comprises the following steps: first according to the network parameter of moment t a certain in transient state process of electric power system, generation system admittance matrix Y; Secondly, all the other load buses beyond reciprocity value node process, and obtain update the system admittance matrix Y '.

3. method according to claim 2, is characterized in that, described processing method is: constant power load model does not deal with; Constant current load is considered as constant power load model; Induction motor load is as the process of constant-impedance load; Then constant-impedance loads all in system are integrated in the corresponding element of system admittance matrix Y and obtain update the system admittance matrix Y '.

4. method according to claim 3, is characterized in that, the described computing formula be integrated in system admittance matrix Y by constant-impedance load is Y ' _jj=Y _jj+ 1/Z _lj, wherein, subscript L represents load, and j is the numbering of the constant-impedance load bus except waiting value node; Y ' _jjand Y _jjbe respectively the element in system admittance matrix Y and update the system admittance matrix Y '; Z _ljfor the resistance value of constant-impedance load.

5. method according to claim 1 and 2, is characterized in that, in the transient process described in judgement, the mode of voltage stability is: work as CRI ⁱduring >0, Transient Voltage Stability; Work as CRI ⁱwhen=0, transient voltage neutrality; Work as CRI ⁱduring <0, transient voltage unstability.