You are on page 1of 11

512 IEEE Transactions on Power Systems, Vol. 10, No. 1.

February 1995

ON BIFURCATIONS, VOLTAGE COLLAPSE AND LOAD MODELING
Claudio A. Caiiizares, Member
University of Waterloo
Elec. & Comp. Eng. Dep.
Waterloo, ON, Canada N2L 3G1

Abstract: This paper discusses the relation between bifurcations
and power systems stability through a thorough analysis of sev-
eral examples, to clarify some ideas regarding the usefulness and
limitations of bifurcation theory in network studies and operation,
particularly in voltage stability related issues. Different types of
1L6 VLOO n

U
Load 1
load models are used in a sample system to analyze their effect on
system stability and bifurcation. Finally, the Ecuadorian National
Interconnected System (SNI in Spanish) is used to depict and dis- Fig. 1: Sample s y s f e m f o r bifurcation analysis.
cuss the effect of load modeling in saddle-node bifurcation analysis
of real power systems.
Keywords: bifurcations, voltage collapse, load modeling, stability,
eigenvalue analysis. AVRS [12]-[15]. In some of these cases, but not all, the voltage con-
trols force the eigenvaluesto instantaneonslyjump into the unstable
region, making the system immediately unstable. This phenomena
1 INTRODUCTION is not directly associated to a bifurcation, since the eigenvaluesdo
not go through zero or the imaginary axis. Nevertheless, transcriti-
During the past decade utilities have reported serious compli-
cal bifurcation theory can be used to explain the phenomena when
cations in maintaining network stability in their power systems,
AVR limits are assumed to apply gradually [13].
particularly voltage stability, as some events occur and parameters
change in the system [1]-[3]. The study of several cases has led Recent literatureshows that there is not a consensus in the power
system community regarding the importance and usefulness of bi-
the power systems community to identify different causes for this
problem. furcation analysis of power systems (e.g., [16]). Therefore, this pa-
per concentrates mainly on clarifying some issues regarding local
Hopf and saddle-nodebifurcations have been recognized as some
saddle-node bifurcations, by clearly depicting their effect and rela-
of the reasons, albeit not the only ones, for voltage stability prob-
tionship to the voltage collapse phenomena for different load mod-
lems in a variety of power system models [4]-[9]. Local bifurcations
els in several sample systems. This does not rule out the fact that
are detected by monitoring the eigenvaluesof the current operating
other typeof bifurcations, i.e., transcriticalor Hopf, do also occur in
point. As certain parameters in the system change slowly, allow-
real systems, with similar catastrophic results for system (voltage)
ing the system to quickly recover and maintain a stable operating
stability. Although bifurcations cannot be considered as the only
point, the system eventually turns unstable, either due to one of
cause for voltage stability problems, as discussed above, the effect
the eigenvaluesbecoming zero (saddle-node, transcritical, pitchfork
of saddle-nodebifurcations in system stability cannot be overlooked
bifurcations), or due to a pair of complex conjugate eigenvalues
based on this sole argument.
crossing the imaginary axes of the complex plane (Hopf hifurca-
tion). The instability of the system is reflected on the state vari- The final section of this paper concentrates on applying the re-
ables, usually represented by frequency, angles and voltages, by an sults obtainedfor the samplesystems to the 115 bus SNI Ecuadorian
oscillatory behavior or a continuos change (voltage decrease, i.e., system [lq, to discuss some of the applications and shortcomings
collapse, and frequency and angle increase, i.e., loss of synchrw of bifurcation theory in a more realistic environment.
nism). In some cases these bifurcations can be associated to the
power transfer limit of the transmission system; in other instances 2 EXAMPLES
the bifurcations appear due to voltage control problems, like fast
acting automatic voltage regulators (AVR) in the generator [IO], or In this section the stability of four sample systems is thoroughly
voltage dependent current order limiters (VDCOL) in HVDC links analyzed. Local saddle-node bifurcations are depicted and discussed
[7].In all cases these bifurcations occur on very stressed systems, to highlight some of their main characteristics. All four cases dis-
i.e., the region of stability for the current operating point (stable cussed here are based on the system depicted in Fig. 1.
equilibrium point or s.e.p.) is small, hence, the system is not able The generator is modeled with the classical second order me-
to withstand small perturbations and becomes unstable. Although chanical differential equations. The damping torques are assumed
there are reports of these bifurcations occurring in unstressed sys- to have a relatively large value to indirectly simulate the effect of
tems [ll], this camiot be considered typical, since power system damper windings in the generator. The AVR is modeled by keep
controls are designed so that eigenvaluesof several operating points ing the generator terminal voltage constant at 1 pu, but no limits
are well into the left half complex plane. are included in this simple model. Although this is uot a detailed
Some voltage collapse problems can also be associated to voltage model of the generatorby any means, it gives a good starting point
control devices like under-load tap changer (ULTC) transformers or for analyzing the significanceof saddle-node bifurcations in system
stability, without adding unnecessary complexities that ouly ob-
scure some of the questions that are being addressed in tlus paper.
94 SM 512-4 PWRS A paper recommended and approved For an analysis of bifurcations in detailed generator models review
by t h e IEEE Power System Fngineering Committee of t h e reference [I]].
IEEE Power Engineering S o c i e t y for p r e s e n t a t i o n a t The transinissionsystem is simply modeled as a constant lumped
t h e IEEEIPES 1994 Summer Meeting, San Francisco, CA, reactance, which is a typical representation of these elements in
-
July 24 28, 1994. Manuscript submitted December 23, transient stability studies, where the system frequency oscillates
1993; made a v a i l a b l e f o r p r i n t i n g May 3, 1994. slowly around the nominal value (e.g., 60Hz), allowing for a quasi-
steady state or phasor analysis of the system. Consequently, this
type of methodology cannot be used to simulated high frequency
voltage problems in the system.
0885-8950/95/$04.00 0 1994 IEEE

~ ~~

hence.1 Generator-Infinite Bus Fig. and one u.e. p . due to the quasi- one eigenvalue of J becomes zero ( J is singular) and no equilibria linear profile of this measure. the s.0.e. 2.. It is interesting Figures 2.So. i.S0.) eigenvalues. making can be observed in larger and more detailed modeled systems [4. and 4 were obtained from equations (1) by setting to observe that this particular energy function V ( .e. 0) is a s. ) is the integral M = D = O. and the u. e .p.) corresponds to a saddle-node can give the user this information.e. C .. all the eigenvaluesand eigenvectorsassociated to these = iMwZ+ l60 hT(zL. differential equations ( 1 ) represent the system.s for each or singular values for measuring proximity to bifurcation (see refer.O) are unstable equilibrium points (u. 2: ( a ) M i n i m u m magnitude of the s . 19.p.e.cos(S)] X From the eigenvaluesof J it can be readily shown that (So.e. By using the potential energy distance V($. and ( b ) m i n i m u m singular value of the s .O).Spu.. making these measures inadequate to monitor system proximity instability.p..s.p.p. the maximum power trausfer point for the the stability region. When P = P m a Z . X = k = O.p.O) presents the same profiles.p.p.p. 0 ) and (-3rr/2.P) (3) with 60 = sin-'(PX/V) (other equilibrium points are not of inter- est). only one eigenvalueis on the right half plane.e. 3(a) and minimum magnitude of the complex eigenvaluesand the minimum (b).8.p. and also visualize the relative size of exist. In particular. due to the elimination of the bifurcation. note that v i s tangent to the bifurcation diagram.e.p. (rr . (S0. Similar profiles of the stability region as the system approaches bifurcation.. 2 the of the area of the bifurcation diagrams depicted in Figs.e. i. equal area criterion). including all system nonlinearities. Other tednuques have been proposed to avoid using eigenvalues and the dashed lines the correspondmg "closest" u.6 node The first load model is an infinite bus.. and ( f r r .P ] du X = -1M w Z 2 + P(60 . The point (rr/2. yielding equilibria (So. have values or linearizations of the system around equilibrium poiiits merged. In Fig.. e . and the "closest" u. (u.e. p .. power P > 0 from the generator at a constant voltage V > 0. 9. The latter transmission system.Pm. yielding a typical "nose" curve for angle (TEF) techniques have been shown to have a "linear" profile and variables and a different view of the classical P-6 stability curve. the same example is used here to introduce some basic concepts of bifurcation analysis. yield additional information regarding the stability of the system Some interesting observations can be made in these figures: [23]. AU equilibrium points can be easily found by solving equations (1) with [6 &IT = 0 . which is drawing active curve. by looking . decreasing rapidly after that. P ) azes. and closest U . Transient Energy Function 3(a) in the ( 6 . 3(c). Similar TEF profiles have also been observed for s. Hence. For this simple case. as depicted in Fig.6 ) + - V [cos(So) . then for any value of P > Pmal= V / X . 3.) of type one. for a chosen load level P (i. P ) plane.p.lpu. The s = w (1) potential energy can be approximated in this case by the integral of the power-angle product. one can the system load changes.sin(u) . ( c ) potential energy W p between s . 201. Figure 3(b) is the projection of Fig.P)du + lo[ equilibria can be calculated from the Jacobian matrix (2). whereas ( i r r . i. siuce it meets all the transversality conditions required nonlinearities that characterize the stability of the system for large for this type of bifurcation in state-parameter space (7.e. 513 Fig. Observe that the profile of these highly nonlinear AC/DC system models when close to bifurcation two measures is highly nonlinear aud varies only slightly up to the [25]. all eigenvaluesare in the left halfconiplex plane. p .e. Some of these examples are discussed from a different point of view in [18]. system under discussion.O) V(63~9P) = W ~ ( WW + ) p(6. perturbations like line openings at fixed load levels.So. the euergy can be defmed as a summa- tion of the kinetic and the potential energy as show11 in (3). value of the parameter P. J is also singular and two equilibrium points property is rather important. when assmniiig only small variations of the system frequency (a typical conjecture in transient stability X studies). Hence. this property has increasing P load. being tlus singular value of the s. becomes zero. e .e.e. the u. been demonstrated for other system models and energy functions At the value Pmal= 2pu one eigenvalue (singular value) of the in reference [24]. P . it more sensitive to small perturbations that can easily drive it to 211. hence. 3: Bifurcation diagrams: (a) in ( S . making these methods more suited for measuring proximity e The uormalized left eigenvector v at the bifurcation poilit is to bifurcation but at a higher computational cost.p. Notice the sharp change in the eigenvalues magnitude and singular values when the system approaches bifurcation. The four study cases are obtained by changing the load model at the receiving end of the transmission line. ence [22] for a summary). the energy measurement clearly depict the reduction load level of P x 1. 3(a) y d (b) represent the s. w .e. s.e.e.0) exist. to bifurcation. 6 V x !Mwz . since no technique based on eigeii- ( r / 2 . If P changes. This Right v and left w eigenvectors at the bifurcation point are also is the classical generator-infinite bus model utilized to introduce depicted.O) are plotted for a continuously the reason for the "linear" profile of the TEF..O. basic ideas of angle and frequency stability in power systems. as a function of P .P) between the i. as shown below. get a good idea of the distance to bifurcation. For the sample tangent to the bifurcation diagram [26]. and V = lpu. ( b ) P . The continuos lines in Figs.e.

V . a s. P) of tlus set of equations is for all equilibria).e. 6: Time domain sirnolation f o r a change i n P from 0. This particular example does not present any stability problems that could be characterized Fig.p. and a u. P ) azes.618034. however. This eigenvector also yields information regarding the initial dynamics of the system at the bifurcation point [SI.. between s. Figure equations (1) plus equation (4) below. When r -t 0. so that an operator knows in state and parameter space how close the system is to the in- stability point.e. unless it is used in conjunction with and "energy" measure that takes into consideration the nonlinearities in the system.V / T X sin(&) 0 .p. but it has been shown to yield information regarding P .01 at t = 2s. introducinga dynamic voltage and algebraic constraints cannot be qualified as a "causal" system term in the reactive power load roughly simulates the response of according Kwatny's et al definition [5]. so that corrective measures can be taken to avoid stability problems 19. 4 depicts the dynamics of the system as the load is increased slightly beyond the bifurcation value. this system induction motors to sudden voltage changes under normal voltage presents a degenerate bifurcation point in state-paranleter space operatingconditions [29.Vo. Observe the slow initial change in the state variables (6. eigenvalues as a function of load level P . dynamics of the state variables (6. J = -V/M.)] 2. ( 6 ) grams. 5(a) depicts a saddle-node bifurcation diagram for 6 and V (w = 0 ing to an equilibrium point (60. For a detailed analysis model of the actual system load. coalesce at a singular represented by (5). equation constant active and reactive power model.p.e. corresponds 1 (4) again to the point of maximum power transfer. and the Jacobian correspond.V) plane. .V node curve. and setting T = lins to simulate fast load The differential equations used to model this sample system are dynamicsin the context of a typical transient stability study. at the maximum entries in tlus vector. P. and can be used to devise a technique depicted.e. 271. however.cos(60) -DIM -1/MXsin(60) (5) ..p.w). tlus diagram by itself does not . with equilibria disappearing for values of P > P.2 PQ Dynamic Load Here k is a scalar representing constant power factor.514 Fig. The bifurcation diagram can be used to determine the dis- tance to the bifurcation point. i.e. This cannot be considered an accurate where two stable equilihrinnipointscoalesce.s depicted in it. Nevertheless. u. 5(b)) for the parameter value P. in the previous example.p.. as it is shown in the next section with an R-L load. ( c ) potential energy W .e.30].. b The normalized right eigenvector w at the bifurcation does Fig. 271. Right v and leji w eigenvectors at the 6ifurcation point are cation point [9.99 t o 2. some of the stability problems. particularly voltage collapse. and closest the equations that are most sensitive to changes at the bifur. This model is meant to (4) becomes an algebraic constraint...62 at t = 28. the dynamics of the simulate the behavior of typical static loads when ULTCs are taken system lose physical meaning since this set of differential equations into consideration 1121. = 0. to 1 0 correctly capture all the bifurcation phenomena. to move away from tlus instability [28].0. a rather small system perturbation. ~ . and r is the The infinite bus load model is now replaced by a more "realistic" time constant of the dynamic voltage term. T h e vector w' is the projection of w onto the (6. that Figures 5 and 6 were obtainedusing the same constant values as occur when a saddle-node bifurcation is encountered by the system. point (Fig. with initial slow ji IO . Furthennore. angular-frequency instability would necessarily force the voltages to collapse. ( 6 ) minimum not have a direct geometric interpretation in bifurcation dia.1 / r X [ 2 V .I rr // yield the size of the stability region associated to the different s. magnitude of s. Furthennore. This is the reason why bifurcation studies should be done [ ] considering all system dynamics and not only voltage controls.sin(S. 5: Bijurcation diagrams: ( a ) i n ( 6 . but allows to introduce and depict of these types of system models review references [31]-[33]. one can pinpoint the areas (state variables) that are dianging the most and driving the system to bifurcation. w ) [a]. Notice the separation or lost of synchronism of the generator with respect to the infinite bus.61 to as voltage collapse.e. Observe the fast voltage collapse and the mechanical a bifurcation problem. 4: T i m e domain simulation for a change in P f r o m 1. Finally. Fig. and cannot give the operator a sense of the type of perturbations that the system c a n withstand at some specific load level. it clearly depicts an instability due to 0.p. and if voltages were allowed to change the instalility reflected in 6 and w. Tlus eigenvector de- pends on the time constants ( M in the sample system) of the differential equations [26]. Figure 5(c) depicts the classical . whereas v is independent of them.

due to the presence of a resistive term in the model [35]. due to the small voltage time constant ( c ) P .6 equations (1) and (4) for a slight increase in load power P . differential equations (7) are used to model this particular system. point Pmas. Hence. the being stable. observe that no bifurcations occur readily used as in previous examples to obtain a relative measure in this case. xsin(6) X .5 0. s = w (7) & = 1 M [V'G.e. Observe that all state variables ( S . Furthermore. ( b ) minimum the state variables start cliaiging slowly to then rapidly render the magnitude of s.G) azes. and X E W an approximate TEF could be constructed in tlus case. 7(c)). braic constraints (S).s in the bifurcation di- agram of Fig. ignoring frequency dynamics. 7(d).1 0. The tip of this nose curve is not a bifurcation point. Fig.6 ) + --[c0s(60)-cos(6)] . 5(d).W')] Figures 6 depict the time domain simulation of the differential (6) :::v . 515 P .. Notice fhat all equilibrium as compared to the machine inertia.e. even though the rather quickly when compared to the angle and frequency. s Q 0. wludi is a projection of the bifurcation diagram in the ( P . ( d ) P .p. each value of G 2 0. 5(a) are depicted in Fig. A relatively large The foregoing discussion can be directly extended to larger and change in G. Tlus P .3 0.. model.. In this case.kP[ln(Vo) .e. taking t y p i d y done on a balanced phasor model of the network. w . i.78885.4 0. the system load is changed to a static RL load model. However. only one s.s and corresponding u. with all equilibria in tlus curve to what happens in the case of the PQ load model when r -+ 0.Dw v cos(6) I Fig. The results of evaluating the energy difference between the s. wlucli takes the system around the maximum power more complex power system models. 7 and 8.p.5 at t = 25. since G is the parameter that changes in the 3 APPLICATIONS system and not P. Nevertheless.2 0. i. this setting T = 0. Figure 7(a) depicts the bifurcation diagram for changes in the parameter G.V nose curve. 3(b).-+ - v2 x x 1 point P. the tlurd differential equation becomes an alge- system presents exactly the same P-V nose curve as in the case braic constraint.4 0. faking the systems around the maximum power transfer V = 1 [-kVzG . with the same dynamic voltage term as in the previous case. in Fig. V ) plane. = 0.3 RL Loads Now the PQ dynamic load is replaced by a11 RL dynamic load. it cannot be is a parameter used to simulate slow changing conditions (small .6) plane.5 0 . The nose curve in this case cannot be considered a bifurcation diagram. one obtains Figs. as shown by the eigenvaluesplot of the system equilibria of the stability region as the conductance G changes. This sample system is a classical bendunark for voltage stability studies.w. with G representing the load conductance and the slow varying parameter in tlus example. IG(1 +jk)( = 1/X (G = 4/& = 1. can be represented by the set of nonlinear differential and alge- A n exact energy function cannot be found for tlus sample system. due to the relation between elinunating the dynamic voltage dependence in equations (7) by conductance G and the active power P depicted in Fig.e. i. the voltage collapses at t x 5s points are stable f o r any value of the parameter G. a P-6 nose curve.e.e. 7 : Bifurcation diagrams: ( a ) in (6. have led many researchers to only consider voltage dynanucs for the analysis of bifurcations problems. Transient stability analysis is transfer point P m a z .p. Figures 8 corroborate the previous discussion. 8 : Time domain simulation f o r a change in G f r o m 1. i.* 0.. representing the reduction of the stability region as the system approaches bifurcation.V z ) .. can be obtained for Finally. Observe that the right eigenvector v is also tangent to the bifurcation diagram at the saddle-node bifurcation point.p. The energy function for this particular system is represented by equation (6) [34].V nose curve i s ezacily the same as in the case with the P Q load particular voltage behavior due to the differences in time constants.e. eigenvalues as a function of the parameter C:.P) = +&(Vz X V l M w z + P ( 6 0 . where x E 8" stands for the state variables. In this plots the same patterns of the previous example appear.618034). system nnstable. due to the characteristic voltage behavior discussed in this section. V(S. and Pmaz= 0.e.e. in tlus case the bifurcation occurs at a value of 6 ES 30' << 90'. the load is represented by a constant power factor impedance. does not render the system unstable..618034pu. results remain exactly the same with the only difference that V is although it corresponds to the system's maximum power transfer allowed to change instantaneously. i.1 o. the previous example clearly shows that this assumption is not completely justifiable.e. 7(b)..G curve.5 to 2. a i d contrary of the PQ load model (Fig. V ) remain sfable. however.L \ 0.V. with the entries in this vector pointing to a voltage problem regardless of the value of r > 0. can also be obtainedhere by projecting the bifurcation diagrani in the ( P . Despite this change in load model. Notice once more the quasi-linear profile of the TEF. A similar result as the one depicted in Fig. wlud~ it to a new s. Although y E 8"' portrays a set of iniplicitly defiled variables.V nose curve. in spite of the large perturbation. 2.V.p. Using the same constant values as in the previous examples.1 0.

X) = 0 . QI = -(Qio + AQi.e.e. lyzing stability of these models.. fully matching the operationalknowledgeof the actualsystem. transcritical bifurcations can also be encountered. without having to actually calculate and since saddle-node bifurcations occur generically in differential equa..yo. . [36]). APl. rendering Hence. ~ ) be invertible. 10 and 11 [15. Nevertheless. a city in the south-west of assumed that the pattern of load change can be represented with Ecuador) and 34 (a 138KV bus located in Manabi. one-phase impedances without a n y frequency dependence.yo. The problem with this approach is that it cannot detect remote voltage control. fork bifurcations.X) (Vtlxo)2 for PQ load models and typically small resistive losses in the trans- mission system. 9..p.261. APi. which give the system equations a special Hopf bifurcations. one needs to run a costly transient -(p12+A~. Nevertheless. V ) planes ( P .) and the use of reactive powers throughout the equa. The generators For a detailed discussion on this issues and their relation to bifur- in this case are simply model using the second order differential cation analysis in power systems review [7.g.. which is an acceptable model to address the main issues in and eigenvectors of the system Jacobian (IO) must be monitored. Hopf bifurcations are unlikely to occur [lo].. the output and cost of t h e e studies are closely [31? 321. For fast For equations ( 8 ) . the algebraic constraints must meet certain load dynamics a very detailed model of the whole system is needed.20. since it has been equilibria eigenvalues to detect the crossing of the imaginary axis. Apt.V. areusedtomodeladirectionofloadchange system models.(~~. the system can be represented by the set of differential equations The SNI Ecuadorian system is used in this paper due to its par- ticular voltage stability problems.X) . a province in the one degree of freedom (A). yielding a close approximation for a singularity of this Jacobian and test for the corresponding of the bifurcation point (singularity point) and the corresponding transversality conditions [7. This method allows to trace bifurcation manifolds (diagrams) for any type of system model.X) . . and to study some possible solutions.X) and SQi(6. This is an old discussion that has not been resolved to the total satisfaction of all the Dower system commuuity.+ Api.261. very fast dynamics quantities. since load.Pi.s in these hal figures. however. the present discussion. For this paper a mixed costly task. yshould o . the system.. due to the P Q load model used in this case. Nev- ertheless. and that this evolution of load drives the central west coast of Ecuador). yielding a maximum loadability margin of . it is a very valid question. Due to system size. since these tools have been tions. and bifurcation analysis is not techniques must be used to analyze the dynamics of the system an exception. which is a ferential equations representing the load. the eigenvalues amples.V nose curves) for various is the load phasor voltage at bus 1. a compromise and a critical view of the results yielded by the system “causal” [5]. X).X) are the active and reactive the area on the diagrams can be associated to an energy measure power mismatches at the load buses. the load model must be & i d in-the context of tce tools This model has the difficulty of not fully representingreal physical that are used for stability studies. and is son the author has chosen not to differentiate between s. like thyristor switching of motor drives cannot be fully represented straints g(. 19. Notice that differential equations (11) can system nose curves. The results shown here have been used to c o i f i n sus- This reduction is actually unnecessary. P i . Qil.7ifQ WnG) 7tVQ Vi Figures 9 and 10 depict the projections of the bifurcation di- Here Pi and Qi are the powers injected by the load.. however. balanced pha- any degree of complexity in modeling and system size. nor differentiate between transcritical and pitch- symmetry. this set of equations can be used to represent designed to deal with quasi-steady state models. at equilibrium points (xo. In all these figures. ~ n G ) . giving an idea of how Although the issue of whether this load model is a “realistic” stable the system is at a particular load level. Furthermore. since as demonstrated tem load (1292MVA). the technique is particularly suited For saddle-node bifurcations in particular. For the latter. the Jacobian very expensive.. Bus 34 was particularly chosen since the right be rewritten as and left eigenvectors at the bifurcation point.y-’(x. A continuation method was used to obtain the bifurcation diagrams in Figs.. trace the system eigenvalues and eigenvectors as the parameter X tions (9) (191. these types of studies are such that f(xo.e. Particularly. Qi. one needs only to look to detect saddle-node bifurcations. since one only needs to check picions regarding the possible causes of the voltage problems in for and remove singularities in the algebraic equations Jacobian..p.(QI] + AQi. with costs growing exponentially with system size. sality conditions of the system at the bifurcation point. if D.O)-T~.yo.~ l ~ p ~ simulation or use special techniques that take into considerations additional nonlinear terms in the system [8. and system models. one typically expects to be set to zero to represent static load models.(61 .s and the system reference frequency. particularly load (8) modeling [371. The fonner needs of the computation of the system load model represented by equations (11) is used. were plotted to depict two distinct system to bifurcation. Pi1. in typical transient stability program. have stable equilibria above the saddle-node bifurcation for these AQi. where 6Pi(6. of these constraints D y g ~ ~ l o . The latter is usually unnecessary right eigenvector v [%I. q f P . shown to adequately represent a variety of system loads in typical since the Jacobian J in (10) does not become singular in this case.. Furthennore. respectively. 31. Furthermore. + . AQI. 26. and also to depict the effect of X = f(x.A) (Vt/xo). the results of stabil- ity studies are closely related to system modeling.. and hL6t agram onto different ( P . a very expensive procedure when dealing with relatively large sys- tems. .( Q I ~ AQi. equations with constant terminal voltage used in the previous ex- For bifurcation analysis of this system model. Although the bifurcation diagram projections in Fig.V.+ A f i l x ) ( K / V ? ) 2 (11) (subcritical or supercritical). For this rea- the frequency and voltage dependent dynamic terms. 371. in this paper and others before (e. nevertheless. detecting possible occurrences Qi. transient stability studies [29. 201.q V p 71fQ. otherwise singular perturbations or noise some of these programs is needed. 221.g(. i. within the sor models that treat transformersand transmission lines as lumped bounds of the assumptions used to build the model. For most bifurcation studies it is 15 (a 138KV bus located in Machala. under certain operating conditions like changes. the voltage profiles in buses as the parameter X slowly varies. nevertheless. the representation of the actual system load does not pertain to this parameter X in this case can be associated to a total MVA sys- particular paper.x)(KIII. 7.. note that these time constants can u. and riVQ represent the time constants of Hopf bifurcations are beyond the scope of this paper. i. which is reflected in the presence of the algebraic con. to discern wluch type of Hopf has taken place 9 = -(Pi. 9(a) do not yield mudi information regarding the size of the stability region by themselves.( 4 .. are constant weighting factors that define the steady state base of Hopf bifurcations before a saddle-node was not pursued.e. 321.X) (9) different load models in saddle-node bifurcation analysis of real systems.X). system trajectories.) is invertible (nonsingular) along related to the models used. one needs to check for the transver- Embedded in this set of equations are the algebraic and/or dif.X) = F(x. for instance.X) = 0 and g(Xo. 30. andAQ1.516 perturbations) in the system. pinpoint that partic- ular area as the one with the largest reactive power deficits. conditions in order for the model to approximate physical reality being toolslike the EMTP more appropriate enviromnents for ana- [5.

981-991.p. Based on the results of the bifurcation analyses of the previous cases.” ZEEE Trans. Nevertheless. 33. which is one of the possible solutions under consideration at the Ecuadorian utility (INECEL in Spanish) to improve the stability lama lambda of the SNI system. vapor turbines in El Salitral (located near ena. The latter comment should be regarded with perience. Proceedings: Bulk PoWeT System voltage node bifurcations. 253-262. February 1990.132 = 170. Circuits and Syst. pletely losing voltage control of the system. Chiang. Bifurcation diagrams of the SNZ Ecuadorian system when ( a ) PQ and (b) mixed load models. “Towards a Theory of Voltage loadability margin of the system in Fig. Phadke.6 I 0. 11.” ZEEE Trans. one cannot use the simple analogy of the bifurcation dia- as the results for a real system show when compared to the ac- gram area to convey an idea of the size of the stability region for tual operational knowledge of it. 9: Bifurcation diagvams of the SNZ Ecuadorian system f o r Fig. Due to the relatively large resistive terms in this system that these analyses and the related methods are thoroughly valid. . since the loadiiig level indicated by X does iiot have a one. ters. and figure (b) t o mixed load la) (b) models. Furthermore. station (a relatively large generation station in the central Andean 1. complex located in the south central Ecuadorian Andes) and com.0615 = 79. H.60 0. Draft 3. ed. Nevertheless.” Sysiems U Conirol Let- problems have been detected during the actual system operation. Pasrija. the generators. Figure ( a ) corresponds to PQ load models. which is not an accurate model. “Static Bifurcations due to some generators readcng maximum reactive limits at the in Electric Power Networks: Loss of Steady-State Stability and largest Ecuadorian generating station (Paute. as with any other each value of A.1 0.. since major generating facilities reach maximum re. ed. since now the loads reduced their power demand model are not “realistic.2 0. Q limits are enforced in the generators. K.A. 10 depict the effect of generator Q limits in saddle- [l] L. 5. 11. but the approach allows for a first uuderstanding of the general effects of AVR limits in saddle-node [2] L. G. for several system models. given load level. Power Systems.05 0. observe the significant improvement in system loadability. The relation between these In Fig. G. pp. tive power support that is the characteristic of this system. Begovic and A. the coastal city of Guayaquil) [4] M. have demonstrated the validity and limitatious caution.15 0. [3] Modelling of Voltage Collapse Including Dynamic Phenom- active power limits.417712 is practically twice as mush as the one in not be dismissed based solely on the idea that the load and system the previous case. the effect of differeut load Notice again the wide changes in voltage magnitude at bus 34 for models in saddle-node bifurcations has been addressed in a variety chauging loading conditions. i. Fig.J 0 0.229133).5MVA. in both cases. Notice the rather small [GI I. allowing the system to handle “heavier” used to ignore the output of Power Flow studies. Y. EPRI EL-6183.05109 = 66MVA in Fig. Dobson and H. VA.. Particularly. June 1992. L. of these types of analyses. 517 0. The new system yields Figs. Fink. These Q limits were roughly modeled by fix- Phenomena-Voltage Stability and Security. The The results from bifurcation studies presented in the paper cam value of .1 0. Vol.2 0.p and a u. These particular Collapse in Electric Power Systems.4 0. = 0. REFERENCES Figures. model. Kwatny. Ecc. and the Agoyan hydro Voltage Collapse. 9(b). In Figs.. including a real 115 bus system. and deinoustrates figure. Fink.e. pp. a 70 MVAR synchronous compensator is placed on bus 5’4. however. 10. Vol. 198-203. indicating once more the lack of reac.. Proceedings: Bulk Power System Voltage bifurcations for different load models.. 4 CONCLUSIONS The paper has thoroughly discussed the limitations and typi- 296MVA (A. cal uses of saddle-node bifurcations. January 1989. close to the saddle-node bifurcation oue expects the system to have a small stability region due to the merging of the s.” Observe that similar arguments can be as the voltages decrease. a 70MVAR synchronous compeusator is placed 011 bus 34. [5] H. Baliar. a 1000 MVA hydro Voltage Collapse. Task Force 38-02-10.3 lama lambda lama lambda Fig. the user must discern the usefulness of can yield the robustness of the system to large perturbations at a these results based on the methods and models used.1747 in Fig. “Dynamic Simulation of at X = 0.2 0. ’m 0. = 0. 10: k?ijurcation diagrams of the SNZ Ecuadorian system f o r to a maximum reactive limit in the hydro generating statio11 of ( a ) P Q and (b) mized load models when Q limits are considered in Agoyan. lO(a) de- picts an innnediate system instability at A = 0. 10 one can clearly Phenomena-Voltage Stability and Security. 1989.40 o. 13. Inc.e. of test systems.. region) at X = 0. Fig.9 and they are the reason for the repetitive voltage problems through- out the network. the largest load center in Ecuador. due Fig. 11(a) still presents a Q limit instability at X = 0.3 0 0. ECC/NSF Work- observe that the Q limits significantlyreduce the loadabilitymargin shop.5MVA. Vol. ing the reactive power output of the generator at its limits [12]. No. Fairfax. October 1986.e. research and ex- loading conditions. A. August 1991. 10(b). H. M. CIGRE Report. D. No. lO(a). the mixed load model (11) was used to obtain particular bifurcations and voltage collapse bas also been analyzed new bifurcations diagrams and a different saddle-node bifurcation. lO(a). for this particular load model only time simulations system analysis technique. pp. However.l 0. This paper answers the latter two ques- to-one correspondence to MVA loading levels as in the previous tions pertaining saddle-node bifurcation studies.

Vol. Gao. Dy. No. No. 1. Circuits E. Morrison. Walve. C. 243-249.” Proc. May 1992. Holmes. pp. Nortli- Holland. Power Systems. “Voltage Stability Eval. 54-64. August 1986. pp. “A Tax- C. Perturbation Techniques to Power System Transient Stability P. Caiiizares and F. 8. “Dynamic Siinnlationof Voltage Col. February 1992. Dr. 1529-1542. ‘‘Techniquesfor Detecting Proximity to Voltage Collapse in AClDC Systems. 6. DeMarco. “Application of Singular Systems. Power Systems. Pai. paper WM 184-2-PWRS. Systems. F. 9-52. pp. Circuits and Syst. 181. “Fast Calculation of a Voltage Stability Index. Stubbe. a t the Electrical and Computer J. No. I. Varaiya. T. L. Verlag. “Point of Collapse and Continuation Methods for Large AC/DC Systems. C. 7. P. 10. L. J. and Bifurcations of Vector Fields. No. pp. in Spanish. A. P. puter Studies. Gao. 111 SEPOPE. Tomac. Power Systems. A. OH. 762-766. Vol. [20] R. February 1993. “Q-V Curve Interpretations of Energy Measures for Voltage Security. 1 . Lee. Power Systems. M. DeMarco. “A New Method of Constructing Lyapnnov I. 3. Aspects of Voltage Problems. 3. C. New York. Vol. pp. Pal. 9. Power Systems.” International J o u r n a l of Electric Power U Energy C. August 1991. “Saddle-Node Bifurcations in Power Sys- tems. pp.518 [7] C. C. 1986.“ IEEE Trans. 30. Vu. T. Static Analysis. 14. No. Liu.” in [2]. ISCAS. “A Vector Energy Func- tion Approach for Security Analysis of AC/DC Systems. December 1991. Int. C.J. 1989. 1988.“Voltage CollapsePrecipitatedby thehn- Functions for Power Systems. 37. Vol. F. No. L.” IEEE Trans.” IEEE Trans. M’91) was bornin Mexico. J. [21] P. L. 2 . 7. August PhD (1991) degrees in Electrical Engineeringfrom the University of 1993. pp. February 1992. Febru. S. 207-216. October 1992. . L. Belo Horizonte. EPRI Report EL-5003. Pai. S . Schattler.” 1960. Anderson. I.-l. 3. A. in Electric Power Systems. T. paper IP-18. Vol. uation Using Modal Analysis. Dobson. Alvarado. 8. “Nonlinear Oscillations in Power and Syst. CaFFzares. pp. 3. No. 1001-1011. Sauer. Vol. H. and the MS (1988) and lapses. No. in “National Interconnected System Load Flow: Peak Demand. A. pp. where he was a Professor for 9 years. A. L. Sastry. November 1992. Syst.” IEEE Trans.-CIGRE.” in [2]. C. Lesieutre. Engineering Department. Overbye anf C. B. A. K.. Alvarado. K. August 1993. Zaborszky. 9. 133-141. 424-431. Circuits and Syst. Abed and P. J. Bergen. 2. “Voltage Stability Coiiditiois Considering Load Power Systems. Ecuador. K. pp.F. K. August 1992. Long. froin the Escuela PolitCcnica Nacional (EPN). Kundur. onomy of the Dynamics of the Large Power Systems with Em- pp. “Point of Collapse Methods Applied to AC/DC Space to Avoid Saddle Node Bifurcation and Voltage Collpase Power Systems. 1196-1202. Overbye. Springer. 7. 651-663. Vol. 3. C. pp. 7. Liu and K. C.” IEEE/PES Winter Meeting. Ajjarapu and B. pp. T. M. No. P. Kluwer Academic Publishers. 15. 1 . January 1993. M. Dobson. Load Models for Voltage Collapse Analysis. Morrison.” IEEE Trans. ISCAS. C d e a r e s . D.” Proc. phasis on Its Voltage Stability. M. C. Vol. ary 1992. Limits are Encountered. May 1984. “Dynamic vs.” IEEE Load Modeling f o r Power Flow and Transient Stability Com- Trans. Columbus. 39.” IEEE TTans. L. 1. D. Smed.” IEEE Transactions Power Systems.” IEEE Trnas. 1984. Jimma. July 1993. Power Systems. F. Claudio A. 8. “Computing and Optimum Direction in Control W. Pal. pp. Seydel. F. paper 38-18. “Modelling of Power System Components at Severe cation to Noillinear Dynanical Phenomena in an Electrical Disturbances. K. he received the Electrical Engineer diploma INECEL. pp. Large High Voltage Efechic Power System. 39. September 1992. W. Vol. May 1992. 1988. 9. G . L. 240-243. Caiizares. pp. DeMarco. Dobson and L.” Proc. 1-8. 1159-1171. Vol. “A Study of Dynamic B. Vol. Power Systems. L. No. Power Systems. 7. March 1992.” Proc. Vol. Vol. 894-904. Nonlinear Oscillations.No. 6.” in [2]. C. 673-683. September 1983. 905- mediate Change in Stability When Generator Reactive Power 908. 4.” Pror. “Voltage Stability Anal- ysis Using Static and Dynamic Approadies. B. Vu. Alvarado. “Voltage Security Enhance- ment Using Energy Based Sensitivities. DeMarco.-C. pp.January 1987. Lu. No. A. CGzares. DeMarco. lined Nonlinear Dynanucal Systems. “The Effects of Sinall Noise on Implicitly De- Vol. Deuse and M. pp. “Bifurcation Theory and its Appli. 222-229. C. Automatic Control. V. DeMarco and A.” IEEE Trans. Quito.” IEEE Trans. and Ius research activities are mostly con- namical Systems. Guckenheinier and P. centrated in the analysis of stability issues in ac/dc systems. H. In April 1984. DeMarco and C. Conf. Brazil.” in [2]. J. 37-43. A. G. G.” IEEE Trans. Long. Venkatasubramanian.” IEEE Trans. Wisconsin-Madison. Elsevier Science Publishers. W.-I. Discussions to [7. 597-601. No. L. K. Kundur. Vol. “Types of Voltage Collapse. pp. Energy Function Analysis f o r Power System Sta- bility. 7.” IEEE Trans. K. Vol. pp. R. From Equilibrium to Chaos-Practical Bifurcation and Stability Analysis. CGzares is currently an Assistant Pro- fessor at the University of Waterloo. Quito-Ecuador. Dobson. 1616-1620. Dobson. “Observations on the Geometry of Saddle Node Bi- furcations and Voltage Collapse in Electrical Power Systems. [SI V. Caaizares (53‘87. Characteristics. pp. JIEE. 1. Laf. Hill. A.

where is the internal generator reactance. it is of the generator must be. INDIA): The discussers would like to commend the author for his insightful voltage angle maintained at zero? presentation on bifurcation. The be no shaft dynamics. the real power which is drawn by this threat of voltage collapse due to lack of generation and transmission facility coupled infinite bus normally cannot be constant during with environmental and economic pressure. angle become a dynamic state. With this that many a utility is facing the possible constraint. It synchronous generator. If the classical dynamic model 1 of the paper shows the generator terminal voltage is used with constant input power then there will to have magnitude 1. What assumption has been made is modeled by a reactance X which is used in the about the turbine/governor that has led to this dynamic model.1 describes a load model as an infinite power system networks. and how is the generator terminal voltage angle related to the generator internal angle which is normally related to the speed integral? S. Figure the line is lossless. Since the load to be the load. Application of bifurcation theory in power system has added bus which normally means constant voltage a new dimension to the study of power system magnitude and angle (this would be a valid way to stability issues. a few clarifications/queries as follows are in order and the discussers constant by proper turbinelgovernor assumptions.0. SINGH and S. the generator "internal" equation (4).N.C. would appreciate author's clarifications on but the infinite bus power will change with time them. S. IL): This paper presents Section 2. it is during any transient. Urbana. maintain constant terminal voltage. The angle Was this error present in the results shown? associated with the internal voltage is normally the dynamic state which is related to the integral of the Section 2. 1. 1994. The dynamic model generator shaft speed.3 voltage magnitude must vary instantaneously to entry of equation (5) (i. The paper does not mention the model? If this voltage dependent power is meant internal reactance of the machine. the dynamic generator dynamic model in this way.3 is titled RL loads. machine. The shaft torque can be made However. Thus it does not seem found that the active power load 'P' being appropriate to say that "the first load model is an drawn by the infinite bus has been taken same as the mechanical power input to the infinite bus.0 is consistent with the system explain this load model further by giving a dynamic model description of section 2 which complete mathematical description of the load in states that the AVR maintains this constant at 1. SRIVASTAVA. Sauer (University of Illinois at Urbana- Champaign. terms of reallreactive power and voltage? From In order to achieve this. dynamic analysis. The author's paper assumes greater significance in view of the fact maintain the load voltage angle at zero). Could the author please magnitude of 1. cos rather than sin). then the output understand and appreciate the results. constant real power since necessary to clearly understand the model. 519 Discussion after any disturbance.N. Under what will do so at an initial equilibrium point. and how is the load Kanpur. SRIVASTAVA. JOSH1 (Department of Electrical That is. Specifically. but not assumptions this will be valid? . K. Would the author please clarify this power P? Peter W. information on the relationship between figure 1 and the dynamic model of equation (I)? Manuscript received August 4. angle delta must be some type of difference angle which makes the load voltage angle zero at all A clarification of these modeling issues would times. The section 2 description of equation (7) shows a voltage dependent shaft and figure 1 indicate that the transmission system input torque. which is drawing active power P". then I do not see how it enters the voltage angle is shown to be zero.K.0 with angle delta. how did the generator terminal voltage Engineering.2 introduces a PQ dynamic load model results for a system which is shown as a single stated to be "constant active and reactive power". Referring to eq (1) of the paper. single line and single load. In order to If the load is constant real power. Indian Institute of Technology.e. there appears to be an error in the 3. voltage stability and effect of load modeling in Section 2. Would the author please provide additional greatly enhance the papers contribution.

All instances of Hopf generator terminal bus and the infinite bus is 90". In the system chosen for bifurcation analysis (Fig. representation be included in order for the results to be valid. A careful reading of the paper In the section on applications to the SNI Ecuadorian system. pp. Manuscript received September 2. since there is only one generator and no reference bus (e. l). In section 3.2. it will be of time constant as small as lms (used in the author's example). Hopf bifurcation will only be observed when an example illustrates a purely angle stability problem. In most of the literature on voltage stability. in [A]. unless rendered dynamic by some means. K. e. it is as pointed out earlier.600-611. on of transformation is applied. thus.. Modelling form may also be derived from field measurements. as in the Power Systems. in ref [ A ] . the loads) were considered. Hopf bifurcations are authors have stated. transcritical bifurcation can also infinite bus). As regards the two examples using RL loads. just the opposite of what the transmission system. power flow Jacobian was identified as the bifurcation point. has been shown that even with generators' being modeled by their swing equations and stability limit occurs at the point of maximum power transfer. cannot cause voltage instability.140. clarified in [B]. NJ): In the introduction it is LTCs. Hopf bifurcation. the author has made a Our concern here is. pp. and be encountered. Discussion of "An Investigation of Voltage may be encountered in general simulation studies unless some kind Instability Problem." IEE Proc.This is well bifurcation reported in the literature on voltage stability can be established. However. For small signal analyses.K. Vo1. 1994. beyond the stability limit in a real power system. an which give the system equations a special symmetry. Pal (Consultant. however. the author further made a justification for not considering angle and other generator internal statement that for PQ load models and dynamics in voltage stability studies lies in the overall slow typically small resistive losses in the dynamics of the composite load. the issues. due to the invalid system model (constant traced to unrealistic load modelling.4. studying the generator-infinite bus system . P. the points the author set out to make in this In the second example using the P.e. However.7. we should point out that static loads. on Decision and Control. We hope this should save the generator terminal voltage). In conclusion.. the mathematical steps shown in equations @)-(lo)are irrelevant.. in somewhat [AI C Rajagopalan.g.e. e. However. M. The 3. the analysis readily extends to larger systems. even Tampa. the stability conditions probability of large number of complex would be completely different if the load had a different eigenvlues and. Pal. a static constant power load model is bifurcation (stability limit) occurs when the angle between the used in a dynamic system formulation. This leads to an increased loads with small time constants. We would like to point out that. this model would be as good as any other first- order load model. Pai. unlikely to occur. The transmission system having small resistance bifurcation point corresponds to the singular point of the power the jacobian matrix becomes highly flow Jacobian. we paper were already made in a more rigorous and straightforward comment that the load model given in equation (4) is realistic in manner.W. The myth was Manuscript received August 22. IEEE Trans.g. In section 3. No. While a transcritical therefore the swing equation is redundant. Although a simple system was used for the purpose of illustration. Constant pertaining to a few load buses are equal or SlfP/~lf= P T i p /TIq ? generator terminal voltage is admissible only when the load dynamics is slow compared with AVR response. 'Analysis of voltage control system voltage stability limit has been (arbitrarily) associated with the exhibiting Hopf bifurcation'. characteristic. of singularity of the power flow Jacobian. for the first time.279-286. with a load bifurcation is not generic. in Reference 36 of the paper. more detail. present example.A. i. and therefore can have discontinuities.g. A load model in this M. in the context of voltage load. "Voltage Stability: Analysis Needs. This paper only adds to the confusion that exists in the the sense that any first order dynamic load model derived from literature. physical laws can be reduced to this form. The third equation in (7) stated that the main purpose of the paper is to clarify the therefore has no significance. the analyses presented in the paper. suggests just the opposite. The singular point of the 28th IEEE Conf. (It is not clear why the infinite bus has been classified as paper. May 1992. we note that conditions like remote voltage control. The unrealistic load model. July 1993. particularly voltage stability. Computational difficulties M. Yorino. no known load behaves like an infinite bus. Sauer and M. when no dynamics of the system components responsible for voltage instability (i. The model Requirement and Modelling Adequacy. it is interest if author could elucidate it with imperative that generator flux dynamics along with AVR the help of an example. uses time derivative of a variable which is not a true state variable." by N. Vo1. 1994. it For the system and load models used in the second example.Q dynamic load model. the system model is not valid for asymmetric. We note that this is not an admissible model for for the voltage dependent parts of the load. pp. We support this by reviewing some of since the generators were modelled by constant terminal voltage. Pal. Dec 19989. No. 130-136. the author's interpretation of the remark that under certain operating effect of the magnitude of the time constant. Also. Proc. and its inclusion only helps confuse relationship between bifurcations and power system stability. Edison.K. FL. ~~ . et al.) This particular stability. First.the first example of the The author mentions Hopf bifurcation in several places in the paper. The main 4.520 2. The author would have obtained identical stated that the AVR is modelled by keeping the generator terminal results from the power flow Jacobian after modifying it to account voltage constant. and again. As noted above. How will the load model of eq (11) authors may wish to verify the truth of the above statement by affect the study if all the time constants repeating the analysis using a detailed generator model. the question of angle instability does not arise. the stability limit thus obtained is well author future work in the pursuit of something nonexistent. USA. These have been shown in a rigorous way. C.

makes some additional assumptions regarding generator loading ing that P is changing gradually. the steady state power demand of the infinite bus. TI. since very crude network demand in steady-state. and talk about a constant internal generator voltage.. one should not be con. be associated to the following matrix defined on the paper: Regarding the meaning of P in the generator-infinite bus model of section 2.g. Sauer mentions in [ :. Abed and Varaiyashow that Hopf bifurcations Pr. In [lo]. cations before. and TI. and due to limitations on is the angle that defines the power flowing through the line.e. Notice that for one-parameter sality conditions at bifurcations of power system models. equation representation of the mechanical dynamics of the syn. equilibria until the bifurcation point is encountered. = P + D L ~ L for these system models occur only when transfer resistances are QL = k P + T V . i.Sauer: Prof. particularly Profs. Sauer for their interest and constructive discus. i. the load stants TI. the load model in this case strated in [lo]. Since the analyses presented and rating. . ity. the generator outbut in this lossless system. ::. would like to thank the discussers. which is With regard to transcritical bifurcations in power systems. which yond the scope of the present paper. the eigenvalues reach the imagi. and view the paper the author has published a paper illustrating this problem in examples and results in that context. this name is misleading and it is not descriptive of the load tions and avoid associating these to the degree of asymmetry of model used in the example of section 2. The idea in power systems modeled using swing equations is highly asym. The intention here was not to duel on modeling issues. changes to this torque implicit assumption in the paper is obviously that of ignoring produce “small” oscillations of the generator angle 6 according the internal generator impedance. comments on basic modeling issues are very valid. and a clarifi- nary axis of the complex plane. Srivastava one “generically” expects to encounter saddle-node bifurcations et al and Prof. system models are used. and Prof. Prof. for the The load model represented by equation (11) does not present load models used in the examples and without loss of general- any difficulties during simulation for any values of the time con.e. one is mostly worried about tracing demonstrated in references [26] and [C]. In On the issue of modeling the AVR and the internal genera- other words. since gradual changes of P shift system cerned during simulations with particular values of these con. but in such a coarse system model this certainly does in the paper concentrate only in the system equilibria. where the generated power is the same as the as teztbook examples of bifurcation analysis. 6 corresponds to the angle the cause of this particular bifurcation. rather than Hopfs. is reached. and other articles that such lossless systems is assumed to be drawing active and reactive powers cannot have Hopf bifurcations.e. Srivastava et al: It is important to start this discussion by repeating some of The equations used in the paper to crudely model the gener. e. assum. The author only wanted to demon- sume that a system parameter changes very slowly. as local bifurcation analysis.1. ~ / T ! . infinite bus demand V / X sin(6) would certainly change accord- tation of the load model that is useful when studding transver..^. since these models correspond to quasi-steady-state phasor rep- cussers to reference [B] for a detail analysis of this problem.. it has been formally demon. ~Since .. difference between the transmission line terminal voltages. For the reader uncomfortable is the standard way of locating and studying local bifurcation with this assumption. and that strate the effect of different load types in the same simple gener- the system responds to these changes by eventually reaching a ator model.?. the author would kindly refer the dis. the three rather simple systems books. Since this topic is be. . P corresponds to the generator mechanical power in the lossless system under study. one has to be careful when studying Hopf bifurca. stants when using robust numerical integration techniques. study only the characteristics of the equilibrium points. (1) to dynamically change. that the load voltage is the system reference. the generator shaft torque is the changing param. until the system equilibria become non-hyperbolic obscured by model complexities. tor impedances in the second order generator model chosen. demand P. In the textbook examples shown cation of certain assumptions made by the author ought to be in the paper. the eter at synchronous speeds in this case. the author must recognize that Finally. Cafiizares (University of Waterloo): The author included. The T ~ .. Sauer’s the Jacobian matrix only. since that has been discussed at length in many other publi- chronous machine. ing to the dynamic changes in 6. ] his discussion. Nevertheless. the disclaimers made throughout the paper regarding the mod- ator is a slight modification of the typical second order or swing els used.e. 1. C. as described in standard power systems text. The discussers’ concern seems to voltage angle is 0’. TI. regardless of the structure of the Jacobian. Furthermore. and the length of this closure. certainly the classical swing equation model. allow the voltage V in equation equations (2) and (5) in the paper. the system equilibria. Notice that for all three an ac/dc system where remote voltage control is shown to be system models represented by Fig. 521 Claudio A. Hence. W.. resentations of the original equations. to each one of the discussers’ remarks and questions follow: Prof. even for zero transfer resistance as shown in Jacobian used in the previous section. S. this parameter is aswmed to be the power system addressed in order to make more sense of the examples shown. which not represent a significant issue. This model assumes that the mechanical torque illustrated in section 2 of the paper are meant to be taken only remains constant. one can assume. this is used only in a mathematical represen. since these resistances are typically small. Sauer’s at the bifurcation point. one can go back to the idea of lumping problems. For the PQ dynamic model. this case is to add one degree of complexity to the load model metric. P. However. Notice that although the Jacobian for comments offer an opportunity to clarify this point. [A].2. i. as Prof. Comments latter. not a realistic one unless one to equations (1)until and equilibrium point 6. so that the points brought up in the paper are not steady state. i. [D]. although technically one cannot rule out the sion which should certainly help enhance the paper. Hence. it represents. Local bifurcation analyses as. one needs not to worry about the oscillations of 6 and the generator reactance with the transmission line impedance. This power is the system parameter which would be singular for the given condition T I ~ ~ = / T ~ ~ ~ changed for the bifurcation studies illustrated in the paper.

These corrections will the discusser is not familiar with the vast literature in bifur- help explain the voltage dependency of the shaft torque in (7). D. the results shown in 1 1 the paper for the very real SNI Ecuadorian system have been V(6. August 1994. should be of no concern to the power systems community. where it is shown that the phenomenon These typos are not part of the simulations and results shown has very little to do with load modeling. He suggests that bifurca- and differential-algebraic systems. The active powers are only constant in steady state. because it seems that the to allow for direct comparisons between the different sample ”myth” which the discusser says to have clarified in one of his systems. so that the torque remains constant for system variations other than those associated to the [A] A. Finally.522 with large damping DL.Vz) . do not know about bi.w. the active and re. Fair- exists in the literature. ECC Inc. with any dynamical systems. 1988. “Transcritical and Hopf characteristics of the changing system equilibria for bifurcation bifurcations in ac/dc systems. missing the point that these are very simple systems resulting simple system presents certain futures that make it meant to be used in a teaching environment. the the paper. output is adjusted to the system demand changes. he would realize that his statement is incorrect. as he sates on and the second equation in (7) should read: his discussion. [16]. 1986. 1994 certainly convinced that the system stability problems described . pp. once more the idea is to make an additional change to the community can be enlighten on the unique views that he has in system equations. which seems to be his style as one can see from previous discussions to many [D] H. e.V. and therefore these sumed to remain constant at 0”. such as Lyapunov functions. ”Conditions for saddle-node bifurcations in ac/dc power systems. ir the paper. A. tems described by a second-order vector differential equa- From his remarks one gathers that either he knows something tion. Although this paper was not written to demonstrate the existence of bifur- cations in “real” power systems (there are many other articles the Lyapunov function defined in equation (6) should be: which concentrate on this particular issue). furcation related problems in nonlinear dynamical systems. June many researchers.3. Jacobian equation (5) should read: but many other engineering and scientific areas.” IEEE Trans. e Mr. of Electric Power d Energy Systems.h ( V ) ] (6) The discusser mentions that Hopf bifurcations “will only be observed when an unrealistic load model” is used. He would be surprise on how generalized this phenomenon is. Reading his comments. without significantly affecting the final results this particular area of system stability. [C] C. K. Proc.. not only in power systems. Wu. bifurcations.” in [E]. Cafizares. function of the steady state voltage. Chiang and F.. this author would like to encourage the discusser to Now. coming back to the assumptions in the case of the “RL” submit his results for publication. phenomena. values for the models time constants. -. 6. one concludes that the discusser is Manuscript received October 24. R. “Stability of nonlinear sys- papers in bifurcation theory applied to power systems. Obviously equations (5). (6). if the discusser would read some of the ref- . since the particular modeling Before commenting on the RL load example in section 2. Circuits and Systems. Thus.V C O S ( ~ ) ] certainly corroborated in practice. July 1994. Although this discusser writes several paragraphs criticizing the models used in load can be hardly justified and labeled in practical terms. H.(Vi . Prentice-Hall. as it was done in all previous cases. Well. In equation (7b) above. Evidently.cos(ao)] (5) no reason for the power system to be the exception. since one is mostly interested in the [B] C. in this paper will not occur in “real” systems. I DW erences cited in this paper.-sln(6) v . including this author. F. and [D]. VA.g. This discusser comments are surprisingly blunt. and duplicated very appealing to demonstrate certain issues in the studies of with little effort by a y o n e interested in the subject. vol. editor. Power Systems Analysis. Bergen. tion analysis is a waste of time.6) + -[VOCOS(&). and there is 1 0 1 0 [ J = -(Vo/MX)cos(60) -DIM -(VO/rX)sin(60) 0 -(I/MX)sin(60) -(1/7X)[ZVo . w = IV~~G 1 . where there is no need for X 1 any modeling assumptions.” or he is just talking about different fax. studies. the shaft torque is a quadratic many discussions is still in several researchers minds. assumptions used in this paper and others before seem to be the the author would like to correct some typographical errors in only reason for the appearance of the phenomena. Hranilovic.” accepted for publication in the Int. and (7) of the paper. [B]. X . 703-711. New slow varying parameter G. the load voltage angle 6~ is as.P) = ZMwZ + P(& . Cafiizares and S. so that the rest of the research load.. The author would like to invite this discusser to read papers [lo]. cation theory of nonlinear systems. as [E] L.e. Bulk Power System Voltage Phe- one gathers from his final comment about the ”confusion that nomena Ill-Voltage Stability and Security. M.kP[ln(Vo). this way the generator steady state Jersey. 35. no.i. Pal: J.. A. Fink.