Energy xxx (2013) 1e8

Contents lists available at SciVerse ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Modeling, nonlinear dynamical analysis of a novel power system with random

wind power and it’s control
Diyi Chen a, b, *, Si Liu c, Xiaoyi Ma a
a
Department of Electrical Engineering, Northwest A&F University, Shaanxi Yangling 712100, China
b
School of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, AZ 85287, USA
c
College of Electrical Engineering, Shandong University, Shandong Jinan 250061, China

a r t i c l e i n f o a b s t r a c t

Article history: The stability problem of power networks becomes increasingly important since the rapid development
Received 17 September 2012 and broad application of wind energy technology. We continue to lack a well-established mathematical
Received in revised form model to describe and characterize power systems associated with wind power in essence of random
1 February 2013
properties. Here, we present a novel model to solve this significant problem by incorporating the
Accepted 3 February 2013
Available online xxx
nonlinear dynamical theory and the fluctuation nature of wind energy. The model produces some
interesting dynamical phenomena, e.g., routes to chaos. To eliminate the chaotic behavior which is
disadvantageous to the stability and normal functioning of the whole system, we offer a fuzzy control
Keywords:
Power system
approach to drive the system with uncertain parameters from chaotic states to steady states. The control
Stability method is validated by both numerical simulations and theoretical analysis. Our theory and control
Random energy scheme can be expected to be potentially applicable in a variety of power systems with wind sources.
Nonlinear dynamics
2013 Elsevier Ltd. All rights reserved.
Fuzzy control

1. Introduction Similarly, Kumar and Gokulakrishnan [14] studied the impact of

Wind power sources have been increasingly brought into the
conventional power system recently [1e6]. At present, the gener-
ation capability of wind power sources relies on the conditions of
natural environment, such as wind speed. Their grid-connection
introduces significant stochastic fluctuation of generation which
could be potentially spatial dependent. The stochastically variable
wind sources on power system stability have significant impact on
the dynamics of power systems, operational characteristics and
stability of the whole power system, which has been given much
attention [7e10]. Some effort has been dedicated to the small signal
stability of power system with wind turbine generators. For
example, Gautam et al. [11] developed an approach to analyze the
impact of increased penetration of DFIG-based wind turbines on
transient and small signal stability of a large power system. Bu et al.
[12] proposed a method of probabilistic analysis to investigate the
impact of stochastic uncertainty of grid-connected wind generation
on power system small-signal stability. Other contributions to dy-
namic stability are major in frequency-domain approach [13].
* Corresponding author. Department of Electrical Engineering, Northwest A&F
University, Shaanxi Yangling 712100, China. Tel.: þ1 480 577 4579.
E-mail address: diyi.chen@asu.edu (D. Chen).
FACTS controllers on the stability of power systems connected with
wind energy conversion system. In particular, some papers mainly
focus on voltage stability [15,16], angle stability [17,18], and thermal
stability [19]. In addition, some publications are about transient
stability [20,21], steady-state stability [22]. Despite existent con-
tributions to the dynamic behaviors of wind turbines [23e26],
there are few literature addressing the issue of the stability of the
power system with random wind power based on nonlinear dy-
namics theory.
The control problem of nonlinear systems with uncertain pa-
rameters has been a well-established area [27e29]. Aghababa M.
and Aghababa H. [30] investigated the problem of finite-time chaos
synchronization between two different uncertain chaotic systems
with unknown parameters and input nonlinearities. A study of
sliding mode control of uncertain dynamical systems with time
delay was presented by Song and Sun [31]. Sakthivel et al. [32]
offered a robust H-infinity control design approach for a class of
uncertain discrete-time stochastic neural networks with time-
varying delays. In particular, due to the theoretical value and
practical significance, fuzzy control has been employed in con-
trolling nonlinear dynamics [33e35]. Yet, to the best of our
knowledge relatively little attention has been given to imple-
menting fuzzy control to power system associated with random
wind energy [36].

0360-5442/$ e see front matter
2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.energy.2013.02.013

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
2 D. Chen et al. / Energy xxx (2013) 1e8

Regarding the lacking of mathematical approach for power where mechanical power Pm, electromagnetic power Pe and
systems with wind sources in essence of random properties, we damping power PD are expressed as per unit values (Pm, Pe, PD).
model a power system consisting of a wind power generator and a Their base values are in kV A for the unit.
normal generator. The dynamic characters of the system are stud- Actually, damping power PD is approximated as
ied in detail by the aid of nonlinear dynamical theory. Some
interesting phenomena are observed, such as the routes to chaos. In D dd
PD ¼ (7)
order to eliminate chaos, an adaptive fuzzy control is offered to 2pf0 dt
drive the system with uncertain parameters to a steady state.
The rest of the article is organized as follows. A new mathe- where D is generator damping coefficients, which is a constant.
matical model of a typical power system with random wind energy Considering Pe is proportional to sind and time base value is
is presented in Sec. 2. Sec. 3 studies the nonlinear dynamical be- TB ¼ 1/2pf0, Eq. (6) can be simplified as
haviors in detail. An adaptive fuzzy control is proposed in Sec. 4.
Moreover, the numerical simulations are in agreement with theory d2 d 1 1 D
¼ Pm  Pe sindðtÞ  wðtÞ (8)
analysis. The conclusion is drawn in Sec. 5. dt 2 H H H
The electrical power system with random wind energy is
2. System model described as

The simple power system with random wind power is illus- d2 d 1 1 D 1

trated in Fig. 1 which can be viewed as a generalized case. ¼ Pm  Pe sindðtÞ  wðtÞ þ Ps cosbt (9)
dt 2 H H H H
Here, 1 is the equivalent generator of electrical power system, 2
is equivalent wind generator of electrical power system, 3 and 4 are where Ps and b are parameters of the random wind energy, they
equivalent main transformers, 5 is load, 6 are circuit breakers and 7 represent its amplitude and frequency, respectively. Here, one sets
is tie line of the power systems. d1 < Ps < d2.
Rotor equations for the wind turbine model is From Eqs. (1) and (9), the state equations of the electrical power
dw d2 d system with random wind energy could be obtained as follows
a ¼ ¼ ; (1) 8
dt dt 2
>
>
> d_ ðtÞ ¼ wðtÞ
>
<
where a is the angular acceleration of wind turbine rotor, w is 1 1 D 1
_
wðtÞ ¼ Pm  Pe sindðtÞ  wðtÞ þ Ps cosbt (10)
generator angular velocity and d is the angle. Thus, we can get >
> H H H H
>
>
:
J a ¼ Mm  Me  MD (2) yðtÞ ¼ dðtÞ  d0 ðtÞ

where J is the inertia moment of generators rotor including prime where Pm is mechanical power and Pe is electromagnetic power of
mover and generator rotor, Mm is mechanical torque which acts on the wind power system, d(t) and w(t) are system state variables, y(t)
the spindle of generator, Me is braking electromagnetism torque of is output variable of the system.
machines and MD is damping torque. For that system 1 in Fig. 1 is similar to system 2 in Fig. 1, Ps in
From Eqs. (1) and (2), one has system 1 in Fig. 1 represents the amplitude of the random wind
generator.
d2 d If one denotes a ¼ Pe/H, g ¼ D/H,r ¼ Pm/H, and F ¼ Ps/H, Eq. (10)
J ¼ Mm  Me  MD (3) can be rewritten as
dt 2

In industrial practices, inertia moment of generator rotor J can
d_ ðtÞ ¼ wðtÞ (11)
be represented by H customarily. Thus, Eq. (3) is rewritten as _
wðtÞ ¼ r  asindðtÞ  gwðtÞ þ Fcosbt
H d2 d
¼ Mm  Me  MD (4) where F is a random quantity, which is d1/H < F < d2/H.
2pf0 dt 2
3. Nonlinear dynamical analysis
where H and t are in seconds for the unit; fundamental frequency is
50 Hz f0 ¼ 50 Hz, d is in radian for the unit, Mm, Me and MD are
When the power system operates in rated conditions, a random
expressed as per unit values (Mm, Me, MD).
energy disturbance, such as the fluctuation of the random energy
Taking industrial practices into consideration, torque M can be
frequency or the random energy amplitude, will affect the stability
replaced by power P, which is given by
of the system. Subsequently, the system will enter a new operating
P state. Because of the nonlinearity of every part, there will be a
M ¼ : (5) complex transient process. This paper only focuses on the nonlinear
w
dynamical behaviors along with the fluctuation of the amplitude of
One has
the power system. This is because of the fact that the method of
d2 d analyzing nonlinear dynamical behaviors in different random en-
H ¼ Pm  Pe  PD ; (6) ergy interferences is approximately same. Therefore, here we only
dt 2
study the differential adjustment of Ps.

3.1. Bifurcation map

The Bifurcation map is used to analyze the dynamical charac-

teristics of the nonlinear system as the system parameter varies.
Fig. 1. Simple interconnected power system with random wind power system. Bifurcation is the main route to chaos from a stable state.

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
 D. Chen et al. / Energy xxx (2013) 1e8 3

Apparently, it is known that as the random parameter Ps changes all
the time, the running state of the system changes accordingly,
which demonstrates that the system is disturbed by the random
energy disturbances. To be concrete, we study the Bifurcation di-
agram by plotting the maxima of the coordinate dt with parameter
Ps varying as shown in Fig. 2. Here, one sets Pe ¼ 100, H ¼ 100, D ¼ 2,
Pm ¼ 20, and the initial state are chosen as [0, 0.3]. The bifurcation
control parameter is monitored in the range 27 < Ps < 29.
3.2. Phase diagrams and power spectrum
Remark 1. Assumption 1 is usually required in system theory.
Assumption 2 is a standard assumption in an adaptive control
literature.
Remark 2. Note that many practical systems can be expressed or
transformed in the form Eq. (12). Examples include the single force
arm manipulator and so on. It is worth noting that the positive
constant b guarantees the controllability of the system Eq. (12).
Define the tracking error vector as
h iT
_ /; eðn1Þ ˛Rn
e ¼ e; e; (13)

In accordance with the parameter settings in subsection 3.1, where e ¼ ymy ¼ ymx
various numerical computations of the phase portraits, timing di- And the reference signal vector ym is
agrams and their corresponding power spectra were obtained,
h iT
which further confirm the occurrence of routes to chaos depicted
ym ¼ ym ; y_ m ; /; yðn1Þ
previously from Fig. 3. For a periodic steady state, all spikes in the
power spectrum are harmonically related to the fundamental ðnÞ
By using eðnÞ ¼ ym  xðnÞ , one has
whereas a broadband noise-like power spectrum is the signature of
a chaotic behavior. ðnÞ
0 ¼ ym þ eðnÞ þ f ðxÞ þ dðtÞ þ bu (14)
From Figs. 2 and 3, we know that the nonlinear dynamical be-
haviors of the power system with random wind power are complex Based on the feedback linearization approach, when the func-
and the system goes through a series of operating states when the tion f(x) and b are known, d(t) ¼ 0 and the state x is available for
random parameter Ps increases. Although the chaos in the system is measurement, there exists some ideal controllers
a transient process, it is still disadvantage to the stability of the
1h ðnÞ
i
whole system. u* ¼  f ðxÞ þ ym þ K T e (15)
Due to the existence of chaotic dynamics, when the power b
system operates on the chaotic condition, the stability and security
where K ¼ (kn,/,k1)T ˛ Rn satisfies that all rotes for polynomial
of the whole power system may be damaged in the sense that chaos
s(n)þk1s(n1)þ/þkn distribute on the left semi-open complex
is sensitive to initial values and quasi-random. It is thus very
plane.
necessary to control the chaotic system to a steady state, in order to
Substituting (15) into (14), one yields
ensure the system stability.
eðnÞ þ k1 eðn1Þ þ / þ kn e ¼ 0
4. Adaptive fuzzy control
Thus, it can be obtained that e(t) / 0 with t / N, which means
4.1. Problem formulation that system output y gradually converge to ideal output ym.
However, the ideal control law (15) is inapplicable, for that the
We consider the nth order nonlinear dynamical system as nonlinear function f(x) and b are not known, and the system states x
follows are not available. Therefore, to overcome such problems, a fuzzy
  system will be used to approximate directly the ideal control u* and
xðnÞ ¼ f x; x;
_ /; xðn1Þ þ dðtÞ þ bu; an observer will be designed to estimate the system states.
(12)
y ¼ x
4.2. Fuzzy logic system description and fuzzy adaptive controller
where u ˛ R is the control input, x ¼ ½x; x; _ /; xðn1Þ T ¼
T n design
½x1 ; x2 ; /; xn  ˛R is the vector of unmeasured states, y ˛ R is the
measured output, f is a smooth function, b is positive constant, d(t)
The basic configuration of a fuzzy logic system consists of a
is the external disturbance.
fuzzifier, some fuzzy IFeTHEN rules, a fuzzy inference engine and a
Assumption 1. There exists an unknown positive constant d*,
defuzzifier. The fuzzy inference engine uses the fuzzy IFeTHEN
such that jdðtÞj  d* .
rules to perform a mapping from an input vector xT ¼
Assumption 2. The reference signals ym ; y_ m ; /; yðn1Þ and their
[x1,x2,/,xn] ˛ Rn to an output uD ˛ R.
time derivatives y(n) are assumed to be continuous and bounded.
The ith fuzzy rule is written as

RðiÞ : IF x1 is Al11 and / and xn is Alnn ; THEN uD is Sl1 /ln :

where l1 ¼ 1,2,/,mi, i ¼ 1,2,/,n.Al11 ,Al22 ,/,Alnn are fuzzy variables and

Sl1 /ln is the fuzzy singleton for the output in the ith rule.
By using the singleton fuzzifier, product inference, and center-
average defuzzifier, the output of the fuzzy system can be expressed as

!
Pm1 Pmn Y
n

l1 ¼ 1
/ yl1 /ln
ln ¼ 1 u
mAli ðxi Þ
i
i¼1 T
u ¼ uD ðxjqÞ ¼ ! ¼ q xðxÞ
Pm1 Pmn Y
n

l1 ¼ 1
/ ln ¼ 1
mAli ðxi Þ
i
i¼1
Fig. 2. Bifurcation map for system Eq. (11) with Ps varying from 27 to 29. (16)

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
4 D. Chen et al. / Energy xxx (2013) 1e8

Fig. 3. Phase diagrams, time domain diagrams and power spectrum for different Ps (a) Ps ¼ 27.685; (b) Ps ¼ 27.686; (c) Ps ¼ 29.0.

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
 D. Chen et al. / Energy xxx (2013) 1e8 5

Fig. 3. (continued).

Y
n " #
 
where m li ðxi Þ is the membership function of the fuzzy set Alii , mi *  
Ai q ¼ arg min supuD ðxjqÞ  u* 
i¼1 Y
n
x˛Rn
½y1u ; y2u ; /; ylu1 /ln  is the adjustable
T mi
is the number of fuzzy rules, q ¼
T q˛Ri ¼ 1
parameter vector (composed of consequent parameters), and x ¼
Y
n X
m1 X
mn Yn
1 2 l /l Define the minimum approximation error as
½x ; x ; /x 1 2  with xl1 /l2 ðxÞ ¼ m li ðxi Þ / ð m li
Ai Ai
i¼1 l1 ¼ 1 ln ¼ 1 i ¼ 1
1
ðxi ÞÞ being the fuzzy basis function. In this paper, it is assumed that w ¼ uD ðxjqÞ  u* þ dðtÞ;
b
there exists always at least one active rule.
Remark 3. The fuzzy system (16) is used most frequently in which represents the lumped disturbance term generated by the
control applications and is able to approximate any nonlinear state estimation error, the fuzzy approximation error and external
smooth function ym on a compact operating space to any degree of disturbances.
accuracy following the universal approximation results. Now, the output tracking error dynamics of system (18) can be
Importantly, it is assumed that the structure of the fuzzy system given by
(i.e. the pertinent inputs, the number of membership functions for
each input and the number of rules) and the membership function 

 

 *  * 1
parameters are properly specified beforehand. The consequent pa- e_ ¼ Le þ b uD xq  uD ðxjqÞ  b uD xq  u* þ dðtÞ
b
rameters q are then determined by appropriate adaptation algorithms.
(19)
4.3. Adaptive law and stability analysis From Eq. (16), Eq. (19) can be obtained as

Substitute Eq. (15) and (16) into Eq. (12), one has

T
*
e_ ¼ Le þ b q  q xðxÞ  bw (20)
h i
ðnÞ T *
e ¼ K e þ b u  uD ðxjqÞ  dðtÞ (17) Theorem. Consider the system (12) under assumptions 1e2 and
fuzzy controller (16). According to Lyapunov stability theory, the
Eq. (17) can be rewritten as
fuzzy system is controllable as long as the adaptive law is designed
h i as q_ ¼ geT pn xðxÞ and appropriate uD ðxjqÞ with enough fuzzy rules
e_ ¼ Le þ b u*  uD ðxjqÞ  dðtÞ (18) is designed to make w small enough.

0
1 0 0 / 0 0
 0 Proof. Define Lyapunov function as

/ 1 T b *
T

0 0 1 0 0 0 0 V ¼ e Pe þ q q q*  q
where L ¼ « « « « 1 « « ; b ¼ « : 2 2 g
0 0 0 0 / 0 1 0
Kn Kn1 / K1 b where g is a positive constant, and positive definite matrix P
Define the optimal value of q as satisfies

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
6 D. Chen et al. / Energy xxx (2013) 1e8

LT P þ PL ¼ Q
Here, Q is an arbitrary n  n positive definite matrix, and L is
given by Eq. (18).
Define V1 ¼ 1/2eTPe, V2 ¼ b/2g(q*q)T(q*q), M ¼ b(q*q)Tx(x)
bw, and Eq. (20) can be rewritten as
e_ ¼ Le þ M
1 1 1 T T
1
V_ 1 ¼ e_ T Pe þ eT P e_ ¼ e L þ M T Pe þ eT PðLe þ MÞ
2 2 2 2
1 T T
1 T 1 T
¼ e L P þ PL e þ M Pe þ e PM
2 2 2
1 T 1 T T

¼  e Qe þ M Pe þ e PM
2 2
1
¼  eT Qe þ eT PM
2
One has
Fig. 4. Direct adaptive fuzzy control system.
1 h
T i
V_ 1 ¼  eT Qe þ eT Pb q  q xðxÞ  w
*
1
2
(21)
b *
T
0.9

V_ 2 ¼  q  q q_
Membership function degree
0.8

g 0.7

0.6
From Eq. (21), one has
0.5

1 h
T i b *
T 0.4
V_ ¼  eT Qe þ eT Pb q  q xðxÞ  w  q  q q_
*
(22) 0.3
2 g
0.2

Consider b ¼ [0,/,0,b]T, and pn is the last column of P, one 0.1

obtains 0
-3 -2 -1 0 1 2 3
x
T T
e Pb ¼ e pn b
Fig. 5. Membership functions of x.
Eq. (22) can be written as
We define the reference signal as xd(t) ¼ sin(pt). Moreover, the
1 b *
T h i membership functions are specified as follows
V_ ¼  eT Qe þ q  q geT pn xðxÞ  q_  eT pn bw
2 g 1
mN3 ðxÞ ¼ ; m ðxÞ
According to the theorem, q_ ¼ geT pn xðxÞ, one has 1 þ expð5ðx þ 2ÞÞ N2

1 ¼ exp  ðx þ 1:5Þ2 ; mN1 ðxÞ ¼ exp  ðx þ 0:5Þ2 ;

V_ ¼  eT Qe  eT pn bw
2

Consider the theorem above and Q > 0, as minimum approxi- mp1 ðxÞ ¼ exp  ðx  0:5Þ2 ; mp2 ðxÞ
mation error w is small enough, one has jeT pn bwjh1=2eT Q , thus

1
¼ exp  ðx  1:5Þ2 ; mp3 ðxÞ ¼
V_ < 0 1 þ expð5ðx  2ÞÞ
 
150 0
The structure diagram of direct adaptive fuzzy control system is We choose system ,parameters as Q ¼
0 150
shown in Fig. 4. T T
K ¼ [k2 k1] ¼ [10 1] and the adaptive parameter g ¼ 50. Similarly,
the initial state is [0, 0.3].
Numerical simulation results from Figs. 5 to 8 show the effec-
4.4. Numerical simulation
tiveness of the developed adaptive fuzzy control schemes for the

Supposing [x1 x2]T ¼ [d(t) w(t)]T and d0 ¼ 0, the fuzzy controller is 1.5

added into the power system (10). Thus, one has

8 1

>
> x_ 1 ¼ x2
>
<
Position tracking

0.5

1 1 D 1
> x_ 2 ¼ Pm  Pe sinðx1 Þ  x2 þ Ps cosbt þ bu (23) 0
>
> H H H H
:
y ¼ x1 -0.5

-1
where Pm ¼ 20, Pe ¼ 100, D ¼ 2, H ¼ 100, b ¼ 1, b ¼ 1, 27 < Ps < 29.
Substituting them into Eq. (23) yields -1.5
0 5 10 15 20 25 30 35 40 45 50
Time t

Fig. 6. The curve of position tracking y (blue line) and ym (red line). (For interpretation
€
x ¼ sinx  0:02x_ þ 0:2 þ Fcost þ u of the references to colour in this figure legend, the reader is referred to the web
y ¼ x version of this article.)

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
 D. Chen et al. / Energy xxx (2013) 1e8 7

3
Science & Technology of P. R. of China (2011BAD29B08) and the
2 “111” Project from the Ministry of Education of P. R. of China and
the State Administration of Foreign Experts Affairs of P. R. of China
Position tracking error

1
(B12007).
0

-1
References
-2
[1] Herzog AV, Lipman TE, Kammen DM. Renewable energy sources. available
-3
0 5 10 15 20 25 30 35 40 45 50
online: http://www-fa.upc.es/personals/fluids/oriol/ale/eolss.pdf; 29.08.2012.
Time t [2] Sun XJ, Huang DG, Wu GQ. The current state of offshore wind energy tech-
nology development. Energy 2012;41:298e312.
Fig. 7. The curve of position tracking error. [3] Li X, Hubacek K, Siu YL. Wind power in China e dream or reality? Energy
2012;37:51e60.
[4] Abbes M, Belhadj J. Wind resource estimation and wind park design in El-Kef
15
region, Tunisia. Energy 2012;40:348e57.
[5] Xu JZ, He DX, Zhao XL. Status and prospects of Chinese wind energy. Energy
10 2010;35:4439e44.
[6] Harijan K, Uqaili MA, Memon M, Mirza UK. Forecasting the diffusion of wind
5 power in Pakistan. Energy 2011;36:6068e73.
Control input

0 augmentation of power system including wind farms by using ECS. IEEE
-5
-10
-15
0 5 10 15 20 25 30
Time t
Fig. 8. Control input signal.
system (10). From Figs. 5 and 6, it can be seen that good tracking
performance is obtained. The roundedness of the variables e and
control input u are illustrated in Figs. 7 and 8, respectively.
5. Conclusions and discussion
The challenge of integrating renewable energy power into
electricity grids cannot be treated as a traditional issue. Random
characteristics of most of the renewable energy exist, especially for
wind power, should be intensively studied, although some
achievements have been made in the study of the ability of
different energy systems to integrate wind power and the planning
of wind power from the point of industrial application [37e39].
This issue can be addressed by relying on the stability analysis of a
complex nonlinear system with random disturbances based on
nonlinear dynamics theory.
In this regard, we selected a basically model, a classical power
system with a normal generator and a random renewable energy-
wind power system, as an object to study. Firstly, we rebuilt the
mathematical model of the electric power system with random
wind power from basic materials. Secondly, we introduced
nonlinear dynamics theory to analyze the stability of the complex
nonlinear system with random parts including bifurcation map,
phase diagrams and power spectrum. These numerical results
clearly show that d(t) and w(t) will reach infinity or chaotic state
with certain value of random wind power. In order to control the
system to a steady state, adaptive fuzzy control is offered to guar-
antee the stable and security of the whole system operation. Finally,
numerical simulations are carried out to examine the applicability
and feasibility of the proposed scheme, which are in good agree-
ment with the theoretical analysis.
Acknowledgments
This wok was supported by the scientific research foundation
of National Natural Science Foundation (51109180), the National
Science & Technology Supporting Plan from the Ministry of
[7] Muyeen SM, Takahashi R, Ali MH, Murata T, Tamura J. Transient stability
Transactions on Power Systems 2008;23:1179e87.
[8] Liu W, Lund H, Mathiesen BV. Large-scale integration of wind power into the
existing Chinese energy system. Energy 2011;36:4753e60.
[9] Hocine L, Mounira M. Effect of nonlinear energy on wind farm generators
connected to a distribution grid. Energy 2011;36:3255e61.
35 40 45 50 [10] Shi LB, Wang C, Yao LZ, Wang LM, Ni YX. Analysis of impact of grid-connected
wind power on small signal stability. Wind Energy 2011;4:517e37.
[11] Gautam D, Vittal V, Harbour T. Impact of increased penetration of DFIG-based
wind turbine generators on transient and small signal stability of power
systems. IEEE Transactions on Power Systems 2009;24:1426e34.
[12] Bu SQ, Du W, Wang HF, Chen Z, Xiao LY, Li HF. Probabilistic analysis of small-
signal stability of large-scale power systems as affected by penetration of
wind generation. IEEE Transactions on Power Systems 2012;27:762e70.
[13] Wang L, Wang KH. Dynamic stability analysis of a DFIG-based offshore wind
farm connected to a power grid through an HVDC link. IEEE Transactions on
Power Systems 2011;26:1501e10.
[14] Kumar NS, Gokulakrishnan J. Impact of FACTS controllers on the stability of
power systems connected with doubly fed induction generators. Electrical
Power and Energy Systems 2011;33:1172e84.
[15] Lahacani NA, Aouzellag D, Mendil B. Static compensator for maintaining
voltage stability of wind farm integration to a distribution network. Renew-
able Energy 2010;35:2476e82.
[16] Le HT, Santoso S, Nguyen TQ. Augmenting wind power penetration and grid
voltage stability limits using ESS: application design, sizing and a case study.
IEEE Transactions on Power Systems 2012;27:161e71.
[17] Hossain MJ, Hemanshu RP, Mahmud MA, Rodrigo AR. Investigation of the
impacts of large-scale wind power penetration on the angle and voltage
stability of power systems. IEEE Systems Journal 2012;6:76e84.
[18] Vittal E, O’Malley M, Keane A. Rotor angle stability with high penetrations of
wind generation. IEEE Transactions on Power Systems 2012;27:353e62.
[19] Perninge M, Lindskog F, Soder L. Importance sampling of injected powers for
electric power system security analysis. IEEE Transactions on Power Systems
2012;27:3e11.
[20] Faried SO, Billinton R, Aboreshaid S. Probabilistic evaluation of transient sta-
bility of a power system incorporating wind farms. IET Renewable Power
Generation 2010;4:299e307.
[21] Sarkhanloo MS, Yazdankhah AS, Kazemzadeh R. A new control strategy for
small wind farm with capabilities of supplying required reactive power and
transient stability improvement. Renewable Energy 2012;44:32e9.
[22] Kim EH, Kim JH, Kim SH, Choi J, Lee KY, Kim HC. Impact analysis of wind farms
in the Jeju island power system. IEEE Systems Journal 2012;6:134e9.
[23] Wang L, Li CN. Dynamic stability analysis of a tidal power generation system
connected to an onshore distribution system. IEEE Transactions on Energy
Conversion 2011;26:1191e7.
[24] Andres EL, Marcelo FF, Pedro EB, Jorge AS, Maria IV. Control strategy of a DVR
to improve stability in wind farms using squirrel-cage induction generators.
IEEE Transactions on Power Systems 2011;26:1609e17.
[25] Hiskens IA. Dynamics of type-3 wind turbine generator models. IEEE Trans-
actions on Power Systems 2012;27:465e74.
[26] Rahimi M, Parniani M. Dynamic behavior analysis of doubly-fed induction
generator wind turbines-the influence of rotor and speed controller param-
eters. Electrical Power and Energy Systems 2010;32:464e77.
[27] Yang CC. One input control for exponential synchronization in generalized
Lorenz systems with uncertain parameters. Journal of The Franklin Institute-
Engineering and Applied Mathematics 2012;349:349e65.
[28] Chen DY, Zhao WL, Ma XY. No-chattering sliding mode control chaos in
Hindmarsh-Rose neurons with uncertain parameters. Computers and Math-
ematics with Applications 2011;61:3161e71.
[29] Chen DY, Zhao WL, Ma XY, Wang J. Control for a class of 4D chaotic systems
with random-varying parameters and noise disturbance. Journal of Vibration
and Control 2012, http://dx.doi.org/10.1177/1077546312441657.

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013
8 D. Chen et al. / Energy xxx (2013) 1e8

[30] Aghababa M, Aghababa H. A general nonlinear adaptive control scheme for
finite-time synchronization of chaotic systems with uncertain parameters and
nonlinear inputs. Nonlinear Dynamics 2012;69:1903e14.
[31] Song B, Sun JQ. Sliding mode control of uncertain dynamical systems with
time delay using the continuous time approximation method. Journal of Vi-
bration and Control 2012;18:1254e60.
[32] Sakthivel R, Mathiyalagan K, Anthoni SM. Robust H-infinity control for un-
certain discrete-time stochastic neural networks with time-varying delays.
IET Control Theory and Applications 2012;6:1220e8.
[33] Liu YJ, Tong SC, Li TS. Observer-based adaptive fuzzy tracking control for a class of
uncertain nonlinear MIMO systems. Fuzzy Sets and Systems 2011;164:25e44.
[34] Boulkroune A, M’Saad M. A fuzzy adaptive variable-structure control scheme
for uncertain chaotic MIMO systems with sector nonlinearities and dead-
zones. Expert Systems with Applications 2011;38:14744e50.
[35] Behn C, Loepelmann P. Adaptive versus fuzzy control of uncertain mechanical
systems. International Journal of Applied Mechanics 2012;4:1250019.
[36] Sun JQ, Fujimoto K, Dou CX, Zhong JQ, Yuan SW. H-infinity fuzzy tracking
control for multi-machine interconnected power system with model un-
certainties. International Journal of Innovative Computing Information and
Control 2006;2:61e8.
[37] Lund H. Large-scale integration of wind power into different energy systems.
Energy 2005;30(13):2402e12.
[38] Pillai JR, Heussen K, Ostergaard PA. Comparative analysis of hourly and dy-
namic power balancing models for validating future energy scenarios. Energy
2011;36(5):3233e43.
[39] Lund H, Andersen AN, Ostergaard PA, Mathiesen BV, Connolly D. From elec-
tricity smart grids to smart energy systems-A market operation based
approach and understanding. Energy 2012;42(1):96e102.

Please cite this article in press as: Chen D, et al., Modeling, nonlinear dynamical analysis of a novel power system with random wind power and
it’s control, Energy (2013), http://dx.doi.org/10.1016/j.energy.2013.02.013