Load Frequency Control Based on Equivalent-input-disturbance Method for Isolated Small Hybrid Power System
Chunsheng Wang1, Jiamin Li1, Yukun Hu2, Xin Zhang3, Liz Varga2
1School of Automation, Central South University, Changsha, 410083, China
2Department of Civil, Environmental & Geomatic Engineering, University College London, London W C1E 6BT, UK
3School of Water, Energy and Environment, Cranfield University, Cranfield MK43 0AL, UK
Frequency stability is critical to power quality. In order to ensure frequency stability for stable operation, many attempts to assure frequency control of the isolated hybrid power system have been made. Various advanced intelligent methods have been introduced recently, such as sliding mode control [4], artificial neural network [5], and fuzzy logic [6]. However, these methods are complicated in structure and large in calculation. The traditional PI and PID control [7] methods are still the most widely employed methods in practical system due to their practical structure and low cost. Yet, PI method can’t achieve satisfactory control effect in the face of complex system uncertainties and external disturbances. For isolated power system connected to RE, improving the disturbance-rejection ability is a key point since varied RE is the disturbance during rated operation of the system to some extent.
In fact, adding new equipment to the isolated power system can also regulate frequency. Such equipment might include energy storage systems [8], plug-in hybrid electric vehicles (EVs) [9] and electrolyzers [10]. With the development of EVs and the support of government policies, it has become a more flexible choice than batteries which are costly. The charging and discharging power control of EVs batteries can reduce the power imbalance and improve the frequency control as batteries do. Access of EVs can also meet people's EVs charging needs while adjusting the frequency, but at the same time, the scheduling and availability of the EVs should be considered.
Therefore, to improve the disturbance-rejection of isolated power system, equivalent-input-disturbance (EID) method has been applied to frequency control in this paper. EID, simple in structure, has superior disturbance-rejection performance in addressing RE variations. To reduce the impact of RE access, power balance can be achieved by mainly controlling the output of diesel generator. In addition, EVs with fast response, are also installed in this system to help frequency regulation. By regarding the load variation
Abstract
In any isolated small power system employing a mix of renewable energy (RE) technology and legacy technology, frequency deviation is inevitable due to RE fluctuations and load demand variations. Therefore, an appropriate frequency control scheme is essential. This paper presents an Equivalent-Input-Disturbance (EID) control for load frequency control of isolated small hybrid renewable energy power system. EID method is applied to regulate the output power of the diesel generator and electric vehicle (EV) to maintain the power balance and ultimately maintain the frequency within normal range.
Simulation studies show that this strategy has higher flexibility and control performance than the traditional PI method.
Keywords: load frequency control, isolated small hybrid power system, equivalent-input-disturbance, renewable energy, electric vehicle.
Nomenclature
Abbreviation
EID Equivalent-Input-Disturbance LFC Load Frequency Control
WP Wind Power
SP Solar Power
EV Electric Vehicle
RE Renewable Energy
1. Introduction
Renewable energy (RE), such as wind power (WP) and solar power (SP) are forward-looking options [1] for distributed energy resource utilization in hybrid energy power systems [2]. However, WP and SP are intermittent as they are highly dependent on environmental conditions such as weather and seasons [3]. Thus, frequency variations will be caused when RE connect to isolated small power system.
and the RE intermittency as disturbances during the rated operation of the system, EID method can compensate the disturbances to maintain the frequency in the normal range. The MATLAB simulation results show that the EID method has better control effect compared with PI in reducing the impact of RE.
2. System description and modeling
2.1 Description of the proposed isolated small hybrid power system
The hybrid energy power system [11] studied consists of RE (WP+SP), EV, diesel generator with governor and load demand is shown in Fig.1. ,f PG,Pload, PEVC,
PEVD
,u representing deviation of frequency, power fluctuation of diesel generator, load, charge and discharge power of EV and control input, respectively.
Fig.1. The block diagram of isolated hybrid power system
The power supplied to the load demand is the sum of the output power of WP, SP, diesel generator and EVs.
Such RE connection and load variation will lead to an imbalance in the supply and demand of active power.
The power deviation can be expressed as:
Re
G EVD EVC load
P P P P P P
(1)
P representing the power supply of RE generations. Re
The control strategy, controlling the output of the diesel generator and EVs to compensate for the power deviation based on EID method, can maintain the frequency within the normal range.
2.2 Renewable energy generation model
This paper will focus on the verification and validation of EID method. Therefore, the simple linear first order lag is chosen to be the transfer function model of SP system and WP system. They can be expressed as
( ) 1
PV PV
PV
PV
P K
G s
sT (2) where is solar irradiation. KPVis gain constant and
( )
1
WP WP
WP
W WP
P K
G s
V sT
(3) where VW is wind speed. KWP is gain constant and TWPis time constant.
2.3 EV model
When a large number of EVs participate in frequency regulation, their discharging characteristic is similar to batteries’, and their charging characteristic is similar to controllable loads.
The discharge transfer function can be ( )
1
EVD EVD
EVD
EVD
P K
G s
f sT
(4) The charge transfer function can be
( ) 1
EVC EVC
EVC
EVC
P K
G s
f sT
(5) where KEVD，KEVC is gain constant. TEVD,TEVCis time constant.
This model is intended to involve the EVs fleet as a battery storage in frequency regulation, which is not a real model of the EVs.
2.4 Load frequency control system model
The EID-LFC is designed to control the output power of diesel generator and EVs to compensate for the power fluctuation from load demand and RE. The system can then maintain frequency balance. The state space-equation for the entire power system model is( ) ( ) ( ) [ Re( ) load( )]
x t Ax t Bu t H P t P t (6) where
( ) ( ) G( ) G( ) EVC( ) EVD( )T x t f t P t X t P t P t ,
1 0
1 1
0 0 0
1 1
0 0 0
0 0 1 0
0 0 0 1
P P P
P P P P
T T
G G
EVC
EVC EVC
EVD
EVD EVD
K K K
T T T T
T T
A RT T
K
T T
K
T T
,
0 0 1
0 0
G
B T
,
0 0 0 0
P P
K T H
. PR E is RE variation.
The EID-LFC controller for LFC is designed in next section.
3. Proposed frequency control method based on EID
3.1 Theory of EID method
The key idea of EID is to find an equivalent value of the real disturbance that has the same effect on output when input it reversely in the control input channel. In this paper, the variations of RE and load demand are the disturbance to be rejected. As shown in Fig. 2, the EID control system consists of an internal model controller, an EID estimator, a state observer, a controlled plant, and a state feedback controller.
Fig.2 The EID control system The state feedback control law is：
u tf( )K x tpˆ( )K x tR R( ) (7) ˆ( )
x t is observer state, x tR( )is internal model state.
As explained in [12], we can obtain an estimated EID as follows
ˆd t( )B LC x t [ ( )x tˆ( )]u tf( )u t( ) (8) where B (B BT )1BT.
Since the output y t contains measurement noise, ( ) ˆ( )
d t needs a filter, which is chosen as a first-order ( )
F s for simplicity:
( ) 1 F s 1
Ts (9) So the modified imposed control law is：
u t( )u tf( ) d s( ) (10) According to the separation theorem, as long as stability is the only concern, the state-feedback control law K and the state observer gain L can be designed independently. As explained in [12], the transfer function
d( )
G s from ( )d s to ˆ( )d s can be expressed as
( ) ( )[ ( )]1
G sd B sI A sI ALC B (11) Based on the small-gain theory, the closed-loop EID- based system under the control law (10) is stable if the system satisfies,
||FGd|| (12) 1 where ||FGd||supmax[ (F j)Gd(j)] , 0 ,
and max( ) denotes the maximum singular value.
The optimal control method is a common way to obtain the EID control parameters [KP KR] and the state observer gain L, and the performance index needs to be properly selected when applied.
3.2 EID method for load frequency control
The overall control strategy for the power system based on EID is shown in Fig.3.In this isolated hybrid power system, the RE output are WP and SP which are fluctuating with wind and intensity of illumination. Load demands are also fluctuating, which will lead to power imbalance and frequency variation. To solve this problem, variation of RE and load demand perturbation are considered as external disturbance to system. The EID method is applied to suppress this disturbance to control the output power of the diesel generator and the charging and discharging of EV to maintain the power balance.
Fig.3 The control strategy block diagram for isolated small power system based on EID method
4. Simulation results and analysis
The normal power system operation requires that a frequency deviation within ±0.2 Hz, which is chosen as the standard in this paper. In this section, the effectiveness of the proposed method is verified by the simulation results. The WP, SP and Load variation are simulated by white noise in MATLAB as shown in Fig.4~Fig.6.
Fig.4 WP variation
Fig.5 SP variation
Fig.6 Load variation
Simulation result is shown in Fig.7 under the cases of whether connected EV under PI control. Frequency variation is suppressed in about ±0.06Hz due to the charging and discharging of EV. It is obvious that EV has an effective performance on frequency regulation.
Fig.7 Frequency variation under PI control Then we consider applying the EID control to the power system with RE and EV. The is shown in f Fig.8. It can be seen from the figure, EID control can make f vary in a minimal range which is ±0.015Hz.
Therefore, it can be concluded that frequency fluctuations caused by RE and load variation can be more effectively compensated by EID control than PI control.
Fig.8 Comparison of frequency variation between EID and PI control
5. Conclusions
A new frequency control approach for the isolated small hybrid power system with WP, SP and EV is proposed based on EID method. The EID method is used to reject the disturbances which lead to power imbalance. The EID method is applied to the LFC model of the system to regulate the frequency through maintaining the power balance between diesel generator, RE and EVs. The optimal control method is adopted to calculate the control parameters. The simulation results show that the frequency deviation can be reduced to ±0.015Hz when actively control the output of diesel generator and the EVs under EID control. The proposed control strategy for frequency control of the small hybrid power system can achieve more effective frequency control performance than PI control.
6. Discussion
To verify the effectiveness of EID control method, the real dynamic characteristics of the model such as EVs and WTG, are not well considered in the simulation. And further research should be done on EV scheduling problem. Moreover, the influence of system uncertainty should be taken into account, and actual data should be used for more practical simulation.
Acknowledgement
This work is supported by National Natural Science Foundation of China (61573381).
Reference
[1] Sahu, B.K. Wind energy developments and policies in China: A short review. Renewable and Sustainable Energy Reviews, 2018;81: 1393-1405 [2] Hatziargyriou N, Asano H, Iravani R, et al.
Microgrids. Power & Energy Magazine IEEE, 2007;5(4):78-94.
[3] Bae S, Kwasinski A. Dynamic modeling and
photovoltaic resources. IEEE Transactions on Smart Grid, 2012;3(4):1867-1876.
[4] Wang C, Mi Y, Fu Y, et al. Frequency control of an isolated micro-grid using double sliding mode controllers and disturbance observer. IEEE Transactions on Smart Grid, 2016;9(2): 923 - 930.
[5] Safari M, Sarvi M. Optimal load sharing strategy for a wind/diesel/battery hybrid power system based on imperialist competitive neural network algorithm.
IET Renewable Power Generation, 2014; 8(8):937- 946.
[6] Hasanien HM, Muyeen SM, Tamura J. Frequency control of isolated network with wind and diesel generators by using fuzzy logic controller.
International Conference on Electrical Machines &
Systems. IEEE, 2010.
[7] Cominos P, Munro N. PID controllers: recent tuning methods and design to specification. IEE- Proceedings of Control Theory and Applications, 2002; 149(1):46-53.
[8] Wu D, Tang F, Dragicevic T, et al. Autonomous active power control for islanded AC microgrids with photovoltaic generation and energy storage system. IEEE Transactions on Energy Conversion, 2014;29(4):882-892.
[9] Hajipour E, Bozorg M, Fotuhi-Firuzabad M.
Stochastic capacity expansion planning of remote microgrids with wind farms and energy storage.
IEEE Transactions on Sustainable Energy, 2015;6(2):491-498.
[10] Igualada L, Corchero C, Cruz-Zambrano M, et al.
Optimal energy management for a residential microgrid Including a vehicle-to-grid system. IEEE Transactions on Smart Grid, 2014; 5(4):2163-2172.
[11] Pahasa J, Ngamroo I. Coordinated control of wind turbine blade pitch angle and PHEVs using MPCs for load frequency control of microgrid. IEEE Systems Journal, 2014:1-9.
[12] She J, Fang M, Ohyama Y, et al. Improving disturbance-rejection performance based on an equivalent-input-disturbance approach. IEEE Transactions on Industrial Electronics, 2008;
55(1):380-389.