Abstract—Conventional active islanding detection methods
(IDMs) are designed for the inverter-based distributed gen-eration systems (DGSs) with current control interface. Such strategies can hardly be extended to the DGS with power con-trol interface because the power concon-trol loop can affect the IDMs by enlarging the nondetection zones (NDZs). This paper presents an active IDM based on negative-sequence power injections for the DGS with power control interface. Combining with the IDM, the power control of the DGS is achieved with two control loops. One is the positive-sequence power loop that satisfies the conventional power control requirements. The other is the negative-sequence power/current variation loop for the islanding detection. The pos-itive and negative sequences are separated by a simple strategy based on an all digital phase-locked loop. Due to the differences between the grid impedance and the local load impedance, the percentage of the voltage imbalance (VI) at the point of com-mon coupling is utilized to indicate the islanding operation. For the grid-connected DGS, the VI is dominated by the grid voltage, which is a constant. If the DGS is disconnected from the grid, the VI is determined by the injected negative-sequence power/current. The NDZs of the presented scheme with different system config-urations, such as grid impedances, load quality factors, etc., are also analyzed in this paper. By injecting the negative-sequence power/current periodically, the NDZs resulting from the imbal-ance of the grid voltages or the local loads are further mitigated. For multi-DGSs, the IDM is still effective if combined with the conventional under/overfrequency-protection strategy. The simu-lation and experimental results verify the effectiveness of the IDM.
Index Terms—Distributed generation system (DGS),
island-ing detection, positive- and negative-sequence separation, power control.
I. INTRODUCTION
I
N the last few years, distributed generation systems (DGSs) have gained popularity amongst industry and utilities because of many potential benefits, such as peak shaving, fuelManuscript received September 10, 2010; revised January 16, 2011 and April 25, 2011; accepted June 5, 2011. Date of publication July 22, 2011; date of current version November 23, 2011. This work was supported in part by the National Natural Science Foundation of China under Grant 60974130 and in part by the Power Electronics Science and Education Development Program of the Delta Environmental and Educational Foundation under Grant DREK2010003. Paper no. TEC-00362-2010.
H. Geng and G. Yang are with the Department of Automation, Tsinghua University, Beijing 100084, China (e-mail: genghua@tsinghua. edu.cn; [email protected]).
D. Xu and B. Wu are with the Department of Electrical and Computer Engineering, Ryerson University, Toronto, ON M4X1G6, Canada (e-mail: [email protected]; [email protected]).
Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TEC.2011.2159720
Fig. 1. Schematic diagram of a single inverter-based DGS.
switching, improved power quality and reliability, increased ef-ficiency, and improved environmental performance [1].
Most small- and medium-size DGSs are inverter based and have flexible interfaces with either active or reactive power sup-plies. A typical diagram of the inverter-based DGS is shown in Fig. 1, in which the local load is supported by the grid and the DGS. In Fig. 1, the dc source represents any distributed power sources such as solar panels or wind turbines. Normally, invert-ers have quick responses and fast disturbance rejection capa-bility. Therefore, the inverter-based DGSs have been applied to different applications with corresponding control schemes, such as current control, power control, or voltage control. The cur-rent control scheme contains only curcur-rent control loops which aim to extract constant or desired currents from the DGS [2]. The power control scheme intends to regulate the active and reactive power produced by DGS in order to improve the power factor [3], [4] or balance the power source and the load in the islanding mode [5]. The voltage control can support the grid during the peak load time by injecting reactive power [2], [6].
Aside from satisfying those control requirements, the DGS interface control should perform an additional function which is an islanding detection. A DGS is islanded when it supplies power to some loads while the main utility source is discon-nected. Usually, the protection schemes for the conventional power system are designed under the assumption that power flows from the substations to the end users, and all circuits downstream would be de-energized if a fault takes place and a breaker opens. However, it is not the case when the DGS is inte-grated. Thus, islanding protection is important for the equipment and personnel safety in the utility with DGSs [7]–[9]. In other situations where the intentional islanding operation is permitted, fast islanding detection is also required for appropriate decision making to manage autonomous operation of the island [10].
1064 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 26, NO. 4, DECEMBER 2011
Fig. 2. Block diagram of the power control scheme with a conventional active IDM.
A number of islanding detection methods (IDMs) have been proposed for DGSs with current control schemes [1], [11], [12]. Compared with passive IDMs, active IDMs have smaller non-detection zones (NDZs), and, thus, are more popular. Active IDMs attempt to cause the power mismatches with additional output variations, such as active power variation [12]–[15], re-active power variation [11], [12], [16]–[21] or harmonic power variation [22], [23], so that a certain system parameter is forced to drift once an islanding condition occurs [12]. The active-power-variation-based IDMs are not so popular because they need a huge amount of power for a reasonable voltage drift and are, therefore, not efficient [11]–[13]. It is expected that most of the active IDMs can be extended to the DGS equipped with other control schemes. However, as reported in [2] and [24], the reactive-power-variation-based IDMs have an enlarged NDZs or even fail if the DGS is equipped with power control scheme because the outer-loop power controller can suppress the power mismatches or variations during the islanding operation.
In this paper, an active IDM based on negative power vari-ation is presented for the DGSs with power control interface. With such an IDM, the power controller of the DGS contains a positive-sequence power control loop for conventional power control and a negative-sequence power/current variation loop for islanding detection. The proportional plus resonant (P + R) con-trollers are employed in the current loop to regulate the positive-and negative-sequence currents. An all digital phase-locked loop (ADPLL) is designed for the phase-angle detection of the DGS voltages, and a simple scheme for the positive- and sequence separation is further presented. With the negative-sequence power variation, the negative-negative-sequence voltages at the point of common coupling (PCC) may vary after islanding due to the differences between the grid and load impedances, and it can be the criterion to indicate an islanding operation. As analyzed in this paper, system configurations, such as grid impedances, load quality factors, and imbalance of the grid volt-ages or load parameters, can affect the islanding detection. In order to mitigate the NDZs resulted from the imbalance of the grid voltages or loads, the negative-sequence power/current can be periodically disturbed. For multi-DGSs, the IDM is still ef-fective if combined with the conventional under/overfrequency-protection (UFP-OFP). The simulation and experimental results verify the effectiveness of the IDM.
II. POWERCONTROLWITHREACTIVEPOWERVARIATIONS
Since the reactive-power-variation-based IDMs are more pop-ular than other IDMs in real applications, the power control combined with such IDMs is discussed in this section.
The power control scheme with a conventional IDM is im-plemented in the rotating dq-frame, and the block diagram is given in Fig. 2 [1], [24]. In the power control loop, the in-verter output current references i∗d, i∗q are obtained from the proportional-integral (PI) regulator. Pdg∗ , Q∗dg and Pdg, Qdg are
the power references and the measured DGS output powers.
i∗d, i∗q are then transformed to i∗d, i∗q by applying the matrix in the phase-angle transformation block to achieve the power vari-ations. The angle for the transformation θsis determined by
dif-ferent IDMs, such as a constant (namely active frequency drift (AFD) method [19], [20]) or a positive-feedback function of the voltage frequency at the PCC (slip mode frequency shift (SMS) method [16], [17], or Sandia frequency shift (SFS) method [21]). In the current control loop, vp cc,d, vp cc,q and id, iqare the PCC
voltages and the inverter output currents, respectively. The angle for the dq-frame transformation θ and the voltage frequency for the decoupling control of the current loop ω are obtained with a PLL which is usually software based and implemented in the synchronous reference frame [25].
A. Influence of the Power Control Loop on the IDM NDZs The phase of the impedance of a parallel RLC load θlat an
arbitrary frequency f , as a function of the quality factor Qf and
resonant frequency f0of the load, is given by [11]
θl = tan−1 R1− ω 2LC ωL = tan−1 Qf f0 f − f f0 (1) where Qf = ω0RC = ωR0L = R C L, ω0 = 2πf0 = 1 √ LC.
Following the phase criteria, the frequency of an islanded system in steady-state condition can be obtained when the angle
θlis equal to that of the inverter θdg.
For the current-controlled DGS, θdg = θs if the DGS is
op-erating in unity power factor mode (i∗q = 0). The NDZs of the
reactive power variation IDMs (AFD, SMS, and SFS) can be derived using the phase criteria as discussed in [11].
Fig. 3. Responses of different control interfaced DGSs with reactive power variation IDMs. (a) AFD IDM. (b) SMS IDM. (c) SFS IDM.
In the DGS with power control interface, the phase criteria also holds true but θdg does not equal to θs any more, which
changes the NDZs [2]. Applying the phase criteria, we have
tan θdg =
iq
id
= tan θl. (2)
Usually, the inner current loop is designed to respond much faster than the outer power loop in order to guarantee a stable operation of the DGS. By ignoring the current loop dynamics, the inverter output current is
id = i∗ d = kpp+ kip s Pdg∗ − Pdg cos θs − kpp + kip s Q∗dg − Qdg sin θs iq= i∗ q = kpp+ kip s Pdg∗ − Pdg sin θs + kpp + kip s Q∗dg − Qdg cos θs (3)
where Pdg =32vp cc,did and Qdg = 32vp cc,diq, and kpp and kip
are the proportional and integral coefficients of the PI regulators in the power loop.
Substituting (3) into (2) and considering the unity power fac-tor operation (Q∗dg = 0), the phase criteria for the
power-control-interfaced DGS can be further expressed as
tan θl = tan θs· ks Tss + 1 = Qf f0 f − f f0 (4) where k = 2 cos θs 3vp c c , dki p ≈ 2 3vp c c , dki p and Ts= 2 cos θs 3vp c c , dki p + kp p ki p ≈ 2 3vp c c , dki p + kp p
ki p. θs is a small disturbance determined by the IDMs.
It is noticed that the power control loop may suppress the frequency drifting after islanding by introducing a high-pass filter as indicated in (4) [2]. In view of this, the AFD method is noneffective for the power-control-interfaced DGS since θs is
a constant and the reactive power variation is negligible due to the filtering effect. For SMS and SFS methods, the NDZs are
enlarged because only high-frequency components of tan θscan
contribute to the islanding detection.
B. Verification of the Filtering Effect of the Power Control Loop on the IDM NDZs
In this section, the counteracting effect of the power control loop on the IDM NDZs is investigated with time-domain-based simulations. Different reactive-power-variation-based IDMs, in-cluding AFD, SMS, and SFS methods, are implemented in the DGS with current and power control schemes. The system and control parameters are listed in the Appendix. The grid is discon-nected at t = 0.2 s and the local load is R = Rnorm, f0 = 60 Hz
and Qf = 2.5. The system responses are illustrated in Fig. 3.
Fig. 3 shows that the frequency of the PCC voltage is drift-ing away from its normal range (59.3–60.5 Hz) with the AFD, SMS, or SFS methods for the current-controlled DGS. However, the frequency drifting is negligible if the DGS is power control based, which indicates that the islanding cannot be identified in such cases. Although increasing the IDM control gain can improve the islanding detection, the inverter output power may oscillate as shown in [24] because the low-frequency compo-nents of θs cannot contribute to the islanding detection.
There-fore, such a strategy can lead to power-quality problems in the intentional islanding operation situations.
III. IDM WITHNEGATIVE-SEQUENCEPOWERVARIATIONS
The negative-sequence current injection has been employed to detect the islanding operation of the current-controlled DGSs [23]. Compared with other IDMs, such strategy is very fast (within four cycles of the grid voltage) and a small (< 3%) negative-sequence current is enough for the islanding detection. However, such scheme has not been extended into the power-controlled DGSs, and the NDZs have not been discussed as well as the applications in the multi-DGSs.
The block diagram of the power control scheme with the presented IDM is shown in Fig. 4. The power control loop regulates the positive-sequence power flows, Pdg ,p os, Qdg ,p os,
to approach the references Pdg∗ , Q∗dg with a PI controller. The references of the current loop consist of the positive- and
1066 IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 26, NO. 4, DECEMBER 2011
Fig. 4. Block diagram of the power control scheme with the presented IDM.
negative-sequence components, in which the positive compo-nents i∗d,p os, i∗q ,p os come from the outer power loop and the negative-sequence components i∗d,neg, i∗q ,negare proportional to
i∗d,p os, i∗q ,p os with a small coefficient k. The P+R controllers are adopted to regulate the positive- and negative-sequence inverter currents simultaneously, which avoids multiple rotat-ing frame transformations and PI controllers. The positive- and negative-sequence phase angles θp os, θneg in the control
dia-gram are derived from an ADPLL [26].
With such scheme, the positive-sequence power of the DGS is controlled to satisfy the output power requirements, while the negative-sequence power/current is injected for the islanding de-tection. When the DGS is grid connected, the negative-sequence power/current mainly flows into the grid due to the impedance differences. Under the islanded condition, the power/current flows into the load and it results in the negative-sequence volt-ages at the PCC. The level of voltage imbalance (VI) at the PCC is then detected and qualified. If the unbalance level is beyond a permissible value, e.g., 3% [23], it indicates that islanding has occurred.
A. Sequence Separation With the ADPLL
An ADPLL based on the digital techniques has been proposed in [26] for the sequence separation of the grid-connected power converters in the unbalanced grid applications. Following is the brief introduction of the strategy proposed in [26].
Based on the symmetrical components method, the positive-and negative-sequence voltage vectorsv+,
v−can be expressed as v+ v− = 1 3 1 a2 a 1 a a2 ⎡ ⎢ ⎣ va vb vc ⎤ ⎥ ⎦ (5) where v+ = vα ++ jvβ + and v−= vα−+ jvβ−; va, vb, and
vc are the PCC phase voltage vectors; and
a = e−j23π.
Denoting the instantaneous PCC phase voltage as va, vb, vc
and the phase angles as θa, θb, θc, the positive and negative
components of grid voltage can be deduced from (5) as
vα + = 1 3 va− 1 2(vb+ vc) + √ 3 2 vb tan θb − vc tan θc vβ + = 1 3 − Va tan θa +1 2 vb tan θb + vc tan θc + √ 3 2 (vb− vc) vα−=13 va− 1 2(vb+ vc)− √ 3 2 vb tan θb − vc tan θc vβ−= 13 va tan θa −1 2 vb tan θb + vc tan θc + √ 3 2 (vb− vc) . (6) For the real implementation, the trigonometric values in (6) can be obtained with a lookup table, and a hysteresis limiter is employed to avoid division by zero problems. θa, θb, and θc
are detected by the ADPLL, and the positive- and negative-sequences can be obtained with a simple calculation in a low-cost microprocessor:
θp os= tan−1(vβ +/vα +)
θneg = tan−1(vβ−/vα−) (7)
B. NDZ Analysis and Elimination
Applying the symmetrical component method, the system model of Fig. 1 can be expressed with the positive-, negative-, and zero-sequence variables. If the DGS is grid connected, the following equations hold true:
vg ,pn 0− vl,pn 0 = zgig ,pn 0 vdg ,pn 0− vl,pn 0 = zdgidg ,pn 0 vl,pn 0 = AzlA−1il,pn 0 il,pn 0 = ig ,pn 0+ idg ,pn 0 (8) where A = 13 ⎡ ⎣1 a a 2 1 a2 a 1 1 1 ⎤
⎦; v, i, and z are the voltage,
[7] IEEE Recommended Practice for Industrial and Commercial Power
Sys-tems Analysis. (Color Book Series—Brown Book), IEEE Standard
399-1997, 1998.
[8] IEEE Standard for Interconnecting Distributed Resources With Electric
Power Systems, IEEE Standard 1547-2003, pp. 1–16, 2003.
[9] Static Inverter and Charge Controllers for Use in Photovoltaic Systems, Standard 1741, Underwriters Lab, Inc., Northbrook, IL, 1999.
[10] H. Karimi, H. Nikkhajoei, and R. Iravani, “Control of an electronically-coupled distributed resource unit subsequent to an islanding event,” IEEE
Trans. Power Del., vol. 23, no. 1, pp. 493–501, Jan. 2008.
[11] L. Lopes and H. Sun, “Performance assessment of active frequency drifting islanding detection methods,” IEEE Trans. Energy Convers., vol. 21, no. 1, pp. 171–180, Mar. 2006.
[12] W. Bower and M. Ropp, “Evaluation of islanding detection methods for photovoltaic utility-interactive power systems,” Rep. IEA-PVPS, Tech. Rep. T5-09:2002, Int. Energy Agency, Mar. 2002.
[13] G. Kern, “Sunsine300, utility interactive ac module anti-islanding test results,” in Conf. Record 26th IEEE Photovolt. Spec. Conf., Sep./Oct., 1997, pp. 1265–1268.
[14] P. Mahat, Z. Chen, and B. Bak-Jensen, “A hybrid islanding detection technique using average rate of voltage change and real power shift,”
IEEE Trans Power Del., vol. 24, no. 2, pp. 764–771, Apr. 2009.
[15] H. Zeineldin and J. Kirtley, “A simple technique for islanding detection with negligible nondetection zone,” IEEE Trans Power Del., vol. 24, no. 2, pp. 779–786, Apr. 2009.
[16] G. Smith, P. Onions, and D. Infield, “Predicting islanding operation of grid connected PV inverters,” IEE Proc.—Electr. Power Appl., vol. 147, no. 1, pp. 1–6, Jan. 2000.
[17] M. Ropp, M. Begovic, A. Rohatgi, G. Kern, S. Bonn, R. H., and S. Gonza-lez, “Determining the relative effectiveness of islanding detection methods using phase criteria and nondetection zones,” IEEE Trans. Energy
Con-vers., vol. 15, no. 3, pp. 290–296, Sep. 2000.
[18] M. Ropp, M. Begovic, and A. Rohatgi, “Analysis and performance assess-ment of the active frequency drift method of islanding prevention,” IEEE
Trans Energy Convers., vol. 14, no. 3, pp. 810–816, Sep. 1999.
[19] G.-K. Hung, C.-C. Chang, and C.-L. Chen, “Automatic phase-shift method for islanding detection of grid-connected photovoltaic inverters,” IEEE
Trans Energy Convers., vol. 18, no. 1, pp. 169–173, Mar. 2003.
[20] R. A. Jones, T. R. Sims, and A. Imece, “Investigation of potential islanding of dispersed photovoltaic systems,” Sandia Natl. Lab., Albuquerque, NM, Tech. Rep. SAND87-7024, 1988.
[21] J. Stevens, R. Bonn, J. Ginn, S. Gonzalez, and G. Kern, “Development and testing of an approach to anti-islanding in utility-interconnected pho-tovoltaic systems,” Sandia Natl. Lab., Albuquerque, NM, Tech. Rep. SAND2000-1939, 2000.
[22] G. Hernandez-Gonzalez and R. Iravani, “Current injection for active is-landing detection of electronically-interfaced distributed resources,” IEEE
Trans Power Del., vol. 21, no. 3, pp. 1698–1705, Jul. 2006.
[23] H. Karimi, A. Yazdani, and R. Iravani, “Negative-sequence current in-jection for fast islanding detection of a distributed resource unit,” IEEE
Trans Power Electron., vol. 23, no. 1, pp. 298–307, Jan. 2008.
[24] X. Wang, W. Freitas, W. Xu, and V. Dinavahi, “Impact of DG interface controls on the sandia frequency shift antiislanding method,” IEEE Trans
Energy Convers., vol. 22, no. 3, pp. 792–794, Sep. 2007.
[25] R. Teodorescu and F. Blaabjerg, “Flexible control of small wind turbines with grid failure detection operating in stand-alone and grid-connected mode,” IEEE Trans. Power Electron., vol. 19, no. 5, pp. 1323–1332, Sep. 2004.
[26] H. Geng, D. Xu, and B. Wu, “A novel hardware-based all-digital phase-locked loop applied to grid-connected power converters,” IEEE Trans.
Ind. Electron., vol. 58, no. 5, pp. 1737–1745, May 2011.
[27] IEEE Recommended Practice for Monitoring Electric Power Quality, IEEE Standard 1159-1995, 1995.
sity, Beijing, China, in 2008.
From 2008 to 2010, he was a Postdoctoral Research Fellow in the Depart-ment of Electrical and Computer Engineering, Ryerson University, Toronto, ON, Canada. Since May 2010, he has been an Assistant Professor in the De-partment of Automation, Tsinghua University. His current research interests include distribution generation systems, renewable energy conversion systems, and digital control systems.
Dewei (David) Xu(S’99–M’01) received the B.Sc., M.A.Sc., and Ph.D. degrees in electrical engineering from Tsinghua University, Beijing, China, in 1996, 1998, and 2001, respectively.
He has been with Ryerson University, Toronto, ON, Canada, since 2001, where he is currently an Associate Professor. His research interests include renewable energy system, high-power converters, electric motor drives, and advanced digital control for power electronics.
Bin Wu(S’89–M’92–SM’99–F’08) received the M.A.Sc. and Ph.D. degrees in electrical and computer engineering from the University of Toronto, Toronto, ON, Canada, in 1989 and 1993, respectively.
After being with Rockwell Automation Canada as a Senior Engineer, he joined Ryerson University, Toronto, where he is currently a Professor and NSERC/Rockwell Industrial Research Chair in power electronics and electric Drives. He has published more than 200 technical papers, authored a Wiley-IEEE Press book, and holds more than 20 issued and pending patents in the field of power conversion, advanced controls, adjustable-speed drives, and renewable energy systems.
Dr. Wu received the Gold Medal of the Governor General of Canada, the Premier’s Research Excellence Award, Ryerson Distinguished Scholar Award, Ryerson Research Chair Award, and the NSERC Synergy Award for Innovation. He is a fellow of the Engineering Institute of Canada and the Canadian Academy of Engineering. He is an Associate Editor of IEEE TRANSACTIONS ONPOWER
ELECTRONICSand IEEE CANADIANREVIEW.
Geng Yang(M’¯01–SM’¯02) received the B.S. and M.S. degrees in electrical engineering from the Xian University of Science and Technology, Xian, China, in 1982 and 1984, respectively, and the Ph.D. degree in electrical engineering from Sophia University, Tokyo, Japan, in 1992.
From 1985 to 1987, he was an Assistant at the Xian University of Science and Technology. He was a Visiting Scientist at Fukui University, Fukui, Japan, and at Sophia University, in 1987 and 1988, respectively. From 1992 to 1994, he was a Senior Researcher with Kasuga Electrical Works, Ltd. From 1995 to 1999, he was engaged as a Lecturer, Associate Professor, and Professor at the Xian University of Science and Technology. Since 2000, he has been a Professor in the Department of Automation, Tsinghua University, Beijing, China. He is the author or coauthor of more than 80 technical papers, and a Tsinghua-University Press book, and has filed two Japan and five China patents in the field of power conversion, advanced control, and adjustable-speed drives systems. His current research interests include electrical drive and system, power electronic equipment, and control technology of wind and photovoltaic energy conversion systems.
Dr. Yang is a member of the Institute of Electrical Engineers of Japan and the Director of the Board of China Power Electric Institute. He is a member of the China Electrotechnical Society (CES) and served as the Vice Director of Education Committee of Electrical Automation Branch in CES.
