Produção Nacional
Revisado por pares
Hybrid fuzzy anti‐islanding for grid‐connected and islanding operation in distributed generation systems
2015; Institution of Engineering and Technology; Volume: 9; Issue: 3 Linguagem: Inglês
10.1049/iet-pel.2014.0717
ISSN1755-4543
AutoresCassius R. Aguiar, Guilherme Henrique Favaro Fuzato, Renan F. Bastos, Amílcar F. Q. Gonçalves, Ricardo Q. Machado,
Tópico(s)Power Systems Fault Detection
ResumoIET Power ElectronicsVolume 9, Issue 3 p. 512-518 Research ArticlesFree Access Hybrid fuzzy anti-islanding for grid-connected and islanding operation in distributed generation systems Cassius Rossi Aguiar, Corresponding Author Cassius Rossi Aguiar caroaguiar@gmail.com Federal University of Technology-Paraná, Toledo, PR, 85902-490 BrazilSearch for more papers by this authorGuilherme Fuzato, Guilherme Fuzato Department of Electrical and Computer Engineering, University of São Paulo, São Carlos, SP, 13566-590 BrazilSearch for more papers by this authorRenan Fernandes Bastos, Renan Fernandes Bastos Department of Electrical and Computer Engineering, University of São Paulo, São Carlos, SP, 13566-590 BrazilSearch for more papers by this authorAmilcar F. Querubini Gonçalves, Amilcar F. Querubini Gonçalves Federal University of Technology-Paraná, Medianeira, PR, 85884-000 BrazilSearch for more papers by this authorRicardo Quadros Machado, Ricardo Quadros Machado Department of Electrical and Computer Engineering, University of São Paulo, São Carlos, SP, 13566-590 BrazilSearch for more papers by this author Cassius Rossi Aguiar, Corresponding Author Cassius Rossi Aguiar caroaguiar@gmail.com Federal University of Technology-Paraná, Toledo, PR, 85902-490 BrazilSearch for more papers by this authorGuilherme Fuzato, Guilherme Fuzato Department of Electrical and Computer Engineering, University of São Paulo, São Carlos, SP, 13566-590 BrazilSearch for more papers by this authorRenan Fernandes Bastos, Renan Fernandes Bastos Department of Electrical and Computer Engineering, University of São Paulo, São Carlos, SP, 13566-590 BrazilSearch for more papers by this authorAmilcar F. Querubini Gonçalves, Amilcar F. Querubini Gonçalves Federal University of Technology-Paraná, Medianeira, PR, 85884-000 BrazilSearch for more papers by this authorRicardo Quadros Machado, Ricardo Quadros Machado Department of Electrical and Computer Engineering, University of São Paulo, São Carlos, SP, 13566-590 BrazilSearch for more papers by this author First published: 01 March 2016 https://doi.org/10.1049/iet-pel.2014.0717Citations: 8AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract This study proposes a method for islanding detection in distributed generation systems, namely, hybrid fuzzy positive feedback (PF). The aim is to reduce the injection of disturbances during the grid-connected operation from the anti-islanding methods, particularly at the instant of load connection and, do not require adjustment of the PF gain and frequency filters. This method is designed from fuzzy inference rules and the frequency and voltage inputs are also taken into account in the design procedure. Since the classical active methods do not identify if the voltage variation arises from the islanding condition or from a power quality issue in the grid, the fuzzy PF enables the proposed methods to differentiate the islanding condition from other grid events. In contrast to classical methods that have a fixed PF gain, a dynamic fuzzy PF gain is obtained by means of the fuzzy inference rules. The classical active technique is compared with the proposed approach in terms of detection time and injected disturbances during the grid-connected operation. The simulation and experimental results are presented to demonstrate the effectiveness of the proposed approach. 1 Introduction The connection of the distributed generation (DG) using DC–AC voltage source inverter (VSI) has been the subject of intense research in order to promote solutions to the problems of instability and detection of contingencies when DG systems operate connected to the main grid. In this context, an important requirement for the integration of DG structures in the electric power system is the ability of DG to detect the islanding condition. Islanding can be defined as a portion of the utility system that contains both load and distributed resources which remain energised while isolated from the remaining utility system [1, 2]. It may lead to the reduction of power quality levels, equipment damage, malfunction of the protection devices and even safety hazards to the operating personnel [3]. In this context, islanding is a major issue to DG users and utility, since in recent years the possibility of islanding occurrence has increased with the DG penetration [4, 5]. In general, anti-islanding methods can be classified as passive or active methods. Active methods insert a continuous effect through positive feedback (PF) from voltage or frequency variation in order to force the DG system to an unstable operation when the islanding condition occurs. The main drawbacks of these methods are related to the injection of disturbances during the grid-connected operation, which can reduce the power quality and may destabilise the DG system [6-9]. This paper proposes an intelligent anti-islanding method with fuzzy PF, in order to reduce the injected disturbances on the current reference control of the DG during the grid-connected operation. The main difference between the proposed and classical methods is that the fuzzy PF gain continuously varies and does not remain constant while in operation. This is in contrast to classical methods, where the PF gain is fixed and the output is proportional to input variation. This paper is organised as follows: Section 2 describes the control strategy; Section 3 analyses the voltage PF method; Section 4 presents the proposed method; Sections 5 and 6 present the simulation and experimental results in order to confirm the proposed approach; whereas the conclusions are presented in Section 7. 2 Control strategy The control method used in a DG system has two distinct operation modes: when it is operating while connected to the grid, and when it is in islanding mode. In terms of control methodologies, while connected to the grid the VSI operates with a dq synchronous reference frame where the output current is controlled. As it is isolated from the grid, the VSI controls the terminal voltage in order to keep the DG stable with high levels of power quality, which is carried out by using a cascade (current/voltage) control in αβ stationary reference frame. Under normal operation, the DG is connected to the grid through a current controller used to regulate the current flow in the grid with unity power factor. However, when the islanding moment is detected, the DG control is changed to operate in islanded mode by means of a switched 'control'. This switched control method is managed by the anti-islanding technique as shown in Fig. 1. Fig. 1Open in figure viewerPowerPoint Schematic diagram of the DG system The use of αβ stationary reference frame is justified, as the local load in the islanding mode may be unbalanced or at least non-linear, which requires the voltage compensation of positive and negative sequences in dq synchronous reference frame. On the basis of the arguments above, αβ stationary reference frame is preferable in order to avoid additional sequence compensation. In Park's dq reference frame, the synchronous angular frequency ωpll is determined by the phase-locked loop (PLL) algorithm, while the power circuit dynamics of a DG can be represented by (1) [10] (1)where vd, vq, id and iq are dq axes inverter terminal voltages and currents; and are dq axes output voltages; Rdg and Ldg are per-phase resistance and inductance, respectively, of the L filter. 3 Analysis of classical anti-islanding algorithms Classical active methods with PF can be applied using two approaches which are well-documented in the literature: one employs voltage [11, 12] and the other employs frequency PF [13-16]. The voltage anti-islanding uses the variation on the voltage magnitude and applies a PF to the direct-axis current reference to destabilise the system, in order to modify the inverter's output voltage. 3.1 Voltage PF (VPF) The VPF method detects the voltage variation to produce a PF output Δid, which alters the DG direct-axis current reference as shown in Fig. 2 [11, 12]. Fig. 2Open in figure viewerPowerPoint Block diagram of the VPF The transfer function of the VPF is displayed in (2) [8, 11] (2)The VPF consists of a PF gain (kVPF) and a band-pass filter, which is a combination of a washout filter (Tω) and a first-order low-pass filter (Tω). The main difficulty is finding a PF gain to ensure fast detection and minimal impact on the DG power quality. Similar to the frequency method, there is no specific methodology for obtaining the PF gain and the filter gains, which are typically obtained through a heuristic technique or practice test. 4 Proposed method In the classical methods, the main problem is to reduce power quality and insert disturbances during the grid-connected operation [17-20]. Accordingly, the aim in the proposed method is to reduce the injection of disturbances by means of the dynamic fuzzy PF gain which do not require adjustment. 4.1 Design of fuzzy PF gain (output) The input fuzzification is the method by which the input is assigned membership of different membership sets [21]. Due to their simplicity, triangular and trapezoidal membership sets are used for the direct-axis voltage (vd) and the frequency (ωpll) inputs. The fuzzification of inputs design is based on the standard requirements as [1, 2]. Local loads with a high quality factor (2.5) and zero power flow through the grid (power produced by the DG is completely consumed by the local load), are used as critical events in order to parameterise the membership sets. The fuzzification also takes into account the dynamic behaviour of the variables (ωpll and vd) when the islanding moment occurs, making it possible to design the intervals for each membership set. The fuzzy output or the membership sets for the fuzzy PF gain, is activated by combining the membership functions of the input variables. The membership sets for the fuzzy PF gain represent the output of the hybrid fuzzy PF (HFPF) graphically shown in Fig. 3. In Fig. 3, linguistic label S (small) is associated with a near-zero PF gain, meaning that the technique does not inject PF into the DG control. The same analysis can be related to the negative medium (NM) and positive medium (PM), where both proposed methods will insert a medium PF. Negative large (NL) and positive large (PL) produce a high PF gain, which means that the DG system is close to the islanding and the fuzzy PF gain will be increased in order to detect the island condition. Fig. 3Open in figure viewerPowerPoint Membership functions for the fuzzy PF outputs The sets are dispersed over the universe of discourse mapped inside the interval [−1, 1]. The set (S) has its geometrical point around zero to indicate the DG system in a stable operation mode, under which any type of disturbance will be injected into the DG control structure. The membership sets for the fuzzy PF gain are defuzzified (converted from linguistic labels to an analogue signal), using the centre of area method [22], which is given by (3)In (3) μi is the degree of membership, si represents the normalised controller output (expressed in per unit) and n is the number of discrete elements. 4.2 HFPF (anti-islanding) In the proposed HFPF method, the input variables are the direct-axis voltage (vd) and the frequency derivative (ωpll) as shown in Fig. 4. According to (4), the fuzzy output PF ) is used to change the direct-axis current reference and hence to drift the voltage beyond the boundaries. Fig. 4Open in figure viewerPowerPoint Block diagram of the HFPF HFPF is an intelligent method proposed to detect the islanding condition by varying the voltage magnitude at the PCC, and the derivative (ωpll) is used to improve the method performance and especially reduce the power quality issue related to the injection of disturbances into the DG (4)Unlike classical methods that use a single input to determine the contingency, the anti-islanding techniques proposed in this paper detect the event from the combination of the voltage magnitude and the frequency derivative. Since the HFPF method combines vd and ωpll, it allows the voltage deviation, followed by a proportional variation in the derivative of ωpll, to be sufficient for the direct-axis current reference feedback, which increases the efficiency of the hybrid method. However, disturbances in the voltage that are not followed by frequency variations will not inject PF into the current reference. Accordingly, it is possible to differentiate power quality issues such as voltage sags and fluctuations, for example and not inject disturbances into the DG control. The combination of vd and ωpll maximises its performance around these deficiencies. 4.3 Design and operation of HFPF The HFPF design is based on building membership sets for each input, which are found in Fig. 5. The membership sets of the vd and derivative of ωpll are associated with the fuzzy output PF, and are represented in Figs. 5a and b, respectively. Thus, when the voltage is represented by the linguistic label RV (rated voltage), the DG system is stable in relation to the voltage standards, and therefore the fuzzy output PF gain is associated with linguistic label small (S), independent of the frequency derivative value. Fig. 5Open in figure viewerPowerPoint Membership functions for the HFPF inputs a Direct-axis voltage b Derivative of frequency c HFPF surface Another situation is observed when the direct-axis voltage is related to the linguistics labels VNM (voltage NM) or VPM (voltage PM), and the frequency derivative is dWS (frequency derivative small). In this case, the fuzzy output is small (S) in order to avoid the injection of disturbances primarily at the load connection moment. Load connection can cause voltage variation, and is not necessarily an islanding situation. In this context, if the voltage is VNM and the frequency derivative is dWNL (frequency derivative NL) or dWPL (frequency derivative PL), the fuzzy output is defined as negative medium, because the voltage deviation is medium and the frequency is large, which means that the system is moving towards exceeding the voltage lower limits. In the same way, if the voltage is VPM and the frequency derivative is dWNL or dWPL, the fuzzy output is defined as positive medium, because the voltage deviation is medium and the frequency is large, which means that the system is moving towards exceeding the voltage upper limits. The same idea is applied when the voltage is VPL (voltage PL) or VNL (voltage NL) as presented in Table 1. Table 1. Fuzzy rules Derivative of frequency dWNL dWS dWPL voltage, p.u. VNL NL NM NL VNM NM S NM RV S S S VPM PM S PM VPL PL PM PL To illustrate the rules, the surface of the HFPF method is graphically represented in Fig. 5c. 5 Simulation results To evaluate the injected disturbances using the classical active method and the proposed method, three types of load connections are performed: no controlled rectifier, RLC (resistance, inductance and capacitance) load and an induction motor. These load types are typically found in the DG system operation and have been used to demonstrate the dynamic anti-islanding methods when the DG system is submitted to load connections. The levels of injected disturbances are analysed by calculating the variation of the current reference by the action of the anti-islanding algorithms. The non-linear load, RLC, and motor are connected in 2, 3 and 4 s in the simulation time, respectively. Fig. 6 shows the results of injected disturbances by the VPF and HFPF methods. Fig. 6a analyses the effect of the VPF method in the direct-axis of the DG system control. The results demonstrate that the current reference achieves more than 6% of deviation for RLC load connection (worst case). For the motor start-up the calculated peak rises up to 4%, whereas for the non-controlled rectifier load connection the current error cannot be observed in the direct-axis. Fig. 6Open in figure viewerPowerPoint Injection disturbances tests for different PF gains and types of load a VPF output on the quadrature-axis b HFPF output on the direct-axis Considering that the frequencies of the washout and low-pass filter are adjusted at 5 and 10 Hz, respectively, the injection of disturbances is proportional to the PF gain as observed in Fig. 6a. The disturbance of the HFPF is represented in Fig. 6b, and is analysed for the same group of loads applied to the VPF. The findings show that the disturbances are approximately zero (0.5 × 10−16). Compared with the active method, HFPF eliminates the disturbances and makes the grid-connected operation less sensitive to the issue of power quality. The very low disturbances are achieved through a dynamic fuzzy gain, which is obtained by the membership sets proposed. The detection time is also analysed and the simulations are carried out with the RLC load consuming the total power produced by the DG, with Qf equal to 2.5 and resonant frequency equal to 60 Hz. In this context, the islanding takes place in 3 s in the simulation period. For the VPF method shown in Fig. 7a, the result exhibits the voltage variation at the PCC for the PF gain equal to 8. As can be observed in Fig. 7b, for gains smaller than 8 the detection time is not consistent with the standards [1, 2], which demonstrates the requirement for high PF gains to acquire fast detection. Fig. 7Open in figure viewerPowerPoint Islanding detection for Qf = 2.5 a VPF for kVPF = 8 b VPF for kVPF = 5 c HFPF detection d HFPF output With respect to the HFPF, Fig. 7c illustrates a fast voltage variation until the islanding situation is detected in 29.2 ms. The voltage variation is a result of the dynamic fuzzy output as seen in Fig. 7d. Compared with the VPF active method, HFPF reduces the detection time by 62%. In addition, none of the proposed fuzzy techniques require PF gains nor filters to be designed, which make them attractive for islanding detection. 6 Experimental results The DG system and the anti-islanding methods presented in this paper were experimentally tested in laboratory. The anti-islanding methods, PLL and the controller are implemented using a Texas Instruments TMS320F28335 digital signal processor (DSP). Figs. 8a and b show the photo and a diagram of the experimental laboratory setup, respectively. The DG and the grid are tested under the conditions presented in Table 2. Table 2. Experimental setup parameters Parameters Value rated line-to-line grid voltage 110 V frequency 60 Hz DC supply voltage 300 V switching frequency 10 kHz transformer ratio 2 : 1 inverter filter inductance (Ldg) 2 mH inverter filter capacitance (Cf) 10 µF capacitance resistance (Rf) 10 Ω grid filter inductance (Lgrid) 0.5 mH output capacity 1 kW Fig. 8Open in figure viewerPowerPoint Experimental setup a Photo of the experimental setup b Diagram of the experimental setup For all experimental results, the DG is started up in the islanding operation mode, then synchronised and connected to the grid. In contrast to the simulation results, when the DG detects the islanding condition, the control is switched to the island mode. The islanding condition is performed manually by means of the 'Manual' switch, represented in Fig. 8b. Finally, the load consumes the total power produced by the DG source. With the DG and grid synchronised, the DG control is changed by means of a switched 'control' to operate in grid-connected mode as shown in Fig. 9. This figure also shows that the transition from island to the grid-connected mode is smooth and has no voltage oscillations. Fig. 9Open in figure viewerPowerPoint Instant of the connection in the grid. Ch1: current phase "a" of DG (10 A/div). Ch2: current phase "a" of grid (5 A/div). Ch3: voltage phase "a" of DG (40 V/div). Ch4: voltage phase "a" of grid (40 V/div). For the VPF method, the washout and low-pass filter frequencies are set at 5 and 10 Hz, respectively, and two PF gains are used to illustrate the dynamic of the method. For PF gain equal to 1.8, the detection time is not according to standard requirements [1, 2] because the voltage oscillation caused by the VPF is insufficient to detect the islanding condition as shown in Fig. 10a. However, by increasing the PF gain from 1.8 to 2.0 the detection time is reduced to 5.5 cycles as shown in Fig. 10b. The experimental results indicate that the detection time is directly related to the PF gain, i.e. a high PF gain must be set to ensure fast detection time. Fig. 10Open in figure viewerPowerPoint Islanding tests using VPF method. Ch1: current phase "a" of DG (10 A/div). Ch2: current phase "a" of grid (5 A/div). Ch3: relay (20 V/div). Ch4: voltage phase "a" of grid (40 V/div) a kVPF = 1.8 b kVPF = 2.0 HFPF methods are implemented using a matrix with 400 points to represent the fuzzy PF output in the TMS320F28335. The algorithm reads the inputs (ωpll and vd) and calculates the matrix indices to generate the fuzzy PF output. Using HFPF the time detection is approximately 1.25 cycles, or 77.3% less than the time needed for the VPF method to detect the contingency as shown in Fig. 11a. The frequency and rms line voltage of the experimental result are represented in Figs. 11b and c, respectively. It is observed that as the voltage and frequency variation increase, the HFPF output is increased towards the voltage variation in order to detect the islanding condition. Similarly, as the voltage is returning to its nominal value faster than the frequency, the HFPF is decreased following the voltage deviation as shown in Fig. 11d. Fig. 11Open in figure viewerPowerPoint Islanding test using the HFPF method. Ch1: current phase "a" of DG (10 A/div). Ch2: current phase "a" of grid (5 A/div). Ch3: relay (20 V/div). Ch4: voltage phase "a" of grid (40 V/div) a Islanding detection for HFPF b Frequency deviation (experimental) c RMS line voltage (experimental) d HFPF output (experimental) Fig. 12 illustrates the dynamic fuzzy PF gain for the experimental result shown in Fig. 11a. The high gains are achieved when voltage and frequency deviations are close to the upper or lower limits. Fig. 12Open in figure viewerPowerPoint HFPF gain (experimental) 7 Conclusion This paper developed a HFPF anti-islanding method based on fuzzy inference rules. In the classical method, to obtain an adequate detection time requires high PF gains and consequently high injection disturbances on the DG control. As shown in the results, the proposed method eliminates the injection disturbances and also does not require setting a PF gain and frequency filters. The simulation results show that, for the RLC load, the VPF method achieves more than 6% of deviation in the direct-axis of the DG control. While for HFPF the disturbances are approximately zero (0.5 × 10−16). The experimental results show that the proposed method reduces the detection time by 77.3% in relation to the classical method and illustrated by the dynamic fuzzy PF gain, which is responsible for the fast detection and low injected disturbances. 8 References 1 IEEE Std. 929-2000: ' IEEE recommended practice for utility interface of photovoltaic (PV) systems', 2000 2 IEEE Std. 1547.1-2005: ' IEEE standard conformance test procedures for equipment interconnecting distributed resources with electric power systems', 2005 3Yu B., Matsui M., and Yu G.: 'A review of current anti-islanding methods for photovoltaic power system', Sol. Energy, 2010, 84, (5), pp. 745– 754 (doi: http://doi.org/10.1016/j.solener.2010.01.018) 4Ye Z., Walling R., and Garces L. et al.: ' Study and development of anti-islanding control for grid-connected inverters' ( National Renewable Energy Laboratory, 2004) 5Seul-Ki K., Jin-Hong J., and Heung-Kwan C. et al.: 'Voltage shift acceleration control for anti-islanding of distributed generation inverters', IEEE Trans. Power Deliv., 2011, 26, (4), pp. 2223– 2234 (doi: http://doi.org/10.1109/TPWRD.2011.2132153) 6Liu F., Kang Y., and Zhang Y. et al.: 'Improved SMS islanding detection method for grid-connected converters', IET Renew. Power Gener., 2010, 4, (1), pp. 36– 42 (doi: http://doi.org/10.1049/iet-rpg.2009.0019) 7Estébanez E.J., Moreno V.M., and Pigazo A. et al.: 'Performance evaluation of active islanding-detection algorithms in distributed-generation photovoltaic systems: two inverters case', IEEE Trans. Ind. Electron., 2011, 58, (4), pp. 1185– 1193 (doi: http://doi.org/10.1109/TIE.2010.2044132) 8Ye Z., Dame M., and Kroposki B.: ' Grid-connected inverter anti-islanding test results for general electric inverter-based interconnection technology' ( National Renewable Energy Laboratory, 2005) 9Chung-Chuan H., and Yu-Chun C.: 'Active anti-islanding detection based on pulse current injection for distributed generation systems', IET Power Electron., 2013, 6, (8), pp. 1658– 1667 (doi: http://doi.org/10.1049/iet-pel.2012.0542) 10Mohamed Y.A.R.I., Zeineldin H.H., and Salama M.M.A. et al.: 'Seamless formation and robust control of distributed generation microgrids via direct voltage control and optimized dynamic power sharing', IEEE Trans. Power Electron., 2012, 27, (3), pp. 1283– 1294 (doi: http://doi.org/10.1109/TPEL.2011.2164939) 11Pengwei D., Zhihong Y., and Aponte E.E. et al.: 'Positive-feedback-based active anti-islanding schemes for inverter-based distributed generators: Basic principle, design guideline and performance analysis', IEEE Trans. Power Electron., 2010, 25, (12), pp. 2941– 2948 (doi: http://doi.org/10.1109/TPEL.2010.2057446) 12Bower W., and Ropp M.: ' Evaluation of islanding detection methods for utility-interactive inverters in photovoltaic systems' ( SANDIA REPORT, 2002) 13Seul-Ki K., Jin-Hong J., and Jong-Bo A. et al.: 'Frequency-shift acceleration control for anti-islanding of a distributed-generation inverter', IEEE Trans. Ind. Electron., 2010, 57, (2), pp. 494– 504 (doi: http://doi.org/10.1109/TIE.2009.2036028) 14Yafaoui A., Bin W., and Kouro S.: 'Improved active frequency drift anti-islanding detection method for grid connected photovoltaic systems', IEEE Trans. Power Electron., 2012, 27, (5), pp. 2367– 2375 (doi: http://doi.org/10.1109/TPEL.2011.2171997) 15Vahedi H., and Karrari M.: 'Adaptive fuzzy sandia frequency-shift method for islanding protection of inverter-based distributed generation', IEEE Trans. Power Deliv., 2013, 28, (1), pp. 84– 92 (doi: http://doi.org/10.1109/TPWRD.2012.2219628) 16Liu F., Zhang Y., and Xue M. et al.: 'Investigation and evaluation of active frequency drifting methods in multiple grid-connected inverters', IET Power Electron., 2012, 5, (4), pp. 485– 492 (doi: http://doi.org/10.1049/iet-pel.2011.0158) 17Jae-Hyung K., Jun-Gu K., and Young-Hyok J. et al.: 'An islanding detection method for a grid-connected system based on the Goertzel algorithm', IEEE Trans. Power Electron., 2011, 26, (4), pp. 1049– 1055 (doi: http://doi.org/10.1109/TPEL.2011.2107751) 18Velasco D., Trujillo C., and Garcera G. et al.: 'An active anti-islanding method based on phase-pll perturbation', IEEE Trans. Power Electron., 2011, 26, (4), pp. 1056– 1066 (doi: http://doi.org/10.1109/TPEL.2010.2089643) 19Ye Z., Dehong X., and Ning H. et al.: 'A novel RPV (reactive-power-variation) anti-islanding method based on adapted reactive power perturbation', IEEE Trans. Power Electron., 2013, 28, (11), pp. 4998– 5012 (doi: http://doi.org/10.1109/TPEL.2013.2245918) 20Trindade F.C.L., do Nascimento K.V., and Vieira J.C.M.: 'Investigation on voltage sags caused by DG anti-islanding protection', IEEE Trans. Power Deliv., 2013, 28, (2), pp. 972– 980 (doi: http://doi.org/10.1109/TPWRD.2012.2237419) 21Jantzen J.: ' Foundations of fuzzy control: A practical approach' ( Wiley, 2007, 1st edn.) 22Weisheng C., Licheng J., and Ruihong L. et al.: 'Adaptive backstepping fuzzy control for nonlinearly parameterized systems with periodic disturbances', IEEE Trans. Fuzzy Syst., 2010, 18, (4), pp. 674– 685 (doi: http://doi.org/10.1109/TFUZZ.2010.2046329) Citing Literature Volume9, Issue3March 2016Pages 512-518 FiguresReferencesRelatedInformation
Referência(s)