- Open Access
- Total Downloads : 232
- Authors : B. Ashok Kumar, C. Kamal Basha
- Paper ID : IJERTV3IS120219
- Volume & Issue : Volume 03, Issue 12 (December 2014)
- Published (First Online): 09-12-2014
- ISSN (Online) : 2278-0181
- Publisher Name : IJERT
- License: This work is licensed under a Creative Commons Attribution 4.0 International License
Design of AACC for Cascade Dc Power System to Improve the Stability with Fuzzylogic Controller
1B. Ashok Kumar,
1 PG Student,
Department of Electrical and Electronics Engineering, Madanapalle Institute of Technology and Science, Madanapalle, India
2C. Kamal Basha
2 Associate Professor,
Department of Electrical and Electronics Engineering, Madanapalle Institute of Technology and Science, Madanapalle, India.
Abstract: The fundamental configuration of the dc distributed power system (DPS) is cascade connection of converters. In cascaded system impedance problem occurs due to independent design of converters and may make the system unstable. In the previous strategies to minimize impedance problem they have adopted alteration of the converters at both source and load end. In this paper a new method of mitigating impedance problem which is in parallel with the cascaded system's middle bus and it just needs to know the bus voltage with all the existing sub-systems remained same. The AACC acts as a proportional bus capacitor to mitigate the converter impedance at source side. In this method FUZZYLOGIC CONTROLLER is used to improve the stability. Due to the bus capacitor of the loss of force is minimized. The dynamic response of the system is also enhanced compared to normal capacitor. Since this system requires no electrolytic condenser the cascaded system's lifetime is drawn out. In this paper, a 480Wcascaded system with fuzzy logic controller is built and performance is studied. And the results are verified by using MATLAB/SIMULINK.
The flexible system design, good energy conversion efficiency, and tremendous power deliver capability of dc DPS make it presence in many areas such as Defense, Research and Industries -. Modularity design is one of the good attributes of dc DPS in which every subsystem is initially designed exclusively. System's advancement cycles and expenses are successfully reduced with modulization model of DPS . It was additionally called attention to that if both the converters at sending end and load are stable exclusively, Zo<Zin in the Varity of frequencies, the cascaded system stability will be assured. The method is alleged Center brook model. Hence, different impedance criteria going for a more precise and down to earth expectation of the subsystem collaboration had been created in the most recent two decades .
Fig.1. Cascaded power supply system
Answers for tackling the instability issue are suggested and can be extensively ordered into passive and active strategies. Passive systems utilize passive parts like R, L and C to enhance framework stability. A R-load was added to alter dynamic qualities of load and in this manner enhancing framework performance . Both R-C and R-L dampers were familiar with minimization of the output impedance peak of the converter at source end, which ensures Zo<Zin in the whole range of frequencies.. Active techniques for balancing out the framework are focused around adjusting the control of the converters at source and load end or including a power cradle amid the sending and load subsystems. The previous methodology, then again, is typically intricate in usage and in some cases clashing with other control goals. In the last method, the power cushion is joined in arrangement in the middle of the sub-systems, and it influences the Z connection that may not be adequate in a few applications .
All the previously stated arrangements are required to change the inside configuration, which also includes the principle AND/OR control circuits, of the sub- systems of dc DPS, prompting upgrade of the sub-systems that have as of now been separately composed. This negates with the destination of the measured quality outline of dc DPS and expands the framework's improvement cycles.
In this paper, an Adaptive Active Capacitor Converter (AACC) FUZZY logic controller is arranged and shunted with the middle bus of the cascaded system is presented . The difference between AACC and an adaptive bus capacitor is the output power cascaded-system that decreases the Zo of the converter at source end to abstain from associating with Zin of the converter at load side. As an issue, the cascaded-system gets to be steady. The AACC just needs to distinguish the moderate Vbus with no alternation of the current sub-systems; consequently, it acts as an issue stabilizer for dc DPS. In the mean time, the proportional condenser of the AACC is suitable, guaranteeing a negligible extra power misfortune and a finer element reaction of the system than that utilizing a detached condenser. Moreover, as no condenser of electrolyte type is needed in the AACC, the cascaded systems life-time is delayed. In the present paper we are utilizing a fuzzy logic controller rather than pi controller to enhance the stability.
INSTABILITY PROBLEM OF CASCADED SYSTEM The cascaded-system shown in Figures (1) and (2)
will be unstable if Zo is converged with Zin and fcs is short of what fc_L, For this situation, the oscillating frequency is fc-s which is independent of system's power. The above examination shows that Zo is independent of Po, and when f
< fc_L, the extent of Zin is contrarily relative to Po. In this way, the cascaded-system is destined to be un-stable at rated load in light of the fact that Zin is negligible and effortlessly crossed with Zo at this condition. Note-that the above concluded points are general and material to all DC- DC converters .
Fig.2.Graphical representation of effect of Impedance in cascaded system
response is also enhanced. For the cascaded-systems with different load converters also AACC is applicable. Fin this case the AACC has the similar process guideline as structure of Figure.4, which reduces source converter's Zo lower than impedance of the aggregate information of the various converters at load end .
Fig. 4. Cascaded system with AACC
In cascaded system AACC with PI controller having distortions and the settling time is more, to overcome this we are adopting FUZZY logic controller instead of PI controller.
PROPOSED AACC WITH FUZZY LOGIC CONTROLLER
A. DESIGN OF AACC OUTPUT FILTER CAPACITOR CA
In AACC, when switching harmonics are ignored for the bus voltage and current of the cascaded systems and can be expressed as
ia (t) = Cbus(Dvbus/dt) (1)
There is nothing more attractive than an aggregate
vbus Vbus Vbus _ allow sint
detachment in the middle of Zo and Zin to guarantee to the cascaded-system is steady. Since source converter s output capacitor is defiantly |Zo-peak|, one natural route is to decrease the source converter' output impedance by including a halfway transport capacitor Cbus to the cascaded system, as demonstrated in Figure3. Here, Cbus can be dealt
Where Vbus is the average value of vbus, and is the angular frequency of the ripple in Vbus, i.e.,
The instantaneous input power of AACC can be obtained by (1) and (2), i.e.,
Pa (t) vbusia =
with as an extra output filter condenser of the sending end
converter, and the proportionate L-C output Z model of source converter with Cbus.
Fig .3. Intermediate bus system of cascaded system
A bigger Cbus brings about a littler band-width of the sending-end converter that is as of now separately planned, prompting a poor dynamic execution. Since the obliged Cbus is moderately expansive, it unavoidably embraces electrolytic capacito, demonstrating a critical lessening of the lifetime. To conquer this poor dynamic execution we are utilizing Versatile Dynamic Capacitor Converter with PI controller. The basic components of AACC are QA1 and QA2 switches, inductor La, and condenser Ca. By varying value of La properly, transport area of AACC will exhibit an adaptively fluctuating Cbus which guarantees solidness of the cascaded-system in the whole load run and active
bus bus _ allow bus bus _ allow
Generally Vbus _ allow Vbus ; thus, (3) can be expressed as
Pa (t) Vbus Vbus _ allow cost . (4)
From the equations (1) and (4), the instantaneous Pin, iA, and Va waveforms of AACC are shown in Figure. 5. It can be derived that discharging of Ca starts from Tos /4 to 3Tos /4, and Va decreases and charging of Ca starts from 3Tos /4 to 5Tos /4, and Va increase. Due to this peak and least value of Va occur at Tos/4 and 3Tos /4 respectively .
The power charge of Ca from 3Tos /4 to 5Tos /4 is
We keep Vamin = Vbus and Vbus _= at 1%*Vbus, and the normalized Vamax and Vo_dc with base of Vbus are
Va max 4%
4 V a max 1
VbusCbus Vbus _ allow costdt
V 1 1% 1
V adc (12)
4 adc V
2 C 4
2V C V sin2 t
We know that C* is normalize C with bottom of C
a a bus
bus bus bus _ allow
a a bus
From the equations (11) and (12), Va* (max) and
Va*(dc) as function of C *
are drawn in Figure. 8, Where
Fig. 5. Waveforms of instantaneous Pin, La, and Va of the AACC.
And Ea (t) is given by
Vamax increase with reduction of Ca. To take layer capacitors or clay capacitors in the place of electrolytic condensers, the charge of Ca has to be small enough. This will give elevated Vamax which results in elevated voltage pressure on Qa1 and Qa2. Hence Ca have to be chosen at rated weight as it is the most horrible situation for the cascaded-system and having necessary peak value Ca.
E (t) 1 C v2 (t) 1 C V 2
a 2 a a 2 a a min
Where Vamin = minimum voltage of the condenser Ca, substituting (5) in (6) gives
1 C v2 (t) V 2 2V C V
Fig.6. Graphs of V
* and V * of Condenser C
2 a a a min
bus bus bus _ allow
4V C V
SLECTION OF QA1 AND QA2
bus bus bus _ allow
From the Figure (4), the peak value of Qa1 and Qa2 is the voltage-stress VA,
Substituting t=5Tos/4 into (14), the Peak voltage of
i.e. VQa1 VQa2 Va max
condenser Ca be able to be given like
The rated current of La of Qa1 and Qa2 is the current-stress,
4VbusCbus Vbus _ allow V 2
and be able to be find out from (1) and (2),
i.e. IQa1 IQa2 Cbus max Vbusallow
at rated-load Cbus
is the desired value. The
The average voltage of Ca can be approximated as
power devices for Qa1 and Qa2 are selected based on (13)
(Va min Va max )
4VbusCbus Vbus _ allow V 2
a min )
Two things have to be taken into account when selecting the La value. One is ensuring the current through
inductor is able of track the I ref, and inductor current will keeps small for second ripple. Here the AACCs inductor
For assurance of good process of AACC, the Cas immediate voltage has to be higher than the Vin of AACC, which is given as
va (t) Vbus
current Ia requires to path the oscillating wave, whose oscillating frequency is the cutoff-frequency of voltage loop gain of sending end converter. The oscillation frequency fcs is more than switching frequency fsa of AACC. Assurance of tracking speed of Ia and will be enough to prefer the value of La by single concern specified
to the inductor current ripple. As the two control switches of AACC function in a balancing mode, the AACC is in service in incessant current conduction form.
Hence the duty cycle of Qa1 is
dQa1 (t) = 1 Vbus. Va (t) (15) When QA1 is turned ON and QA2 is turned OFF, the voltage across La is Vbus. This voltage causes Ia to increase. The ripple of ia can be expressed as
iLa= Vbus LaÂ· dQa1 (t) Â· fsa. (16) Substituting equation (22) in equation (23) results as
Little, MEDIUM, and Enormous. The control rules of the anticipated controller are resolved from the perspective of commonsense system operation and by experimentation and are demonstrated in Table-I.
Table-I: rule base of FLC
La (Va (t) VbusVa (t)iLa fsa
C. FUZZY INFERENCE
From the above equation the oscillation period which La varies with Va (t).
FUZZY LOGIC CONTROLLER
The fuzzy logic controller (FLC)  dissimilar to the crispy-logic in the Boolean theory that uses just two logic levels (0 to 1), is an extension of logic that concedes unending logic levels (from 0 to 1), to take care of an issue that has vulnerabilities or uncertain circumstances. Once more, a fuzzy control is a methodology control that is focused around fuzzy logic and is ordinarily portrayed by "IF-THEN" runs the show. The configuration of the proposed FLC is portrayed in the accompanying.
The Fuzzification methodology comprises of discovering fitting membership functions to depict crisp information. For the outline of the anticipated FLC, departure of pace of synchronous generator and firing angle of thyristors are chosen as the input & output, individually. Triangular membership functions are indicated in Figure. 5, in which the phonetic variables N, Z, and P stand for negative, zero, and positive, separately. The membership functions are controlled by the experimentation method so as to get the better framework execution. The mathematical statement of the triangular membership capacity used to focus the evaluation of membership in which the estimation of evaluation of membership, " is the width, " is the direction of the time when the evaluation of membership is 1 and " is the estimation of the input variable.
Fig.7. Membership function
FUZZY RULE BASE
The heart of a fuzzy controller is the rule base as the control technique is used for controlling the closed-loop system is put away as an issue of control rules. The particular gimmick of the proposed fuzzy controller is its extremely basic configuration containing two variables. The utilization of the singleinput singleoutput (SISO) variable makes the fuzzy controller exceptionally direct. Fig. 7 demonstrates the membership functions for the output variable comprising of three singleton fuzzy sets
The essential working principle of the deduction motor is it induces, i.e. it finds a sensible solution. Really, the surmising motor is a system which utilizes the rule base and the input data of the controller to make the determination. The finish of the derivation motor is the fuzzy output of the controller, which therefore turns into the input to the defuzzification intrface. For the derivation system of the anticipated FLC, Mamdani's-technique  is used. A fuzzy rule normally has an IF-THEN arrangement as takes after: IF IS And IS THEN where and are fuzzy input variables, is the fuzzy output variable, is the rule number, is the aggregate number of rules and are fuzzy subsets in the universe of talks , and , individually. Consequently, as indicated by Mamdani, the level of similarity, of every fuzzy rule is as per the following: where and are the estimations of the evaluation of membership.
In the previous operation, the fuzzy finish of the induction motor is defuzzified, i.e. it is changed to a crisp signal. The last signal is the last result of the FLC, which is obviously, the crisp control signal to the procedure. The inside of-range system is the most well-known and rather basic defuzzification strategy which is actualized to focus the yield crispy worth. This is given by the accompanying statement: where is the crispy yield work and is now characterized in the past area. To perceive how compelling the fuzzy controlled Fell system in enhancing the stability and its execution is contrasted with that of a traditional PI controlled Fell System plan.
SIMULATION RESULTS AND DISCUSSIONS:
Design of AACC for a cascaded-system is discussed here. The system is consisted of two phase- shifted full-converters of full-bridged type as shown in figure.9. The parameters of both the converters at sending end and load can be derived from the table.
Table-II: parameters of source converter Load converter
Winding turn ratio of Tr1
Table-III: parameters of load converter
Winding turns ratio Tr2
CASCADED SYSTEM WITH AACC
The simulation model of the Cascaded system with AACC of PI controller is shown below figure. The input 360V is given the source converter, and the transformer step downs this voltage as 5:1 ratio of 72V. And the voltage is minimized as 48V with PI controller. This output voltage is given as the input to the load converter. The load converter is also step down the voltage as 3:1 ratio and the output is 12V. The resistive load of 480 is connected at
the end of load converter which gives load current and will be less compares with the passive method.
CASCADED SYSTEM WITH AACC AND FUZZY LOGIC CONTROLLER
The simulation model of the Cascaded system with AACC of FUZZY controller is shown below figure. The input 360V is given the source converter, and the transformer step downs this voltage as 5:1 ratio of 72V. And the voltage is minimized as 48V with PI controller. This output voltage is given as the input to the load converter. The load converter is also step down the voltage as 3:1 ratio and the output is 12V. The resistive load of 480 is connected at the end of load converter which gives load current and will be less compares with the AACC of PI controller method.
Fig.8. Cascaded system with AACC and PI controller
Fig.9. Cascaded system with AACC and FUZZY controller
Fig.10. output current of cascaded system with AACC and PI controller
Fig.11. Bus Voltage of cascaded system with AACC and PI controller
Fig.12. Vo of cascaded-system with AACC and PI controller
Fig.13. Io of cascaded-system with AACC and FUZZY controller
Fig.14. Vbus of cascaded-system with AACC and FUZZY controller
Fig.15. Vo of cascaded -system with AACC and FUZZY controller
The instability problem in dc DPS can be easily solved by this new method which is most efficient method. It provides AACC as an equivalent adaptive bus condenser changed in accordance with cascaded system output power and mitigates impedance interaction in cascaded-system by mitigating the Zo of the sending end converter. Based on the stability margin of the cascaded-system, The AACC produces more energy to give large equivalent capacitor when the cascaded-system oscillations are high. And vice- versa happens when the oscillations are less. Oscillating voltage is within the allowable scope and AACC will be shut down when there is no oscillation. Hence the stability of the cascaded-system can be ensured with the use of AACC. And also no electrolytic condenser is required
TYPE SETTLING TIME
with aacc 0.09
with aacc of fuzzy controller 0.02
The settling time of load current for AACC and FUZZY Logic controller is less when compares with cascaded system with AACC and PI controller.
Xin Zhang, Xinbo Ruan, Senior Member, IEEE, Hyok Kim, and Chi K. Tse, Fellow, Adaptive Active Capacitor Converter for Improving Stability of Cascaded DC Power Supply System, IEEE Trans. Power Electron., vol. 28, no. 4, April 2013.
F. Blaabjerg, A. Consoli, J. A. Ferreira, and J. D. van Wyk,
The future of electronic power processing and conversion, IEEE Trans. Power Electron., vol. 20, no. 3, pp. 715720, May 2005.
J. D. van Wyk and F. C. Lee, Power electronics technologyStatus and future, in Proc. CPES, 1999, pp. 6170.
K. T.Kornegay, Design issues in power electronic building block (PEBB) system integration, in Proc. IEEE VLSI, Apr. 1998, pp. 4852.
T. Ericsen, Power electronic building blocksA systematic approach to power electronics, in Proc. IEEE Power Eng. Soc. Summer Meeting, 2000, pp. 12161218.
D. Boroyevich, I. CvetkoviÂ´c, D. Dong, R. Burgos, F.Wang, and F. C. Lee, Future electronic power distribution systemsA contemplative view, in Proc. IEEE OPTIM, May 2010, pp. 13691380.
S. Luo, A review of distributed power systems part I: DC distributed power system, IEEE Aerosp. Electron Syst. Mag., vol. 20, no. 8, pp. 5 16, Aug. 2005.
R. D. Middlebrook, Input filter considerations in design and application of switching regulators, in Proc. IEEE IAS, 1979, pp. 366382.
X. Wang, D. Vilathgamuwa, and S. Choi, Decoupling load and power system dynamics to improve system stability, in Proc. IEEE PEDS, 2005, pp. 268273.
M. Cespedes, L. Xing, and J. Sun, Constant-power load system stabilization by passive damping, IEEE Trans. Power Electron. Lett, vol. 26, no. 7, pp. 18321836, Jul. 2011.
S. Abe, M. Hirokawa, T. Zaitsu, and T. Ninomiya, Optimal bus capacitance design for system stability in onboard distributed power architecture, in Proc. EPE-PEMC, 2008, pp. 393399.
X. Feng, J. Liu, and F. C. Lee, Impedance specifications for stable DC distributed power systems, IEEE Trans. Power Electron., vol. 17, no. 2, pp. 157162, Mar. 2002.
Mohd. Hasan Ali, Member, IEEE, Toshiaki Murata, and Junji Tamura, Senior Member, A Fuzzy Logic-Controlled Superconducting Magnetic Energy Storage for Transient Stability Augmentation, IEEE Trans. Control system Technology, vol. 15.no. 1 Jan 2007.