(IJEECS) International Journal of Electrical, Electronics and Computer Systems. Vol: 12 Issue: 01, January 2013
Voltage Improvement and Power Loss Reduction for a Typical Sub-transmission region Using Optimally Placed and Sized Distributed Generation H. Musa, S.S. Adamu Department of Electrical Engineering Bayero University, Kano Nigeria Abstract— Recent unbundling of Nigerian power holding companies into small units have created opportunities for involvement of independent power generation companies. Remarkable achievements and improvements are expected in power quality especially in the northern part of the country where power quality low is and serious power losses along the transmission and distribution lines of the power system are the norm. This paper proposes using particle swarm optimization (PSO) for allocation and sizing of Distributed Generation (DG) in Shiroro complex that supplies the northern part of Nigeria based on National Control Centre’s allocation using real data of the network. The study involves simulations for various load conditions in a power flow solutions. Based on the results the presence of DG in the network has greatly improved the voltage profile as well as power loss reduction in the region. Keywords— Distributed Generation, Power losses, Voltage profile, Particle Swarm Optimization, Placement and sizing
I. INTRODUCTION The issue of DG placement and sizing is of great importance in the present days of energy generation crisis, as its installation at non optimal location tends to increase system losses. The increase in these losses means increase in costs and hence has a negative effect on the planner’s desired objective [1]. For this reason it is necessary to employ optimization techniques that can find the best location for DG installation. Research outputs have shown that improper allocation and sizing can result in high power losses and can jeopardize the system performance or results in instability [2]. Many research and investigation have been done in the area of optimal DG unit placement for loss minimization in distribution system. In [3] the authors have shown the potential of DG with power electronic interface to provide ancillary services such as reactive power, voltage sag compensation and harmonic filtering. The work has demonstrated the ability of DG to compensate voltage sag that can occur during fault in the power system. However, the work could not quantify the amount of power loss reduction due to DG installation in the network. Also the authors in [4] have used convolution technique to evaluate the probability of quantifying the benefit of voltage profile improvement involving wind turbine generation. According to [5] noniterative analytical approaches can be used to optimally place DG in both radial and meshed systems for the minimization of power losses. All the approaches above are mathematical in nature and are found to be complex. A similar work in [6] has proposed an approach for the computation of annual energy losses for different penetration levels of DG in a distribution
network. The authors have found that when DG units are dispersed along the network feeders the high losses are reduced up to a certain value of the DG capacity and beyond which the loss tend to increase. This idea was used to optimize DG capacity for minimum power loss. There are many frequently used methods that gives fast results is the soft computing techniques in DG sizing optimizations. One of them is the optimal sizing of DG handled as a security constrained optimization problem solved by genetic algorithm in [7] where sitting is done by using sensitivity analysis of power flow equations. Though the GA methods have been employed successfully to solve complex optimization problems, recent research has identified some deficiencies in GA performance. Many other methods uses heuristic techniques for DG placement in Distribution networks among which are; [8] that employs New Hybrid search approach, and in [9] conventional approach was adopted using simple genetic algorithm and Binary PSO. However, due to the complexity of the GA and the failure to find global solution of other optimization methods this paper uses particle swarm optimization algorithm (PSO) for optimal placement and sizing of DG units in order to improve voltage and minimize power losses in a typical Nigerian grid that supplies the northern part of the country. II. BRIEF OVERVIEW OF NORTHERN NIGERIAN GRID SYSTEM (SHIRORO COMPLEX) The Shiroro power complex is a region of the national grid that generates and link north part of Nigeria with transmits power at 330 KV lines of the grid. The supply covers all the major cities of the region which are; Kaduna, Jos, Kano, Gombe, Birnin Kebbi and other towns. In this network the generation sources are three hydro generating stations located at Kainji, Jebba and Shiroro. The grid injection to this area is also centrally controlled by National Control Center (NCC) at Oshogbo. The single line diagram is as shown in Fig.1 The stand by Generators in Abuja which are meant to generate a certain amount of power during off grid operation are not considered in the network. The maximum system load for the complex is assumed to be 1500MW. The minimum load demand is also assumed to be 60% of maximum demand. These figures are based on data for off-peak demand as reported in [10]. For both maximum and minimum demands the power factor is 0.9 lagging.
©IJEECS
(IJEECS) International Journal of Electrical, Electronics and Computer Systems. Vol: 12 Issue: 01, January 2013
Plosses new i 1 ( I i JI DG ) 2 Ri
Grid feed from Oshogbo
n
~ 1
Jebba TS
Plosses I R 2JIiIDGRi J.IDG Ri new i1 i i n
Jebba
~
2
GS
Shiroro
~
4
~
3
Kainji
DG1
9
2
(5)
PLRi Plossesnew Plosses
Kaduna
5
2
Where J=1 is for feeder with DG, otherwise J=0. Therefore, the PLR value with DG connection at bus ‘i’ is:
GS Birnin Kebbi
( 4)
(6)
Jos
~
Kano
PLR i
6
n i 1
( 2 JI i I DG JI DG ) R i 2
(7 )
7
~
8
Gombe
~
DG2
Fig. 1. One-line diagram for Shiroro complex with DGs allocated DG3 Fig 1. One-line diagram for Shiroro complex with DGs allocated
III. METHODOLOGY This proposed method places the DG based on DG current usage maximization and reduction of current sourced from grid for power loss improvement. The intention is to introduce an approximation method that will identify the size of DGs and the optimal location based on maximization of what is called power loss reduction (PLR) value. Generally losses associated with active current in single-source radial networks cannot be minimized as the source has to supply all the active power. However, if DGs are placed in the network, the active branch current sourced from the single source is reduced due to the active current from the DG that balances the demand. The consequences are loss reduction due to current reduction sourced from the single source (main feeder) and improved voltage profile. The power loss reduction is therefore the difference between loss associated with active current when no DG connection in the network and when DG is connected. Thus the value of the power losses and the DG current that gives maximum loss reduction are as in equations (1) and (2) respectively
Plosses i 1 I i R i n
Equation (7) is the Objective function that maximizes power losses reduction value in the network. The optimization method that uses PSO is developed based on a computational technique as extensively discussed in [11]. In this work the proposed algorithm as shown in Fig. 2 is a revised version of that was used in [12]. The real power losses was randomly generating in an initial population of particles with random positions and velocities in the solution space. Each of the Particles is subjected to constraints specified while running the power flow in order to calculate the power losses. Start
Input network Data and make initial setting for power flow and PSO Determine fitness of (power loss) for each particle Run power flow without DG and calculate the system total load demand Construct
Calculate the DG size for all the buses for i = 2 to n
Is Saving < 1kw ?
Gbest
is
i 1 i 1 ymax ymin
No
?
Run load flow again with load adjusted
(1)
&
Calculate the new position and restively
Determine loss saving and allocate DG based Bus with maximum saving
Yes
2
Pbest
Yes
No
is Iteration = max value ?
Yes Set optimal location
n
DiIaiRi
(2)
IDG i1 n
Randomize the DG size and Run power flow
i1
Print optimal DG sizes and respective locations
End
Ri No Re-generate
However, significant loss reduction will occur at the feeders with DG connection. Thus, the new total power loss in the network with the DG connection is as expressed in the following equations;
Plosses new i 1 I i n
new 2
Ri
(3)
Check constraints
Fig. 2 Proposed algorithm for DG placement and sizing
IV. SIMULATION RESULTS A power flow solution was conducted for base case as shown in Fig. 3 which shows serious voltage drops at Kaduna, Jos Kano and Gombe that requires urgent attention. Also voltage profiles for three different operating scenarios which are for maximum demand and normal generation pattern, maximum
©IJEECS
(IJEECS) International Journal of Electrical, Electronics and Computer Systems. Vol: 12 Issue: 01, January 2013 demand and minimum generation pattern, and minimum demand and normal generation pattern are obtained from power flow solution. Voltage increases above nominal at some buses, while significant voltage drops at Kano and Gombe under minimal loading but normal generation occurs as shown in Fig. 4. The voltage drops under maximum demand and minimum generation are low and are out of the acceptable range of voltage limit.
TABLE I OPTIMAL DG PLACEMENT
LOCATION (BUS NO.) KANO (6) KADUNA (5) GOMBE (8)
OPTIMAL SIZE (MW) 54.346 40.624 36.342
The voltage profile improvement for the base case and the three placements are as shown in Fig. 6.It can be observed that for each placement the voltage profile keep increasing until the third placement when all the buses voltages are above the minimum acceptable value of the network
Fig. 3 Voltage profile for base case
The drops at Jos and Gombe for instant are up to 14%. The drops resulting from maximum demand at normal generation shows a little improvement in the sense that the percentage drop is now 12.8% in this case.
Fig. 6 Voltage Profiles for DG Placement at three different locations
For the purpose of comparison the base case is compared with the third placement in terms of voltage profile improvements as shown in Fig. 7. It may however, be seen that there is significant improvement in voltage profile at Gombe bus which use to suffer from serious low voltage problem when compared with the other buses.
Fig. 4 Voltage profile for different scenarios
Fig 5 shows the line losses for the network base case with the highest losses along the lines between buses 5-6 followed by lines 7-8. Another loss that is significant is line between buses 1-4 that supplies the region.
Fig. 7 Voltage profiles for base case and 3rd placement
Table II shows the percentage improvement for each placement. Generally the placement has improved the voltage by more than 20% after the third DG placement. TABLE II VOLTAGE PROFILE IMPROVEMENT WITH DG PLACEMENT
Fig. 5 Power losses of base case network
Optimal location for the first placement is Kano bus (6) with the optimal size and power loss reduction as shown in Table I. Using the first placement as the base case, the second optimal location is Kaduna bus (5) during the second placement. The process was repeated until the power loss reduction becomes significantly small ©IJEECS
1ST DG PLACEMENT 2ND DG PLACEMENT 3RD DG PLACEMENT
% IMPROVEMENT 10.23 20.45 23.21
(IJEECS) International Journal of Electrical, Electronics and Computer Systems. Vol: 12 Issue: 01, January 2013 Similarly Table III shows the power loss reduction which when compared with the base case, remarkable improvement is recorded interms of power losses along the lines. For any additional DG placement the power losses keeps decreasing which puts the region in as a suitable candidate position for DG placement depending the objectives of the Power Holding company of Nigeria (PHCN).
[3]
[4]
[5]
[6] TABLE III POWER LOSSES REDUCTION WITH DG PLACEMENT
LOCATION OF DG
LOSSES WITHOUT DG (MW) 11.53
LOSSES WITH DG (MW)
PLR (MW)
5.94 4.48 3.77
5.59 7.05 7.76
KANO KADUNA GOMBE
[7]
[8]
[9]
[10]
Joos, G., Ooi, B.T. McGillis, D., Galiana, F.D and Marceau, R., “The Potential of Distributed Generation to provide Ancilliary Services”, IEEE Power Engineering Society Summer Meeting, pp.1762-1767. 2000 Chiradeja, Pathomthat and Ramakumar, R “An Approach to Quantify the technical Benefits of Distributed Generation”, IEEE Transaction on Energy Conversion, Vol. 19, No. 4, pp. 764-773. 2004 Wang Caisheng, and Hashem, M, “Analytical Approaches for Optimal Placement of Distributed Generation Sources in Power System”, IEEE Transaction on Power System, Vo. 19, No. 4, PP. 2068 – 2076 2004 Quezada, Victor H. Mendez Juan River Abbad and Toma‘s Gomez San Roman “Assessment of Energy Distribution Losses for Increasing, Penetration of Distributed generation”, IEEE Transaction on Power System, Vol. 21, No. 2, Pp. 533-540. 2006 N. Mithulananthan, Than Oo and Le Van Phu, “Distributed Generator Placement in Power Distribution System Using Genetic Algorithm to Reduce Losses”, Thammasat Int. J. Sc. Tech., vol. 9, pp.55-62, 2004. Mrs. A. Lakshmi Devi and Prof B. Subramanyam, “A New Hybrid Method For Loss Reduction And Voltage Improvement In Radial Distribution System- A Case Study”, i-manager’s journal on Electrical Engineering, vol1, No.3,pp.64- 68,January-March 2008 Rafiei, M., Zadeh, M.S. “Optimization of a typical biomass fueled power plant using Genetic algorithm and binary particle swarm optimization” 17th Conference on Electrical Power Distribution Networks (EPDC), pp1-10 2nd May 2012 TRANSYSCO, “Performance Management and Capacity Utilization”, 3rd Quarter report; Power Holding Company of Nigeria, Abuja, 2009. Fan, H. Y., and Shi, Y. “Study of Vmax of the Particle Swarm Optimization Algorithm.”Proceedings of the Workshop on Particle Swarm Optimization. Indianapolis, IN: Purdue School of Engineering and Technology, IUPUI (in press). 2001 Musa, H., Adamu, S. S., “PSO Based DG Sizing for Improvement of Voltage Stability Index in Radial Distribution Systems” proceedings of 2nd IASTED Int. Conference on Power and Energy Systems and Application (PESA 2012) pp 175-180, Nov. 2012.
V. CONCLUSIONS This paper has presented an effective approach for determining the optimal locations and sizes of DG units in a typical sub-transmission region of Nigerian grid. The placement of the DG is based on maximum power loss reduction value and the optimal sizing was also based on power loss reduction. The results obtained based on the proposed approach shows that the penetration of the DG units has improved the voltage profile of all the busses in the region and at the same time the power losses have greatly improved. The region which has for long been suffering from these problems due to the long distances from the major generation sources that are located in the southern part coupled with the radial nature of the network are addressed by deployment of DG units in the network. It was also observed that the voltage profile of the network has generally improved by more than 20 % across all the busses. .
[11]
ACKNOWLEDGMENT H. Musa acknowledges with gratitude the financial support, in form of research fellowship, offered by Bayero University, Kano Nigeria.
Sunusi Sani Adamu receivesthe B.Eng degree from Bayero University Kano, Nigeria in 1985; the MSc degree in electrical power and machines from Ahmadu Bello University, Zaria , Nigeria in 1996; and the PhD in Electrical Engineering, from Bayero University, Kano, Nigeria in 2008. He is a currently a senior lecturer in the Department of Electrical Engineering Bayero University, Kano. His main research area includes power systems simulation and control, and development of microcontroller based industrial retrofits. Dr Sunusi is a member of the Nigerian Society of Engineers and a registered professional engineer in Nigeria.
REFERENCES [1]
[2]
Momoh J., Yan Xia, Boswell G. D. "An approach to determine Distributed Generation (DG) benefits in power networks," Power Symposium, 2008. NAPS pp 1-7, Sept. 2008. Akorede, M. K., Himaz, H., Ari, I., Abkadir, M. Z. A., “Effective method for optimal allocation of distributed generation units in meshed electric power systems” Generation , Transmission & Distribution IET Vol:5, Issue:2 pp 276-287 Feb. 2011
[12]
Haruna Musa receives his B. Eng degree in electrical engineering from Bayero University kano Nigeria in 1988 and M. Sc in electrical engineering in 1991 from University of Lagos in Nigeria. Currently he is a lecturer in the Department of Electrical Engineering Bayero University Kano at the same time pursuing his Ph.D in the same Department. His research area includes power system stability, Distributed Generation and control of Microgrids. He is a member of Nigerian society of Engineers and Registered Engineer with COREN.
©IJEECS
