ABUAD Journal of Engineering Research and Development (AJERD)

Volume 1, Issue 1, 61-66

Optimization of an Expanded Nigeria

Electricity Grid System using Economic
Load Dispatch
Elutunji BURAIMOH1,*, Temitayo Olutimi EJIDOKUN1, Oladimeji Joseph AYAMOLOWO1
1
Department of Electrical, Electronic and Computer Engineering, Afe Babalola University, Ado-Ekiti, Nigeria
[email protected]/[email protected]/[email protected]

*
Corresponding Author: [email protected]

Date of First Submission: 29/09/2017

Date Accepted: 18/10/2017

Abstract: The objective of an economical and reliable power system Economic Load Dispatch (ELD) is a sub problem of the
is to ensure that customer’s load requirements are met at a reduced optimal power flow (OPF) having the objective of fuel cost
cost of generation. Economic Load Dispatch (ELD) determines the minimization. The fuel cost equation of a thermal plant is
power output of each thermal power plant, which minimizes the
generally expressed as continuous quadratic equation. In real
overall cost of fuel needed to serve the entire system load. This study
carried out an optimization analysis on the economical distribution
situations, the fuel cost equations can be discontinuous [5].
of loads on the existing and expanded thermal plants. The Owing to the increasing complexity of interconnections, the
optimization analysis was carried out by formulating ELD problem size of the areas of electric power systems, scarcity of energy
using MATLAB software packages. This was with a view of resources, increasing power generation cost and
developing a dynamic load scheduling between the aforementioned, environmental concern, it is necessary to minimize the
in order to reduce cost, enhance overall performance and reliability expenses through optimal economic dispatch. For the total
of the system. The obtained result shows it is efficient and cost- cost of generation and transmission to be minimum at a
effective to operate and generate power in the expanded grid. forecasted load demand, Economic Dispatch defines the
Correspondingly, the expansion of the grid to include new thermal
production level of each participating generating plant.
plants of Olorunsogo, Omotoso and Geregu eased the load stress on
the old plants of Afam, Egbin, Sapele and Ughelli: thereby mitigating
In Nigeria like other countries, power stations are far from
the effect of power transmission loss and offsetting the total cost of the loads at diverse distances. Consequently, these generating
generation. stations have dissimilar fuel cost functions. Hence for
providing cheaper power, load has to be distributed among
Keywords: Cost Analysis, Economic Load Dispatch, Load Demand, various power stations in a way which results in lowest cost
Optimization and Transmission Loss for generation.
Whether a generator should participate in sharing the load
1. INTRODUCTION at a given interval of time is a problem of unit commitment.
The modern power system around the world has grown in Once the unit commitment problem has been solved, it
complexity of interconnection and power demand. The focus becomes a problem of optimal allocation of the available
has shifted towards enhanced performance, increased generations to meet the forecasted load demand for the current
customer focus, low cost, reliable and clean power [1]. The interval [6]
Economic Load Dispatch involves allotting generation levels In a practical power system, the power plants are not
to the various generating plants in a power system network in located at the same distance from the centre of loads and their
order to ensure optimal and economical supply to the system fuel costs are different. Also, under ideal operating conditions,
load [2]. the generation capacity is more than the total load demand and
In order to maintain a high degree of economy and losses. Thus, there are many options for scheduling generation.
reliability of the power system, economic dispatch is one of In an interconnected power system, the objective is to find the
the options available to the utility companies. Economic real and reactive power scheduling of each power plant in
dispatch allocates the total power demand among the online such a way as to minimize the operating cost. This means that
generating units in order to minimize the cost of generation the generators’ real power and reactive power are allowed to
while satisfying pertinent system constraints [3]. Economic vary within certain limits so as to meet a particular load
Load Dispatch (ELD) problem play a vital role in the demand with minimum fuel cost. [7]. The objective of
operation of power system it is the short-term determination economic load dispatch is to minimise the overall cost of
of the optimal output of the number of electricity generation generation
facilities, to meet the system load, at the lowest possible cost For the purpose of economic dispatch studies, online
subjected to transmission and operational constraints. [4] generators are represented by functions that relate their

http://ajerd.abuad.edu.ng/ 61
ABUAD Journal of Engineering Research and Development (AJERD)
Volume 1, Issue 1, 61-66

production cost to their power output. Quadratic cost The losses on the transmission system are a function of the
functions are used to model generator in order to simplify the generation dispatch
mathematical formulation of the problem and to allow many Losses can be included in the optimization formulation by
of the conventional optimization techniques to be used. The slightly rewriting the Lagrangian:
economic dispatch problem is traditionally solved using
conventional mathematical techniques such as lambda ng ng
iteration and gradient schemes. These approaches require that Ct λ ( PD +PL -∑ Pi ) + ∑ µi (max) ( Pi − Pi (max) ) +
L =+
fuel cost curves should increase monotonically to obtain the =i 1 =i 1
global optimal solution. The input-output of units are (3)
ng
inherently non-linear with valve point loading or ramp rate
limits and having multiple local minimum points in the cost
∑µ
i =1
i (min) ( Pi − Pi (min) )
function [8].
Where L is the Lagrangian function of the power
2. PROBLEM FORMULATION
generated (P i ) and lambda (λ), P i(min) and P i(max) are the
The objective of economic dispatch problem is to operate minimum and maximum generating limits respectively for a
our power system in a manner that minimizes the costs of given plant I and P L is the power loss
generator. In other words, Economic dispatch determines the The minimization of the cost function in the equation (3) is
best way to minimize the current generator operating costs. subject to several restrictions capturing both the power and
It assumes that there are N units already connected to the heat balance of the system, and the operational constraints of
system. The purpose of the economic dispatch problem is to the generator [11].
find the optimum policy for these N units. The lambda- The necessary condition for an optimal economic dispatch
iteration method is a good approach for solving the economic shown in equation (4) is obtained by differentiating equation
dispatch problem: generator limits are easily handled, penalty (3) with respect to P i .
factors are used to consider the impact of losses. Economic
dispatch is not concerned with determining which units to turn
∂L ∂Ct ∂P
on/off. Basic form of economic dispatch ignores the = + λ (0 + L − 1)= 0 (4)
transmission system limitations ∂Pi ∂Pi ∂Pi
In this system, the essential condition is to balance the sum
of the generated powers P i and the load demand P D [9] and Equality constraint imposed by the losses and load demand.
[10]
ng

PD + PL − ∑ Pi =
ng
0
∑P = P
(5)
i D (1) i =1
i =1

Solving equation (4) for λ we get

Electrical power systems are designed and operated to
meet the continuous variation of power demand. In power
dCi ∂P
system minimizing, the operation cost is very important [2] − λ (1 − L ) = 0 (6)
Assume that C t is equal to the total cost for supplying the dPi ∂Pi
indicated load. The problem is to minimize C t subject to the
1 dCi
constraints that are the sum of the power generated in each λ=
unit as in equation (1). In this solution process, operating  ∂PL  dPi
limits of each generation unit must be known. That is, 1 − 
 ∂Pi 
ng ng

C = ∑ C = ∑α + β P + γ P2 (2) Where the penalty factor L i is defined by

t i
=i 1 =i 1
i i i i i
1
C i is the cost rate of the unit. The output of each unit is P i , Li = (7)
 ∂PL 
1 −
which represents the electrical power output of units. Where

α i , β i and γ i are the coefficients of the quadratic cost function  ∂Pi 
of the unit’s power.
When transmission distances are very small and load
density is very high, transmission losses may be neglected and Thus the effect of transmission losses is to introduce a
the optimal dispatch of generation is achieved with all plants penalty factor with a value that depends on the location of the
operating at equal incremental production cost. However, in a plant with respect to the loads. In the equation for lambda (λ)
large interconnected network where power is transmitted over as shown in equation (6) it is clear that the most economic
long distances with low load density areas, transmission losses cost of dispatch is obtained when the incremental cost
are major factor and affect the optimum dispatch of generation. multiplied by the corresponding penalty factor are equal for

http://ajerd.abuad.edu.ng/ 62
ABUAD Journal of Engineering Research and Development (AJERD)
Volume 1, Issue 1, 61-66

all the participating plants. Hence, L i >1; more expensive and Applying the equality constraint in equation (5),
L i <1; less expensive expanding with Taylor's series about operating point λ (k), and
The incremental transmission loss can be found from the neglecting the higher order terms results in
loss equation and is given by
∆P ( k )
∂PL
ng ∆λ ( k ) = (15)
2∑ Bij Pj + B0i
(k )
=  dPi 
∂Pi j =1
(8)
∑  d λ 
And the incremental production cost is given by And therefore,

dCi
= 2γ i Pi + βi (9) λ ( k=
+1)
λ ( k ) + ∆λ ( k ) (16)
dPi
Where
Substituting equations (8) and (9) into
ng

− ∑ Pi (
k)
∂Ci ∂P ∆P (k )
=PD + P (k )
(17)
+λ L =λ
L
(10) i =1
∂Pi ∂Pi

We have equation (11)

ng 3. RESULTS AND DISCUSSION

βi + 2γ i Pi + 2λ ∑ Bij Pj + B0i λ =
λ (11) The thermal plant generation limits used as input for this
j =1 optimization analysis are shown in the Table 1. Hence
indicating the freedom of operation of the thermal plants.
Dividing through by 2λ and pulling out B ii from the
summation, we have equation (12) Table 1: Thermal plant generation limits
Plants P i (min) in P i (max) in MW
MW
 γi β
n
 g
1 

λ
+ Bii  i

P + ∑
=j 1, j ≠ i
Bij Pj = 1 − B0i − i
2 λ  (12)
 Omotoso 60 335
Olorunsogo 60 335
Geregu 80 414
This can be extended for all the plants in linear equations
Afam 150 977
in matrix form to have
Egbin 220 1300
Sapele 190 1020
γ1   β1 
 λ + B11  1 − B01 − λ 
B12  B1ng Ugheli 145 972

   P1   
 B γ2   β The economic load dispatch results of the thermal
+ B22  B2 ng   P2  1  1 − B02 − 2 
 21 λ  =  λ  generating stations of Afam, Egbin, Sapele and Ugheli which
    2  constitute the old grid is shown in the Table 2. Table 2
      P     presents how the four thermal plants handle the total power
 γ ng   ng   β ng  demands between 705 MW 3873 MW.
 Bng1 Bng 2  + Bngng  1 − B0 ng −  The total generation cost increases with the increase in
 λ   λ  overall generation as shown in the figure 1. As the total power
demands tends toward 3873 MW, the cost of generation
[ E ][ P ] = [ D ] (13) increases exponentially as a result of the operation of most of
the thermal plants around their maximum limit. This operation
around limit does not only affect the total fuel cost, it also
To find the optimal dispatch first an estimated guess of λ,
affects the power transmission loss. As total transmission loss
say λ (1) is made. Thus, from equation (13) we have P i at the
is a function of power generated, the power losses in
kth iteration expressed as
transmission are also taken care of by the generating plants.
Hence the increase in transmission loss as a result of operating
λ ( k ) − βi plants around their limit also increases the total cost of fuel.
Pi ( k ) =
2 ( γ i + λ ( k ) Bii )
(14)
The optimization program started operating Egbin around its
limit of 1300 MW in order to meet the total demand of 1893
MW at total generation cost of 178694.78 $/h.

http://ajerd.abuad.edu.ng/ 63
ABUAD Journal of Engineering Research and Development (AJERD)
Volume 1, Issue 1, 61-66

The operation of Egbin at the maximum limit continued efficient to bring these thermal plants into their maximum
even after increasing the total power demand to 2685 MW but operation in the following order; Egbin, Afam, Sapele and
now, Afam is also operated at the limit. Subsequently, Sapele lastly Ugheli in the quest to increase the total demand as
joined in the operation around limit when the total demand presented in Table 2
was increased further. The implication of this is that it’s most

Table 2: Old Thermal Plants Economic Load Dispatch

Plants 705 MW 1101 1497 1893 2289 2685 3081 3477 3873 MW
MW MW MW MW MW MW MW
Afam 150.00 150.00 250.40 419.00 676.90 977.00 977.00 977.00 977.00
Egbin 246.88 682.93 969.48 1215.30 1300.00 1300.00 1300.00 1300.00 1300.00
Sapele 190.00 225.60 340.08 457.50 641.70 914.90 1020.00 1020.00 1020.00
Ugheli 145.00 145.00 145.00 145.00 145.00 145.00 533.90 972.00 972.00
Cost ($/h) 68357.14 88840.9 125909.4 178694.7 254626.3 374522.4 716255.0 1505219. 7101304.2
2 8 6 5 8 65 6

This operation of new thermal plants at the maximum limit

Similarly, the economic load dispatch result of expanded continued all through the generation increase from 1101 MW
grid to include the old thermal generating plants of Afam, to 4665 MW. Consequently, as the load demand increases the
Egbin, Sapele and Ugheli and new plants of Omotoso, old thermal plants are brought gradually into maximum
Olorunsogo and Geregu is presented in Table 3. operation.
Evidently, the total generation cost of the expanded grid The optimization results imply that it is more efficient to
increases with the increase in overall generation as shown in bring these new thermal plants into their maximum operation
the figure 1. All through the increase in the generation, the in order to meet the increasing load demand. Subsequently,
cost of generation increases linearly even with the operation the older thermal plants are brought into their maximum
of the thermal plants around their maximum limit. The operation in this order; Egbin, Afam, Sapele and lastly Ugheli.
optimization program started operating the new thermal plants This is not different from the efficiency order as recorded for
of Omotoso, Olorunsogo and Geregu around their limit even the old grid. This affirms that it Egbin still remains the most
at lower power total demands at a minimum total generation efficient of all the older thermal plants
cost of 117485.66 $/h.

Table 3: Expanded thermal plants economic load dispatch

Plants 1101 1497 1893 2289 2685 3081 3477 3873 4269 4665
MW MW MW MW MW MW MW MW MW MW
Omotoso 60.00 335.00 335.00 335.00 335.00 335.00 335.00 335.00 335.00 335.00
Olorunsogo 335.00 335.00 335.00 335.00 335.00 335.00 335.00 335.00 335.00 335.00
Geregu 80.00 80.00 263.26 414.00 414.00 414.00 414.00 414.00 414.00 414.00
Afam 150.00 150.00 150.00 150.00 365.60 586.10 901.70 977.00 977.00 977.00
Egbin 220.00 394.24 663.93 944.23 1141.50 1300.00 1300.00 1300.00 1300.00 1300.00
Sapele 190.00 190.00 218.72 329.11 420.00 576.20 807.00 1020.00 1020.00 1020.00
Ugheli 145.00 145.00 145.00 145.00 145.00 145.00 145.00 145.00 145.00 972.00
Total Cost 117485. 130440.9 153066 187987. 235184. 302361. 406555. 476045. 476045.0 1579486
($/h) 66 9 .11 27 12 19 88 02 2 .31
demand. For instance, total amounts of 7101304.26 $/h and
The total generation costs of the economic dispatches of 476045.02 $/h are required to generate same power demand of
the old grid and expanded grid are reported in the Figure 1. 3873 MW in the old grid and expanded grid respectively.
Figure 1 is obtained from the plot of the total generation cost Consequently, the total amount of 6625259.24 $/h is saved by
in $/h against the power demand in MW for both the old and the generating 3873 MW of electricity in the expanded grid
expanded (new) electricity grid as obtained from results in Table 2. instead of the old grid
It is fortunate to know that the cost of generating the same
amount of Megawatt of Power in the expanded grid is
significantly less than that of the old grid. The cost has an
exponential and linear relationships with the old and expanded
grid respectively. Initially the cost of generating lower power
of 1101 MW was lower in the old system at a sum of 88840.9
$/h as against 117485.66 $/h. However, the cost of operating
the old grid increased exponentially and became more than
cost of operating expanded system with increase in power

http://ajerd.abuad.edu.ng/ 64
ABUAD Journal of Engineering Research and Development (AJERD)
Volume 1, Issue 1, 61-66

Figure 3: Afam generation output in old and expanded grid

Figure 1: Total Generation cost comparison

The addition of the new thermal plants of Olorunsogo,

Omotoso and Geregu to the grid reduced the load stress on the
old plants thereby offsetting the total cost of generation. The
old thermal plants are not brought into operation around their
maximum until the new plants are not able to meet the
generation demand at the operation around their limits as
indicated in figures 2, 3, 4 and 5. Figures 2, 3, 4, and 5 are
obtained from the plot power generation in Egbin, Afam,
Sapele and Ugheli stations against the total power demand of
the old and expanded grid as obtained from the results in
Table 3. For instance, Egbin starts operating around its
maximum limit at total load demand of 1893 MW in the old
grid and 3081 MW in the expanded grid. Afam starts
operating around its maximum limit at total load demand of Figure 4: Sapele generation output in old and expanded grid
2685 MW in the old grid and 3873 MW in the expanded grid.
Sapele starts operating around its maximum limit at total load
demand of 3081 MW in the old grid and 3873 MW in the
expanded grid. Ugheli starts operating around its maximum
limit at total load demand of 3477 MW in the old grid but kept
operation at the minimum limit in the expanded grid.

Figure 5: Ugheli generation output on old and expanded grid

4. CONCLUSIONS
Economic dispatch optimizes to determine the generation
output of different plants while minimizing the operating cost.
Figure 2: Egbin generation output in old and expanded grid Hence, the operating cost plays an important role in the
economic scheduling. A critical comparison of the total
generation cost of the old and expanded grid system shows
that the latter’s operation is more economical and judicious.
Utilising generating plants close to the load results in
lower losses. This impact of losses was included in the

http://ajerd.abuad.edu.ng/ 65
ABUAD Journal of Engineering Research and Development (AJERD)
Volume 1, Issue 1, 61-66

economic dispatch optimization carried out. This work International Journal of Scientific & Engineering Research,
revealed certain elements which shape power generation vol. 5, pp. 589–595.
pattern at minimum cost. This include the plant operating [9] Saadat, H. (1999). Power System Analysis ,5th ed.. New
efficiency, fuel cost and certainly transmission losses. York: McGraw-Hill.
It is discovered that the proximity of thermal generating [10] Kothari, D. P., & Nagrath, I. J. (2003). Modern Power
plants to gas supply does not guarantee minimum cost of System Analysis, 3rd ed. New Delhi: Tata McGraw Hill
generation as they are sometimes located far from the load. Education Private Limited.
And, if the plant is located far from the load centre, [11] Rist, J. F., Dias, M. F., Palman, M., Zelazo, D., &
transmission losses may be considerably higher and hence the Cukurel, B. (2017). Economic dispatch of a single micro-gas
plant may be overly uneconomical. turbine under CHP operation. Applied Energy, vol. 200, pp. 1–
New generating plants (Omotoso in Ondo State, 18.
Olorunsogo in Ogun State and Geregu in Kogi State) are more
efficient than older generating plants (Afam, Egbin, Ugheli
and Sapele) due to wearing and proximity to the load.
Consequently, it is more important and in economic interest to
continue to carry out grid expansion to include more
generating stations around Nigeria.. In general, operating
generating plants closer to the load results in lower losses.
This impact on losses must therefore be accommodated in
economic dispatch optimization problem formulation.

ACKNOWLEDGMENT
The authors appreciate God Almighty who is The Fountain
of all knowledge.
REFERENCES
[1] Gaurav, G., & Shrivastava, J. (2015). Analysis of
Economic Load Dispatch and Unit Commitment Using
Dynamic Programming. International Journal of Advanced
Research in Electrical, Electronics and Instrumentation
Engineering, vol. 4, pp. 5476–5483.
[2] Kaur, A., Singh, H. P., & Bhardwaj, A. (2014). Analysis
of Economic Load Dispatch Using Genetic Algorithm.
International Journal of Application or Innovation in
Engineering & Management (IJAIEM), vol. 3, pp. 240–246.
[3] Bakare, G. A., Aliyu, U. O., Venayagamoorthy, G. K., &
Shu, Y. K. (2005). Genetic Algorithms Based Economic
Dispatch with Application to Coordination of Nigerian
Thermal Power Plants. IEEE Transactions on Power Systems,
vol. 78, pp 1–6..
[4] Chaturvedi, N., & Walkey, A. S. (2014). A Survey on
Economic Load Dispatch Problem Using Particle Swarm
Optimization Technique. International Journal of Emerging
Technology and Advanced Engineering, vol. 4, pp. 188–193.
[5] Roshni, S., Pradhan, C. R., & Mohapatra, B. (2015). A
Comparitive Study of Economic Load Dispatch Problems
Using Classical method and Artificial Intelligence Method.
International Journal of Advanced Research in Electrical,
Electronics and Instrumentation Engineering, vol. 4, pp.
1564–1569.
[6] Krishnarayalu, M. S. (2015). Unit Commitment with
Economic Dispatch. International Electrical Engineering
Journal, vol. 6, pp. 1913–1916.
[7] Hosseinnezhad, V., & Babaei, E. (2013). Economic Load
Dispatch Using Genetic Algorithm and Pattern Search
Methods. International Journal of Electrical Power & Energy
Systems, vol. 49, pp. 160–169.
[8] Awodiji, O. O., Bakare, G. A., & Aliyu, U. O. (2014).
Short-Term Economic Load Dispatch of Nigerian Thermal
Power Plants Based on Differential Evolution Approach.

http://ajerd.abuad.edu.ng/ 66