time tablr scheduer in henetic algorithm.

1. See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/286610160
Genetic algorithm based Generator Scheduling-a review
Article in World Applied Sciences Journal · January 2014
DOI: 10.5829/idosi.wasj.2014.30.12.1892
CITATIONS
4
READS
745
4 authors, including:
Syed Basit Ali Bukhari
The University of Azad Jammu and Kashmir
56 PUBLICATIONS 760 CITATIONS
SEE PROFILE
Atta Ul munim Zaki
Shanghai Jiao Tong University
8 PUBLICATIONS 74 CITATIONS
SEE PROFILE
All content following this page was uploaded by Syed Basit Ali Bukhari on 19 November 2016.
The user has requested enhancement of the downloaded file.

2. World Applied Sciences Journal 30 (12): 1826-1833, 2014
ISSN 1818-4952
© IDOSI Publications, 2014
DOI: 10.5829/idosi.wasj.2014.30.12.1892
Corressponding Author: Syed Basit Ali Bukhari, Department of Electrical Engineering, University of Engineering and Technology
Taxila, Pakistan. Cell: +923000926502.
1826
Genetic Algorithm Based Generator Scheduling-A Review
Syed Basit Ali Bukhari, Aftab Ahmad, Syed Auon Raza and Atta Ul Munim Zaki
Department of Electrical Engineering, University of Engineering and Technology Taxila, Pakistan
Abstract: Generator Scheduling (GS) problemis a non-linear, mixed integer, combinatorial and constrained
optimization problem. The main objective of generator scheduling is to prepare the most economic startup and
shutdown schedule for generators to meet the forecasted load demand and spinning reservefor a scheduled
time horizon while satisfying various constraints. GS problemcan be realized as two interlinked optimization
problem, the on/off scheduling problem of generators and real power allocation problem. A feasible generators
schedule must satisfy various system and plant constraints.Variousoptimization approaches have been
developed for generator scheduling problem. Natural evolution based techniques are widelyusedto obtain
global optimal solution. Genetic algorithm is one of the evolutionary technique. This paper presents a
comprehensive review on genetic algorithm based generator scheduling problemsolution. A surveyof all of
research papers up to 2013 on this topic is given.
Key words: Unit commitment (UC) Optimization methods Genetic Algorithm (GA) Generation
scheduling Constraint satisfaction
INTRODUCTION computer memory. So different heuristic approaches
Generator scheduling problem (GSP) in the power relaxation (LR) [7, 8] have been developed to reduce the
system determines hourly on/off schedules for the computational time and search space. But these
generators with their power output over a specified time techniques find solution that is far away from global
horizon. The objective of generator scheduling is minimize optimal solution.
the total operating cost of system while satisfying To obtain a global optimal solution in reasonable
different system, envoirment and unit constraints. computational time meta-heuristic techniques have
Mathematically it is a complex non-linear combinatorial been developed and most widely used for generator
optimization problem.The exponential increasein search scheduling. Simulated annealing is a stochastic
space with number of generating units and various approach and finds global optimum solution but in
system, environmentaland unit constraint make the GS more computing time. Artificial neural network, Particle
problem a complex optimization problem. Swarm optimization, Ant colony, evolutionary
The exact optimal solution for GSP can be obtained programming, tabu search and Genetic algorithm are
by complete enumeration but this is not applicable to some random search technique and give more feasible and
large systems because it requires exhaustive near-optimal solutions [9-13].
computational time. So different techniques have been Hybrid techniques are the combination of
developed to solve this problem [1, 2]. These techniques deterministic and meta-heuristic approaches and are
can broadly classified as deterministic, meta-heuristic and extensively used to solve UCP [12, 14-16].
hybridized approaches.
Deterministic approaches such as Integer Generator Scheduling Mathematical Formulation
programming, dynamic programming (DP), branch and Objective Function: The objective of the generator
bound method have a problem of dimensionality [3-5]. scheduling is to minimize the total production cost and
These method requires high computational time and mathematically given as follow.
such as priority list [6], modified DP and Langrangian

3. ( ) ( ) ( ) ( )
( ) ( )
1 1
( 1 1
T N
off
i i i i i
t i
MinTC u t F p t u t u t C t
= =
= + − −
∑∑
( )
( ) ( ) ( )
2
i i i i i i
F P t P t P t
=
+ +
1
: ( ). ( ) ( )
N
i i d
i
t u t P t P t
=
∀ =
∑
( ) ( ) ( )
max
1
:
N
i i d R
i
t u t P P t P t
=
∀ ≥ +
∑
( )
min max
i i i
P P t P
≤ ≤
on on
i i
X T
≥
off off
i i
X T
≥
World Appl. Sci. J., 30 (12): 1826-1833, 2014
1827
(1)
Here T defines the scheduling period, N defines no.
of generating units, u(t) is the status of unit at specified
i
period, SC(t ) represents the transition cost andF(Pi) is
i
off
the operating cost of a unit for a specified time interval Fig. 1: No. of papers published each year on UCP
and it is given as follow. basedon GA
(2) Genetic Algorithm: Genetic algorithm isa stochastic
where , and are machine constants. principle. Genetic algorithm is a particularclass of
i i i
Constraints evolutionary biology such as inheritance, mutation,
Real Power Balance Constraint: Total power generated selection and recombination. A typical genetic algorithm
at time t must be equal to the load demand at that time. requires two things to be defined:
(3) A genetic representation of the solution domain.
Here P (t) represents the demand at time t. Initial population is randomly generated and Genetic
d
Spinning Reserve: To increase reliability of system there produce the off springs. A crossover operator exchanges
must be some extra power available in the system for information between two different chromosome and
emergency conditions such as load forecast error, mutation randomly changes the value of one or more
generator failure and generator shortage. strings and is used to search theunexplored search space.
(4) the performance of GA. Although the work on genetic
Here P is the maximum power that a unit generate commitment problem has been found from year 1993.
i
max
andP (t) represents the spinning reserve at time t. Figure 1 indicates the no. of research papers on unit
R
Generation Output Limits: Genetic algorithm techniques for solving
(5) categorized in:
where P and P are the minimum and maximum power Traditional Genetic algorithm
i i
min max
generation limits. Hybridized genetic algorithm approaches
Minimum Up-Time Constraints(MUT): techniques)
(6) Application of Genetic Algorithm for Generator
Minimum Down-Time Constraints(MDT): genetic algorithm for determining the near optimal
(7) function is used in this approach for constraint violation.
search technique established on natural evolution
evolutionary algorithms that use techniques inspired by
A fitness function to evaluate the solution domain.
operators such as cross over, mutation are used to
Several other operators have been developed to enhance
algorithm based optimization started in 80’s but Research
papers regarding genetic algorithm for solving unit
commitment using GA from 1993 to present.
generatorscheduling/unit commitment problem can be
(combination of genetic algorithm and other
Scheduling: D. Dasgupta, et al. [17] proposed simple
solution for the unit commitment planning. Penalty

4. World Appl. Sci. J., 30 (12): 1826-1833, 2014
1828
An elitism scheme is proposed so that the best solution domain specific recombination and permutation
are copied in a group and passed to next step. The operators were proposed. A new decoding scheme is
scheme was implemented on 10 generators over time base designed to reduce the probability of generating
of 24-hours. The work of Dasgupta was extended byX. infeasible solution. Penalized value depending upon the
Ma, et al. [18]. The results of genetic algorithm approach degree of violation is added in the objective function for
to solveGenerator scheduling problem were improved by constraint handling. Results show superiority of
using two point crossover.A forced mutation operator proposed method compared with non-varying fitness
was proposed forhandling constraints. Two types of function strategies.
coding scheme with different string length were used. T. Senjyu, et al. [24] presented a new GA, in which he
The scheme having smaller string length was found best developed unit characteristics based operator and
for 10 unit system. The introduction of forced mutation intelligent technique for producing initial population.
operator improved the performance of the algorithm. The initial population was produced base on load curve
In [19] a new coding scheme for UC variables was data for the feasible initial population generation. New
developed in GA. In the proposed coding scheme shift, intelligent mutation and crossover operator were
MUT/MDT constraints were directly integrated in the introduced. Unit were classified in different groups based
binary string which was used torepresent the on/off on MUT/MDT constraint. The scheduling is obtained up
status of unit. Other constraints like splining reserves, to 100-generators over a time of 24-hours.
power balance etc. were handled by adding a fixed Penalty In [25] the suitability of parallel repair GA for
term in objective function. Test results for 38-units generation planning were discussed in detail. The
practical system showthe proficiency of proposed approach gave a model framework that was less restrictive
approach. compared to DP and LR. A hybrid parallel model was
The search space and results of traditional genetic developed to avoid local convergence and to minimize
algorithm were improved by T.M. Field [20]. A new Computing time. A comparison of results of LR method
domain specific operator for flipping of a bit and multi UC and proposed genetic algorithm, demonstrate the
scheduling was proposed. Multi scheduling with different performance of proposed strategy.
reserve help the utility while choosing economical A new scheme was applied in [26] for chromosome
scheduling. representation and encoding the variables for solving
S. Orero et al. [21] developed the genetic algorithm large scale commitment problem. To ensure the feasible
to handle unit ramps rate limits while solving UCP. solution genetic operators were applied after the
Ramping was considered only in generation scheduling satisfaction of power balance constraint. Remaining
and not in economic dispatch sub problem. Quadratic loss constraints were added in the objective function.
factor for MUT/MDT handling and an absolute value Schedule for 10-generators was obtained over a time span
penalty factor for the spinning reserve/load demand of 24-hour.
constraint violation are used in proposed strategy. I.G. Damousis, et al. [27] proposed a new solution for
The robustness of the algorithm depends on the selection UCP based on the integer coded GA in which string size
and grading of the penalty factors. The scheduling was and computational time is reduced as compared to binary
obtained for 26 generators. coded GA. To avert distortion in the search space created
In [22] the author used a mixed binary and decimal from the penalty function method used for constraint
coding scheme in genetic algorithm to solve UCP. Total violation, MUT/MDT were directly coded in the
operating cost based penalty less fitness function has chromosome. Unit swapping operator and Excessive-
been used in this strategy. The combinatorial problem was reserve elimination operator were proposed to improve the
solved by GA and quadratic programming was used for performance of ICGA.
generation allocation. Simulation results show that the In [28] some new search operators were introduced
scheme produced global optimal solution in a reasonable in genetic algorithm to obtain generator scheduling.
CPU time. The proposed strategy uses Production cost, Unit startup
For constraint optimization in genetic algorithm a cost and load demand for defining mutation probability.
dynamic fitness function method was proposed by Repair algorithm was used to handle system and unit
V. Petridis [23]. To generate only feasible solution constraints. The schedule was obtained for a practical

5. World Appl. Sci. J., 30 (12): 1826-1833, 2014
1829
system consisting of 12-generators. The proposed A new genetic algorithm using parallel structure was
algorithm produced near optimal solution but consumed proposed to handle the constraint violation and solution
more time. infeasibility [35]. The proposed approach used unit
In [29]a genetic algorithm for solving large scale UCP characteristics classification and computational method to
was presented. To increase computational speed and to attain better results. By using parallel structure the
handle the MUT/MDT constraint, units having similar computation time was much reduced compared to
characteristics were clustered in a group and then genetic traditional GA method.
algorithm was applied to obtain the generator schedule. In [36] the author proposed a genetic algorithm based
Test results had showed an improvement in the cost approach to solve the UCP in which Optimal flow of
compared with cost obtained traditional GA. power with line constraints is also included. In first step
T. Wei, et al. [30] updated the result of genetic UC scheduling was obtained with prevailing constraint
algorithm by using adaptive crossover rate that varies and in the second step line constraint violations were
with the maximum colony adaptation and the average reduced using genetic algorithm based optimal power
colony adaptation degree of each generation. Rate of flow. A practical system with 12-generators, 66 buses and
variation was adjusted by evolutionary generation and 93 transmission lines was used to test the effectiveness of
colony adaptation degree. The approach was found to be proposed approach.
more precise and good convergent than simple GA. In [37] the author used deterministic genetic
In [31] a real coded GA and hybrid Taguchi GA algorithm to solve unit commitment. A deterministic
method was applied to solve the UCP. The Taguchi selection procedure and an annular crossover operator
technique was used to enhance the offspring’s quality was proposed to avoid premature convergence. Increased
created from crossover and mutation operation. The possibility of genetic information exchange was achieved
proposed strategy not only enhanced search space but by proposed annular crossover operator. A repair
also produced optimal solution with improved algorithm was used for constraint handling. A better
convergence. Results indicated that the HTGA had convergence was obtained by proposed strategy when
improved efficiency as compared real genetic algorithm. compared with traditional GA approaches.
An optimal unit commitment schedule using genetic A parallel structure integrated with improved and
algorithm was developed by the author in [32]. A hill optimized genetic algorithm was proposed in [38].
climbing method is applied to improve the performance of The proposed approach effectively handles the
genetic algorithm. The proposed strategy combined the infeasibility of the solution. An intelligent mutation and a
features of accelerated search and easy convergence. scaling function for selection in each generation was
Test results revealed that the approach has better proposed. The strategy gave better economy, speedy GA
practical value for optimal generator unit commitment performance and increased probability to find global
problem with good sharpness and high flexibility. optimal feasible solution.
In [33]a three dimensional matrix based GA In [39] author applied genetic algorithm to solve multi
approachwas proposed for representation of chromo objective function optimization problem. Unit commitment
some and unit commitment schedule. The strategy used problem along with constraint emission was discussed in
Binary coded GA for UCP and real coded GA for EDP. the proposed strategy. To handle minimum up time and
To ensure the feasibility of solution power balance minimum down time constraints in genetic algorithm the
constraint was satisfied prior to genetic operation. Tests integer base coding method was proposed for generating
on 10-generators revealed the feasibility of proposed initial feasible solution. No penalty function was used for
method. MUT/MDT constraints. Constrained Emission was also
In [34] a new integer coded GA was applied for considered in proposed approach.Commitment schedule
generation planning. A hybrid crossover based on over 24 hours was obtained for 10-generators. Test results
average modified bound and swapped operator and a provide evidence for the effectiveness of the proposed
hybrid uniform and non-uniformed mutation operator approach.
were proposed. Test results for system up to 300- An intelligent coding scheme for genetic algorithm
generators are given and justify the efficiency of was proposed by D. Sundararajan, et al. [40] to solve the
proposed approach. scheduling problem. The Minimum up time and minimum

6. World Appl. Sci. J., 30 (12): 1826-1833, 2014
1830
down time constraint were satisfied by proposed T. Takata, et al. [46] presented a hybrid genetic
intelligent coding scheme. Penalty parameter constraint algorithm and langrangian relaxation technique for solving
handling technique was used to obtain a satisfactory unit commitment problem. To overcome the limitations of
balance of power constraints. The scheme produced much Langrangian Relaxation in handling constraint genetic
improved results for 10 and 26-generators over 24 hours. algorithm was employed. In Genetic algorithm constraint
The startup and shut down time in genetic algorithm satisfaction can easily be obtainedby simply adding
were represented in the binary strings of variables in [41]. penalty factor in the objective function. Moreover, the
Penalty function method was used to handle other introduction of heuristics to facilitate genetic
constraint. A new operator known as transportation manipulation of the string, which improved the efficiency
operator was proposed. Exchange of information takes of the optimization. Simulation results had shown that this
place between chromosomes of two randomly chosen method was effective in solving practical UCP.
units. The proposed approach helped in improving the In [47] a hybrid fuzzy logic and genetic algorithm was
computational efficiency genetic algorithm. developed to solve the generator scheduling problem.
An improved genetic algorithm with an approach to The fuzzy logic based model was proposed to deal with
solve both real and integer parts of the UCP was load balance and spinning reserve constraint and to
proposed in [42]. Ramp rate constraints are also calculate penalty term for other constraint handling. GA
incorporated in the proposed strategy. Multi point was then used for generation planning. A high quality
variable crossover technique was used. Experiments were solution was obtained from proposed approach compared
performed up to 100-units and performance of proposed to results obtained from traditional GA, LR and integer
strategy was found satisfactory. programming.
Application of Hybridized GA for Generator Scheduling: simulated annealing was developed to obtain the unit
X-Qiang, et al. [43] developed a dynamic programming on/off scheduling in [48]. The proposed approach
crossover operator to create off springs. The DP is increases the speed of SA and improves the performance
included in place of crossover parameter without affecting of genetic algorithm. Scheduling was obtained for a
the GA and DP algorithm. The proposed approach uses practical system of 40-generators.
dynamic programming on genetic algorithm based parent A chaos search immune algorithm embedded in the
population to produce new chromosome. The penalty genetic algorithm was proposed by C.C. Liao in [49].
function or repair algorithm was not used for handling Initially immune and genetic algorithm were nested then
constraints as DP crossover generates the feasible off the fuzzy logic and chaos search was implemented to
springs if the initial population satisfy the constraints. solve the UC problem. Crossover and mutation
The approach worked well for non-linear optimization probability are changed from constant value to varying
problem. value and calculated by using fuzzy logic. The proposed
A genetic based neural network and dynamic strategy guaranteed Global optimal convergence of
programming (DP) strategy was proposed by H. Shyh-Jier solution. Test results showed the proficiency of proposed
[44]. At initial phase a genetic based enhanced neural method.
network was applied to generate the initial commitment A fuzzy logic to make the fuzzy based unit
schedule and then DP was used to ameliorate this commitment model and genetic based approach for
schedule. Computation performance of proposed method scheduling was proposed by A.H. Mantawy [50]. The
was more compared to other approaches. model takes the uncertainty in the expected load demand
A.H. Mantawy, et al. [45] developed hybrid method and spinning reserve in the framework of fuzzy. A penalty
using genetic algorithm and tabu search method. The term is calculated by the proposed fuzzy model to
solution was coded as combination of binary number and converge the solution to more optimal solution. The
decimal number to save memory and to reduce the economic dispatch part of problem was solved by
computation time for the GA. The approach uses genetic dynamic programming.
algorithm for generation of initial population and tabu A two-layer strategy consisting of genetic algorithm
search algorithm in the reproduction phase. The proposed and improved lambda iteration was presented by the
method reduces the probability of premature convergence author in [51]. Unit level and population level crossover
of genetic algorithm. were introduced to increase the search space. A swap
A hybrid approach containing genetic algorithm and

7. World Appl. Sci. J., 30 (12): 1826-1833, 2014
1831
mutation operator was proposed for committing the unit 7. Ouyang, Z. and S. Shahidehpour, 1991. An intelligent
on/off based on unit’s FLAPC. Lambda iteration approach
was used for power allocation.
J. Zhang, et al. [52] developed a hybrid genetic
algorithm and particle swarm optimization. The approach
uses genetic algorithm for optimizing unit commitment
problem and particle swarm optimization for economic
load dispatch problem. Probability of Local optimal
convergence is prevented by using PSO. Feasible UC
scheduling is obtained for 10-generator over a time
horizon of 24-hours.
CONCLUSION
Generation scheduling is very critical in daily
operation of power system. The optimal Generator
scheduling gives significant production cost savings.
A comprehensive review genetic algorithm based
generator scheduling is discussed in this paper. The GA
approach has been found very effective to find an optimal
solution. The GA based techniques not only produce
better solution but also reduces the computational time
and search effort. Various GA based schemes are
discussed in this paper and it is found that intelligent
genetic algorithm, parallel structure genetic algorithm and
hybrid GA techniques appear to be best among all
proposed GA strategies.
REFERENCES
1. Yamin, H., 2004. Review on methods of generation
scheduling in electric power systems. Electric Power
Systems Research, 69(2): 227-248.
2. Padhy, N.P., 2004. Unit commitment-a bibliographical
survey. Power Systems, IEEE Transactions on,
19(2): 1196-1205.
3. Lowery, P., 1966. Generating unit commitment by
dynamic programming. Power Apparatus and
Systems, IEEE Transactions on, 5: 422-426.
4. Takriti, S. and J.R. Birge, 2000. Using integer
programming to refine Lagrangian-based unit
commitment solutions. Power Systems, IEEE
Transactions on, 15(1): 151-156.
5. Chen, C.L. and S.C. Wang, 1993. Branch-and-bound
scheduling for thermal generating units. Energy
Conversion, IEEE Transactions on, 8(2): 184-189.
6. Senjyu, T., K. Shimabukuro, K. Uezato and T.
Funabashi, 2003. A fast technique for unit
commitment problem by extended priority list. Power
Systems, IEEE Transactions on, 18(2): 882-888.
dynamic programming for unit commitment
application. Power Systems, IEEE Transactions on,
6(3): 1203-1209.
8. Ongsakul, W. and N. Petcharaks, 2004. Unit
commitment by enhanced adaptive Lagrangian
relaxation. Power Systems, IEEE Transactions on,
19(1): 620-628.
9. Zhao, B., C. Guo, B. Bai and Y. Cao, 2006. An
improved particle swarm optimization algorithm for
unit commitment. International Journal of Electrical
Power & Energy Systems, 28(7): 482-490.
10. Sum-Im, T. and W. Ongsakul, 2003. Ant colony
search algorithm for unit commitment. In the
proceedings the 2003 Industrial Technology, 2003
IEEE International Conference on, pp: 72-77.
11. Juste, K., H. Kita, E. Tanaka and J. Hasegawa, 1999.
An evolutionary programming solution to the unit
commitment problem. Power Systems, IEEE
Transactions on, 14(4): 1452-1459.
12. Victoire, T. andA. Jeyakumar, 2005. Unit commitment
by a tabu-search-based hybrid-optimisation
technique. In the proceedings the 2005 Generation,
Transmission and Distribution, IEE Proceedings,
pp: 563-574.
13. Mantawy, A., Y.L. Abdel-Magid and S.Z. Selim, 1998.
A simulated annealing algorithm for unit commitment.
PowerSystems,IEEETransactions on, 13(1): 197-204.
14. Nayak, R. and J. Sharma, 2000. A hybrid neural
network and simulated annealing approach to the
unit commitment problem. Computers & Electrical
Engineering, 26(6): 461-477.
15. Rajan, C.C.A. and M. Mohan, 2004. An evolutionary
programming-based tabu search method for solving
the unit commitment problem. Power Systems, IEEE
Transactions on, 19(1): 577-585.
16. Rajan, C.A., M. Mohan and K. Manivannan, 2003.
Neural-based tabu search method for solving unit
commitment problem. IEE Proceedings-Generation,
Transmission and Distribution, 150(4): 469-474.
17. Dasgupta, D. and D.R. McGregor, 1993. Short term
unit-commitment using genetic algorithms. In the
proceedings the 1993 Tools with Artificial
Intelligence, 1993. TAI'93. Proceedings., Fifth
International Conference on, pp: 240-247.
18. Ma, X., A. El-Keib, R. Smith and H. Ma, 1995. A
genetic algorithm based approach to thermal unit
commitment of electric power systems. Electric Power
Systems Research, 34(1): 29-36.

8. World Appl. Sci. J., 30 (12): 1826-1833, 2014
1832
19. Yang, H.T., P.C. Yang and C.L. Huang, 1995. 30. Wei, T., 2006. Unit Commitment of Power System
Applications of the genetic algorithm to the unit Based on Improved Genetic Algorithm.
commitment problem in power generation industry. Electrotechnical Application, 1 023.
In the proceedings the 1995 Fuzzy Systems, 1995. 31. Chen, C.Y., 2006. Integration of Genetic Algorithm
International Joint Conference of the Fourth IEEE and Taguchi Method for Thermal Unit Commitment.
International Conference on Fuzzy Systems and The 32. Jia, J. and J.F. Zhang, 2007. Research on Generator
Second International Fuzzy Engineering Symposium., Unit Optimal Commitment Based on the Genetic
Proceedings of 1995 IEEE International Conference Algorithm. Journal of Hohai University Changzhou,
on, pp: 267-274. 1 003.
20. Maifeld, T.T. and G.B. Sheble, 1996. Genetic-based 33. Wen-Ping, C. and L. Xian-Jue, 2008. Application of
unit commitment algorithm. Power Systems, IEEE double matrix hybrid coded genetic algorithm in unit
Transactions on, 11(3): 1359-1370. commitment. In the proceedings the 2008 Electricity
21. Orero, S. and M. Irving, 1996. A genetic algorithm for Distribution, 2008. CICED 2008. China International
generator scheduling in power systems. International Conference on, pp: 1-6.
Journal of Electrical Power & Energy Systems, 34. Amjady, N. and A. Shirzadi, 2009. Unit commitment
18(1): 19-26. using a new integer coded genetic algorithm.
22. Mantawy, A., Y.L. Abdel-Magid and S.Z. Selim, 1997. European Transactions on Electrical Power,
A New Genetic Algorithm Approach for Unit 19(8): 1161-1176.
Commitment. 35. Abookazemi, K., M.W. Mustafa and H. Ahmad, 2009.
23. Petridis,V.,S. KazarlisandA.Bakirtzis,1998. Varying Structured genetic algorithm technique for unit
fitness functions in genetic algorithm constrained commitment problem. International Journal of Recent
optimization: the cutting stock and unit commitment Trends in Engineering, 1(3): 135-139.
problems. Systems, Man and Cybernetics, Part B: 36. Senthil, Kumar V. and M. Mohan, 2010.
Cybernetics, IEEE Transactions on, 28(5): 629-640. Solution to security constrained unit commitment
24. Senjyu, T., H. Yamashiro, K. Uezato and T. problem using genetic algorithm. International
Funabashi, 2002. A unit commitment problem by Journal of Electrical Power and Energy Systems,
using genetic algorithm based on unit characteristic 32(2): 117-125.
classification. In the proceedings the 2002 Power 37. Pavez-Lazo, B. and J. Soto-Cartes, 2011. A
Engineering Society Winter Meeting, 2002. IEEE, deterministic annular crossover genetic algorithm
pp: 58-63. optimisation for the unit commitment problem.
25. Arroyo, J.M. and A.J. Conejo, 2002. A parallel repair Expert Systems with Applications, 38(6): 6523-6529.
genetic algorithm to solve the unit commitment 38. Abookazemi, K., H. Ahmad, A. Tavakolpour and
problem. Power Systems, IEEE Transactions on, M.Y. Hassan, 2011. Unit commitment solution using
17(4): 1216-1224. an optimized genetic system. International Journal of
26. Swarup, K. and S. Yamashiro, 2002. Unit commitment Electrical Power & Energy Systems, 33(4): 969-975.
solution methodology using genetic algorithm. 39. Shobana, S. and R. Janani, Optimization of Unit
Power Systems, IEEE Transactions on, 17(1): 87-91. Commitment Problem and Constrained Emission
27. Damousis, I.G., A.G. Bakirtzis and P.S. Dokopoulos, using Genetic Algorithm.
2004. A solution to the unit-commitment problem 40. Dhanalakshmi, S., S. Baskar, S. Kannan and
using integer-coded genetic algorithm. Power K. Mahadevan, 2013. Generation scheduling problem
Systems, IEEE Transactions on, 19(2): 1165-1172. by intelligent genetic algorithm. Computers &
28. Dudek, G., 2004. Unit commitment by genetic Electrical Engineering, 39(1): 79-88.
algorithm with specialized search operators. Electric 41. Dudek, G., 2013. Genetic algorithm with binary
Power Systems Research, 72(3): 299-308. representation of generating unit start-up and
29. Senjyu, T., A. Saber, T. Miyagi, K. Shimabukuro, shut-down times for the unit commitment problem.
N. Urasaki and T. Funabashi, 2005. Fast technique Expert Systems with Applications.
for unit commitment by genetic algorithm based on 42. Datta, D., 2013. Unit commitment problem with ramp
unit clustering. IEE Proceedings-Generation, rate constraint using a binary-real-coded genetic
Transmission and Distribution, 152(5): 705-713. algorithm. Applied Soft Computing.

9. World Appl. Sci. J., 30 (12): 1826-1833, 2014
1833
43. Cai, X.Q. and K.M. Lo, 1997. Unit commitment by a 48. Cheng, C.P., C.W. Liu and C.C. Liu, 2002. Unit
genetic algorithm. Nonlinear Analysis: Theory, commitment by annealing-genetic algorithm.
Methods & Applications, 30(7): 4289-4299. International Journal of Electrical Power & Energy
44. Huang, C.L., 1997. Application of genetic-based Systems, 24(2): 149-158.
neural networks to thermal unit commitment. Power 49. Liao, G.C. and T.P. Tsao, 2004. Application
Systems, IEEE Transactions on, 12(2): 654-660. embedded chaos search immune genetic algorithm
45. Mantawy, A., Y.L. Abdel-Magid and S.Z. Selim, 1999. for short-term unit commitment. Electric Power
A new genetic-based tabu search algorithm for unit Systems Research, 71(2): 135-144.
commitment problem. Electric Power Systems 50. Mantawy, A., 2004. A genetic algorithm solution to
Research, 49(2): 71-78. a new fuzzy unit commitment model. Electric Power
46. Takata, T., J. Takahashi, H. Yokoi, H. Nakano, Systems Research, 72(2): 171-178.
M. Aoyagi, M. Kato, K. Shimada and J. Arai, 1999. 51. Kumar, V.S. and M. Mohan, 2008. Genetic Algorithm
Advanced method for unit commitment problem with Improved Lambda-Iteration Technique to solve
using genetic algorithm and mathematical the Unit Commitment Problem. proc. of Intl. Journal
programming. Transactions-institute of Electrical on Electrical and Power Engineering, 2(2): 85-91.
Engineers of Japan B, 119: 333-343. 52. Zhang, J., T.Q. Liu, P. Su and X. Zhang, 2009. Unit
47. Mantawy, A., 2000. A genetic-based algorithm for commitment optimization based on genetic algorithm
fuzzy unit commitment model. In the proceedings the and particle swarm optimization hybrid algorithm [J].
2000 Power Engineering Society Summer Meeting, Power System Protection and Control, pp: 9.
2000. IEEE, pp: 250-254.
View publication stats