Research Article  Open Access
Chengfen Zhang, Xikui Ma, "NSGAII Algorithm with a Local Search Strategy for Multiobjective Optimal Design of DryType AirCore Reactor", Mathematical Problems in Engineering, vol. 2015, Article ID 839035, 9 pages, 2015. https://doi.org/10.1155/2015/839035
NSGAII Algorithm with a Local Search Strategy for Multiobjective Optimal Design of DryType AirCore Reactor
Abstract
Drytype aircore reactor is now widely applied in electrical power distribution systems, for which the optimization design is a crucial issue. In the optimization design problem of drytype aircore reactor, the objectives of minimizing the production cost and minimizing the operation cost are both important. In this paper, a multiobjective optimal model is established considering simultaneously the two objectives of minimizing the production cost and minimizing the operation cost. To solve the multiobjective optimization problem, a memetic evolutionary algorithm is proposed, which combines elitist nondominated sorting genetic algorithm version II (NSGAII) with a local search strategy based on the covariance matrix adaptation evolution strategy (CMAES). NSGAII can provide decision maker with flexible choices among the different tradeoff solutions, while the localsearch strategy, which is applied to nondominated individuals randomly selected from the current population in a given generation and quantity, can accelerate the convergence speed. Furthermore, another modification is that an external archive is set in the proposed algorithm for increasing the evolutionary efficiency. The proposed algorithm is tested on a drytype aircore reactor made of rectangular crosssection litzwire. Simulation results show that the proposed algorithm has high efficiency and it converges to a better Pareto front.
1. Introduction
As an important apparatus applied to harmonic filtering, short circuit current limiting, and reactive power compensation, drytype aircore reactor plays a vital role in reducing failure and improving security of power system operation. In the past two decades, the design problem for reducing the production cost and the operation cost of drytype aircore reactor has received considerable attention. Most literatures about the design problem of drytype aircore reactors are mainly focused on the following aspects: the fast and accurate computation of the inductance, magnetic field, and winding power loss [1–4]; the complete model for impulse voltage distribution and electrical strength [5, 6]; and acoustic noise evaluation or controls [7, 8]. Up to now, however, only a few works have been published on the optimal design problem of drytype aircore reactor. For example, Liu et al. [9] employed the genetic algorithm to minimize the operation cost measured by the power loss of drytype aircore reactor, and obtained the same results as the traditional design method only. In [10], based on the balance of the additional equality constraint conditions, a hybrid method of genetic algorithm and simplex method is proposed for the optimum design of the roundwire drytype aircore reactor to enhance the local searching ability and improve the optimization efficiency. Unfortunately, the optimum design of drytype aircore reactor is considered as a single objective problem with several constraints, although it involves more than one objective.
Actually, it is well known that the optimum design of drytype aircore reactor is a very complex problem as it requires the simultaneous minimization of two objective functions. The first objective deals with the minimization of the production cost by reducing aluminum wire weight. The second objective minimizes the operation cost by the minimization of power loss. And the voltage, height, package temperature rise and current density may be considered as constraint conditions. In theory, these two objectives are in conflict with each other. No single solution can be claimed as an optimum solution to multiple conflicting objectives, the resulting multiobjective optimization problem involves a number of tradeoff solutions. In order to provide a means to assess tradeoff between two conflicting objectives, one way is to formulate the optimum design of drytype aircore reactor as a multiobjective optimization problem.
This paper is concerned with the optimum design of drytype aircore reactor made of rectangular crosssection litzwire as a multiobjective optimization problem. Due to the nondifferentiality, nonlinearity, highconstraint, and highdiscretion of the optimum design problem, most traditional optimization or search approaches (e.g., Sequential Search, Tabu Search, etc.) not only often fall into local maxima and minima, but also are not easily applied to solve the optimization problem with multiobjectives. As one of the elegant approaches, the elitist nondominated sorting genetic algorithm II (NSGAII) [11] has been demonstrated to be one of the efficient algorithms for solving multiobjective optimization problems such as generation expansion planning problem [12, 13], economic and environmental power dispatch [14], multipurpose water reservoir management [15], and the periodic operation of the naphtha pyrolysis process [16]. Although NSGAII is a powerful algorithm that captures a global search space and obtains welldistributed Pareto front, it is very timeconsuming to solve the complex design problem; that is, it suffers from the lower convergence speed, especially at the early stage of the optimization. Therefore, it is of great importance to enhance the algorithm’s search ability. One of our aims is to makes use of a local search strategy to speed up the convergence of the NSGAII.
In this study, the NSGAII algorithm with certain modifications is applied to settle the multiobjective problem. As an approach to improve the NSGAII algorithm, a local search strategy based on the Covariance Matrix Adaptation Evolution Strategy (CMAES) [17] is incorporated into the original NSGAII algorithm to promote algorithm’s exploiting capability and then to accelerate the algorithm convergence for the multiobjective optimum design of drytype aircore reactor made of rectangular crosssection litzwire, hereafter designated as the NSGACMA algorithm. In practice, the local search is applied to the certain selected nondominated individuals of a given generation. Furthermore, an external archive is set in the proposed algorithm for improving the evolutionary efficiency especially at the early stage. To validate the performance of the proposed algorithm, the performance measures of the convergence metric and the two set coverage measure are considered. The effectiveness of the proposed algorithm is tested on a 50 kVar (i.e., 317.5 V, 157.5 A, 6.42 mH) drytype aircore reactor. The results show that the proposed algorithm not only converges to a better Pareto front, but also is of very high efficiency.
The paper is organized as follows. The optimum design of drytype aircore reactor is formulated as a multiobjective optimization problem with a number of equality constraints and inequality constraints in Section 2. Then, the details of the proposed NSGACMA is described in Section 3 and its implementation to the optimum design of drytype aircore reactor is outlined in Section 4, the experiment results and the comparisons with the original NSGAII are also presented. Finally, some conclusions are drawn in Section 5.
2. Problem Formulation
The drytype aircore reactor, the diagram of which is shown in Figure 1, consists of several parallel packages; adjacent packages are connected by spacer bars; each package is composed of parallel multilayer coil windings; both ends of the coil windings are welded in spider frame made of aluminum. In this paper, four mutually independent variable parameters defining the structure of drytype aircore reactor are subjected to optimization. The design variables are the number of winding package , the number of winding layer , the inner diameter , and the average side length of litzwire . In these design variables, the inner diameter and the average side length are taken as continuous variables, whereas the number of winding package and the number of winding layer are taken as discrete values.
The optimal design of drytype aircore reactor is equivalent to finding a set of best decision vectors that minimizes the two competing objective functions, the production, and the operation cost, subject to a number of equality constraints and inequality constraints. Apparently, the multiobjective formulation is a choice way to treat the nonlinear and constrained optimal problem. Its mathematical model can be described as follows.
2.1. The Objective Functions
Since minimizing the aluminum wire weight is equivalent to minimizing the production cost, the first objective function is to minimize the aluminum wire weight, usually defined aswhere is the aluminum wire weight, the mass density of aluminum, and the radialside length and the axialside length of the th layer litzwire, respectively, the middle diameter (the average value of inner diameter and outer diameter) of the th layer litzwire, the turns of the th layer litzwire.
It is well known that the smaller power loss results in the less operation cost, so the second objective function can be represented by the following expression:where is the power loss, the conductivity of aluminum, the current through the th layer litzwire, the angular frequency, the magnetic flux density’s axial component of the th layer litzwire at the th turn, and the magnetic flux density’s radial component of the th layer litzwire at the th turn.
2.2. Design Constraints
The objective functions are subjected to following some equality constraints and inequality constraints.
2.2.1. Equality Constraints
In order to design a high performance drytype aircore reactor, some additional equality constraints are proposed. There are two most common additional constraints which are layer resistance drop balance constraint and package temperature rise balance constraint. Layer resistance drop balance can assure that the aircore reactor has the minimum power loss; package temperature rise balance can assure the best heat sinking effect. In addition, in order to make the aircore reactor have more compact structure and reduce the manufacturing cost, the package height balance constraint should be considered too. The three additional equality constraints are expressed as follows.Layer resistance drop balance constraint:Package height balance constraint: where is the height of the th package.Package temperature rise balance constraint: where is the expectant degree of users, is the temperature rise of the th package and is defined as where is unit area heat load of the th package.
2.2.2. Inequality Constraints
(1) Current Density Limits. For stable operation, the current density of each layer litzwire must fail into a region of operation defined by lower and upper limits:where is the current density of the th layer litzwire and and are the lower and the upper limits of the current density, respectively. And is defined as follows:
(2) Design Variable Limits. The limits on these design variables are given by the following inequalities:where the superscripts “min” and “max” denote the corresponding lower and upper limits, respectively, is defined as
It is now evident from the above discussion that the optimal design problem of drytype aircore reactor is a nonlinear engineering optimization problem without analytical expression, that is, the objective functions cannot be expressed as the functions of the design variables in analytical form. However, the variable values in the objective functions can be obtained from the above four mutually independent design variables in the process of design.
3. The Proposed NSGACMA Algorithm
3.1. Analysis of NSGAII Algorithm
NSGAII algorithm has been demonstrated as one of the most efficient and famous algorithms for multiobjective optimization. It uses the fast nondominated sorting to rank the population fronts and a parameter called crowding distance is calculated in the same front. Then, tournament selection is made between two individuals randomly selected from parent population. The individual with lower front number is selected if the two individuals come from different fronts. The individual with higher crowding distance is selected if the two individuals are from the same front. Then, both the crossover and the mutation operators are used to generate a new offspring population. Finally, the parent and offspring populations are combined together where a fast nondominated sorting and crowding distance assignment procedure is used to rank the combined population and only the best individuals are selected as the new parent population.
NSGAII is proposed on basis of nondominated sorting genetic algorithm (NSGA) [18], and the main advantages of NSGAII compared with NSGA are as follows: it uses a fast nondominated sorting approach, it has no sharing parameter, and it uses an elitist strategy. NSGAII has good global search ability and well distributed nondominated solutions in the Pareto front. But the convergence capability of NSGAII is limited when solving the engineering optimal problem such as the multiobjective optimum design of drytype aircore reactor, and its elite strategy may result in the loss of the nondominated solutions in the Pareto front.
Memetic algorithms perform global exploration by evolutionary algorithms and local exploitation by a local search strategy, respectively. It has global search ability of evolutionary algorithm and local search ability of local search strategy simultaneously. It is reported that memetic algorithm converges to high quality solutions more efficiently than evolution algorithms [19, 20]. To improve the efficiency of NSGAII, in this paper, a local search strategy based on the CMAES is introduced into the algorithm to accelerate the speed of convergence. In addition, an external archive is set in NSGAII to increase the evolutionary efficiency.
3.2. Local Search Strategy Based on the CMAES
Local search can enhance the search capability of evolutionary algorithms by carrying out local exploitation, and the global and local searches may be well balanced. The CMAES is probably one of the most powerful selfadaptation mechanisms for continuous search spaces. It uses a covariance matrix to construct the mutation distribution and adapts this covariance matrix from cumulative paths of successful mutations. Firstly, it samples a number of new candidate solutions from a multivariate normal distribution and then updates the distribution by means of two major strategies: step size update and covariance matrix adaptation. The process in five steps is given and explained as follows.
Step 1. Initialize , , , , and as follows. , , (unity matrix). and are given as inputs,where is the generation counter and the superscript “” represents the th generation, is the dimension of searching space and also the number of the variable for local search, is the population size, is the object parameter vector of the th individual, is the covariance matrix, is the evolution path used to speed up the adaptation of the covariance matrix , is also the evolution path that is similar to , and is the global step size.
Step 2. At each generation , compute the object parameter vector as follows:where is the number of the selected search points in the population; represents the weighted mean of the best individuals of generation ; is the weight coefficient; the subscript denotes the th best individual; is the distributed random vector of the th individual; the components of are independent distribution; determines and ; the singular value decomposition of can be expressed as .
Step 3. Update the covariance matrix by means of the evolution path :where determines the cumulative time for , is the change rate of , and the expressions of and are given as follows:respectively.
Step 4. Update the global step size by using the evolution path :where determines the cumulative time for ; determines the possible change rate of ; is the expectation of the length of the distributed random vector, which is approximated by .
Step 5. The computation is repeated until the maximum number of generations is met.
It should be pointed out that the local search strategy based on CMAES cannot be directly used to handle multiple objectives. To apply the local search strategy based on CMAES to the constrained twoobjective optimization problem in this paper, the following overall objective is calculated by linear combination of different objectives as a weighted sum:where and are the weight coefficients and and is the constraint violation value used as the penalty function. Here, to avoid prioritybased weight, the values of the two weight coefficients are set to be equal, that is, , and the normalized objective function values are to be used.
Although the local search strategy may improve the performance of the original NSGAII, it implies an additional cost, which may become prohibitive in the optimization design of drytype aircore reactor. When the quality of an individual is very poor, the application of local search strategy also seems to be waste of the computing time. Thus, the local search (i.e., CMAES) is applied only to nondominated individuals randomly selected from the current population obtained from elitism of a given generation. Actually, our experience has shown that in drytype aircore reactor the numbers of winding package and winding layer have a great influence upon the results. Therefore, in order to get good search results, the local search is applied only to the inner diameter and average side length.
Since the overall objective function is used in the CMAES method, the points obtained from the CMAES should be checked for nondominance before they are accepted in the current population. If the point obtained by the CMAES dominates any point in the current population, the dominated point must be replaced by the new point.
3.3. External Archive
In each generation of the original NSGAII, a new offspring population is generated. The parent and offspring populations are combined together where a fast nondominated sorting procedure is used to sort the combined population and only the best individuals are selected.
In order to increase the evolutionary efficiency especially at the early generation, an external archive of size is introduced into NSGAII, and individuals of the archive participate in the tournament selection together with individuals of the parent population. Individuals in the first front of the combined population are stored in the archive, when the archive exceeds the maximal size , the individuals with the lowest crowding distance are to be discarded.
3.4. Procedures of the Proposed Algorithm
Let and denote the generation number and the maximum generation number, respectively. The local search is applied to nondominated individuals each generation. The application of NSGACMA algorithm to the multiobjective optimization design of drytype aircore reactor can be written as follows. The corresponding flowchart is given in Figure 2.
Step 1 (initialization). Input the initial parameters of the 50 kVar drytype aircore reactor and the proposed algorithm; randomly generate an initial population; set .
Step 2 (evaluation). Calculate the values of the objective functions and constraints for each individual in the current population.
Step 3 (fast nondominated sorting and crowding distance assignment). Apply nondominated sorting to classify the current population into different nondominated fronts and then calculate the crowding distance of each individual in the same front.
Step 4 (selection). Perform tournament selection between two individuals randomly selected from the parent population and external archive.
Step 5 (crossover and mutation). Apply a crossover operator to each of the randomly selected pairs of parent individuals and a new population is generated. Then, apply a mutation operator to the new population and a new offspring population is obtained.
Step 6 (elitist strategy). Combine parent and offspring population to generate a combined population and then apply fast nondominated sorting and crowding distance assignment to the combined population. Select best individuals as a new parent population according to nondominated front and crowding distance.
Step 7 (local search). If , apply local search strategy based on CMAES to nondominated individuals randomly selected from the new parent population. Then, update the new parent population. Otherwise, go to Step 8.
Step 8 (external archive). In the new generated parent population, individuals in the first nondominated front are stored in the external archive. When the archive exceeds the maximal size , individuals with lower crowding distance are to be discarded.
Step 9 (termination). If , end the algorithm. Otherwise, return to Step 2.
4. Test Results and Discussion
To verify the effectiveness of the proposed algorithm, NSGAII and NSGACMA are used for the optimization design of a 50 kVar (i.e., 317.5 V, 157.5 A, and 6.42 mH) drytype aircore reactor made of rectangular crosssection litzwire, and comparisons are made between their results.
4.1. Mixed Variable Handling Method
There are four mixed design variables in the optimization design of drytype aircore reactor. The number of winding package and the number of winding layer are integer variables and the inner diameter and the average side length of the litzwire are continuous variables. Here, the hybrid code is used, that is, the binary code for the integer variables and the real code for the continuous variables.
4.2. Parameter Settings
The main parameters of a 50 kVar drytype aircore reactor are the extreme values of design variables and the constraint conditions, which are given in Table 1.

The parameter settings of CMAES, which are discussed in [17], are summarized in Table 2. The parameters and are comparatively uncritical and can be chosen in a wide range.

In the proposed algorithm, the crossover probability is denoted as , the mutation probabilities as for binary code and as for real code; the distribution indexes are expressed as for crossover operator and as for mutation operator. In the local search procedure, let denote the number of generation for local search and and the neighborhood size of the inner diameter and the average side length of the selected nondominated individuals, respectively. The parameters for NSGACMA are given in Table 3. Here, the parameters for NSGAII are set at the same values as NSGACMA to ensure the comparison is fair.

4.3. Performance Measurements
In order to evaluate the performance of the proposed algorithm, the convergence metric [21] and the two set coverage measure [22] are used as evaluation indexes.
The convergence metric measures the extent of convergence towards a reference set, and lower values of the metric represent good convergence ability. Let be the population of the th generation, the reference or target set, in which the points can be either a set of Pareto optimal points (if known) or the nondominated set of points in a combined pool of all generationwise populations, and the final approximate Pareto optimal set obtained from . Then, the smallest normalized Euclidean distance to can be calculated from each point in as follows:where is the number of objective functions, is the th objective function value of individual , and are the maximum and minimum values of the th objective function in , respectively. The convergence metric is the average value of the normalized distance for all points in :
To keep the convergence metric within , we normalize the values according to .
Let and be two approximate Pareto optimal sets. The twoset coverage measure is defined aswhere means dominate or equal (also called weakly dominate); and are the elements of and , respectively. The value means that all the elements of are weakly dominated by . implies that no element of is weakly dominated by . Since the two set coverage measure is not always symmetric, both values and have to be considered.
4.4. Results and Discussion
To compare the proposed algorithm with NSGAII, the two algorithms run for 20 times, respectively, with 20 initial populations of random solutions (one for each run). For the optimization problem discussed in this paper, there is no known true Pareto front, so a reference set is used to calculate the performance metric. The reference set represented as is obtained by adopting all the Pareto optimal solutions from all of the 40 runs.
To demonstrate the effectiveness of the algorithm, the parameter function evaluation is used, and its budget is set to 10000. Since the function evaluation always consumes most of the time of the algorithm for the engineering optimization problems, and the overall number of function evaluations is high.
Figure 3 shows the changes in the mean values of the convergence metric of NSGAII and NSGACMA versus the number of function evaluation. It can be seen that the average convergence metric values of the two algorithms are similar in the early iterative process. The average convergence metric values of NSGACMA are significantly decreased when the local search strategy based on CMAES is performed. After that, the average convergence metric values of the two algorithms are all decreased slowly, and the convergence metric values of NSGACMA are always smaller than those of NSGAII algorithm. It means that the nondominated solutions obtained by NSGACMA are obviously much more converged to the reference set than the original NSGAII especially in early generation.
Figure 4 presents the mean values of the two set coverage metric and against the number of function evaluation; here and are the Pareto optimal sets of NSGAII and NSGACMA, respectively. It is seen that the two set coverage metrics and are similar in the initial stage, is increased evidently when the local search strategy is performed, then the two set coverage metrics and decrease gradually and is always lower than . It means that NSGACMA is more efficient than the original NSGAII, and it has more nondominated solutions which dominate the nondominated solutions obtained by NSGAII.
In order to intuitively compare the distribution of the solutions obtained by the two algorithms, Figures 5 and 6 show the Pareto fronts obtained from the two algorithms at the 2500th and 10000th function evaluations, respectively. It can be seen from Figure 5 that all the solutions do not convergence to the reference set at this time, but the Pareto optimal solutions obtained by NSGACMA are more close to the reference set . It can be seen from Figure 6 that the Pareto optimal solutions obtained by NSGACMA are almost the same as the reference set , however, many solutions among the Pareto optimal solutions obtained by NSGAII are dominated by the solution in the reference set . We can conclude that the improved algorithm is more efficiency than the original NSGAII and it can obtain a better nondominated Pareto front which provides a spectrum of tradeoff among the competing objectives.
It can be seen from Table 4 that the aluminum wire weight and the power loss obtained by NSGACMA can be reduced at 16.31% and 20.32% at most, respectively, compared with the computer aided design result (aluminum wire weight: 83.28 kg, power loss: 1686.00 W) [23].

5. Conclusions
In this paper, the multiobjective optimization design of drytype aircore reactor made of rectangular crosssection litzwire is studied considering the minimal production cost as well as the minimal operation cost simultaneously. A modified nondominated sorting genetic algorithmII (called NSGACMA) is then proposed to solve this multiobjective optimal problem. Within the NSGACMA, the global exploration is done by NSGAII and the local exploitation by the local search strategy based on CMAES to improve the search ability and accelerate the convergence speed. In order to guarantee the search efficiency, the local search is applied to nondominated individuals in a given generation and quantity. In addition, an external archive is established for improving the evolutionary efficiency. From the simulation results, it is clearly seen that NSGACMA has smaller convergence metric value and more nondominated solutions which dominate the nondominated solutions obtained by NSGAII. This implies that NSGACMA has better search ability and it obtains better Pareto front than NSGAII. For its promising performance, the NSGACMA algorithm is certainly more suitable and effective than NSGAII for solving the multiobjective design problems of drytype aircore reactor. Also, it is one of the efficient potential candidates in solving other complicated multiobjective problems.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
References
 S. I. Babic and C. Akyel, “New analyticnumerical solutions for the mutual inductance of two coaxial circular coils with rectangular cross section in air,” IEEE Transactions on Magnetics, vol. 42, no. 6, pp. 1661–1669, 2006. View at: Publisher Site  Google Scholar
 S. I. Babic and C. Akyel, “An improvement in the calculation of the magnetic field for an arbitrary geometry coil with rectangular cross section,” International Journal of Numerical Modelling: Electronic Networks, Devices and Fields, vol. 18, no. 6, pp. 493–504, 2005. View at: Publisher Site  Google Scholar
 J. Morozionkov and J. A. Virbalis, “Magnetic field of power plant air core reactor,” Elektronika Ir Elektrotechnika, vol. 79, no. 7, pp. 67–70, 2007. View at: Google Scholar
 B. S. Ram, “Inductance and winding eddy loss of aircored reactors,” IEE Proceedings—Generation Transmission and Distribution, vol. 146, no. 5, pp. 416–420, 1999. View at: Google Scholar
 A. A. Dahab, P. E. Burke, and T. H. Fawzi, “A complete model of a single layer aircored reactor for impulse voltage distribution,” IEEE Transactions on Power Delivery, vol. 3, no. 4, pp. 1745–1753, 1988. View at: Publisher Site  Google Scholar
 S. L. Varricchio and N. H. C. Santiago, “Electrical strength in air core reactors,” in Proceedings of the 4th IEEE International Conference on Properties and Application of Dielectric Materials, pp. 876–879, 1994. View at: Google Scholar
 J.L. Lilien, “Acoustic noise generated by air power reactor in openair substation,” European Transactions on Electrical Power, vol. 16, no. 3, pp. 297–310, 2006. View at: Publisher Site  Google Scholar
 Q. Wang, Q. Chen, and M. Zeng, “A CFDtaguchi combined method for numerical investigation of natural convection cooling performance of aircore reactor with noise reducing cover,” Numerical Heat Transfer Part A: Applications, vol. 55, no. 12, pp. 1116–1130, 2009. View at: Publisher Site  Google Scholar
 Z.G. Liu, Y.S. Geng, J.H. Wang, and S. Ouyang, “Optimum design of drytype aircore series reactor based on modified adaptive genetic algorithm,” Proceedings of the Chinese Society of Electrical Engineering, vol. 23, no. 9, pp. 103–106, 2003. View at: Google Scholar
 Y. Zhao, F. Chen, B. Kang, and X. Ma, “Optimum design of drytype aircore reactor based on the additional constraints balance and hybrid genetic algorithm,” International Journal of Applied Electromagnetics and Mechanics, vol. 33, no. 12, pp. 279–284, 2010. View at: Publisher Site  Google Scholar
 K. Deb, A. Pratap, S. Agarwal, and T. Meyarivan, “A fast and elitist multiobjective genetic algorithm: NSGAII,” IEEE Transactions on Evolutionary Computation, vol. 6, no. 2, pp. 182–197, 2002. View at: Publisher Site  Google Scholar
 P. Murugan, S. Kannan, and S. Baskar, “NSGAII algorithm for multiobjective generation expansion planning problem,” Electric Power Systems Research, vol. 79, no. 4, pp. 622–628, 2009. View at: Publisher Site  Google Scholar
 S. Kannan, S. Baskar, J. D. McCalley, and P. Murugan, “Application of NSGAII algorithm to generation expansion planning,” IEEE Transactions on Power Systems, vol. 24, no. 1, pp. 454–461, 2009. View at: Publisher Site  Google Scholar
 L. Xuebin, “Study of multiobjective optimization and multiattribute decisionmaking for economic and environmental power dispatch,” Electric Power Systems Research, vol. 79, no. 5, pp. 789–795, 2009. View at: Publisher Site  Google Scholar
 L. A. Scola, R. H. C. Takahashi, and S. A. A. G. Cerqueira, “Multipurpose water reservoir management: an evolutionary multiobjective optimization approach,” Mathematical Problems in Engineering, vol. 2014, Article ID 638259, 14 pages, 2014. View at: Publisher Site  Google Scholar
 X. Gao, B. Chen, X. He et al., “Multiobjective optimization for the periodic operation of the naphtha pyrolysis process using a new parallel hybrid algorithm combining NSGAII with SQP,” Computers and Chemical Engineering, vol. 32, no. 11, pp. 2801–2811, 2008. View at: Publisher Site  Google Scholar
 N. Hansen and A. Ostermeier, “Completely derandomized selfadaptation in evolution strategies,” Evolutionary Computation, vol. 9, no. 2, pp. 159–195, 2001. View at: Publisher Site  Google Scholar
 N. Srinivas and K. Deb, “Multiobjective function optimization using nondominated sorting genetic algorithms,” Evolutionary Computation, vol. 2, no. 3, pp. 221–248, 1995. View at: Google Scholar
 N. Krasnogor and J. Smith, “A tutorial for competent memetic algorithms: model, taxonomy, and design issues,” IEEE Transactions on Evolutionary Computation, vol. 9, no. 5, pp. 474–488, 2005. View at: Publisher Site  Google Scholar
 H. Ishibuchi, T. Yoshida, and T. Murata, “Balance between genetic search and local search in memetic algorithms for multiobjective permutation flowshop scheduling,” IEEE Transactions on Evolutionary Computation, vol. 7, no. 2, pp. 204–223, 2003. View at: Publisher Site  Google Scholar
 K. Deb and S. Jain, “Running performance metrics for evolutionary multiobjective optimization,” Tech. Rep. 2002004, Indian Institute of Technology, Kanpur, India, 2004. View at: Google Scholar
 E. Zitzler, K. Deb, and L. Thiele, “Comparison of multiobjective evolutionary algorithms: empirical results,” Evolutionary computation, vol. 8, no. 2, pp. 173–195, 2000. View at: Publisher Site  Google Scholar
 B. Kang, Optimization model of drytype aircore reactor and optimization design based on hybrid genetic algorithm [M.S. thesis], Xi’an Jiaotong University, Xi’an, China, 2009.
Copyright
Copyright © 2015 Chengfen Zhang and Xikui Ma. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.