Metaheuristics for Intelligent Electrical Networks

Héliodore, Frédéric; Nakib, Amir; Ismail, Boussaad; Ouchraa, Salma; Schmitt, Laurent

doi:10.1002/9781119136736

Cited by 6 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Matrix (5) implies that C is the winner under the 'C2+' strategy, because C performs the best on 5 courses, while A and B perform the best on 4 courses. However, if we adopt the 'C2' strategy to analyze the data in Table III, we obtain the following three matrices,    …”

Section: A Examples Of the Two Paradoxesmentioning

confidence: 99%

“…Numerical comparison is an important means to testing and benchmarking the performance of metaheuristic algorithms [1][2][3][4][5][6][7][8], especially nondeterministic algorithms such as evolutionary algorithms [1][2][3][4]9]. In order to assess the performance of new optimization algorithms, competitions are held annually, e.g., the CEC Real-Parameter Optimization Competition [10,11] and the GECCO Black-Box Optimization Competition [12,13].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Paradox-Free Analysis for Comparing the Performance of Optimization Algorithms

Yuan

Liu

2023

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Numerical comparison serves as a major tool in evaluating the performance of optimization algorithms, especially nondeterministic algorithms, but existing methods may suffer from a 'cycle ranking' paradox and/or a 'survival of the nonfittest' paradox. This paper searches for paradox-free data analysis methods for numerical comparison. It is discovered that a class of sufficient conditions exist for designing paradox-free analysis. Rigorous modeling and deduction are applied to a class of profile methods employing a filter. It is thus further discovered and proven that algorithm-independent filter conditions can prevent cycle ranking and survival of non-fittest paradoxes from occurring. By adopting an algorithm-independent filter, popular profile methods such as the 'modified data profile method', 'the accuracy profile method', and 'the operational characteristics zones method' can be paradox free in comparing or benchmarking the performance of optimization algorithms.

show abstract

Section: A Examples Of the Two Paradoxesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Paradox-Free Analysis for Comparing the Performance of Optimization Algorithms

Yuan

Liu

2023

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

show abstract

“…The descent method is among the earliest and simplest method for local search [13] where the solution is chosen from the agent's neighbour. This simple method anyhow causes the agent to be easily trapped in a local optimal [14]. As single-agent metaheuristic algorithm has higher possibility to trap to the local optima [15], many researchers came out with strategies for local optimal avoidance.…”

Section: Introductionmentioning

confidence: 99%

“…The possibility to escape from local optima in SA is good due to the stochastic cooling factor [21]. However, there is still a possibility for SA to find the same local optimal again throughout the optimization process [14]. TS algorithm was developed by Glover in 1986 considering the use of local memory where the recent history of the search is memorized and stored in Tabu list, and prohibited to be revisited [19] [22].…”

Section: Introductionmentioning

confidence: 99%

“…TS algorithm was developed by Glover in 1986 considering the use of local memory where the recent history of the search is memorized and stored in Tabu list, and prohibited to be revisited [19] [22]. TS is biased to exploration if the size of Tabu list is increased, whereas it is biased to exploitation if the size of Tabu list is reduced [14]. A local search is applied to move from the current solution to the improved solution in the neighborhood of the current solution.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Study on the Effect of Local Neighbourhood Parameter towards the Performance of SAFIRO

Rahman¹,

Ibrahim²,

Azlina³

et al. 2018

IJET

View full text Add to dashboard Cite

Single-agent Finite Impulse Response Optimizer (SAFIRO) is a recently proposed metaheuristic optimization algorithm which adopted the procedure of the ultimate unbiased finite impulse response filter (UFIR) in state estimation. In SAFIRO, a random mutation with shrinking local neighborhood method is employed during measurement phase to balance the exploration and the exploitation process. Beta, β, is one of the parameters used in the local neighborhood to control the step size. In this study, the effect of β towards the performance of SAFIRO is observed by assigning the value of 1, 5, 10, 15, and 20. The best setting of β for SAFIRO is also determined. The CEC2014 Benchmark Test Suite is used to evaluate the SAFIRO performance with different β values. Results show that the performance of β is depending on the problems to be optimized. 17 out of 30 functions show the best performance of SAFIRO by setting β = 10. Statistical analysis using Friedman test and Holm post hoc test were performed to rank the performance. β = 10 has the highest rank where its performance is significantly better than other values, but equivalent to β = 5 and β = 15. Hence, it is recommended to tune the β for best performance, however, β = 10 is a good value to be used in SAFIRO for solving optimization problems.

show abstract

Performance Evaluation of Crow Search Algorithm on Capacitated Vehicle Routing Problem

Dhanya

Kanmani

2018

Soft Computing Systems

View full text Add to dashboard Cite

Metaheuristics for Intelligent Electrical Networks

Cited by 6 publications

References 0 publications

Paradox-Free Analysis for Comparing the Performance of Optimization Algorithms

Paradox-Free Analysis for Comparing the Performance of Optimization Algorithms

A Study on the Effect of Local Neighbourhood Parameter towards the Performance of SAFIRO

Performance Evaluation of Crow Search Algorithm on Capacitated Vehicle Routing Problem

Contact Info

Product

Resources

About