A Perturbed Particle Swarm Optimization using Harmony Search and Mutation

Phuchan, Wikrom; Kruatrachue, Boontee

doi:10.1109/iceast50382.2020.9165425

Cited by 1 publication

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This has led to the application of this algorithm over wide range of optimization problems. This algorithm has been really effective in terms of performance in solving difficult optimization problems, which has led to development of different versions of this algorithm 31–38 . In HSA, the possible solution is termed as harmony which is made up of decision variable.…”

Section: Harmony Search Algorithmmentioning

confidence: 99%

“…To minimize CF, harmonic search algorithm can be explained through below five steps (Figure 2) Initially HM is generated with random values within the solution space 33–40

x_{italicmn var} = l_{n var} + italicrand \times ((), u_{n var} - l_{n var})

where m = 1,2.…”

Section: Harmony Search Algorithmmentioning

confidence: 99%

“…It consists of two stages: Stage I: A random number rand is generated, if rand > HMCR (HM considering rate) then variable of new harmony is generated by Equation . If rand ≤ HMCR , then one of the variable from already existing harmony is assigned to new variable 33–40

x_{italicnew, n var} = x_{italicnew, n var} + italicrand \times (||, u_{n var} - l_{n var})

Stage II: pitch adjustment mechanism is used in this stage to avoid the solution which is local minima.…”

Section: Harmony Search Algorithmmentioning

confidence: 99%

“…A new parameter called PAR (pitch adjustment rate) is taken in the interval (0,1) and after Stage I, a random number rand is generated. If rand ≤ PAR , the value of generated variable is changed by using Equation , otherwise, it remains unchanged 33–40

x_{italicnew, n var} = x_{italicnew, n var} + ((), italicrand - 0.5) \times (||, u_{n var} - l_{n var})

The new harmony is verified, and through CF value, and it is decided whether to include or exclude newly generated harmony.…”

Section: Harmony Search Algorithmmentioning

confidence: 99%

“…The cost functions (CF) are used, and the results are compared to get a relative comparison of the effect of PSO and HSA. A detailed study of HSA is presented in other studies 33–40 . Power radiated is calculated analytically, and the necessary reduction in power is obtained after optimization.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Optimization for the reduction of power in sidebands and side lobes of time‐modulated antenna array

Ram

2021

Int J Communication

View full text Add to dashboard Cite

Summary Optimization for the reduction of power losses in sidebands and side lobes of time‐modulated antenna array (TMLAA) is dealt in this work. Optimization for the reduction of power in TMLAA is analyzed by considering two ways of the power loss analysis: one with the minimization of power loss with optimization of Toff instant with Ton instant as zero for each element along with optimal inter‐element spacing with uniform amplitude and second, by taking optimization of Ton and Toff instant of each element of TMLAA along with optimal inter‐element spacing with uniform amplitude. Distance between any two consecutive antennas and switch ON and OFF instants are considered as optimizing variables for reduction of power due to sidebands at harmonic frequencies and side lobes at the fundamental harmonic. Towards this approach, an evolutionary algorithm‐based approach called harmony search algorithm (HSA) is applied. Optimal values of the variable parameters of the TMLAA are obtained after optimization by using particle swarm optimization (PSO) and HSA. MATLAB simulation results are presented in this work for different array sets of 16‐, 20‐, 24‐, 28‐, and 32‐element antenna arrays. To validate in electromagnetic (EM)‐simulated environment, a 16‐element TMLAA is designed in CST‐MWS with parameters obtained after optimization, and these results are validated with those of MATLAB results.

show abstract

Section: Harmony Search Algorithmmentioning

confidence: 99%

“…To minimize CF, harmonic search algorithm can be explained through below five steps (Figure 2) Initially HM is generated with random values within the solution space 33–40