scholarly journals Analysis of Particle Swarm Optimization Algorithms with Multiswarms and Time-Varying Parameters

2011 ◽  
Vol 2011 (0) ◽  
pp. 277-282
Author(s):  
Yuji Wakasa ◽  
Shuichi Ohno ◽  
Kanya Tanaka ◽  
Yuki Nishimura
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Algorithms ◽  
Particle Swarm ◽  
Time Varying ◽  
Swarm Optimization ◽  
Varying Parameters ◽  
Time Varying Parameters

Inverse estimation of time-varying heat transfer coefficients for a hollow cylinder by using self-learning particle swarm optimization

2021 ◽  
Vol 0 (0) ◽  
Author(s):  
Kun-Yung Chen ◽  
Te-Wen Tu
Keyword(s):  
Heat Transfer ◽  
Particle Swarm Optimization ◽  
Objective Function ◽  
Hollow Cylinder ◽  
General Type ◽  
Particle Swarm ◽  
Time History ◽  
Time Varying ◽  
Swarm Optimization ◽  
Self Learning

Abstract An inverse methodology is proposed to estimate a time-varying heat transfer coefficient (HTC) for a hollow cylinder with time-dependent boundary conditions of different kinds on inner and outer surfaces. The temperatures at both the inner surface and the interior domain are measured for the hollow cylinder, while the time history of HTC of the outer surface will be inversely determined. This work first expressed the unknown function of HTC in a general form with unknown coefficients, and then regarded these unknown coefficients as the estimated parameters which can be randomly searched and found by the self-learning particle swarm optimization (SLPSO) method. The objective function which wants to be minimized was found with the absolute errors between the measured and estimated temperatures at several measurement times. If the objective function converges toward the null, the inverse solution of the estimated HTC will be found eventually. From numerical experiments, when the function of HTC with exponential type is performed, the unknown coefficients of the HTC function can be accurately estimated. On the contrary, when the function of HTC with a general type is conducted, the unknown coefficients of HTC are poorly estimated. However, the estimated coefficients of an HTC function with the general type can be regarded as the equivalent coefficients for the real function of HTC.

Sign in / Sign up

Export Citation Format

Share Document