Evolutionary computation;
parameter estimation;
solar cell model;
teaching learning based optimization;
DIFFERENTIAL EVOLUTION;
DISPATCH PROBLEM;
ALGORITHM;
IDENTIFICATION;
STRATEGY;
PATH;
D O I:
10.31209/2018.100000042
中图分类号:
TP [自动化技术、计算机技术];
学科分类号:
0812 ;
摘要:
Weak global exploration capability is one of the primary drawbacks in teaching learning based optimization (TLBO). To enhance the search capability of TLBO, an improved TLBO (ITLBO) is introduced in this study. In ITLBO, a uniform random number is replaced by a normal random number, and a weighted average position of the current population is chosen as the other teacher. The performance of ITLBO is compared with that of five meta-heuristic algorithms on a well-known test suite. Results demonstrate that the average performance of ITLBO is superior to that of the compared algorithms. Finally, ITLBO is employed to estimate parameters of two solar cell models. Experiments verify that ITLBO can provide competitive results.