篇名 | An Improved Memetic Algorithm for Traction Characteristic Curve Fitting of Urban Rail Vehicle |
---|---|
卷期 | 31:1 |
作者 | Kai-Wei Liu 、 Xing-Cheng Wang 、 Long-Da Wang 、 Gang Liu |
頁次 | 094-105 |
關鍵字 | fruit fly algorithm 、 genetic algorithm 、 memetic algorithm 、 reverse learning 、 traction characteristic curve 、 EI 、 MEDLINE 、 Scopus |
出刊日期 | 202002 |
DOI | 10.3966/199115992020023101008 |
An Improved Memetic Algorithm (IMA) for the traction characteristic curve fitting of urban rail vehicle is studied in this paper. Combining strong global search ability of Genetic Algorithm (GA), quick convergence ability of Fruit Fly Algorithm (FFA) and strong local search ability of Hill Climbing Algorithm (HCA), this paper constructs a hybrid algorithm to improve the search efficiency for MA. A univariate equation with n orders of the instantaneous speed is adopted for segmented curve fitting because of its simple structure. Besides, this paper proposes a new learning mechanism that enables them to learn from each other between elite individuals and non-elite individuals for MA. Finally, In the Matlab2016a GUI platform, by using several different curve fitting optimization algorithms to carry out simulation experiments, simulation results show that the IMA can find more accurate traction characteristics fitting curve at the same condition.