A new multi-strategy ensemble artificial bee colony algorithm for water demand prediction

Abstract

Artificial bee colony (ABC) is an efficient global optimizer, which has bee successfully used to solve various optimization problems. Recently, multi-strategy ensemble technique was embedded to ABC to make a good trade-off between exploration and exploitation. In this paper, a new multi-strategy ensemble ABC (NMEABC) is proposed. In our approach, each food source is assigned a probability to control the frequency of dimension perturbation. Experimental results show that NMEABC is superior to the original multi-strategy ensemble ABC (MEABC). Finally, NMEABC is applied to predict the water demand in Nanchang city. Simulation results demonstrate that NMEABC can achieve a good prediction accuracy.