Redundant Instruments Placement Using ACO

Abstract

In this paper, Ant Colony Optimization (ACO) Algorithm is introduced to redundant instruments placement for optimum process variable estimation accuracy. It is proved that additional redundancy measurement will enhance estimation accuracy if the measurements relate the process variables in a different way, whereas the quantity of accuracy improvement is determined by the measurements structure. To find the optimal redundant instruments placement is substantially combinatorial optimization problem, Ant Colony System (ACS) can perform this perfectly. In this paper, ACO based redundant instruments placement algorithm is proposed. Simulation shows the proposed outperform the GA algorithm, which is a prevailing algorithm to combinatorial problem.

Sensor network optimization of gearbox based on dependence matrix and improved discrete shuffled frog leaping algorithm

“…Some researchers have focused on various solving algorithms, such as the principal component analysis (Li and Upadhyaya 2011;Li et al 2012), ANNs (Chow et al 2011;Martin et al 2005), and Bayesian networks (Jin et al 2012;Pourali and Mosleh 2013). Recently, some swarm intelligent optimization algorithms, e.g., particle swarm optimization (Pan and Wei 2010;Kulkarni and Venayagamoorthy 2010), genetic algorithm (GA) (Liu et al 2008;Casillas et al 2013), ant colony optimization (Fu 2009), artificial bee colony algorithm (Mini et al 2014), and artificial fish-swarm algorithm (Tao et al 2013) have been applied to solve the problem and some promising results have been successfully achieved. As a novel swarm intelligent algorithm, the shuffled frog leaping algorithm (SFLA) is a genetic-based, heuristically cooperative search algorithm (Eusuff and Lansey 2003), and it has been widely employed in various optimization fields, such as the water distribution network design (Eusuff and Lansey 2003), parameter identification (Ahandani 2014), unit commitment problem (Barati and Farsangi 2014), distribution network reconfiguration problem (Jazebi et al 2014), and so on.…”