Drift Operator for States of Matter Search Algorithm Yuxiang Zhou, Yongquan Zhou ( ✉ ) , Qifang Luo, Shilei Qiao, and Rui Wang College of Information Science and Engineering, Guangxi University for Nationalities, Nanning Guangxi 530006, China yongquanzhou@126.com Abstract. States of matter search (SMS) algorithm is based on the simulation of the states of matter phenomenon. In SMS, individuals emulate molecules which interact to each other by using evolutionary operations which are based on the physical prin‐ ciple of the thermal-energy motion mechanism. Although the SMS algorithms have been used to solve many optimization problems, there still slow convergence and easy to fall into local optimum in some applications. In this paper, a novel drift operator-based states of matter search algorithm (DSMS) is proposed. The main idea involves using drift operator to keep the concept of location and abandon the concept of velocity for accelerate the convergence speed while simplifying algorithm, mean‐ while a new variable differential evolution (DE) strategy is introduced to diversify the individuals in the search space for escape from the local optima. The proposed method is applied to several benchmark problems and is compared to four modern meta-heuristic algorithms. The experimental results show that the proposed algo‐ rithm outperforms other peer algorithms. Keywords: States of matter search algorithm · Drift operator · Thermal-energy motion mechanism · Differential evolution strategy · Meta-heuristic algorithms 1 Introduction Recently, global optimization problem in real-world is more and more complex and has attracted a lot of researchers to search for efficient problem-solving methods. Evolu‐ tionary algorithm is the better solution to solve the global optimization problems and widely applied in various areas of science, engineering, economics and others, where mathematical modeling is used. In general, the goal is to find a goal optimum for an objective function which is defined over a given search space. During the last few decades, several evolutionary algorithms have been suggested that mimics some natural phenomena. Such phenomena include animal-behavior phenomena such as the Particle Swarm Optimization (PSO) algorithm [3], the Bat (BA) algorithm proposed by Yang [4]. Some other methods which are based on physical processes, for example, the Elec‐ tromagnetism-like Algorithm [5], the Gravitational Search Algorithm (GSA) [6] and the States of Matter Search (SMS) algorithm [1, 2]. The SMS algorithm is based on the simulation of the states of matter phenom‐ enon. In SMS, individuals emulate molecules which interact to each other by using © Springer International Publishing Switzerland 2015 D.-S. Huang and K. Han (Eds.): ICIC 2015, Part III, LNAI 9227, pp. 65–71, 2015. DOI: 10.1007/978-3-319-22053-6_7