Search

Complex system optimization is an emerging research topic in the field of evolutionary computation, whose goal is to handle complex systems with multiple coupled subsystems, each including multiple objectives and multiple constraints in real-world applications. This paper proposes a multi-system genetic algorithm (MSGA), stemming from implicit parallelism in population-based search algorithms, to solve multiple coupled subsystems simultaneously in a complex system. The proposed MSGA is composed of within-subsystem evolution and cross-subsystem migration operators. The objective of the former is to optimize each subsystem by appropriate search strategies, and the objective of the latter is to exchange information between multiple subsystems by migration, which is based on the similarity probability of objectives and constraints, and the intersection probability of solutions in different subsystems. During migration across subsystems, three statistical approaches of measuring similarity and three metrics of solution intersection in information theory are used to calculate these probabilities. Performance is tested on a set of multi-subsystem benchmark functions, and the simulation results show that cross-subsystem migration plays the key role for the performance of MSGA. Furthermore, the proposed MSGA is compared with other competitive algorithms, and results show that it is a promising multi-system optimization algorithm. In summary, the contribution of this paper is the introduction of multi-system optimization to the EA community. [ABSTRACT FROM AUTHOR]

Abstract Multi-population based nature-inspired optimization algorithms have attracted wide research interests in the last decade, and become one of the frequently used methods to handle real-world optimization problems. Considering the importance and value of multi-population methods and its applications, we believe it is the right time to provide a comprehensive survey of the published work, and also to discuss several aspects for the future research. The purpose of this paper is to summarize the published techniques related to the multi-population methods in nature-inspired optimization algorithms. Beginning with the concept of multi-population optimization, we review basic and important issues in the multi-population methods and discuss their applications in science and engineering. Finally, this paper presents several interesting open problems with future research directions for multi-population optimization methods. [ABSTRACT FROM AUTHOR]

Swarm intelligence (SI) optimization algorithms are fast and robust global optimization methods, and have attracted significant attention due to their ability to solve complex optimization problems. The underlying idea behind all SI algorithms is similar, and various SI algorithms differ only in their details. In this paper we discuss the algorithmic equivalence of particle swarm optimization (PSO) and various other newer SI algorithms, including the shuffled frog leaping algorithm (SFLA), the group search optimizer (GSO), the firefly algorithm (FA), artificial bee colony algorithm (ABC) and the gravitational search algorithm (GSA). We find that the original versions of SFLA, GSO, FA, ABC, and GSA, are all algorithmically identical to PSO under certain conditions. We discuss their diverse biological motivations and algorithmic details as typically implemented, and show how their differences enhance the diversity of SI research and application. Then we numerically compare SFLA, GSO, FA, ABC, and GSA, with basic and advanced versions on some continuous benchmark functions and combinatorial knapsack problems. Empirical results show that an advanced version of ABC performs best on the continuous benchmark functions, and advanced versions of SFLA and GSA perform best on the combinatorial knapsack problems. We conclude that although these SI algorithms are conceptually equivalent, their implementation details result in notably different performance levels. [ABSTRACT FROM AUTHOR]

This paper introduces a Markov model for evolutionary algorithms (EAs) that is based on interactions among individuals in the population. This interactive Markov model has the potential to provide tractable models for optimization problems of realistic size. We propose two simple discrete optimization search strategies with population-proportion-based selection and a modified mutation operator. The probability of selection is linearly proportional to the number of individuals at each point of the search space. The mutation operator randomly modifies an entire individual rather than a single decision variable. We exactly model these optimization search strategies with interactive Markov models. We present simulation results to confirm the interactive Markov model theory. We show that genetic algorithms and biogeography-based optimization perform better with the addition of population-proportion-based selection on a set of real-world benchmarks. We note that many other EAs, both new and old, might be able to be improved with this addition, or modeled with this method. [ABSTRACT FROM PUBLISHER]

Identifying promising compounds from a vast collection of potential compounds is an important and yet challenging problem in chemical engineering. An efficient solution to this problem will help to reduce the expenditure at the early states of chemical process. In an attempt to solve this problem, the industry is looking for predictive tools that would be useful in testing optimal properties of a candidate compound earlier. This paper explores the application of biogeography-based optimization (BBO) to achieve such predictive work. BBO is a new evolutionary algorithm that is based on the science of biogeography. BBO is a population-based search method that achieves information sharing by species migration. The performance of BBO is compared with genetic algorithm (GA) and particle swarm optimization (PSO) on a set of test functions and the cases of identifying promising compounds. Simulation results show that BBO is a competitive method in determining an optimal solution to the optimization of promising compounds. [ABSTRACT FROM AUTHOR]

Evolutionary algorithms are robust optimization methods that have been used in many engineering applications. However, real-world fitness evaluations can be computationally expensive, so it may be necessary to estimate the fitness with an approximate model. This article reviews design and analysis of computer experiments (DACE) as an approximation method that combines a global polynomial with a local Gaussian model to estimate continuous fitness functions. The article incorporates DACE in various evolutionary algorithms, to test unconstrained and constrained benchmarks, both with and without fitness function evaluation noise. The article also introduces a new evolution control strategy called update-based control that estimates the fitness of certain individuals of each generation based on the exact fitness values of other individuals during that same generation. The results show that update-based evolution control outperforms other strategies on noise-free, noisy, constrained and unconstrained benchmarks. The results also show that update-based evolution control can compensate for fitness evaluation noise. [ABSTRACT FROM AUTHOR]

Biogeography-based optimization (BBO) is a new evolutionary optimization algorithm that is based on the science of biogeography. In this paper, BBO is applied to the optimization of problems in which the fitness function is corrupted by random noise. Noise interferes with the BBO immigration rate and emigration rate, and adversely affects optimization performance. We analyse the effect of noise on BBO using a Markov model. We also incorporate re-sampling in BBO, which samples the fitness of each candidate solution several times and calculates the average to alleviate the effects of noise. BBO performance on noisy benchmark functions is compared with particle swarm optimization (PSO), differential evolution (DE), self-adaptive DE (SaDE) and PSO with constriction (CPSO). The results show that SaDE performs best and BBO performs second best. In addition, BBO with re-sampling is compared with Kalman filter-based BBO (KBBO). The results show that BBO with re-sampling achieves almost the same performance as KBBO but consumes less computational time. [ABSTRACT FROM AUTHOR]

This paper proposes a biogeography-based optimization (BBO) method augmented with a Kalman filter, which is called KFBBO, for PID parameter tuning in a wireless networked learning control system (WNLCS). Because of unreliable transmission of data and commands in wireless networks, the control system is noisy and prone to errors, which results in poor performance by the conventional PID method for wireless networked control in real-world applications. BBO as a new evolutionary optimization is proposed to solve this problem by dynamically optimizing the PID control parameters. Because the wireless network environment is noisy, we also use a Kalman filter to counteract the negative effects of noise and to improve the optimization ability of BBO. Simulation experiments are conducted to evaluate our proposed KFBBO, and the results indicate that the control performance obtained by the improved PID method is better than the conventional PID method. Furthermore, this proposed method is applied to a steam turbine power generation system based on a WNLCS, and the results show its feasibility and effectiveness. [ABSTRACT FROM PUBLISHER]

Abstract: Hybrid evolutionary algorithms (EAs) are effective optimization methods that combine multiple EAs. We propose several hybrid EAs by combining some recently-developed EAs with a biogeography-based hybridization strategy. We test our hybrid EAs on the continuous optimization benchmarks from the 2013 Congress on Evolutionary Computation (CEC) and on some real-world traveling salesman problems. The new hybrid EAs include two approaches to hybridization: (1) iteration-level hybridization, in which various EAs and BBO are executed in sequence; and (2) algorithm-level hybridization, which runs various EAs independently and then exchanges information between them using ideas from biogeography. Our empirical study shows that the new hybrid EAs significantly outperforms their constituent algorithms with the selected tuning parameters and generation limits, and algorithm-level hybridization is generally better than iteration-level hybridization. Results also show that the best new hybrid algorithm in this paper is competitive with the algorithms from the 2013 CEC competition. In addition, we show that the new hybrid EAs are generally robust to tuning parameters. In summary, the contribution of this paper is the introduction of biogeography-based hybridization strategies to the EA community. [Copyright &y& Elsevier]

Abstract: Biogeography-based optimization (BBO) is a new evolutionary algorithm inspired by biogeography, which involves the study of the migration of biological species between habitats. Previous work has shown that various migration models of BBO result in significant changes in performance. Sinusoidal migration models have been shown to provide the best performance so far. Motivated by biogeography theory and previous results, in this paper a generalized sinusoidal migration model curve is proposed. A previously derived BBO Markov model is used to analyze the effect of migration models on optimization performance, and new theoretical results which are confirmed with simulation results are obtained. The results show that the generalized sinusoidal migration model is significantly better than other models for simple but representative problems, including a unimodal one-max problem, a multimodal problem, and a deceptive problem. In addition, performance comparison is further investigated through 23 benchmark functions with a wide range of dimensions and diverse complexities, to verify the superiority of the generalized sinusoidal migration model. [Copyright &y& Elsevier]

Abstract: Biogeography-based optimization (BBO) is a new evolutionary optimization method that is based on the science of biogeography. We propose two extensions to BBO. First, we propose a blended migration operator. Benchmark results show that blended BBO outperforms standard BBO. Second, we employ blended BBO to solve constrained optimization problems. Constraints are handled by modifying the BBO immigration and emigration procedures. The approach that we use does not require any additional tuning parameters beyond those that are required for unconstrained problems. The constrained blended BBO algorithm is compared with solutions based on a stud genetic algorithm (SGA) and standard particle swarm optimization 2007 (SPSO 07). The numerical results demonstrate that constrained blended BBO outperforms SGA and performs similarly to SPSO 07 for constrained single-objective optimization problems. [Copyright &y& Elsevier]