A whale optimization algorithm based on quadratic interpolation for high-dimensional global optimization problems

Whale Optimization Algorithm (WOA), as a new population-based optimization algorithm, performs well in solving optimization problems. However, when tackling high-dimensional global optimization problems, WOA tends to fall into local optimal solutions and has slow convergence rate and low solution accuracy. To address these problems, a whale optimization algorithm based on quadratic interpolation (QIWOA) is presented. On the one hand, a modified exploration process by introducing a new parameter is proposed to efficiently search the regions and deal with the premature convergence problem. On the other hand, quadratic interpolation around the best search agent helps QIWOA to improve the exploitation ability and the solution accuracy. Moreover, the algorithm tries to make a balance between exploitation and exploration. QIWOA is compared with several state-of-the-art algorithms on 30 high-dimensional benchmark functions with dimensions ranging from 100 to 2000. The experimental results show that QIWOA has faster convergence rate and higher solution accuracy than both WOA and other population-based algorithms. For functions with a flat or sharp bottom, QIWOA is difficult to find the global optimum, but it still performs best compared with other algorithms.     

A harmony search algorithm for high-dimensional multimodal optimization problems

Harmony search (HS) and its variants have been found successful applications, however with poor solution accuracy and convergence performance for high-dimensional (≥200) multimodal optimization problems. The reason is mainly huge search space and multiple local minima. To tackle the problem, we present a new HS algorithm called DIHS, which is based on Dynamic-Dimensionality-Reduction-Adjustment (DDRA) and dynamic fret width (fw) strategy. The former is for avoiding generating invalid solutions and the latter is to balance global exploration and local exploitation. Theoretical analysis on the DDRA strategy for success rate of update operation is given and influence of related parameters on solution accuracy is investigated. Our experiments include comparison on solution accuracy and CPU time with seven typical HS algorithms and four widely used evolutionary algorithms (SaDE, CoDE, CMAES and CLPSO) and statistical comparison by the Wilcoxon Signed-Rank Test with the seven HS algorithms and four evolutionary algorithms. The problems in experiments include twelve multimodal and four complex uni-modal functions with high-dimensionality.Experimental results indicate that the proposed approach can provide significant improvement on solution accuracy with less CPU time in solving high-dimensional multimodal optimization problems, and the more dimensionality that the optimization problem is, the more benefits it provides.     

Large scale evolutionary optimization using cooperative coevolution

Evolutionary algorithms (EAs) have been applied with success to many numerical and combinatorial optimization problems in recent years. However, they often lose their effectiveness and advantages when applied to large and complex problems, e.g., those with high dimensions. Although cooperative coevolution has been proposed as a promising framework for tackling high-dimensional optimization problems, only limited studies were reported by decomposing a high-dimensional problem into single variables (dimensions). Such methods of decomposition often failed to solve nonseparable problems, for which tight interactions exist among different decision variables. In this paper, we propose a new cooperative coevolution framework that is capable of optimizing large scale nonseparable problems. A random grouping scheme and adaptive weighting are introduced in problem decomposition and coevolution. Instead of conventional evolutionary algorithms, a novel differential evolution algorithm is adopted. Theoretical analysis is presented in this paper to show why and how the new framework can be effective for optimizing large nonseparable problems. Extensive computational studies are also carried out to evaluate the performance of newly proposed algorithm on a large number of benchmark functions with up to 1000 dimensions. The results show clearly that our framework and algorithm are effective as well as efficient for large scale evolutionary optimisation problems. We are unaware of any other evolutionary algorithms that can optimize 1000-dimension nonseparable problems as effectively and efficiently as we have done.     

Handling boundary constraints for particle swarm optimization in high-dimensional search space

Despite the fact that the popular particle swarm optimizer (PSO) is currently being extensively applied to many real-world problems that often have high-dimensional and complex fitness landscapes, the effects of boundary constraints on PSO have not attracted adequate attention in the literature. However, in accordance with the theoretical analysis in [11], our numerical experiments show that particles tend to fly outside of the boundary in the first few iterations at a very high probability in high-dimensional search spaces. Consequently, the method used to handle boundary violations is critical to the performance of PSO. In this study, we reveal that the widely used random and absorbing bound-handling schemes may paralyze PSO for high-dimensional and complex problems. We also explore in detail the distinct mechanisms responsible for the failures of these two bound-handling schemes. Finally, we suggest that using high-dimensional and complex benchmark functions, such as the composition functions in [19], is a prerequisite to identifying the potential problems in applying PSO to many real-world applications because certain properties of standard benchmark functions make problems inexplicit.     

A hybrid shuffled complex evolution approach with pattern search for unconstrained optimization

The difficulties associated with using classical mathematical programming methods on complex optimization problems have contributed to the development of alternative and efficient numerical approaches. Recently, to overcome the limitations of classical optimization methods, researchers have proposed a wide variety of meta-heuristics for searching near-optimum solutions to problems. Among the existing meta-heuristic algorithms, a relatively new optimization paradigm is the Shuffled Complex Evolution at the University of Arizona (SCE-UA) which is a global optimization strategy that combines concepts of the competition evolution theory, downhill simplex procedure of Nelder-Mead, controlled random search and complex shuffling. In an attempt to reduce processing time and improve the quality of solutions, particularly to avoid being trapped in local optima, in this paper is proposed a hybrid SCE-UA approach. The proposed hybrid algorithm is the combination of SCE-UA (without Nelder-Mead downhill simplex procedure) and a pattern search approach, called SCE-PS, for unconstrained optimization. Pattern search methods are derivative-free, meaning that they do not use explicit or approximate derivatives. Moreover, pattern search algorithms are direct search methods well suitable for the global optimization of highly nonlinear, multiparameter, and multimodal objective functions. The proposed SCE-PS method is tested with six benchmark optimization problems. Simulation results show that the proposed SCE-PS improves the searching performance when compared with the classical SCE-UA and a genetic algorithm with floating-point representation for all the tested problems. As evidenced by the performance indices based on the mean performance of objective function in 30 runs and mean of computational time, the SCE-PS algorithm has demonstrated to be effective and efficient at locating best-practice optimal solutions for unconstrained optimization.     

Coupled eagle strategy and differential evolution for unconstrained and constrained global optimization

The performance of an optimization tool is largely determined by the efficiency of the search algorithm used in the process. The fundamental nature of a search algorithm will essentially determine its search efficiency and thus the types of problems it can solve. Modern metaheuristic algorithms are generally more suitable for global optimization. This paper carries out extensive global optimization of unconstrained and constrained problems using the recently developed eagle strategy by Yang and Deb in combination with the efficient differential evolution. After a detailed formulation and explanation of its implementation, the proposed algorithm is first verified using twenty unconstrained optimization problems or benchmarks. For the validation against constrained problems, this algorithm is subsequently applied to thirteen classical benchmarks and three benchmark engineering problems reported in the engineering literature. The performance of the proposed algorithm is further compared with various, state-of-the-art algorithms in the area. The optimal solutions obtained in this study are better than the best solutions obtained by the existing methods. The unique search features used in the proposed algorithm are analyzed, and their implications for future research are also discussed in detail.     

The spherical search algorithm for bound-constrained global optimization problems

In this paper, a new optimization algorithm called Spherical Search (SS) is proposed to solve the bound-constrained non-linear global optimization problems. The main operations of SS are the calculation of spherical boundary and generation of new trial solution on the surface of the spherical boundary. These operations are mathematically modeled with some more basic level operators: Initialization of solution, greedy selection and parameter adaptation, and are employed on the 30 black-box bound constrained global optimization problems. This study also analyzes the applicability of the proposed algorithm on a set of real-life optimization problems. Meanwhile, to show the robustness and proficiency of SS, the obtained results of the proposed algorithm are compared with the results of other well-known optimization algorithms and their advanced variants: Particle Swarm Optimization (PSO), Differential Evolution (DE), and Covariance Matrix Adapted Evolution Strategy (CMA-ES). The comparative analysis reveals that the performance of SS is quite competitive with respect to the other peer algorithms.     

A new improved whale optimization algorithm with joint search mechanisms for high-dimensional global optimization problems

Similar to other swarm-based algorithms, the recently developed whale optimization algorithm (WOA) has the problems of low accuracy and slow convergence. It is also easy to fall into local optimum. Moreover, WOA and its variants cannot perform well enough in solving high-dimensional optimization problems. This paper puts forward a new improved WOA with joint search mechanisms called JSWOA for solving the above disadvantages. First, the improved algorithm uses tent chaotic map to maintain the diversity of the initial population for global search. Second, a new adaptive inertia weight is given to improve the convergence accuracy and speed, together with jump out from local optimum. Finally, to enhance the quality and diversity of the whale population, as well as increase the probability of obtaining global optimal solution, opposition-based learning mechanism is used to update the individuals of the whale population continuously during each iteration process. The performance of the proposed JSWOA is tested by twenty-three benchmark functions of various types and dimensions. Then, the results are compared with the basic WOA, several variants of WOA and other swarm-based intelligent algorithms. The experimental results show that the proposed JSWOA algorithm with multi-mechanisms is superior to WOA and the other state-of-the-art algorithms in the competition, exhibiting remarkable advantages in the solution accuracy and convergence speed. It is also suitable for dealing with high-dimensional global optimization problems.

     

An evolutionary method for complex-process optimization

In this paper we present a new evolutionary method for complex-process optimization. It is partially based on the principles of the scatter search methodology, but it makes use of innovative strategies to be more effective in the context of complex-process optimization using a small number of tuning parameters. In particular, we introduce a new combination method based on path relinking, which considers a broader area around the population members than previous combination methods. We also use a population-update method which improves the balance between intensification and diversification. New strategies to intensify the search and to escape from suboptimal solutions are also presented. The application of the proposed evolutionary algorithm to different sets of both state-of-the-art continuous global optimization and complex-process optimization problems reveals that it is robust and efficient for the type of problems intended to solve, outperforming the results obtained with other methods found in the literature.     

A multi-start central force optimization for global optimization

Central force optimization (CFO) is an efficient and powerful population-based intelligence algorithm for optimization problems. CFO is deterministic in nature, unlike the most widely used metaheuristics. CFO, however, is not completely free from the problems of premature convergence. One way to overcome local optimality is to utilize the multi-start strategy. By combining the respective advantages of CFO and the multi-start strategy, a multi-start central force optimization (MCFO) algorithm is proposed in this paper. The performance of the MCFO approach is evaluated on a comprehensive set of benchmark functions. The experimental results demonstrate that MCFO not only saves the computational cost, but also performs better than some state-of-the-art CFO algorithms. MCFO is also compared with representative evolutionary algorithms. The results show that MCFO is highly competitive, achieving promising performance.     

一种求解非线性约束优化全局最优的新方法

本文提出了一种求解非线性约束优化的全局最优的新方法—它是基于利用非线性互补函数和不断增加新的约束来重复解库恩-塔克条件的非线性方程组的新方法。因为库恩-塔克条件是非线性约束优化的必要条件,得到的解未必是非线性约束优化的全局最优解,为此,本文首次给出了通过利用该优化问题的先验知识,不断地增加约束来限制全局最优解范围的方法,一些仿真例子表明提出的方法和理论有效的,并且可行的。     

Parallel differential evolution with self-adapting control parameters and generalized opposition-based learning for solving high-dimensional optimization problems

Solving high-dimensional global optimization problems is a time-consuming task because of the high complexity of the problems. To reduce the computational time for high-dimensional problems, this paper presents a parallel differential evolution (DE) based on Graphics Processing Units (GPUs). The proposed approach is called GOjDE, which employs self-adapting control parameters and generalized opposition-based learning (GOBL). The adapting parameters strategy is helpful to avoid manually adjusting the control parameters, and GOBL is beneficial for improving the quality of candidate solutions. Simulation experiments are conducted on a set of recently proposed high-dimensional benchmark problems with dimensions of 100, 200, 500 and 1,000. Simulation results demonstrate that GjODE is better than, or at least comparable to, six other algorithms, and employing GPU can effectively reduce computational time. The obtained maximum speedup is up to 75.     

适用于高维优化问题的改进进化策略

针对高维连续函数优化问题,研究了CES(classical evo lution strateg ies)的变异方式、繁殖方式,提出了全基因变异与单基因变异的概念,通过理论分析和仿真计算论证了单基因变异比全基因变异具有更好的局部搜索能力和少的计算开销;针对CES策略参数(变异幅度)随机性过强,不能很好地跟踪进化过程的问题,提出了随着进化过程递减的策略参数.最后,建立了单基因Gauss变异与均匀变异相结合、使用精英繁殖、递减型策略参数、小种群规模的(μ λ k)-ES,给出了一组100维典型测试函数的仿真计算结果.     

A novel differential evolution algorithm using local abstract convex underestimate strategy for global optimization

Two main challenges in differential evolution (DE) are reducing the number of function evaluations required to obtain optimal solutions and balancing the exploration and exploitation. In this paper, a local abstract convex underestimate strategy based on abstract convexity theory is proposed to address these two problems. First, the supporting hyperplanes are constructed for the neighboring individuals of the trial individual. Consequently, the underestimate value of the trial individual can be obtained by the supporting hyperplanes of its neighboring individuals. Through the guidance of the underestimate value in the select operation, the number of function evaluations can be reduced obviously. Second, some invalid regions of the domain where the global optimum cannot be found are safely excluded according to the underestimate information to improve reliability and exploration efficiency. Finally, the descent directions of supporting hyperplanes are employed for local enhancement to enhance exploitation capability. Accordingly, a novel DE algorithm using local abstract convex underestimate strategy (DELU) is proposed. Numerical experiments on 23 bound-constrained benchmark functions show that the proposed DELU is significantly better than, or at least comparable to several state-of-the art DE variants, non-DE algorithms, and surrogate-assisted evolutionary algorithms.     

一种求解双目标job shop问题的混合进化算法

提出一种求解双目标job shop排序问题的混合进化算法．该算法采用改进的精英复制策略,降低了计算复杂性;通过引入递进进化模式,避免了算法的早熟;通过递进过程中的非劣解邻域搜索,增强了算法局部搜索性能．采用该算法和代表性算法NSGA-Ⅱ,MOGLS对82个标准双目标job shop算例进行优化对比,所得结果验证了该算法求解双目标job shop排序问题的有效性．     

A self-adaptive linear evolutionary algorithm for solving constrained optimization problems

In many real-world applications of evolutionary algorithms, the fitness of an individual requires a quantitative measure. This paper proposes a self-adaptive linear evolutionary algorithm (ALEA) in which we introduce a novel strategy for evaluating individual’s relative strengths and weaknesses. Based on this strategy, searching space of constrained optimization problems with high dimensions for design variables is compressed into two-dimensional performance space in which it is possible to quickly identify ‘good’ individuals of the performance for a multiobjective optimization application, regardless of original space complexity. This is considered as our main contribution. In addition, the proposed new evolutionary algorithm combines two basic operators with modification in reproduction phase, namely, crossover and mutation. Simulation results over a comprehensive set of benchmark functions show that the proposed strategy is feasible and effective, and provides good performance in terms of uniformity and diversity of solutions.     

An improved differential evolution algorithm with triangular mutation for global numerical optimization

This paper presents an Improved Differential Evolution (IDE) algorithm for solving global numerical optimization problems over continuous space. The proposed algorithm introduces a new triangular mutation rule based on the convex combination vector of the triplet defined by the three randomly chosen vectors and the difference vector between the best and the worst individuals among the three randomly selected vectors. The mutation rule is combined with the basic mutation strategy through a non-linear decreasing probability rule. Furthermore, a restart mechanism is also proposed to avoid premature convergence. IDE is tested on a well-known set of unconstrained problems and shows its superiority to state-of-the-art differential evolution variants.     

An improved evolutionary algorithm for handling many-objective optimization problems

It has been shown that the multi-objective evolutionary algorithms (MOEAs) act poorly in solving many-objective optimization problems which include more than three objectives. The research emphasis, in recent years, has been put into improving the MOEAs to enable them to solve many-objective optimization problems efficiently. In this paper, we propose a new composite fitness evaluation function, in a novel way, to select quality solutions from the objective space of a many-objective optimization problem. Using this composite function, we develop a new algorithm on a well-known NSGA-II and call it FR-NSGA-II, a fast reference point based NSGA-II. The algorithm is evaluated for producing quality solutions measured in terms of proximity, diversity and computational time. The working logic of the algorithm is explained using a bi-objective linear programming problem. Then we test the algorithm using experiments with benchmark problems from DTLZ family. We also compare FR-NSGA-II with four competitive algorithms from the extant literature to show that FR-NSGA-II will produce quality solutions even if the number of objectives is as high as 20.     

An efficient dynamical evolutionary algorithm for global optimization

In this paper, we introduce a new dynamical evolutionary algorithm (DEA) that aims to find the global optimum and give the theoretical explanation from statistical mechanics. The algorithm has been evaluated numerically using a wide set of test functions which are nonlinear, multimodal and multidimensional. The numerical results show that it is possible to obtain global optimum or more accurate solutions than other methods for the investigated hard problems.     

Multi-modal function optimization with the ZEDS (zoomed evolutionary dual strategy) algorithmzoomed evolutionary algorithm

This article introduces a new evolutionary algorithm for multi-modal function optimization called ZEDS (zoomed evolutionary dual strategy). ZEDS employs a two-step, zoomed (global to local), evolutionary approach. In the first (global) step, an improved ‘GT algorithm’ is employed to perform a global recombinatory search that divides the search space into niches according to the positions of its approximate solutions. In the second (local) step, a ‘niche evolutionary strategy’ performs a local search in the niches obtained from the first step, which is repeated until acceptable solutions are found. The ZEDS algorithm was applied to some challenging problems with good results, as shown in this article.