Interval algorithm for global numerical optimization

The interval optimization algorithm shows great advantages in bound constrained global optimization. An interval algorithm is presented in this article based on a new selection criterion. The selection criterion is proposed based on numerical experiments and the parameter pf* designed by Casado, Garcia and Csendes in 2000. The proposed criterion at each iteration selects some intervals of which the number is not greater than a constant so that the possible memory problem during the implementation of the algorithm is avoided and the running time of the algorithm is decreased, when the dimension of the problem is increasing. Based on the selection criterion, the proposed algorithm is implemented for a wide set of tested functions which includes easy and hard problems. Numerical experiments show that the proposed algorithm is efficient.     

Parallel global optimization with the particle swarm algorithm

Present day engineering optimization problems often impose large computational demands, resulting in long solution times even on a modern high-end processor. To obtain enhanced computational throughput and global search capability, we detail the coarse-grained parallelization of an increasingly popular global search method, the particle swarm optimization (PSO) algorithm. Parallel PSO performance was evaluated using two categories of optimization problems possessing multiple local minima-large-scale analytical test problems with computationally cheap function evaluations and medium-scale biomechanical system identification problems with computationally expensive function evaluations. For load-balanced analytical test problems formulated using 128 design variables, speedup was close to ideal and parallel efficiency above 95% for up to 32 nodes on a Beowulf cluster. In contrast, for load-imbalanced biomechanical system identification problems with 12 design variables, speedup plateaued and parallel efficiency decreased almost linearly with increasing number of nodes. The primary factor affecting parallel performance was the synchronization requirement of the parallel algorithm, which dictated that each iteration must wait for completion of the slowest fitness evaluation. When the analytical problems were solved using a fixed number of swarm iterations, a single population of 128 particles produced a better convergence rate than did multiple independent runs performed using sub-populations (8 runs with 16 particles, 4 runs with 32 particles, or 2 runs with 64 particles). These results suggest that (1) parallel PSO exhibits excellent parallel performance under load-balanced conditions, (2) an asynchronous implementation would be valuable for real-life problems subject to load imbalance, and (3) larger population sizes should be considered when multiple processors are available.     

混合整数双层线性规划的全局优化算法

本文讨论了上层决策变量为整数变量、下层决策变量为连续变量的混合整数双层线性规划问题,利用其可行解均落在约束域边界上的性质,提出了一种求解混合整数双层线性规划全局最优解的算法,并举例说明了算法的执行过程。     

Residual Correction Procedure with Bernstein Polynomials for Solving Important Systems of Ordinary Differential Equations

One of the most attractive subjects in applied sciences is to obtain exact or approximate solutions for different types of linear and nonlinear systems. Systems of ordinary differential equations like systems of second-order boundary value problems (BVPs), Brusselator system and stiff system are significant in science and engineering. One of the most challenge problems in applied science is to construct methods to approximate solutions of such systems of differential equations which pose great challenges for numerical simulations. Bernstein polynomials method with residual correction procedure is used to treat those challenges. The aim of this paper is to present a technique to approximate solutions of such differential equations in optimal way. In it, we introduce a method called residual correction procedure, to correct some previous approximate solutions for such systems. We study the error analysis of our given method. We first introduce a new result to approximate the absolute solution by using the residual correction procedure. Second, we introduce a new result to get appropriate bound for the absolute error. The collocation method is used and the collocation points can be found by applying Chebyshev roots. Both techniques are explained briefly with illustrative examples to demonstrate the applicability, efficiency and accuracy of the techniques. By using a small number of Bernstein polynomials and correction procedure we achieve some significant results. We present some examples to show the efficiency of our method by comparing the solution of such problems obtained by our method with the solution obtained by Runge-Kutta method, continuous genetic algorithm, rational homotopy perturbation method and adomian decomposition method.     

An evolutionary algorithm for global optimization based on self-organizing maps

In this article, a new population-based algorithm for real-parameter global optimization is presented, which is denoted as self-organizing centroids optimization (SOC-opt). The proposed method uses a stochastic approach which is based on the sequential learning paradigm for self-organizing maps (SOMs). A modified version of the SOM is proposed where each cell contains an individual, which performs a search for a locally optimal solution and it is affected by the search for a global optimum. The movement of the individuals in the search space is based on a discrete-time dynamic filter, and various choices of this filter are possible to obtain different dynamics of the centroids. In this way, a general framework is defined where well-known algorithms represent a particular case. The proposed algorithm is validated through a set of problems, which include non-separable problems, and compared with state-of-the-art algorithms for global optimization.     

A surrogate-assisted particle swarm optimization algorithm based on efficient global optimization for expensive black-box problems

Evolutionary algorithms cannot effectively handle computationally expensive problems because of the unaffordable computational cost brought by a large number of fitness evaluations. Therefore, surrogates are widely used to assist evolutionary algorithms in solving these problems. This article proposes an improved surrogate-assisted particle swarm optimization (ISAPSO) algorithm, in which a hybrid particle swarm optimization (PSO) is combined with global and local surrogates. The global surrogate is not only used to predict fitness values for reducing computational burden but also regarded as a global searcher to speed up the global search process of PSO by using an efficient global optimization algorithm, while the local one is constructed for a local search in the neighbourhood of the current optimal solution by finding the predicted optimal solution of the local surrogate. Empirical studies on 10 widely used benchmark problems and a real-world structural design optimization problem of a driving axle show that the ISAPSO algorithm is effective and highly competitive.     

An adaptive metamodel-based global optimization algorithm for black-box type problems

In this article, an adaptive metamodel-based global optimization (AMGO) algorithm is presented to solve unconstrained black-box problems. In the AMGO algorithm, a type of hybrid model composed of kriging and augmented radial basis function (RBF) is used as the surrogate model. The weight factors of hybrid model are adaptively selected in the optimization process. To balance the local and global search, a sub-optimization problem is constructed during each iteration to determine the new iterative points. As numerical experiments, six standard two-dimensional test functions are selected to show the distributions of iterative points. The AMGO algorithm is also tested on seven well-known benchmark optimization problems and contrasted with three representative metamodel-based optimization methods: efficient global optimization (EGO), GutmannRBF and hybrid and adaptive metamodel (HAM). The test results demonstrate the efficiency and robustness of the proposed method. The AMGO algorithm is finally applied to the structural design of the import and export chamber of a cycloid gear pump, achieving satisfactory results.     

Global optimization for the three-dimensional open-dimension rectangular packing problem

This article addresses the three-dimensional open-dimension rectangular packing problem (3D-ODRPP), which aims to pack a given set of unequal-size rectangular boxes within an enveloping rectangular space such that the volume of the occupied space is minimized. Even though the studied 3D-ODRPP is NP hard, the development of sophisticated global optimization methods has been stimulated. The mathematical programming formulation for the 3D-ODRPP has evolved into an effective and efficient mixed-integer linear programming (MILP) model. This study proposes an advanced exact scheme yielding a guaranteed global optimal solution given that all the instance data are non-negative rational numbers. The developed MILP retains not only fewer variables but also fewer constraints than the state-of-the-art models. The superior effectiveness and efficiency of the developed scheme are demonstrated with numerical experiments, where two sets of benchmark instances from references, real-world instances and instances with rational data are included.     

A global optimization algorithm for simulation-based problems via the extended DIRECT scheme

This article presents a global optimization algorithm via the extension of the DIviding RECTangles (DIRECT) scheme to handle problems with computationally expensive simulations efficiently. The new optimization strategy improves the regular partition scheme of DIRECT to a flexible irregular partition scheme in order to utilize information from irregular points. The metamodelling technique is introduced to work with the flexible partition scheme to speed up the convergence, which is meaningful for simulation-based problems. Comparative results on eight representative benchmark problems and an engineering application with some existing global optimization algorithms indicate that the proposed global optimization strategy is promising for simulation-based problems in terms of efficiency and accuracy.     

A framework for parallelized efficient global optimization with application to vehicle crashworthiness optimization

This article presents a framework for simulation-based design optimization of computationally expensive problems, where economizing the generation of sample designs is highly desirable. One popular approach for such problems is efficient global optimization (EGO), where an initial set of design samples is used to construct a kriging model, which is then used to generate new ‘infill’ sample designs at regions of the search space where there is high expectancy of improvement. This article attempts to address one of the limitations of EGO, where generation of infill samples can become a difficult optimization problem in its own right, as well as allow the generation of multiple samples at a time in order to take advantage of parallel computing in the evaluation of the new samples. The proposed approach is tested on analytical functions, and then applied to the vehicle crashworthiness design of a full Geo Metro model undergoing frontal crash conditions.     

IFQP: A hybrid optimization method for filter management in fluid power systems under uncertainty

An interval-fuzzy quadratic programming (IFQP) method is developed for the assessment of filter allocation and replacement strategies in fluid power systems (FPS) under uncertainty. It can directly handle uncertainties expressed as interval values and/or fuzzy sets that exist in the left-hand and right-hand sides of constraints, as well as in the objective function. Multiple control variables are used to tackle independent uncertainties in the model's right-hand sides and thus optimize the overall satisfaction of the system performance. The IFQP method is applied to a case of planning filter allocation and replacement strategies under uncertainty for an FPS with a single circuit. A piecewise linearization approach is firstly employed to convert the nonlinear FPS problem into a linear one. The generated decision alternatives can help decision makers to identify desired policies for contamination control under various total costs, satisfaction degrees, and system-failure risks under different contaminant-ingression/generation rates.     

Multi-agent genetic algorithm with controllable mutation probability utilizing back propagation neural network for global optimization of trajectory design

A Controllable Mutation Probability (CMP) strategy is proposed and applied to a Multi-Agent Genetic Algorithm (MAGA) to deal with the global optimization of trajectory design in deep space, which is called MGA-CMP. MAGA-CMP is an algorithm setting all the individuals (or agents) on a grid and having two controlling functions to adjust the performance probability of a mutation operator. It pays more attention to global search in the earlier part of the process, and devotes more effort to local search at later stages. Four experiments are implemented to illustrate the efficiency of MAGA-CMP, and results show that MGA-CMP performs better in most examined cases than other well-known search algorithms.     

LQG optimal controller synthesis with robust optimization in multivariable systems

Abstract

This paper presents an algorithm to synthesize a controller for treating the problem of robustness optimization in an LQG (linear‐quadratic‐Gaussian) control system. The controller not only maximizes the excess stability margin in perturbed system but also minimizes the cost functional J in LQG problems by specifying two frequency dependent weighting matrices Q(s) and R(s) in the cost functional.

Our approach is based on Wiener‐Hopf's technique (frequency domain approach), and two weighting matrices in cost functional are shaped by inverse LQG method. The feature of this paper is that the plant of the system has no stable, proper, square, and minimum phase constraints.     

Hybridizing a genetic algorithm with an artificial immune system for global optimization

This paper proposes an algorithm based on a model of the immune system to handle constraints of all types (linear, nonlinear, equality, and inequality) in a genetic algorithm used for global optimization. The approach is implemented both in serial and parallel forms, and it is validated using several test functions taken from the specialized literature. Our results indicate that the proposed approach is highly competitive with respect to penalty-based techniques and with respect to other constraint-handling techniques which are considerably more complex to implement.     

An efficient hybrid algorithm for the optimization of problems with several local minima

The problem of design optimization is of high industrial interest, and has been extensively studied for years, with excellent results. However, there is the well‐known issue of a reasonable balance between the computational effort usually required by stochastic methods, and the fact that deterministic optimizers, even though much more efficient, are not guaranteed to localize a good minimum, as they can remain trapped in the first found local one. To overcome these problems, the authors developed a hybrid strategy, which gave good results in terms of speed and reliability of the obtained optima, especially when the objective function is obtained through a finite element analysis, due, for example, to the absence of an analytical solution of the problem, and the direct use of a stochastic method would be unfeasible for practical purposes, because of the intolerable processing time required. Copyright © 2001 John Wiley & Sons, Ltd.     

Finding all global optima of engineering design problems with discrete signomial terms

ABSTRACT

Many engineering design problems are frequently modelled as nonlinear programming problems with discrete signomial terms. In general, signomial programs are very difficult to solve for obtaining the globally optimal solution. This study reformulates the engineering design problem with discrete signomial terms as a mixed-integer linear program and finds all alternative global optima. Compared with existing exact methods, the proposed method uses fewer variables and constraints in the reformulated model and therefore efficiently solves the engineering problem to derive all global optima. Illustrative examples from the literature are solved to demonstrate the usefulness and efficiency of the proposed method.     

Drilling path optimization by the particle swarm optimization algorithm with global convergence characteristics

Drilling path optimization is one of the key problems in holes-machining. This paper presents a new approach to solve the drilling path optimization problem belonging to discrete space, based on the particle swarm optimization (PSO) algorithm. Since the standard PSO algorithm is not guaranteed to be global convergent or local convergent, based on the mathematical model, the algorithm is improved by adopting the method to generate the stop evolution particle once again to obtain the ability of convergence on the global optimization solution. Also, the operators are proposed by establishing the Order Exchange Unit (OEU) and the Order Exchange List (OEL) to satisfy the need of integer coding in drilling path optimization. The experimentations indicate that the improved algorithm has the characteristics of easy realization, fast convergence speed, and better global convergence capability. Hence the new PSO can play a role in solving the problem of drilling path optimization.     

Semi‐deterministic and genetic algorithms for global optimization of microfluidic protein‐folding devices

In this paper, we reformulate global optimization problems in terms of boundary‐value problems (BVP). This allows us to introduce a new class of optimization algorithms. Indeed, current optimization methods, including non‐deterministic ones, can be seen as discretizations of initial value problems for differential equations, or systems of differential equations. Furthermore, in order to reduce computational time approximate state and sensitivity evaluations are introduced during optimization. Lastly, we demonstrated the efficacy of two algorithms, included in the former class, on two academic test cases and on the design of a fast microfluidic protein‐folding device. The aim of the latter design is to reduce mixing times of proteins to microsecond time scales. Results are compared with those obtained with a classical genetic algorithm. Copyright © 2005 John Wiley & Sons, Ltd.     

A simple and robust Monte Carlo hybrid local search algorithm for the facility location problem

In this article, a new Monte Carlo hybrid local search algorithm (Hyb-LS) is proposed for solving the uncapacitated facility location problem. Hyb-LS is based on repeated sampling using two local search strategies based on best improvement and randomized neighbourhood search. A major advantage of Hyb-LS for its practical use is that the number of restarts is its only parameter to tune. The algorithm is also simple to reimplement, scalable and robust to changes in coefficients within a problem instance. The stopping criterion for local search is learned automatically. Experimental results are presented for four representative and contrasting cost and distance models. The results obtained by Hyb-LS are compared to the optimal or near-optimal solutions found by a mixed integer linear programming (MILP) solver with a generous time limit. For three out of the four models, Hyb-LS obtains better solutions than the upper bound found by the MILP solver for at least one instance.