Papers by Keyword: Constrained Optimization

Abstract: An improved grey wolf optimization (IGWO) algorithm is proposed for solving constrained mechanical design problems in this paper. In proposed IGWO algorithm, a novel nonlinearly update equation of convergence factor based on sines function is presented to balance the exploration ability and exploitation ability. The feasibility-based rules based on tournament selection was introduced to handle constrains. Simulation results and comparisons with other state-of-the-art algorithms using three constrained mechanical design problems are provided.

Abstract: The implementation of the Differential Evolutionary Algorithm for the solution of a multi-pass turning optimization problem is presented in this paper. The optimization of a multi-pass turning process is a highly demanding problem due to the number of constraints imposed. A specific variation of the Differential Evolutionary Algorithm appropriate for the treatment of constrained problems is used. It is based on the separation of candidate solutions into those that satisfy all constraints and those that do not and the simultaneous execution of two optimization algorithms. Numerical results from the minimization of the production cost of a popular multi-pass turning problem with six degrees of freedom, namely cutting speed, feed rate and depth of cut of rough machining and finishing, validate the methodology. The selected problem is characterized by a number of equally stepped roughing passes and a final finishing pass of the cutting tool in order to obtain the desired metal part removal from the initial workpiece.

Abstract: Compression spring is one of the most common mechanical componet being used in most mechanisms. Many criteria and constraints should be considered in designing and specifying the spring dimensions. Therefore, it has been one of the standard case studies considered to test a new optimisation algorithm. This paper introduced an optimization method named Gravitational search Algorithm (GSA) to solve the problem of weight minimization of spring. From previous studies, weight minimization of a spring has been investigated by many researcher using various optimization algorithm technique. The result of this study were compared to one of the previous studies using Particle Swarm Optimization (PSO) algorithm. Also, parametric studies were conducted to select the best values of GSA parameters, beta and epsilon. From the results obtained, it was observed that the optimum dimensions and weight obtained by GSA are better than the values obtained by PSO. The best values of beta and epsilon was found to be 0.6 and 0.01 respectively.

Abstract: A particle swarm algorithm (PSO) based on boundary buffering-natural evolution was proposed for solving constrained optimization problems. By buffering the particles that cross boundaries, the diversity of populations was intensified; to accelerate the convergence speed and avoid local optimum of PSO, natural evolution was introduced. In other words, particle hybridization and mutation strategies were applied; and by combining the modified feasible rules, the constrained optimization problems were solved. The simulation results proved that the method was effective in solving this kind of problems.

Abstract: Most constraint-handling methods in constrained evolutionary optimization usually take advantage of only the valuable information of feasible solutions, while they don’t exploit adequately the information from infeasible ones. In this paper, a concept of “feasible component” is introduced to recognize the characteristics of diverse information extracted from infeasible solutions. Then a component-based ranking strategy is proposed for evolutionary optimization with sparse constraints by integrating feasible components and the idea of stochastic ranking. Experimental results on several problems with sparse constraints show that the component-based ranking strategy performs better than the stochastic ranking.

Abstract: PID (proportional+integral+derivative) controller is well known as a simple and easy-to-implement controller. However, the design procedure is not straightforward for multi-input multi-output (MIMO) systems. It is even more complicated when robustness criterion must be handled. In this paper, a stable robust PID controller for anti-swing control of automatic gantry crane is proposed. The proposed method employs an automatic tuning using DE (differential evolution) to search for a set of PID controller gains that satisfy Kharitonovs polynomials robust stability criterion. This robust stability criterion is used to deal with parametric uncertainty occurs in gantry crane model. The simulation results show that a satisfactory robust PID control performance can be achieved. The PID controller is able to quickly move the cart of the crane while suppressing the swing of the payload for various conditions, i.e. payload mass and cable length variations.

Abstract: Considering that the particle swarm optimization (PSO) algorithm has a tendency to get stuck at the local solutions, an improved PSO algorithm is proposed in this paper to solve constrained optimization problems. In this algorithm, the initial particle population is generated using good point set method such that the initial particles are uniformly distributed in the optimization domain. Then, during the optimization process, the particle population is divided into two sub-populations including feasible sub-population and infeasible sub-population. Finally, different crossover operations and mutation operations are applied for updating the particles in each of the two sub-populations. The effectiveness of the improved PSO algorithm is demonstrated on three benchmark functions.

Abstract: Network lifetime is a critical metric in the design of energy-constrained wireless sensor networks. In this paper, we consider the joint cross layer optimization of the physical layer, medium access control layer and routing layer to maximize network lifetime of a multi-sources and single-sink wireless sensor network with energy constraints. We focus on synchronous small-scale sensor network with interference-free link scheduling and practical MPSK link transmission scheme. As the network lifetime maximization problem is a constrained non-convex optimization problem that is difficult to be solved, and the particle swarm optimization algorithm is a good intelligent algorithm, we employ it to solve the problem mentioned above effectively in this paper. The penalty function technique is brought in to work out the constrained optimization problem by converting it to an unconstrained optimization problem. Simulation results show the effectiveness of the proposed algorithm in energy saving and network lifetime maximization, and the particle swarm optimization can solve the network lifetime maximization problem fast and efficiently.

Abstract: Due to the disadvantages of genetic algorithm such as the weaker ability for local search, premature convergence, random walk and problems related, and so on , the design and improvement of the algorithm is an important research direction of genetic algorithm. And evaluating the performance of algorithm systematically and scientifically is the key to test algorithm whether good or bad .The common method used to evaluate algorithm is test function, however, the existing literature on the optimization algorithm has different methods to evaluate the performance of algorithm, and there is no uniform test criteria. As for those questions above, This paper studies test functions of genetic algorithm, and analyses characteristics of the main test functions, which can be used as the basis of selection algorithm test functions.

Abstract: There are many constrained optimization problems in engineering. Bio-inspired optimization algorithms have been widely used to solve various engineering problems. This paper presents a novel optimization algorithm called Lifecycle-based Swarm Optimization, inspired by biology life cycle. LSO algorithm imitates biologic life cycle process through six optimization operators: chemotactic, assimilation, transposition, crossover, selection and mutation. In addition, the spatial distribution of initialization population meets clumped distribution. Experiments were conducted on a Vehicle Routing Problem with Time Windows for demonstration the effectiveness and stability. The results demonstrate remarkable performance of the LSO algorithm on chosen case when compared to two successful optimization techniques.