Optimal design of an autonomous underwater helicopter's shape based on combinatorial optimization strategy

Based on the system reserve energy index as objective optimization function of the structure vibration system, sets up the multi-objective optimal disposition mathematical model which refer to the system control gain and the sensors and actuators placements, quantities, lengths of the active vibration control system, adopts the multi-island genetic algorithm (MIGA) as optimization strategy. Experiments with intelligent cantilever beam as an example, the vibration control effect is very distinct. Hence, approves the method correctness and feasibility.

The problem of the optimal power dispatch of dispersed generators in direct-current networks under the presence of nonlinear loads (constant power terminals) is addressed through a combinatorial optimization strategy by using a master-slave solution methodology. The optimal power generation in the dispersed is solved in the master optimization stage through the application of the vortex-search algorithm. Each combination of the power outputs at the dispersed generation sources is provided to a power flow methodology known as the hyperbolic power flow approach for direct current networks. The main advantage of the proposed optimization method corresponds to the possibility of solving a complex nonlinear programming problem via sequential quadratic programming, which can be easily implemented at any programming language with low computational effort and high-quality results. The computational tests of the master-slave optimization proposal are evaluated in a 21-bus system, and the numerical results are compared with the implementation of the exact nonlinear programming model in the General Algebraic Modeling System (i.e., GAMS). All the computational results are conducted through the MATLAB programming environment licensed by Universidad Tecnologica de Pereira for academic usage.

Superpixel segmentation aims at grouping discretizing pixels into high-level correlative units and reducing the complexity of subsequent tasks, e.g., saliency detection and object tracking. Existing superpixel segmentation algorithms mainly focus on maintaining the geometrical information, while neglecting the irregular structure of superpixels. In this paper, a superpixel segmentation method is proposed to generate approximately structural superpixels with sharp boundary adherence and comprehensive semantic information. The superpixel segmentation is formulated as a square-wise asymmetric partition problem, where the semantic perceptual superpixels are recorded in a square level to preserve abundant semantic information and save storage simultaneously. Moreover, in order to achieve regular-shape superpixel units to better adhere to image boundaries and contours, a combinatorial optimization strategy is devised to achieve an optimal combination of squares and isolated pixels. Experimental comparisons with some state-of-the-art superpixel segmentation methods on the public benchmarks demonstrate the effectiveness of the proposed method quantitatively and qualitatively. In addition, we have applied the method to brain tissue segmentation to illustrate superior performance.

Optimization of complex mechanics simulations with object-oriented software designs

Aiming at the conceptual design stage of semi-submersible support platform, the optimal primary scale scheme is designed. Latin hypercube design is adopted to sample the sample space. At the same time, SESAM software is used for parametric modeling based on sample point parameters to obtain the model of all schemes. According to the existing data of semi-submersible platforms, the main scale constraints of platforms are determined, and the mathematical model with the minimum heave, trim and roll response as the objective function is established. The multi-island genetic algorithm was used to optimize the main scale of the supporting platform model, and the scheme with the highest stability and suitable displacement was selected. The results show that the model and algorithm are feasible, and the smallest scale scheme of each response can be obtained under the constraint conditions. The multi-island genetic algorithm is proposed to optimize the primary scale of the platform, which has a strong reference value for the determination of the primary scale scheme in the conceptual design stage of the platform.

Training a neural network is a difficult optimization problem because of the nonconvex objective function. Therefore, as an alternative to local search algorithms, many global search algorithms have been used to train neural networks. However, local search algorithms are more efficient with computational resources, and therefore numerous random restarts with a local algorithm may be more effective than a global algorithm at obtaining a low value of the objective function. This study examines, through Monte-Carlo simulations, the relative efficiency of a local search algorithm to 8 stochastic global algorithms: 2 simulated annealing algorithms, 1 simple random stochastic algorithm, 1 genetic algorithm and 4 evolutionary strategy algorithms. The results show that even ignoring the computational requirements of the global algorithms, there is little evidence to support the use of the global algorithms examined for training neural networks.

Non-convex optimization problems are prevalent in scientific and engineering applications, often characterized by multiple local minima that challenge standard optimization techniques. Global optimization methods, such as Genetic Algorithms (GA) (1992) [4] and Particle Swarm Optimization (PSO) (1995) [6], provide robust global exploration but tend to converge slowly. Conversely, local search methods, such as Newton’s method (2004) [1], (2006) [7], and the Conjugate Gradient (CG) (1964) [3] method, offer rapid convergence but are susceptible to local minima. In this paper, we present four hybrid optimization algorithms that integrate the exploratory strength of global algorithms with the refinement ability of local methods. The proposed hybrids, Genetic Algorithms (GA) + Newton, GA + CG, PSO + Newton, and PSO + CG, are evaluated against their standalone counterparts on standard benchmark functions (2013) [5], including Rosenbrock, Rastrigin, Ackley, and Himmelblau. This work builds upon our previous research on the hybridization of two global search algorithms, the Nelder–Mead (simplex) algorithm and the Bat algorithm, which resulted in the Hybrid Simplex Bat Algorithm (HSBA) [2], recently Published. Results demonstrate that the hybrid approaches consistently outperform traditional methods in terms of accuracy, convergence rate, and computational efficiency.

Elastic wheel is an important component of the low floor tram using in urban rail transit, so it is of significance to optimize its structure and predict its service lifetime. Firstly, the fatigue characteristics of rubber material are studied based on the crack energy theory, and a finite element model of an elastic wheel for fatigue lifetime prediction is constructed. The numerical simulation results show that the fatigue lifetime of such a elastic wheel does not match with the expected one. Secondly, based on the orthogonal numerical test and the Sobol model, an improved sensitivity analysis method is proposed. Taking the minimum strain as an optimization objective, and selecting the structural parameters determined by the sensitivity analysis method as the design variables, the second-order response surface model is established, and the multi-island genetic algorithm is used to carry out the single objective and multi-factor optimization design of the rubber bodies of elastic wheel under the condition of stiffness constraint. The simulation results of the elastic wheel show that the fatigue lifetime of the optimized structure is 30% higher than that of the original one. Finally, the optimized elastic wheel is tested, and the results show that the stiffness coefficient, and the fatigue lifetime as well, are all satisfy with the design requirements.

Optimizing Design of Powertrain Transmission Ratio of Heavy Duty Truck

In order to explore the optimization and design method of the rough surface on a vehicle body, a pitted rough surface, a convex rough surface and a grooved rough surface are respectively arranged at the rear of the MIRA model. Computational fluid dynamics and wind tunnel tests are adopted to analyse the relationships between the aerodynamic drag and the factors of the rough surface. The distance L and the height H of the surface roughness are considered as the design variables, and the aerodynamic drag coefficient CD is the objective evaluation function. Sample points are obtained by the Latin hypercube sampling method, and the kriging approximate model is set up according to the sample points and the response values. Then the multi-island genetic algorithm is used as the optimization algorithm to find the optimum solution. The aerodynamic effects of the rough surface model, with optimization and without optimization, are compared and analysed. Experimental and simulation results show that the aerodynamic drag coefficient further decreases after parameter optimization and that the optimization method is valid and general. The results obtained may provide a helpful reference and guidance for future rough surface applications on a vehicle body.

Ship stowage scheduling of container terminal needs to consider the safety of the ship and transport economics. The strategic decisions of this paper are that which bay each stacking unit is assigned and the stacking location of each stacking unit in bay. The goals are minimizing container reshuffle rate on board, minimizing the center of gravity of the ship, holding appropriate trim and ensuring that the heavy containers stacking in the middle of the ship. The classification strategy and stacking strategy in bay are proposed. Based on this, a scheduling model of ship stowage is established. A genetic algorithm which bases on group coding method and the stacking strategy in bay which bases on the greedy algorithm are proposed as the resolution of the model. It can be concluded that combinatorial optimization algorithm and strategy are efficient and effective as the approach to ship stowage scheduling of container terminal from the analysis of the experiment's result.

Distributed Constraint Satisfaction Problems provide a natural mechanism for multiagent coordination and agreement. To date, algorithms for Distributed Constraint Satisfaction Problems have tended to mirror existing non-distributed global-search or local-search algorithms. Unfortunately, existing distributed global-search algorithms derive from classical backtracking search methods and require a total ordering over agents for completeness. Distributed variants of local-search algorithms (such as distributed breakout) inherit the incompleteness properties of their predecessors, or depend on the creation of new communication links between agents. In [5, 4] a new algorithm was presented designed explicitly for distributed environments so that a global ordering is not required, while avoiding the problems of existing local-search algorithms. This paper presents a significant improvement on that algorithm in performance and provability.

As the key technique to build domain-specific search engines, focused crawling has drawn a lot of attention from researchers in the past decade. However, as Web structure analysis techniques advance, several problems in traditional focused crawler design were revealed and they could result in domain-specific collections with low quality. In this work, we studied the problems of focused crawling that are caused by using local search algorithms. We also proposed to use a global search algorithm, the Genetic Algorithm, in focused crawling to address the problems. We conducted evaluation experiments to examine the effectiveness of our approach. The results showed that our approach could build domain-specific collections with higher quality than traditional focused crawling techniques. Furthermore, we used the concept of Web communities to evaluate how comprehensively the focused crawlers could traverse the Web search space, which could be a good complement to the traditional focused crawler evaluation methods.

Search-Based test case prioritization for simulation-Based testing of cyber-Physical system product lines

Optimization Solution Task is a typical and important task in many applications. Many optimization problems have been proved to be NP-hard problems, which cannot be solved by some predefined mathematic formulae. In this case, computer aided method is very helpful. While some local search algorithms are easily to fall into a local optimum solution. On contrast, the population based methods, such as Genetic Algorithms, Artificial Immune System, Autonomy Oriented Computing, are global search algorithms. However, they are not good at the local search. In this paper, an improved method is proposed by combining the local and global search ability, so as to improve the performance in terms of the convergence speed and the convergence reliability. We construct a generic form to deal with the common objective function space or the objective function with the partial derivative. In addition, we present an n-hold method in population based evolution method. The experiments indicate that our approach can effectively improve the convergence reliability, which is much concerned in some applications with the expensive executing expense.