Multi-label classification aims at assigning more than one label to each instance. Many real-world multi-label classification tasks are high dimensional, leading to reduced performance of traditional classifiers. Feature selection is a common approach to tackle this issue by choosing prominent features. Multi-label feature selection is an NP-hard approach, and so far, some swarm intelligence-based strategies and have been proposed to find a near optimal solution within a reasonable time. In this paper, a hybrid intelligence algorithm based on the binary algorithm of particle swarm OPTIMIZATION and a novel local search STRATEGY has been proposed to select a set of prominent features. To this aim, features are divided into two categories based on the extension rate and the relationship between the output and the local search STRATEGY to increase the convergence speed. The first group features have more similarity to class and less similarity to other features, and the second is redundant and less relevant features. Accordingly, a local operator is added to the particle swarm OPTIMIZATION algorithm to reduce redundant features and keep relevant ones among each solution. The aim of this operator leads to enhance the convergence speed of the proposed algorithm compared to other algorithms presented in this field. Evaluation of the proposed solution and the proposed statistical test shows that the proposed approach improves different classification criteria of multi-label classification and outperforms other methods in most cases. Also in cases where achieving higher accuracy is more important than time, it is more appropriate to use this method.

In this paper, recommended spiral passive micromixer was designed and simulated. spiral design has the potential to create and strengthen the centrifugal force and the secondary flow. A series of simulations were carried out to evaluate the effects of channel width, channel depth, the gap between loops, and flowrate on the micromixer performance. These features impact the contact area of the two fluids and ultimately lead to an increment in the quality of the mixture. In this study, for the flow rate of 25 μl/min and molecular diffusion coefficient of 1×10-10 m2/s, mixing efficiency of more than 90% is achieved after 30 (approximately one-third of the total channel length). Finally, the optimized design fabricated using proposed 3D printing method.

Application of the network equivalent concept for external system representation for power system transient analysis is well known. However, the challenge to utilize an equivalent network, approximated by a rational function, is to guarantee the passivity of the corresponding model. In this regard, special techniques are required to enforce the passivity of the equivalent model through a post processing approach that minimizes its impact on the original model characteristics. In this paper, the passivity is enforced by expressing the problem in terms of a convex OPTIMIZATION problem that guarantees the global optimal solution. The convex OPTIMIZATION problem is efficiently solved by recently developed numerical interior–point methods. This passivity enforcement is also global which indicates that the passivity enforcement in one region does not lead to passivity violation in other regions.

Evolution Strategies (ES) are a class of Evolutionary Algorithms based on Gaussian mutation and deterministic selection. Gaussian mutation captures pair-wise dependencies between the variables through a covariance matrix. Covariance Matrix Adaptation (CMA) is a method to update this covariance matrix. In this paper, the CMA-ES, which has found many applications in solving continuous OPTIMIZATION problems, is employed for size OPTIMIZATION of steel space trusses. Design examples revealcompetitive performance of the algorithm compared to the other advanced metaheuristics.

System reliability OPTIMIZATION is one of practical issue in design and engineering. Adding parallel redundant component is a common approach in order to improve the reliability of a specific system. This approach that is known as redundancy allocation problem (RAP), includes discrete component choices with known characteristics such as reliability, cost, weight in order to maximize system reliability. In RAPs, redundancy STRATEGY plays a significant role in increasing reliability of a system or subsystem. Generally, there are three types of redundancy strategies, namely, active, standby and mixed. The latter is a new STRATEGY which is a combination of the two former and have had some noticeable results. This paper presents a new redundancy STRATEGY for system reliability OPTIMIZATION which is called Sync-Mixed. The proposed STRATEGY is a general form of a recently introduced STRATEGY called "Mixed STRATEGY" that effort to decrease number of switching system usages. In order to evaluate the efficiency of the Sync-Mixed STRATEGY with regarding complexity of formulas for new STRATEGY, a single specific subsystem with five components will be used and reliability of the subsystem will be computed by considering all three previous strategies and the new one. In previous studies, some well-known benchmark problems, including series system, series-parallel system, parallel-series system, and complex system, have been taken into account. The performances of all strategies are compared by considering different values of switching and components reliabilities. In RAP, it is assumed that the switch reliability and components reliability are predetermined parameters. However, in this study different combinations of components reliability and switching system reliability are taken into consideration. The main objective is to evaluate the performance of the new STRATEGY compared to traditional strategies in different situations. Obviously, the subsystem reliability has a direct relationship with the components reliability, switch reliability and type of redundancy STRATEGY. This study is an attempt to shed light on the sensitivity of subsystem reliability as a result of switch and component reliability changes in all redundancy strategies.

This paper aims to optimize the transmission and control STRATEGY of a parallel hybrid electric vehicle in order to minimize Fuel Consumption (FC) and emissions, simultaneously. Vehicle transmission is a Power Split Continuously Variable Transmission (PS-CVT), while the employed Torque Coupler (TC) is a two-speed TC. Using this type of TC increases designer ability to create a more efficient transmission. In this vehicle, the electric assist control STRATEGY is used as the control STRATEGY. In this STRATEGY, the engine operates at optimal operation points, obtained using the Global Criterion method (GC). A multi-objective OPTIMIZATION is implemented using GC to minimize vehicle FC and emissions without sacrificing dynamic performance. Finally, results of the conventional method of hybrid vehicle OPTIMIZATION and the results of using the Dynamic Programming method are compared.

An efficient parallel STRATEGY is presented for OPTIMIZATION of the aerodynamic shapes using Genetic Algorithm (GA). The method is a hybrid Parallel Genetic Algorithm (PGA) that combines a multi-population PGA and master-slave PGA using Message Passing Interface. GA parameters are firstly tuned according to the fact that subpopulations evolve independently. The effect of the number of sub-population on the computational time is investigated. Finally, a new STRATEGY is presented based on the load balancing that aims to decrease the idle time of the processors. The algorithm is used for OPTIMIZATION of a transonic airfoil. An unstructured grid finite volume flow solver is utilized for objective function evaluations. For the considered class of problems, the suggested Hierarchical Parallel Genetic Algorithm (HPGA) results in more than 30% reduction in OPTIMIZATION time in comparison to regular master-slave PGA. A semi-liner speed-up is also obtained which indicates that the model is suited for modern cluster work stations.

In this article, the linear quadratic regulator method (LQR) for voltage control of a linear time-varying model of a robot is used to design an online adaptive optimal stable controller to trace the robot arm path. Normally, offline solving of Riccati differential equations in backward with final conditions for linear time-varying system or converting the Riccati differential equation to an algebraic one in linear time-invariant system is inevitable in LQR. However, in this paper, the differential Riccati equations are considered as the adaptation laws along with a voltage control STRATEGY to be solved online in forward method with initial conditions. Choosing a proper Lyapunov function guarantees the asymptotic stability of the tracking. Furthermore, parametric model uncertainties such as mass parameter variation and external disturbances which affect the dynamics of the model are also taken into account. Simulation results show the energy used by dc motors of the voltage optimal control STRATEGY is less than that of the torque control STRATEGY and as good as the classical PID one. The superior performance of the voltage optimal control over torque control STRATEGY is also shown in presence of disturbance.

In most practical cases, structural design variables are linked to a discrete list of sections for optimal design. Cardinality of such a discrete search space is governed by the number of alternatives for each member group. The present work offers an adaptive STRATEGY to detect more efficient alternatives and set aside redundant ones during OPTIMIZATION. In this regard, the difference between the lower and the upper bounds on such variables is gradually reduced by a procedure that adapts history of the selected alternatives in previous iterations. The propsed STRATEGY is implemented on a hybrid paritcle swarm optimizer and imperialist competitive algorithm. The former is a basic swarm intelligent method while the later utilizes subpopulations in its search. Spatial and large-scale structures in various shapes are treated showing successive performance improvement. Variation of a diversity index and resulting band size are traced and discussed to declare behavior merits of the proposed adaptive band STRATEGY.