Multi-Objective Equilibrium Optimizer for Feature Selection in High-Dimensional English Speech Emotion Recognition

Speech emotion recognition(SER)uses acoustic analysis to find features for emotion recognition and examines variations in voice that are caused by emotions.The number of features acquired with acoustic analysis is extremely high,so we introduce a hybrid filter-wrapper feature selection algorithm based on an improved equilibrium optimizer for constructing an emotion recognition system.The proposed algorithm implements multi-objective emotion recognition with the minimum number of selected features and maximum accuracy.First,we use the information gain and Fisher Score to sort the features extracted from signals.Then,we employ a multi-objective ranking method to evaluate these features and assign different importance to them.Features with high rankings have a large probability of being selected.Finally,we propose a repair strategy to address the problem of duplicate solutions in multi-objective feature selection,which can improve the diversity of solutions and avoid falling into local traps.Using random forest and K-nearest neighbor classifiers,four English speech emotion datasets are employed to test the proposed algorithm(MBEO)as well as other multi-objective emotion identification techniques.The results illustrate that it performs well in inverted generational distance,hypervolume,Pareto solutions,and execution time,and MBEO is appropriate for high-dimensional English SER.

Multi-objective Optimization of a Parallel Ankle Rehabilitation Robot Using Modified Differential Evolution Algorithm

Dimensional synthesis is one of the most difficult issues in the field of parallel robots with actuation redundancy. To deal with the optimal design of a redundantly actuated parallel robot used for ankle rehabilitation, a methodology of dimensional synthesis based on multi-objective optimization is presented. First, the dimensional synthesis of the redundant parallel robot is formulated as a nonlinear constrained multi-objective optimization problem. Then four objective functions, separately reflecting occupied space, input/output transmission and torque performances, and multi-criteria constraints, such as dimension, interference and kinematics, are defined. In consideration of the passive exercise of plantar/dorsiflexion requiring large output moment, a torque index is proposed. To cope with the actuation redundancy of the parallel robot, a new output transmission index is defined as well. The multi-objective optimization problem is solved by using a modified Differential Evolution（DE） algorithm, which is characterized by new selection and mutation strategies. Meanwhile, a special penalty method is presented to tackle the multi-criteria constraints. Finally, numerical experiments for different optimization algorithms are implemented. The computation results show that the proposed indices of output transmission and torque, and constraint handling are effective for the redundant parallel robot; the modified DE algorithm is superior to the other tested algorithms, in terms of the ability of global search and the number of non-dominated solutions. The proposed methodology of multi-objective optimization can be also applied to the dimensional synthesis of other redundantly actuated parallel robots only with rotational movements.

Multi-objective Collaborative Optimization for Scheduling Aircraft Landing on Closely Spaced Parallel Runways Based on Genetic Algorithms

A scheduling model of closely spaced parallel runways for arrival aircraft was proposed,with multi-objections of the minimum flight delay cost,the maximum airport capacity,the minimum workload of air traffic controller and the maximum fairness of airlines′scheduling.The time interval between two runways and changes of aircraft landing order were taken as the constraints.Genetic algorithm was used to solve the model,and the model constrained unit delay cost of the aircraft with multiple flight tasks to reduce its delay influence range.Each objective function value or the fitness of particle unsatisfied the constrain condition would be punished.Finally,one domestic airport hub was introduced to verify the algorithm and the model.The results showed that the genetic algorithm presented strong convergence and timeliness for solving constraint multi-objective aircraft landing problem on closely spaced parallel runways,and the optimization results were better than that of actual scheduling.

A Hybrid Parallel Multi-Objective Genetic Algorithm for 0/1 Knapsack Problem

In this paper a hybrid parallel multi-objective genetic algorithm is proposed for solving 0/1 knapsack problem. Multi-objective problems with non-convex and discrete Pareto front can take enormous computation time to converge to the true Pareto front. Hence, the classical multi-objective genetic algorithms (MOGAs) (i.e., non- Parallel MOGAs) may fail to solve such intractable problem in a reasonable amount of time. The proposed hybrid model will combine the best attribute of island and Jakobovic master slave models. We conduct an extensive experimental study in a multi-core system by varying the different size of processors and the result is compared with basic parallel model i.e., master-slave model which is used to parallelize NSGA-II. The experimental results confirm that the hybrid model is showing a clear edge over master-slave model in terms of processing time and approximation to the true Pareto front.

Workspace optimization of parallel robot by using multi-objective genetic algorithm

For the narrow workspace problem of the universal-prismatic-universal(UPU)parallel robotwith fixed orientation,a kind of multi-objective genetic algorithm is studied to optimize the robot’sworkspace.The concept of the effective workspace and its solution method are given.The effectiveworkspace height(EWH)and global condition number index(GCI)of Jacobi matrix are selected asthe optimized objective functions.Setting the robot in two different orientations,the geometric pa-rameters are optimized by the multi-objective genetic algorithm named non-dominated sorting geneticalgorithm II(NSGA-II),and a set of structural parameters is obtained.The optimization results areverified by four indicators with the robot’s moving platform at different orientations.The resultsshow that,after optimization,the fixed-orientation workspace volume,the effective workspace heightand the effective workspace volume increase by 32.4%,17.8%and 72.9%on average,respec-tively.GCI decreases by 6.8%on average.

A Parallel Search System for Dynamic Multi-Objective Traveling Salesman Problem

This paper introduces a parallel search system for dynamic multi-objective traveling salesman problem. We design a multi-objective TSP in a stochastic dynamic environment. This dynamic setting of the problem is very useful for routing in ad-hoc networks. The proposed search system first uses parallel processors to identify the extreme solutions of the search space for each ofk objectives individually at the same time. These solutions are merged into the so-called hit-frequency matrix E. The solutions in E are then searched by parallel processors and evaluated for dominance relationship. The search system is implemented in two different ways master-worker architecture and pipeline architecture.

Sequencing Mixed-model Production Systems by Modified Multi-objective Genetic Algorithms

As two independent problems,scheduling for parts fabrication line and sequencing for mixed-model assembly line have been addressed respectively by many researchers.However,these two problems should be considered simultaneously to improve the efficiency of the whole fabrication/assembly systems.By far,little research effort is devoted to sequencing problems for mixed-model fabrication/assembly systems.This paper is concerned about the sequencing problems in pull production systems which are composed of one mixed-model assembly line with limited intermediate buffers and two flexible parts fabrication flow lines with identical parallel machines and limited intermediate buffers.Two objectives are considered simultaneously：minimizing the total variation in parts consumption in the assembly line and minimizing the total makespan cost in the fabrication/assembly system.The integrated optimization framework,mathematical models and the method to construct the complete schedules for the fabrication lines according to the production sequences for the first stage in fabrication lines are presented.Since the above problems are non-deterministic polynomial-hard（NP-hard）,a modified multi-objective genetic algorithm is proposed for solving the models,in which a method to generate the production sequences for the fabrication lines from the production sequences for the assembly line and a method to generate the initial population are put forward,new selection,crossover and mutation operators are designed,and Pareto ranking method and sharing function method are employed to evaluate the individuals＇ fitness.The feasibility and efficiency of the multi-objective genetic algorithm is shown by computational comparison with a multi-objective simulated annealing algorithm.The sequencing problems for mixed-model production systems can be solved effectively by the proposed modified multi-objective genetic algorithm.

Fast parallel Grad–Shafranov solver for real-time equilibrium reconstruction in EAST tokamak using graphic processing unit

To achieve real-time control of tokamak plasmas, the equilibrium reconstruction has to be completed sufficiently quickly. For the case of an EAST tokamak experiment, real-time equilibrium reconstruction is generally required to provide results within 1ms. A graphic processing unit（GPU） parallel Grad–Shafranov（G-S） solver is developed in P-EFIT code,which is built with the CUDA？ architecture to take advantage of massively parallel GPU cores and significantly accelerate the computation. Optimization and implementation of numerical algorithms for a block tri-diagonal linear system are presented. The solver can complete a calculation within 16 μs with 65×65 grid size and 27 μs with 129×129 grid size, and this solver supports that P-EFIT can fulfill the time feasibility for real-time plasma control with both grid sizes.

A New Definition and Calculation Model for Evolutionary Multi-Objective Optimization

We present a new definition (Evolving Solutions) for Multi-objective Optimization Problem (MOP) to answer the basic question (what's multi-objective optimal solution?) and advance an asynchronous evolutionary model (MINT Model) to solve MOPs. The new theory is based on our understanding of the natural evolution and the analysis of the difference between natural evolution and MOP, thus it is not only different from the Converting Optimization but also different from Pareto Optimization. Some tests prove that our new theory may conquer disadvantages of the upper two methods to some extent.

Scheduling an Energy-Aware Parallel Machine System with Deteriorating and Learning Effects Considering Multiple Optimization Objectives and Stochastic Processing Time

Currently,energy conservation draws wide attention in industrial manufacturing systems.In recent years,many studies have aimed at saving energy consumption in the process of manufacturing and scheduling is regarded as an effective approach.This paper puts forwards a multi-objective stochastic parallel machine scheduling problem with the consideration of deteriorating and learning effects.In it,the real processing time of jobs is calculated by using their processing speed and normal processing time.To describe this problem in a mathematical way,amultiobjective stochastic programming model aiming at realizing makespan and energy consumption minimization is formulated.Furthermore,we develop a multi-objective multi-verse optimization combined with a stochastic simulation method to deal with it.In this approach,the multi-verse optimization is adopted to find favorable solutions from the huge solution domain,while the stochastic simulation method is employed to assess them.By conducting comparison experiments on test problems,it can be verified that the developed approach has better performance in coping with the considered problem,compared to two classic multi-objective evolutionary algorithms.

扩展帝国竞争算法求解分布式不相关并行机车间调度问题

针对考虑加工约束的分布式不相关并行机车间调度问题,以总运输成本、工厂间并行机齐停评价函数和工件种类平均切换次数均衡评价函数为优化目标,提出一种扩展帝国竞争算法进行求解。该算法在原始帝国竞争算法的基础上,增加了适于工厂分配的初始化工厂-工件序列群;根据传统帝国竞争算法容易陷入局部最优的缺点,将较劣序列同化分为了外部同化机制和内部同化机制,采用局部和全局相结合的搜索方式实现扩展帝国竞争算法的智能搜索行为;采用部分匹配交叉和单点变异更新工厂-工件序列群,保证工厂-工件序列的多样性。最后设计3个不同规模12个算例,通过仿真实验验证所提算法的有效性,同时对比相关领域研究成果验证了该算法在求解分布式多目标不相关并行机调度问题方面的优越性。

Optimizing Bidders Selection of Multi-Round Procurement Problem in Software Project Management Using Parallel Max-Min Ant System Algorithm

This paper presents a Game-theoretic optimization via Parallel Min-Max Ant System(PMMAS)algorithm is used in practice to determine the Nash equilibrium value to resolve the confusion in choosing appropriate bidders of multi-round procurement problem in software project management.To this end,we introduce an approach that proposes:(i)A Game-theoretic model of multiround procurement problem(ii)A Nash equilibrium strategy corresponds to multi-round strategy bid(iii)An application of PSO for the determination of global Nash equilibrium.The balance point in Nash Equilibrium can help to maintain a sustainable structure not only in terms of project management but also in terms of future cooperation.As an alternative of procuring entities subjectively,a methodology to support decision making has been studied using Nash equilibrium to create a balance point on benefit in procurement where buyers and suppliers need multiple rounds of bidding.Our goal focus on the balance point in Nash Equilibrium to optimizing bidder selection in multi-round procurement which is the most beneficial for both investors and selected tenderers.Our PMMAS algorithm is implemented based on MPI(message passing interface)to find the approximate optimal solution for the question of how to choose bidders and ensure a path for a win-win relationship of all participants in the procurement process.We also evaluate the speedup ratio and parallel efficiency between our algorithm and other proposed algorithms.As the experiment results,the high feasibility and effectiveness of the PMMAS algorithm are verified.

Closed-loop supply chain network equilibrium strategy model with environmental protection objectives

With the increasing popularity of ecological civilization and sustainable development,enterprises should consider environmental protection measures in their operations in addition to pursue their economic interests.This paper establsihes a closed-loop supply chain network model composed of multiple suppliers,manufacturers,retailers,recyclers,and demand markets—regarding their dual goals of the profit maximization and the minimization of carbon emissions.The conditions necessary for establishing overall equilibrium and an equilibrium model of the entire closed-loop supply chain network are determined by applying variational inequality and dual theory.A modified projection contraction algorithm is used to design a model-solving program.Finally,using numerical examples,the paper conducts a comparative static analysis on important parameters such as the weight coefficients of environmental protection objectives and consumers'awareness of low-carbon environmental protection and attains some beneficial enlightenment on management.The results indicate that when the environmental protection objectives of a certain type of enterprise increases,both the economic benefits and environmental protection performance will improve;when the environmental protection objectives of all enterprises increases simultaneously,environmental protection performance improves significantly,but the changes in economic benefits of different enterprises are inconsistent and profit coordination is more complex.Although consumers’awareness of low-carbon preference could improve environmental performance,it reduces the overall profits of network members and the entire closed-loop supply chain network as a whole.The above conclusions can be used as a reference for the government in designing low-carbon environmental protection policy and in closed-loop supply chain research.

Scale-Up Synthesis of Hydrogen Peroxide from H_2/O_2 with Multiple Parallel DBD Tubes

The scale-up synthesis of H2O2 from H2/O2 via a dielectric barrier discharge （DBD） under ambient conditions was studied. A plasma reactor consisting of multiple parallel DBD tubes was designed to scale up the H2O2 synthesis. The number of tubes had no significant effect on the discharge mode, and no decay occurred in H2O2 selectivity during the scale-up process. These advantages made this technology more stable and efficient. The reactor＇s energy efficiency increased with the number of tubes and reached 136 g H2O2/kWh in the four-tube reaction. The total energy efficiency was limited by the extremely low energy transfer efficiency of power supply, and might be enhanced by optimizing the impedance matching between the power supply and the reactor load. As a result, an assembly of multiple DBD tubes may provide a viable route for the scale-up synthesis of H2O2 by a non-equilibrium plasma.

A Multi-Criteria Decision Making for the Unrelated Parallel Machines Scheduling Problem

In this paper, we propose a multi-criteria machine-schedules decision making method that can be applied to a produc-tion environment involving several unrelated parallel machines and we will focus on three objectives: minimizing makespan, total flow time, and total number of tardy jobs. The decision making method consists of three phases. In the first phase, a mathematical model of a single machine scheduling problem, of which the objective is a weighted sum of the three objectives, is constructed. Such a model will be repeatedly solved by the CPLEX in the proposed Multi-Objective Simulated Annealing (MOSA) algorithm. In the second phase, the MOSA that integrates job clustering method, job group scheduling method, and job group – machine assignment method, is employed to obtain a set of non-dominated group schedules. During this phase, CPLEX software and the bipartite weighted matching algorithm are used repeatedly as parts of the MOSA algorithm. In the last phase, the technique of data envelopment analysis is applied to determine the most preferable schedule. A practical example is then presented in order to demonstrate the applicability of the proposed decision making method.

具有两类平行顾客的M/M/1重试排队系统均衡策略研究

本文研究具有两类平行顾客且服务台可靠的M/M/1重试排队系统的均衡策略.在该排队系统中,两类顾客平行到达,并服从不同参数的负指数分布.当顾客进入系统时,若观察到服务台为空,将立刻开始服务;若观察到服务台处于忙期,则进入重试空间等待重试.在完全可见和几乎可见两种情形下,基于“收益-成本”理论提出合理的效用函数并对两类平行顾客进行均衡分析.此外,建立单位时间的社会收益函数,给出最优社会效益分析.最后运用数值分析直观地表示出随着系统参数的改变,顾客行为策略的变化情况.

Constructal design of a rectangular parallel phase change microchannel in a three-dimensional electronic device

Based on constructal theory,a rectangular parallel phase change microchannel model in a three-dimensional electronic device(TDED)is established with R134a as the cooling fluid.Based on the minimization of a complex function(CF)composed of linear weighting sum of maximum temperature difference and pumping power consumption,constructal design of the TDED is conducted first;and then,maximum temperature difference and pumping power consumption are minimized by non-dominated sorting genetic algorithm-II methods.The results reveal that there exist an optimal mass flow rate(0.0012 kg/s)and a quadratic optimal aspect ratio(AR)(0.39)of the microchannel which lead to quadratic minimum CF(0.817).Compared with the original value,the CF after optimization is reduced by 18.34%.Reducing the inlet temperature of cooling fluid and microchannel number appropriately can help to enhance the overall performance of TDED.By using the artificial neural network and genetic algorithms in the toolboxes of Matlab software,the optimal AR gained in the Pareto solution set is located between 0.2–0.45.The smallest deviation index among three discussed strategies is 0.346,and the corresponding optimal AR is 0.413,which is selected as the optimal design strategy of the microchannel in the TDED under multiple requirements.The findings in this study can serve as theoretical guides for thermal designs of electronic devices.

索杆高度可变的3自由度欠约束并联机器人的力学分析

柔索驱动并联机器人是并联机构中的一种特殊柔性机构,通过使用柔索代替刚性杆驱动末端执行器,其综合了柔性机器人和并联机器人两方面的优点。研究了一种索杆高度可变的3自由度欠约束并联机器人,根据矢量闭环原理,建立该机器人的逆运动学模型;根据柔索的长度,推导该机器人的正运动学模型;采用拉格朗日公式,建立该机器人的动力学模型;利用Monte-Carlo方法,研究该机器人的静力学工作空间;给定末端执行器的运动轨迹,通过机器人的逆运动学模型得出3根柔索的期望长度,以3根柔索的期望长度作为实验输入,得到3根柔索的实验长度曲线和拉力曲线以及机器人的正运动学轨迹。实验结果验证了该机器人的正逆运动学和动力学模型的正确性以及工作空间的有效性。

基于Linux集群的并行计算

对普通用户来说,在专业集群中进行并行计算的处理,存在费用高、管理困难和操作复杂等问题。针对该问题,利用PC和以太网相连的实验环境,采用集群工作框架和并行计算的关键技术,提出基于Linux集群的并行计算系统环境。在该环境上对系统的性能进行测试,结果证明了系统的可行性。

混合网格化学非平衡绕流通量分裂格式及并行算法

讨论了非结构混合网格上的二阶VanLeer逆风矢通量分裂格式,并将其应用于三维高超声速化学非平衡粘性流场的并行计算.高超声速绕流的复杂性要求对N-S方程求解的数值模拟方法应具有较高的计算精度及效率.我们针对混合网格上的有限体积格心格式,引入辅助点方法建立了具有空间二阶精度的VanLeer逆风矢通量分裂格式,提高了数值格式的模拟精度,并采用分布式并行化计算技术用以提高计算效率.粘性通量的计算采用中心格式,化学非平衡动力学模型为7组元空气反应模型,采用考虑了化学反应特征时间的当地时间步长显式Runge-Kutta时间推进格式.对三维双椭球外形的高超声速粘性流场进行了并行计算,获得满意的结果.