基于Agent的谈判模型研究

提出了一种基于Agent技术的谈判模型,应用模糊数学理论建立了谈判论据及谈判解接受度数学模型。采用约束放松的方法对谈判模型进行求解,得到了谈判Agent满意的谈判解。阐述了双边多属性谈判过程,设计并开发了Agent谈判系统原型,实例计算分析表明,该谈判模型是一种双赢的谈判模型。     

混合优化算法在兵力部署优化问题中的应用

首先建立了军事作战中兵方部署的两层非线性整数规划模型,根据这种模型的特点,提出了一种求解大规模问题的基于遗传算法和动态规划、具有递阶结构的混合优化算法,该算法可以快速求得满意解。最后给出几个应用算例。     

基于TODIM的直觉模糊双边公平满意匹配方法

针对直觉模糊偏好信息下的双边匹配问题,考虑匹配主体参照依赖和损失规避的心理行为,提出一种基于TODIM(TOmada de decis\ao interativa multicritério)的双边公平满意匹配方法.首先,对直觉模糊偏好信息下的双边匹配问题进行描述;然后,依据前景理论将双边主体的直觉模糊偏好信息转化为相对于参照点的收益或损失,在此基础上,依据TODIM法计算每个主体的总体优势度,构建满意度计算规则,建立双边公平满意匹配优化模型,求解模型并获得双边匹配解;最后,通过一个算例验证所提出方法的可行性和有效性.     

未知环境下基于行为控制的智能车辆路径规划研究

应用模糊逻辑方法,提出了一种在未知环境下基于行为控制的车辆路径规划算法.利用嵌入式计算机PC104构建了智能车的硬件系统,通过DSP控制多只超声波传感器来获取外部环境.路径规划中设计了避障行为和趋向目标点2个基本行为,采用了行为仲裁和行为融合两种方法进行行为决策.通过仿真对比了两种决策方法的优缺点,同时证明了路径规划算法的有效性.把基于行为仲裁的模糊避障算法应用到试验用车的实际避障中,取得了满意的效果.     

多层服务器集群的可用性能评价

本文对多层服务器机群系统的可用性能评价方法进行了研究,并将可用性能作为系统容量规划评价指标,提出基于边际分析容量规划的MACPA算法,从而有效地降低了规划求 解的计算复杂度。模拟实验显示,在三层服务器条件下,MACPA算法所产生的规划方案可以获得相对最优方案99％以上的可用性能。     

带运输考虑的多阶段动态可重入混合流水车间调度

可重入混合流水车间调度允许一个工件多次进入某些加工阶段,它广泛出现在许多工业制造过程中,如半导体制造、印刷电路板制造等.本文研究了带运输时间的多阶段动态可重入混合流水车间问题,目标是最小化总加权完成时间.针对该问题,建立了整数规划模型,进而基于工件解耦方式提出了两种改进的拉格朗日松弛(LR)算法.在这些算法中,设计了动态规划的改进策略以加速工件级子问题的求解,提出了异步次梯度法以得到有效的乘子更新方向.测试结果说明了所提出的两种改进算法在解的质量和运行时间方面均优于常规LR算法,两种算法都能在可接受的计算时间内得到较好的近优解.     

求解双边加权模糊支持向量机的序贯最小优化算法

高的计算复杂度限制了双边加权模糊支持向量机在实际分类问题中的应用。为了降低计算复杂度,提出了应用序贯最小优化算法（SMO）解该模型,该模型首先将整个二次规划问题分解成一系列规模为2的二次规划子问题,然后求解这些二次规划子问题。为了测试SMO算法的性能,在三个真实数据集和两个人工数据集上进行了数值实验。结果表明：与传统的内点算法相比,在不损失测试精度的情况下,SMO算法明显地降低了模型的计算复杂度,使其在实际中的应用成为可能。     

采用组合优化策略的电子侦察卫星规划方法

对电子侦察卫星任务规划问题进行了分析,建立了问题的多目标规划模型;设计了一种基于带后优化过程MOEO(Multi-objective Extremal Optimization)的多目标规划算法对模型进行求解,该算法包含MOEO主算法过程和基于禁忌搜索(TS)的后优化过程两部分:MOEO主算法中采用插入变异、模式变异及删除变异等算子对解空间进行搜索,基于Pareto最优概念的解排序确保了解在多个目标上的有效优化,精英策略避免了丢失进化过程中产生的非劣解;TS后优化过程中提出了多种邻域结构,使用各种邻域算子或算子的组合,对主算法Pareto最优解进一步优化,以得到更好的解。最后给出了仿真实例证明本文模型及算法对解决电子侦察卫星任务规划问题的有效性。     

加工时间可控的炼钢调度问题两阶段模型及优化算法

炼钢-精炼-连铸是钢铁产品的关键生产工序,其有效的调度对生产过程中减少热能消耗、提高生产效率具有重要意义.根据生产过程中工序加工时间可控性和主要工艺约束提出了分散搜索(scattcr scarch,SS)算法和数学规划相结合的两阶段求解算法.第1阶段应用SS算法基于各阶段正常的加工时间,确定炼钢-精炼生产阶段各设备的加工炉次集和各炉次的加工顺序.第2阶段将SS求得的解转化为时间约束网络图,建立了以炉次等待设备时间和设备等待炉次时间及最大完成时间最小为调度目标,工序加工时间可控的混合整数规划模型,应用CPLEX求解模型确定各炉次的加工时间和开始时间.基于国内某钢铁企业炼钢-精炼-连铸生产过程的实绩生成了14个不同规模的测试案例,对钢厂生产实绩效果与本文两阶段求解算法的优化效果进行了对比,分析了不同等待时间权重对两阶段算法性能的影响,并与采用遗传局域搜索(gcnctic local search,GLS)算法与数学规划相结合的求解算法的优化效果进行了比较.实验结果表明本文给出的模型和两阶段求解算法对加工时间可控的炼钢-精炼-连铸调度问题的优化效果很好.     

竞争环境下B2C平台增值服务投资决策

为提高竞争环境下的平台经济效益,讨论平台企业对双边用户的增值服务投资问题,在考虑用户多归属条件下构建B2C平台竞争模型.通过比较分析发现,在双边单归属或一边多归属条件下,平台企业的最优投资满足一个区间策略:若投资资源小于该区间的下界,则根据边际投资成本小于或大于某一阈值,平台企业选择投资全部或部分资源;若投资资源大于该区间的上界,则最优投资存在两个纯策略均衡;若投资资源位于该区间内,则最优投资存在唯一纳什均衡.此外,在双边多归属条件下,平台企业的最优投资满足一个单阈值策略:根据边际投资成本小于或大于某一阈值,平台企业选择投资全部或部分资源.     

Multi-Dimensional,MultiStep Negotiation

We present a multi-dimensional, multi-step negotiation mechanism for task allocation among cooperative agents based on distributed search. This mechanism uses marginal utility gain and marginal utility cost to structure this search process, so as to find a solution that maximizes the agents’ combined utility. These two utility values together with temporal constraints summarize the agents’ local information and reduce the communication load. This mechanism is anytime in character: by investing more time, the agents increase the likelihood of getting a better solution. We also introduce a multiple attribute utility function into negotiations. This allows agents to negotiate over the multiple attributes of the commitment, which produces more options, making it more likely for agents to find a solution that increases the global utility. A set of protocols are constructed and the experimental result shows a phase transition phenomenon as the complexity of negotiation situation changes. A measure of negotiation complexity is developed that can be used by an agent to choose an appropriate protocol, allowing the agents to explicitly balance the gain from the negotiation and the resource usage of the negotiation.This revised version was published online in August 2005 with a corrected cover date.     

无线充电体域网基于边际效用的网络资源分配方法

射频能量采集技术可以从根本上解决电池容量对无线体域网生存期的限制,为了提高网络资源分配的效率以及公平性,提出一种基于边际效用理论的网络资源分配方法。首先,设计传感器节点的效用函数,将节点所能获得的吞吐量映射成QoS满意的等级;然后,以最大化网络中全部传感器节点整体效用为目标,将多高效、公平的网络资源分配问题构建成效用最大化问题;最后,通过对偶分解方法求得该问题的最优解。仿真结果表明,与总吞吐量最大化和sigmoid效用最大化方法相比,所提出的方法在获得较高系统整体吞吐量的同时,确保了传感器节点个体获得吞吐量的公平性。     

SUTIL – Network selection based on utility function and integer linear programming

This work presents SUTIL, a mechanism for network selection in the context of next generation networks (NGN). SUTIL selection mechanism prioritizes networks with higher relevance to the application and lower energy consumption and it enables full and seamless connectivity to mobile user devices and applications. Consequently, SUTIL contributes to realize the vision of ubiquitous computing, in which services, devices, and sensor-enriched environments interact anytime, anywhere to accomplish human designed tasks. The provided solution is based on utility function and integer linear programming and it aims at: (i) maximizing the user satisfaction while meeting application QoS and (ii) minimizing the energy consumption of devices when connecting to a target network. The solution is global since it considers for a given base station all devices that are simultaneously candidate for handoff. Simulation results showed the benefits of SUTIL usage in NGN environments.     

Neural network approach for allocation with capacity

In this paper we discuss neural network approach for allocation with capacity constraints problem. This problem can be formulated as zero-one integer programming problem. We transform this zero-one integer programming problem into an equivalent nonlinear programming problem by replacing zero-one constraints with quadratic concave equality constraints. We propose two kinds of neural network structures based on penalty function method and augmented Lagrangian multiplier method, and compare them by theoretical analysis and numerical simulation. We show that penalty function based neural network approach is not good to combinatorial optimization problem because it falls in the dilemma whether terminating at an infeasible solution or sticking at any feasible solution, and augmented Lagrangian multiplier method based neural network can alleviate this suffering in some degree.     

A combined neurodynamic approach to optimize the real-time price-based demand response management problem using mixed zero-one programming

This paper presents a microgrid system model considering three types of load and the user’s satisfaction function. The objective function with mixed zero-one programming is used to maximize every user’s profit and satisfaction in the way of the demand response management under real-time price. An energy function is used to transform the constrained problem into an unconstrained problem, and two neural networks are used to find the local optimal solutions of the objective function with different rates of convergence. A neurodynamic approach is used to combine the neural networks with the particle swarm optimization to find the global optimal solution of the objective function. The simulation results show that the combined approach is effective in solving the optimal problem.

     

The multi-level warehouse layout problem with uncertain information: uncertainty theory method

This paper investigates the multi-level warehouse layout problem with indeterminate factors, in which the monthly demands and horizontal transportation distances are described by uncertain variables. We first consider the distribution function of the total cost for transportation. Second, two uncertain models, namely, the chance-constrained programming model and the chance-maximum programming model, are developed to lay out the multi-level warehouse under uncertainty. Some properties of the models are discussed to solve the models. The properties point out that the optimal solution to the chance-constrained programming model is equivalent to a corresponding deterministic model. Additionally, we also discuss the relation between the chance-constrained programming model and the chance-maximum programming model, which leads to an effective approach to search for the optimal solution to the chance-maximum programming model. Finally, a numerical experiment is illustrated to show the ideas of the proposed models.     

Parallel programming in computational science: an introductory practical training course for computer science undergraduates at Aachen University

Parallel programming of high-performance computers has emerged as a key technology for the numerical solution of large-scale problems arising in computational science and engineering (CSE). The authors believe that principles and techniques of parallel programming are among the essential ingredients of any CSE as well as computer science curriculum. Today, opinions on the role and importance of parallel programming are diverse. Rather than seeing it as a marginal beneficial skill optionally taught at the graduate level, we understand parallel programming as crucial basic skill that should be taught as an integral part of the undergraduate computer science curriculum. A practical training course developed for computer science undergraduates at Aachen University is described. Its goal is to introduce young computer science students to different parallel programming paradigms for shared and distributed memory computers as well as to give a first exposition to the field of computational science by simple, yet carefully chosen sample problems.     

风险效用投资组合模型的遗传算法求解

给出一个折衷考虑风险最小化和收益最大化的单目标决策方法,以单位风险收益最大化为决策目标建立了投资组合的非线性分式规划模型,考虑到分式规划问题的求解难度,利用遗传算法求解模型,并给出算法步骤。最后,给出了数值算例,结果表明该算法是简单有效的。     

A macroeconomic model for resource allocation in large-scale distributed systems

In this paper we discuss an economic model for resource sharing in large-scale distributed systems. The model captures traditional concepts such as consumer satisfaction and provider revenues and enables us to analyze the effect of different pricing strategies upon measures of performance important for the consumers and the providers. We show that given a particular set of model parameters the satisfaction reaches an optimum; this value represents the perfect balance between the utility and the price paid for resources. Our results confirm that brokers play a very important role and can influence positively the market. We also show that consumer satisfaction does not track the consumer utility; these two important performance measures for consumers behave differently under different pricing strategies. Pricing strategies also affect the revenues obtained by providers, as well as, the ability to satisfy a larger population of users.     

Utility-based QoS optimisation strategy for multi-criteria scheduling on the grid

The paper presents quality of service (QoS) optimisation strategy for multi-criteria scheduling on the grid, based on a mathematical QoS model and a distributed iterative algorithm. Three QoS criteria are considered, namely payment, deadline and reliability, which are formulated as utility function. The optimisation problem is split into two parts: task optimisation performed on behalf of the user and resource optimisation performed on behalf of the grid. The strategy employs three types of agents: task agents responsible for task optimisation, computation resource and network resource agents responsible for resource optimisation. The agents apply economic models for optimisation purposes. Dynamic programming is used to optimise the total system utility function in terms of an iterative algorithm. The objective of multi-criteria scheduling is to maximise the global utility of the system. This paper proposes an iterative scheduling algorithm that is used to perform QoS optimisation-based multi-criteria scheduling. The proposed QoS optimisation-based multi-criteria scheduling problem solution has been practically examined by simulation experiments.