MEA-BP模型在遥感影像分类中的应用研究

遥感影像分类作为遥感技术的一个重要应用,对遥感技术的发展具有重要作用.针对遥感影像数据特点,在目前的非线性研究方法中主要用到的是BP神经网络模型.但是BP神经网络模型存在对初始权阈值敏感、易陷入局部极小值和收敛速度慢的问题.因此,为了提高模型遥感影像分类精度,提出采用MEA-BP模型进行遥感影像数据分类.首先采用思维进化算法代替BP神经网络算法进行初始寻优,再用改进BP算法对优化的网络模型权阈值进一步精确优化,随后建立基于思维进化算法的BP神经网络分类模型,并将其应用到遥感影像数据分类研究中.仿真结果表明,新模型有效提高了遥感影像分类准确性,为遥感影像分类提出了一种新的方法,具有广泛研究价值.     

次模函数最大化的流算法综述

次模函数优化在计算机科学、数学、经济学等学科得到广泛研究.大数据环境下的次模优化是相对较新的研究领域,受到更多关注.特别地,考虑基于流模型的次模最大化问题.在该问题中,数据以流的形式呈现,其目的是从数据流中抽取满足某些特性的稀疏子集,最大化次模收益函数值.介绍了基于流模型的次模最大化问题的阈值和优先权方法,同时也介绍了若干次模最大化变形的流算法进展.     

Fuzzy optimization approach to component selection of fault-tolerant software system

In developing software systems, a manager’s goal is to design software using limited resources and meet the user requirements. One of the important user requirements concerns the reliability of the software. The decision to choose the right software modules (components) becomes extremely difficult because of the number of parameters to be considered while making the decision. If suitable components are not available, then the decision process is further complicated with build versus buy decisions. In this paper, we have formulated a fuzzy multi-objective approach to optimal decision “build-or-buy” for component selection for a fault-tolerant modular software system under the consensus recovery block scheme. A joint optimization model is formulated where the two objectives are maximization of system reliability and minimization of the system cost with a constraint on delivery time. An example of developing a retail system for small-and-medium-size enterprises is used to illustrate the proposed methodology.     

Performance analysis and optimization of cognitive radio networks with retransmission control

In usual cognitive radio networks with buffer settings, secondary user packets that are interrupted by primary user packets will return to the SU buffer for later retransmission. But this may increase the average delay of the SU packets. In this paper, we propose a novel spectrum sharing strategy (SSS) to dynamically control the retransmission of the SU packets by introducing a returning threshold and a returning probability. This will simultaneously guarantee the Quality of Service for the SUs. In this SSS, when the transmission of an SU packet is interrupted, if the number of SU packets already in SU buffer reaches a returning threshold that is set in advance as a system parameter, an interrupted SU packet is admitted to return to the SU buffer with a dynamic returning probability. This returning probability is inversely proportional to the total number of packets in the system. Based on the SSS with retransmission control proposed in this paper, we build a discrete-time preemptive priority queueing model to comply with digital nature of modern networks. Accordingly, by presenting and analyzing a two-dimensional discrete-time Markov chain that represents the system state transition, we obtain the steady-state distribution of the system and then provide the formulas for several system performance measures. Moreover, with numerical results, we show the influence of the returning threshold on different performance measures. Finally, by building a net benefit function, we optimize the system performance for the returning threshold to balance different system performance measures.     

A novel algorithm for area traffic capacity control with elastic travel demands

A non-linear area traffic control system with limited capacity is considered in this paper. Optimal signal settings and link capacity expansions can be determined while trip distribution and network flow are in equilibrium. This problem can be formulated as a non-linear mathematical program with equilibrium constraints. For the objective function a non-linear constrained optimization program for signal settings and link capacity expansion is determined. For the constraint set the elastic user equilibrium traffic assignment obeying Wardrop’s first principle can be formulated as a variational inequality. Since the constrained optimization problem is non-convex, only local optima can be obtained. In this paper, a novel algorithm using a non-smooth trust region approach is proposed. Numerical tests are performed using a real data city network and various example test networks in which the effectiveness and robustness of the proposed method are confirmed as compared to other well-known solution methods.     

A stackelberg game approach to distributed spectrum management

We consider a cognitive radio system with one primary (licensed) user and multiple secondary (unlicensed) users. Given the interference temperature constraint, the secondary users compete for the available spectrum to fulfill their own communication need. Borrowing the concept of price from market theory, we develop a decentralized Stackelberg game formulation for power allocation. In this scheme, the primary user (leader) announces prices for the available tones such that a system utility is maximized. Using the announced prices, secondary users (followers) compete for the available bandwidth to maximize their own utilities. We show that this Stackelberg game is polynomial time solvable under certain channel conditions. When the individual power constraints of secondary users are inactive (due to strict interference temperature constraint), the proposed distributed power control method is decomposable across the tones and unlike normal water-filling it respects the interference temperature constraints of the primary user. When individual power constraints are active, we propose a distributed approach that solves the problem under an aggregate interference temperature constraint. Moreover, we propose a dual decomposition based power control method and show that it solves the Stackelberg game asymptotically when the number of tones becomes large.     

Lifetime maximization in wireless directional sensor network

This paper addresses two versions of a lifetime maximization problem for target coverage with wireless directional sensor networks. The sensors used in these networks have a maximum sensing range and a limited sensing angle. In the first problem version, predefined sensing directions are assumed to be given, whereas sensing directions can be freely devised in the second problem version. In that case, a polynomial-time algorithm is provided for building sensing directions that allow to maximize the network lifetime. A column generation algorithm is proposed for both problem versions, the subproblem being addressed with a hybrid approach based on a genetic algorithm, and an integer linear programming formulation. Numerical results show that addressing the second problem version allows for significant improvements in terms of network lifetime while the computational effort is comparable for both problem versions.     

多方式弹性需求下公交定价双层规划模型

为描述多方式城市交通网络下公交定价与出行选择行为的相互作用与影响,将出行方式选择与路径选择涵盖于同一网络,建立了上层模型分别以企业利润最大化、乘客出行成本最小化和社会福利最大化为目标函数,下层模型为多方式弹性需求随机用户配流模型的公交定价双层规划模型。运用改进遗传算法对模型整体进行求解,下层模型采用综合对角化算法和MSA算法的组合求解算法。最后,设计了一个算例以说明模型应用。结果表明:运用双层规划模型所确定的公交票价较传统静态票价可使政府、企业及出行者三方都获得更高收益,且上层模型以社会福利最大化为目标函数能代表社会群体中多数人利益,优化效果最为理想。     

Balancing two-sided U-type assembly lines using modified particle swarm optimization algorithm

In this paper, a new two-sided U-type assembly line balancing (TUALB) procedure and a new algorithm based on the particle swarm optimization algorithm to solve the TUALB problem are proposed. The proposed approach minimizes the number of stations for a given cycle time as the primary objective and it minimizes the number of positions as a secondary objective. The proposed approach is illustrated with an example problem. In order to evaluate the efficiency of the proposed algorithm, the test problems available in the literature are used. The experimental results show that the proposed approach performs well.     

Deriving crisp and consistent priorities for fuzzy AHP-based multicriteria systems using non-linear constrained optimization

Fuzzy optimization models are used to derive crisp weights (priority vectors) for the fuzzy analytic hierarchy process (AHP) based multicriteria decision making systems. These optimization models deal with the imprecise judgements of decision makers by formulating the optimization problem as the system of constrained non linear equations. Firstly, a Genetic Algorithm based heuristic solution for this optimization problem is implemented in this paper. It has been found that the crisp weights derived from this solution for fuzzy-AHP system, sometimes lead to less consistent or inconsistent solutions. To deal with this problem, we have proposed a consistency based constraint for the optimization models. A decision maker can set the consistency threshold value and if the solution exists for that threshold value then crisp weights can be derived, otherwise it can be concluded that the fuzzy comparison matrix for AHP is not consistent for the given threshold. Three examples are considered to demonstrate the effectiveness of the proposed method. Results with the proposed constraint based fuzzy optimization model are more consistent than the existing optimization models.     

基于模糊时间序列的传感器网络感知数据预测模型

传感器网络监控系统属于大型复杂系统,由感知节点以一定的时间间隔向sink节点发送感知数据,以实现对应用环境的监控。由于网络本身及应用环境的影响,得到的感知数据往往存在不确定性。此外,周期性报告数据模式影响到实时监控数据的精确性。本文应用时间序列模型预测传感器数据以响应用户查询,可有效降低网络通信量。通过对无线传感器网络的数据分析,引入多属性模糊时间序列预测模型,充分考虑了无线传感器网络时间序列中存在的趋势因素,并提出了适合于传感器网络的修正预测模型。实验结果表明模糊时间序列模型可有效预测传感器网络数据,且能提高预测精度。     

考虑双方感知满意度的应急资源再配置方法

突发事件常诱发次生灾害,在外界大规模应急资源配送前,发挥区域互救优势,共享原生灾害点应急资源进行资源再配置是及时响应次生灾害的重要手段之一。本文尝试将灾民心理因素融入到应急资源的再配置问题中,基于前景理论提出了原生灾害点灾民的感知满意度模型,结合生存概率曲线对次生灾害点灾民感知满意度进行刻画,构建了考虑双方灾民感知满意度的多目标优化模型。此外,引入缩放系数和违反约束检查函数,改进多目标粒子群算法以加快算法初始化,进而提升模型求解效率。最后,本文通过算例验证了模型和算法的可行性和科学性,并与传统资源配置问题进行了对比,为具有连锁反应的突发事件应急管理提供决策支持。     

生态工业共生网络中的鲁棒优化模型

生态工业共生网络(EISN)的稳定可持续运行是生态工业园中各企业实现经济和环境利益最大化的基本保障。在考虑政府激励因素的情况下,建立一个以n个生产者企业和m个分解者企业为核心的EISN利润最大化问题的最优化模型。在此基础上,为更真实反映EISN的运行情况,考虑到分解者企业利用次级原材料的单位成本及次级原材料转化率的不确定性,构建了EISN利润最大化问题的离散鲁棒优化模型,提出了基于网络最小费用流的模型求解方法,证明了模型解的存在性,并给出了相应的案例。     

Sequential Semidefinite Program for Maximum Robustness Design of Structures under Load Uncertainty

A robust structural optimization scheme as well as an optimization algorithm are presented based on the robustness function. Under the uncertainties of the external forces based on the info-gap model, the maximization of the robustness function is formulated as an optimization problem with infinitely many constraints. By using the quadratic embedding technique of uncertainty and the S-procedure, we reformulate the problem into a nonlinear semidefinite programming problem. A sequential semidefinite programming method is proposed which has a global convergent property. It is shown through numerical examples that optimum designs of various linear elastic structures can be found without difficulty.The authors are grateful to the Associate Editor and two anonymous referees for handling the paper efficiently as well as for helpful comments and suggestions.     

Target Level Method in Linear Multi-Criterion Problems

We consider the target level method for solving linear multi-criterion maximization problems. The method finds an efficient (Pareto-optimal) vector estimate that is closest in the Chebyshev metric to the target level point specified by the decision maker. The proposed method describes (parametrizes and approximates) the efficient set. In the linear case the number of scalar optimization problems needed to describe the set of efficient vector estimates is substantially reduced. A formula is derived which, under certain conditions, can be used to compute efficient vector estimates without solving any optimization problems. An algorithm based on these results is proposed for two-criterion problems.     

离散椭球分布下两阶段WCVaR风险利润优化模型及应用

本文研究随机变量非完全分布下的两阶段风险-利润优化问题。采用最坏情况下条件风险(Worst-case Conditional Value-at-Risk:WCVaR) 度量指标,在离散椭球分布下建立了两阶段WCVaR 约束下利润期望最大优化模型,运用优化对偶方法将复杂的Max-Min 结构化简,理论上证明了简化模型和原模型的同解性,以发电商电能分配组合优化为数值实例,验证了模型和计算方法的有效性。     

基于改进的混沌鲸鱼算法优化DBN的广东人口数量预测研究

为了估算广东省人口总数和老龄化人口总数,针对深度置信网络模型的性能易受其权值和阈值选择的影响,提出一种基于改进的混沌鲸鱼算法优化DBN的人口数量预测模型。首先,为提高鲸鱼算法的收敛速度和避免局部最优,将Skew Tent混沌模型和非线性收敛因子引入WOA算法用于初始化WOA种群和改进WOA更新策略。其次为了提高DBN模型的性能,运用ICWOA算法对DBN模型的权值和阈值进行优化选择,融合影响人口数量的多因素特征因子,将不同年龄段的总人数、死亡率、存活率、年生育率、第一产业占比、第二产业占比以及第三产业占比等多因素特征因子作为ICWOA-DBN的输入,人口数量作为ICWOA-DBN的输出,建立ICWOA-DBN人口数量预测模型。通过标准函数测试寻优对比发现,提出的ICWOA具有更快的收敛速度和更小的适应度值。为了验证本文算法ICWOA-DBN的预测性能,以第6次全国人口普查数据为参考依据,选择2005~2016年广东省历年常住人口总数和老龄化人口总数为研究对象,研究结果表明,与IWOA-DBN、WOA-DBN、GA-DBN和DBN相比,提出的ICWOA-DBN的人口数量预测模型的精度最高,为人口数量预测提供新的方法和途径。     

A Lane-Based Optimization Method for Minimizing Delay at Isolated Signal-Controlled Junctions

This paper presents a lane-based optimization method for minimizing delay at isolated signal-controlled junctions. The method integrates the design of lane markings and signal settings, and considers both traffic and pedestrian movements in a unified framework. While the capacity maximization and cycle length minimization problems are formulated as Binary-Mix-Integer-Linear-Programs (BMILPs) that are solvable by standard branch-and-bound routines, the problem of delay minimization is formulated as a Binary-Mix-Integer-Non-Linear Program (BMINLP). A cutting plane algorithm and a heuristic line search algorithm are proposed to solve this difficult BMINLP problem. The integer variables include the permitted movements on traffic lanes and successor functions to govern the order of signal displays, whereas the continuous variables include the assigned lane flows, common flow multiplier, cycle length, and starts and durations of green for traffic movements, lanes and pedestrian crossings. A set of constraints is set up to ensure the feasibility and safety of the resultant optimized lane markings and signal settings. A numerical example is given to demonstrate the effectiveness of the proposed methodology. The heuristic line search algorithm is more cost-effective in terms of both optimality of solution and computing time requirement. This revised version was published online in July 2006 with corrections to the Cover Date.     

Resource contention games in multiclass stochastic flow models

We consider resource contention games in a stochastic hybrid system setting using Stochastic Flow Models (SFM) with multiple classes and class-dependent objectives. We present a general modeling framework for such games, where Infinitesimal Perturbation Analysis (IPA) estimators are derived for the derivatives of various class-dependent objectives. This allows us to study these games from the point of view of system-centric optimization of a performance metric and compare it to the user-centric approach where each user optimizes its own performance metric. We derive explicit solutions for a specific model in which the competing user classes employ threshold control policies and service is provided on a First Come First Serve (FCFS) basis. The unbiasedness of the IPA estimators is established in this case and it is shown that under certain conditions the system-centric and user-centric optimization solutions coincide.     

Multi-group support vector machines with measurement costs: A biobjective approach

Support Vector Machine has shown to have good performance in many practical classification settings. In this paper we propose, for multi-group classification, a biobjective optimization model in which we consider not only the generalization ability (modeled through the margin maximization), but also costs associated with the features. This cost is not limited to an economical payment, but can also refer to risk, computational effort, space requirements, etc. We introduce a Biobjective Mixed Integer Problem, for which Pareto optimal solutions are obtained. Those Pareto optimal solutions correspond to different classification rules, among which the user would choose the one yielding the most appropriate compromise between the cost and the expected misclassification rate.