一种多种群遗传策略的OFDM功率分配算法

针对标准遗传算法在解决比例速率约束下多用户OFDM系统的功率分配出现的收敛速度慢和早熟收敛问题,提出了一种基于多种群遗传策略的功率分配算法。提出算法以业务公平指数为适应度值和以最优个体保持代数为算法终止依据。各个种群使用不同的控制参数,通过移民算子相互联系。仿真结果表明,提出的算法的收敛速度(100代左右)比标准遗传算法的收敛速度(300代左右)快且收敛结果稳定(都基本趋于0),在最大化总容量的同时很好地维持了用户比例速率公平性。     

基于单亲遗传算法的RoboCup动态角色分配

RoboCup的机器人动态角色分配问题是一个典型的组合优化问题。解决这一问题的传统方法是贪心法,但贪心法易陷入局部最优解。提出用针对组合优化问题而构造的序号编码单亲遗传算法解决RoboCup的机器人动态角色分配问题。单亲遗传算法借鉴了传统遗传算法“优胜劣汰”的自然选择机制,但只通过单个体繁殖后代,在解决组合优化问题和复杂工程优化问题方面具有明显的优越性。试验结果显示这种方法的在解决RoboCup机器人动态角色分配问题时的有效性。     

一种改进的BP神经网络算法及其应用

BP算法是目前应用最为广泛的神经网络学习算法,但原始算法收敛速率慢,训练过程易陷入局部极小值以及隐层节点数选择困难.针对这些问题提出了不少改进措施,文中提出了一种改进的BP神经网络算法,在BP算法基础上,从训练算法着手,通过误差的变化趋势,动态调整权值以提高网络的收敛速率;通过数学推导,从理论上验证了该算法的有效性.用MATLAB软件对文中的改进算法进行仿真,并且与其它方法进行比较,结果表明,改进后的算法在收敛速率和抑制噪声等方面有很好的效果,从实验上验证了该算法的有效性.     

融合柯西折射反向学习和变螺旋策略的 WSN象群定位算法

针对现有无线传感器网络(WSN)优化算法在定位过程中收敛速率慢和误差大的问题,提出一种基于柯西折射反向学习和变螺旋机制的象群节点定位算法.首先,利用具有遍历性和随机性的Logistic混沌映射初始化种群,丰富种群多样性,加快算法收敛速率.然后,将折射反向学习机制与柯西变异相融合以随机扰动族长位置,避免算法陷入局部最优.最后,在氏族分离过程中引入自适应变螺旋策略更新病态大象位置,提升算法全局搜索能力.仿真结果表明,与现有WSN优化算法相比,所提出的改进象群优化算法在定位精度和收敛速率方面得到明显提升.     

数据驱动的保证收敛速率最优输出调节

针对具有外部系统扰动的线性离散时间系统的输出调节问题,提出了可保证收敛速率的数据驱动最优输出调节方法,包括状态可在线测量系统的基于状态反馈的算法,与状态不可在线测量系统的基于输出反馈的算法.首先,该问题被分解为输出调节方程求解问题与反馈控制律设计问题,基于输出调节方程的解,通过引入收敛速率参数,建立了可保证收敛速率的最...     

一种连续空间优化问题的蚁群算法及应用

针对随机优化算法收敛困难及搜索时间较长的问题,提出一种求解连续空间优化问题的蚁群算法,为蚁群算法在连续空间中的应用提供了一个可行的方案。给出了该算法的详细定义及实现步骤,并将该算法应用于多变量函数优化及热工控制系统控制器参数优化,仿真结果表明:该算法具有良好的全局优化性能,能加快收敛速率,解决了随机优化算法收敛困难的问题,并提高寻优精度。     

反向传播神经网络收敛性的探讨

标准BP算法采用的最陡梯度下降法使得均方误差达到最小的策略可能存在两大问题：①陷入局部最小而没有收敛到全局最小，即不收敛；②收敛速率慢。本文从训练算法角度方面，比较了标准BP算法、动量算法、可变学习速率算法和Levenberg-Marquardt算法这几种方法的收敛性以及收敛速率，并通过Matlab仿真进行了验证。     

随机ADD算法的不确定网络优化研究

传统的网络优化问题通过对偶梯度下降算法来解决,虽然该算法能够以分布式方式来实现,但其收敛速度较慢.加速对偶下降算法(ADD)通过近似牛顿步长的分布式计算,提高了对偶梯度下降算法的收敛速率.但由于通信网络的不确定性,在约束不确定时,该算法的收敛性难以保证.基于此,提出了一种随机形式的ADD算法来解决该网络优化问题.理论上证明了随机ADD算法当不确定性的均方误差有界时,能以较高概率收敛于最优值的一个误差邻域;当给出更严格的不确定性的约束条件时,算法则可以较高概率收敛于最优值.实验结果表明,随机ADD算法的收敛速率比随机梯度下降算法快两个数量级.     

基于分离度变步长的自然梯度算法的语音分离

为解决固定步长自然梯度算法带来的收敛速率与稳态误差之间矛盾的问题,采用一种自适应步长的自然梯度算法,根据实际分离度自动调整步长,通过设置两个参数作为分离度系数控制收敛速率和稳态误差。仿真实验结果表明相较固定步长算法,该算法具有较好的分离性能,收敛速率提高了1倍,同时稳态误差缩小了3倍。     

一种新的变学习速率自适应独立分量分析算法

学习速率的优选问题是自适应ICA算法中一个重要问题。论文建立了学习速率与相依性测度之间的一种非线性函数关系,以此为基础提出了一种新的变学习速率的自适应ICA算法,并且分析了参数a,b的取值原则及对算法收敛性能和稳态性能的影响。该算法能根据相依性测度所反映的信号分离的状态自适应地调节学习速率,克服了传统算法在稳态阶段步长调整过程中的不足。理论分析和计算机仿真结果都验证了算法的收敛性能和稳态性能。     

基于Volterra模型的预测控制及应用

由于工业实践的需要,非线性预测控制近年来受到广泛地关注.Volterra模型是一类特殊的非线性模型,非常适合描述工业过程中的无记忆非线性对象.传统的基于Volterra模型的控制器合成法及迭代计算预测控制器法计算量大,且不便于处理控制约束.非线性模型预测控制求解是典型的非线性规划问题,序列二次规划(sequential quadratic program,SQP)算法是求解非线性规划问题常用方法之一.针对Volterra非线性模型预测控制求解问题,本文将滤子法与一种信赖域SQP算法相结合,提出一种改进SQP算法用于基于非线性Volterra模型的带控制约束的多步预测控制求解,并分析了所提方法的收敛性.工业实例仿真结果证实了所提方法的可行性与有效性.     

An affine scaling interior trust-region method combining with nonmonotone line search filter technique for linear inequality constrained minimization

This paper proposes an affine scaling interior trust-region method in association with nonmonotone line search filter technique for solving nonlinear optimization problems subject to linear inequality constraints. Based on a Newton step which is derived from the complementarity conditions of linear inequality constrained optimization, a trust-region subproblem subject only to an ellipsoidal constraint is defined by minimizing a quadratic model with an appropriate quadratic function and scaling matrix. The nonmonotone schemes combining with trust-region strategy and line search filter technique can bring about speeding up the convergence progress in the case of high nonlinear. A new backtracking relevance condition is given which assures global convergence without using the switching condition used in the traditional line search filter technique. The fast local convergence rate of the proposed algorithm is achieved which is not depending on any external restoration procedure. The preliminary numerical experiments are reported to show effectiveness of the proposed algorithm.     

A new non-monotone SQP algorithm for the minimax problem

In this paper, a new sequential quadratic programming (SQP) algorithm is proposed to solve the minimax problem which uses the idea of nonmonotonicity. The problem is transformed into an equivalent inequality constrained nonlinear optimization problem. In order to prevent the scaling problem, we do some modifications to the minimization problem. By the non-monotone SQP method, the new algorithm is globally convergent without using a penalty function. Furthermore, it is shown that the proposed method does not suffer from the Maratos effect, so the locally superlinear convergence is achieved. Numerical results suggest that our algorithm for solving the minmax problem is efficient and robust.     

基于变学习率三角基函数神经网络的4型FIR滤波器设计

本文提出一种基于变学习率三角基函数神经网络的线性相位4型FIR滤波器设计方法。该方法根据三角基函数神经网络与线性相位4型FIR滤波器幅频特性之间的关系,构建了一种变学习率三角基函数神经网络模型,在神经网络训练过程中引入变学习率算法自调整学习率取值,解决学习率通常依靠经验或试凑法确定带来的不确定性,提高神经网络的学习效率和收敛速度。通过训练神经网络的权值,使设计的FIR滤波器幅频响应与理想幅频响应在整个通带和阻带内的误差平方和最小。文中利用该方法对FIR高通滤波器和带通滤波器进行了优化设计,仿真结果表明了该方法设计FIR滤波器的有效性和优越性。     

Feasibility refinement in sequential quadratic programming using parametric sensitivity analysis

In this paper we present results that extend the sequential quadratic programming (SQP) algorithm with an additional feasibility refinement based on parametric sensitivity derivatives. The refinement is applicable without restriction on the problem dimensions in sparse SQP solvers. Parametric sensitivity analysis is a tool for post optimality analysis of the solution of a nonlinear optimization problem. For the refinement approach we apply this technique on the quadratic subproblems in order to improve the overall algorithm. The sensitivity derivatives required for this approach can be computed without noticeable computational effort as the system of linear equations to be solved coincides with the system already solved for the search direction computation. For similar algorithms in the context of post optimality analysis a linear rate of convergence has been proven and therefore an extrapolation method is applied to speed up the process. The presented algorithm has been integrated into the nonlinear program (NLP) solver WORHP and we perform a numerical study to evaluate different termination criteria for the proposed algorithm. Furthermore, numerical results on the CUTEst test set are shown.     

A trust region SQP-filter method for nonlinear second-order cone programming

In this paper, we consider the nonlinear second-order cone programming problem. By combining an SQP method and filter technique, we present a trust region SQP-filter method for solving this problem. The proposed algorithm avoids using the classical merit function with penalty term. Furthermore, under standard assumptions, we prove that the iterative sequence generated by the presented algorithm converges globally. Preliminary numerical results indicate that the algorithm is promising.     

结合遗传优化的多结构多尺度形态学消噪

传统的形态滤波以及广义形态滤波、自适应加权广义滤波、基于多结构元素的广义形态滤波、基于多方向的广义滤波和基于多尺度的广义滤波在考虑形态学滤波时基本上只兼顾到某一方面或者说只对某一方面的不足进行了改进,不论哪一种滤波方式都没有完全消除噪声。提出了一种基于自适应遗传算法的多结构多尺度形态学滤波方法,主要考虑了滤波窗口的大小、结构元素的种类和方向以及结构元素的优化选择问题,采用遗传算法对结构元素进行优化,并考虑到遗传算法自身的收敛性,采用了保留精英的策略,另外考虑到遗传算法参数的选择问题采用了自适应策略。同时,结合自适应加权广义形态滤波的思想构建基于遗传优化的多结构多尺度自适应加权形态滤波器,滤波效果比传统的形态滤波、广义形态滤波及在此基础上改进的滤波方法效果均好。     

On the convergence rate performance of the normalizedleast-mean-square adaptation

This paper compares the convergence rate performance of the normalized least-mean-square (NLMS) algorithm to that of the standard least-mean-square (LMS) algorithm, which is based on a well-known interpretation of the NLMS algorithm as a form of the LMS via input normalization. With this interpretation, the analysis is considerably simplified and the difference in rate of parameter convergence can be compared directly by evaluating both the condition number of the normalized and unnormalized input correlation matrix. This paper derives the condition number expressions for the normalized input correlation matrix of which the arbitrary-length filter model is linear with respect to its adaptable parameters and contain only two distinct unnormalized eigenvalues. These expressions, which require that the input samples be statistically stationary and zero-mean Gaussian distributed, provide an important insight into the relative convergence performance of the NLMS algorithm to that of the LMS as a function of filter length. This paper also provides a conjecture which set bounds on the NLMS condition number for any arbitrary number of distinct unnormalized eigenvalues.     

A globally convergent trust region multidimensional filter SQP algorithm for nonlinear programming

We propose a trust region multidimensional filter SQP algorithm. The multidimensional filter technique proposed by Gould et al. [SIAM J. Optim., 15 (2005), pp. 17–38] is extended to solve constrained optimization problems. The constraints are partitioned into p parts. The entry of our filter consists of these different parts. Not only the criteria for accepting a trial step would be relaxed, but also the individual behaviour of each part of the constraints is considered. The filter's entries and the acceptance criteria are different from other filter-related algorithms in the literature. It should be noted that the undesirable link between the objective function and the constraint violation function in the filter acceptance criteria disappears. Our algorithm is also combined with the non-monotone technique for accepting a trial step, which leads to a more flexible acceptance criteria. Under mild conditions, global convergence is proved. Numerical results show the robustness and efficiency of our algorithm.     

改进蚁群算法在路径规划中的应用

针对传统蚁群算法在机器人路径规划时存在收敛速度慢、易陷入局部最优等问题,提出了一种基于自适应归档更新的蚁群算法。根据路径性能指标建立多目标性能评估模型,对最优路径进行多指标优化;采用路径方案归档更新策略进行路径方案的更新和筛选,提高算法的收敛速度;当搜索路径进入不可行区域时,采用自适应路径补偿策略转移不可行路径节点,构造可行路径,减少死锁蚂蚁数量;若算法无法避开障碍或者进入停滞状态,则进行种群重新初始化,增加物种多样性,避免算法陷入局部最优。仿真实验表明,改进后的算法收敛速度更快、收敛精度更高、稳定性更好。