基于果蝇算法优化广义回归神经网络的矢量水听器的DOA估计

为了提高矢量水听器阵列对窄带信号的DOA估计精度,运用果蝇算法优化广义回归神经网络,通过对阵列协方差矩阵实值化,并提取信号子空间的基作为样本特征进行网络训练,构建了果蝇算法优化下的广义回归神经网络,实现了基于矢量水听器阵列的水下声源的DOA估计.仿真实验结果表明,方法泛化性能较好,能解决输入维数过大的问题,且运行时间短,DOA估计精度高,具有较强的工程应用价值.     

基于双混沌优化搜索的改进粒子群算法及应用

针对现有算法在智能电阻箱动态误差校正方面存在的收敛速度慢、计算精度低,且易进入“局部最优”的陷阱等缺点,展开对智能电阻箱动态示数校正过程的重构及设计,并对动态误差校正优化算法进行研究.在双混沌优化系统中添加扰动因子与指数自适应学习方式改进搜索策略;在粒子群算法中将惯性权重因子修正为自适应权重因子,将学习因子修正为异步线性学习因子以优化算法,进而提出一种改进的粒子群优化算法(AL-DCPSO).利用8个经典函数对算法性能进行测试后,将这种算法应用在某型号智能电阻箱动态误差校正的过程中,研究结果表明:改进后的算法具有更高的计算精度(达到0.001)与更强的寻优能力,且在优化过程中呈现出较强的自适应学习能力,计算过程较为稳定,鲁棒性有效提升,耗时在阈值范围内有所增加.其创新性在于将双混沌优化机制的优点与粒子群算法相结合,应用到智能电阻箱动态误差校正的过程中,对动态误差校正方法进行了一定拓展,为粒子群优化算法在具体实际优化过程中的关键参数选取与策略设计,有效提升算法优化性能提供了一些借鉴.     

基于向量值有理插值的最优预测算法研究及陶瓷需求量预测的应用

通过对高维数据整体表达式建模预测方法和分区间等预测算法的缺陷分析,提出基于向量值有理插值的最优预测算法,通过有理向量插值函数和各分量的误差限得到向量之间的相似性,克服了其它很多算法利用向量的整体表达式方法而产生预测的偏差;另外,通过向量的误差限与训练样本所得向量值有理插值函数及迭代仿真方法来确定预测样本向量所对应的最优预测值.通过实例,算法所得预测值的精度比其他算法更高,并且分析了误差限和迭代步长对算法性能的影响.     

基于EWT-PSO-SVM误差校正组合模型的中国股市预测研究

鉴于股市预测的复杂性.遵循"先分解后集成"的总体建模思路.文章基于EWT分解算法和SVM支持向量机模型.同时结合PSO粒子群优化算法和误差校正组合预测方法,构建了一种中国股票市场建模及预测的EWT-PSO-SVM误差校正组合预测模型.先基于EWT算法将原始价格序列分解成若干分量,再根据频率将其重组成高、中、低频3个分量,对它们分别建立PSO-SVM误差校正组合模型.最后集成各个分量的预测结果.与其他预测模型相比较,文章所构建预测模型的MSE、MAE、MAPE、RMSE、Theil不等系数、确定性系数DC和方向性指标DS 7个指标均优于其他基准预测模型,MCS检验结果同样表明本文构建模型的预测性能最优.稳健性检验结果进一步证实了文章构建的模型预测性能所具备的稳健性.     

高斯核正则化学习算法的泛化误差

对广义凸损失函数和变高斯核情形下正则化学习算法的泛化性能展开研究.其目标是给出学习算法泛化误差的一个较为满意上界.泛化误差可以利用正则误差和样本误差来测定.基于高斯核的特性,通过构构建一个径向基函数(简记为RBF)神经网络,给出了正则误差的上界估计,通过投影算子和再生高斯核希尔伯特空间的覆盖数给出样本误差的上界估计.所获结果表明,通过适当选取参数σ和λ,可以提高学习算法的泛化性能.     

自适应隐式有限差分方法求解美式期权定价问题

本文针对美式期权的定价问题设计了基于有限差分方法的预估-校正数值算法.该算法采用显式离散格式先对自由边界条件进行预估,再对经过变量替换后的关于期权价格的偏微分方程采用隐式格式离散,并用Fourier方法分析了此离散格式的稳定性.接下来,引入基于Richardson外推法的后验误差指示子.这个后验误差指示子能够在给定的误差阈值范围内,针对期权价格和自由边界找到合适的网格划分.最后,通过设计多组数值实验并与Fazio^[1]采用显式离散格式算得的数值结果相比较,验证了所提算法的有效性,稳定性和收敛性.     

基于方根分解形式的带衰减因子UKF算法应用

UKF作为一种新的非线性滤波方法已在目标跟踪问题中得到应用,在状态的时间更新阶段直接使用非线性模型,不引入线性化误差,而且不必计算Jacobians矩阵.提出了一种基于方根分解形式的带有衰减因子的UKF算法(SRDMA-UKF),算法的方根形式增加了数字稳定性和状态协方差的半正定性.通过衰减因子的引入加强对当前测量数据的利用,减小历史数据对滤波的影响.仿真实验结果表明,该算法与UKF算法相比具有更好的滤波性能.     

求解病态线性方程组的误差转移法和增广方程组法的机理及算法改进

为获得病态线性方程组的高精度解,建立了一种优化模型,其最优解等价于早先提出的误差转移法和增广方程组法;指出后两者的本质机理是通过极小化解的模来近似极小化解的误差.为使算法适用于数据有污染的情况,进行了正则化改造.证明了新算法理论上与Tikhonov正则化等价.但当正则化参数趋于0时,目标函数的不同使得两者性能迥异,新算法可直接用于数据无污染的情况,而后者仍需选取合适的正则参数.数值算例验证了算法的有效性.     

柯特斯校正公式及其误差估计

提出了一类计算定积分的高精度柯特斯校正公式,通过两种方法进行了推导,给出了它的复化公式及其加速公式,并得到了它们的误差估计和收敛阶.数值实验验证了复化柯特斯校正公式及其加速公式的高效性.     

基于无线通信基站的室内三维定位问题

三维定位问题是现代商用通信网络中对于定位系统存在的一个真正具有技术难度的挑战.根据视距传播环境和非视距传播环境的到达时间的数据集,建立线性误差模型;对于无真实位置的竞赛数据集,定义竞赛数据定位误差评估模型;基于不同的空间场景,提出基于空间单元的定位算法;面对高度误差明显高于平面误差的问题,设计基于高斯加权的误差补偿模型;针对最优定位精度最少基站问题,提出基于贪心策略的基站选择算法;考虑轨迹连续性,设计轨迹准确性验证的10-fold交叉验证方法;基于测量距离有限的真实环境,分析平均"连接度数"与定位精度的关系.实验结果表明,提出的定位算法在有效基站数大于等于5时,能获得较好的定位精度.     

Numerically stable and efficient algorithm for vector channel parameter and frequency offset estimation

In this paper, we present a novel and numerically efficient algorithm for vector channel and calibration vector estimation, which works when frequency offset error caused by either unstable oscillator or Doppler effect is present in Spread Spectrum antenna system. We propose an estimation algorithm based on Gauss–Seidal algorithm rather than using eigen-decomposition or SVD in computing eigenvalues and eigenvectors at each iteration. The algorithm is based on the two-step procedures, one for estimating both channel and frequency offset and the other for estimating the unknown array gain and phase. Consequently, estimates of the DOAs, the multi-path impulse response of the reference signal source, and the carrier frequency offset as well as the calibration of antenna array are provided. The analytic performance improvement in multiplications number is presented. The performance of the proposed algorithm is investigated by means of computer simulations. Throughout the analytic and computer simulation, we show that the proposed algorithm reduces the number of multiplications by order of one.     

A constrained optimization based method for acoustic finite element model updating of cavities using pressure response

A method based on constrained optimization for updating of an acoustic finite element model using pressure response is proposed in this paper. The constrained optimization problem is solved using sequential quadratic programming algorithm. Updating parameters related to the properties of the sound absorbers and the measurement errors are considered. Effectiveness of the method is demonstrated by numerical studies on a 2D rectangular cavity and a car cavity. It is shown that the constrained formulation, that includes lower and upper bounds on the updating parameters in the form of inequality constraints, is important for obtaining a correct updated model. It is seen that the proposed updating method is not only able to effectively update the model to obtain a close match between the finite element model pressure response and the reference pressure response, but is also able to identify the correction factors to the parameters in error with reasonable accuracy.     

基于噪声误差模型的不规则区域室内定位和锚点分布优化

结合实际工业背景,研究了一类在不规则区域且误差不服从高斯分布的室内无线定位问题.给出了噪声误差模型, 在对多个传统定位算法进行性能分析的基础上,研究了待定位区域内锚点阵列的分布, 改进了多锚点阵列下的定位方法,并提出基于Delaunay三角剖分锚点分布优化模型和求解方法.     

State of charge estimation for lithium-ion batteries based on adaptive dual Kalman filter

Accurate estimation of the battery state of charge (SOC) is of great significance for enhancing its service life and safety. In this study, based on the fractional-order equivalent circuit model of lithium-ion battery, the SOC estimation methods using dual Kalman filter (DKF) and dual extended Kalman filter (DEKF) are simulated and compared, in terms of model accuracy and SOC estimation accuracy. Then, combining the advantages of the DKF and DEKF algorithms, an SOC estimation algorithm based on adaptive double Kalman filter is proposed. This algorithm uses the recursive least squares (RLS) method to update the battery model parameters online in real time, and employs the DKF algorithm to filter the SOC twice to reduce the interferences from the battery model error and the current measurement error. In the experimental studies, the measured SOC values are compared with the estimated SOC values produced by the proposed algorithm. The comparison results show that SOC estimation error of the proposed algorithm is within the range of ±0.01 under most test conditions, and it can automatically correct SOC to true value in the presence of system errors. Thus, the validity, accuracy, robustness and adaptability of the proposed algorithm under different operation conditions are verified.     

Minimizing loss probability bounds for portfolio selection

In this paper, we derive a portfolio optimization model by minimizing upper and lower bounds of loss probability. These bounds are obtained under a nonparametric assumption of underlying return distribution by modifying the so-called generalization error bounds for the support vector machine, which has been developed in the field of statistical learning. Based on the bounds, two fractional programs are derived for constructing portfolios, where the numerator of the ratio in the objective includes the value-at-risk (VaR) or conditional value-at-risk (CVaR) while the denominator is any norm of portfolio vector. Depending on the parameter values in the model, the derived formulations can result in a nonconvex constrained optimization, and an algorithm for dealing with such a case is proposed. Some computational experiments are conducted on real stock market data, demonstrating that the CVaR-based fractional programming model outperforms the empirical probability minimization.     

Computing efficient frontiers using estimated parameters

The mean-variance model for portfolio selection requires estimates of many parameters. This paper investigates the effect of errors in parameter estimates on the results of mean-variance analysis. Using a small amount of historical data to estimate parameters exposes the model to estimation errors. However, using a long time horizon to estimate parametes increasers the possibility of nonstationarity in the parameters. This paper investigates the tradeoff between estimation error and stationarity. A simulation study shows that the effects of estimation error can be surprisingly large. The magnitude of the errors increase with the number of securities in the analysis. Due to the error maximization property of mean-variance analysis, estimates of portfolio performance are optimistically biased predictors of actual portfolio performance. It is important for users of mean-variance analysis to recognize and correct for this phenomenon in order to develop more realistic expectations of the future performance of a portfolio. This paper suggests a method for adjusting for the bias. A statistical test is proposed to check for nonstationarity in historical data.     

��ع���Ƶ�������������

In this paper, a vector parameter method for ridge regression is proposed. We choose the negative gradient of mean square error as vector direction and decide vector norm with the expectation constrains both of mean square error and of residual error. We come to conclusions that the mean square error is a decreasing function of vector norm while the residual error a increasing one. It is the monotonicity of the errors that leads to our expectation constrains. Since two conflict constrains are under consideration, our vector parameter ridge regression is expected to bear both satisfactory mean square error and acceptable residual error. Finally, a multi-collinearity model is given as an example.     

Importance Sampling Simulations of Markovian Reliability Systems Using Cross-Entropy

This paper reports simulation experiments, applying the cross entropy method such as the importance sampling algorithm for efficient estimation of rare event probabilities in Markovian reliability systems. The method is compared to various failure biasing schemes that have been proved to give estimators with bounded relative errors. The results from the experiments indicate a considerable improvement of the performance of the importance sampling estimators, where performance is measured by the relative error of the estimate, by the relative error of the estimator, and by the gain of the importance sampling simulation to the normal simulation.     

An Adaptive Newton Algorithm for Optimal Control Problems with Application to Optimal Electrode Design

In this work, we present an adaptive Newton-type method to solve nonlinear constrained optimization problems, in which the constraint is a system of partial differential equations discretized by the finite element method. The adaptive strategy is based on a goal-oriented a posteriori error estimation for the discretization and for the iteration error. The iteration error stems from an inexact solution of the nonlinear system of first-order optimality conditions by the Newton-type method. This strategy allows one to balance the two errors and to derive effective stopping criteria for the Newton iterations. The algorithm proceeds with the search of the optimal point on coarse grids, which are refined only if the discretization error becomes dominant. Using computable error indicators, the mesh is refined locally leading to a highly efficient solution process. The performance of the algorithm is shown with several examples and in particular with an application in the neurosciences: the optimal electrode design for the study of neuronal networks.