IUKF neural network modeling for FOG temperature drift

A novel neural network based on iterated unscented Kalman filter （IUKF） algorithm is established to model and com- pensate for the fiber optic gyro （FOG） bias drift caused by temperature. In the network, FOG temperature and its gradient are set as input and the FOG bias drift is set as the expected output. A 2-5-1 network trained with IUKF algorithm is established. The IUKF algorithm is developed on the basis of the unscented Kalman filter （UKF）. The weight and bias vectors of the hidden layer are set as the state of the UKF and its process and measurement equations are deduced according to the network architecture. To solve the unavoidable estimation deviation of the mean and covariance of the states in the UKF algorithm, iterative computation is introduced into the UKF after the measurement update. While the measure- ment noise R is extended into the state vectors before iteration in order to meet the statistic orthogonality of estimate and mea- surement noise. The IUKF algorithm can provide the optimized estimation for the neural network because of its state expansion and iteration. Temperature rise （-20-20℃） and drop （70-20℃） tests for FOG are carried out in an attemperator. The temperature drift model is built with neural network, and it is trained respectively with BP, UKF and IUKF algorithms. The results prove that the proposed model has higher precision compared with the back- propagation （BP） and UKF network models.     

Application of neural networks for permanent magnet synchronous motor direct torque control

Neural networks require a lot of training to understand the model of a plant or a process. Issues such as learning speed, stability, and weight convergence remain as areas of research and comparison of many training algorithms. The application of neural networks to control interior permanent magnet synchronous motor using direct torque control （DTC） is discussed. A neural network is used to emulate the state selector of the DTC. The neural networks used are the back-propagation and radial basis function. To reduce the training patterns and increase the execution speed of the training process, the inputs of switching table are converted to digital signals, i.e., one bit represent the flux error, one bit the torque error, and three bits the region of stator flux. Computer simulations of the motor and neural-network system using the two approaches are presented and compared. Discussions about the back-propagation and radial basis function as the most promising training techniques are presented, giving its advantages and disadvantages. The system using back-propagation and radial basis function networks controller has quick parallel speed and high torque response.     

Novel approach for identifying Z-axis drift of RLG based on GA-SVR model

This paper describes a novel approach for identifying the Z-axis drift of the ring laser gyroscope （RLG） based on ge-netic algorithm （GA） and support vector regression （SVR） in the single-axis rotation inertial navigation system （SRINS）. GA is used for selecting the optimal parameters of SVR. The latitude error and the temperature variation during the identification stage are adopted as inputs of GA-SVR. The navigation results show that the proposed GA-SVR model can reach an identification accuracy of 0.000 2 （？）/h for the Z-axis drift of RLG. Compared with the ra-dial basis function-neural network （RBF-NN） model, the GA-SVR model is more effective in identification of the Z-axis drift of RLG.     

Building a Tax Predictive Model Based on the Cloud Neural Network

Tax is very important to the whole country, so a scientific tax predictive model is needed. This paper introduces the theory of the cloud model. On this basis, it presents a cloud neural network, and analyzes the main factors which influence the tax revenue. Then if proposes a tax predictive model based on the cloud neural network. The model combines the strongpoints of the cloud model and the neural network. The experiment and simulation results show the ef-     

电力关键基础设施网络仿真模型研究

电力网络模型是关键基础设施模型中的重要组成部分。分析了现实电力网络的拓扑结构和运行原理,根据复杂网络的基本理论,采用基于Agent的建模方法设计实现了关键基础设施中电力网络模型。通过仿真实验,较好的模拟了电力系统输送电能的动态过程和因偶发故障而造成的大规模停电现象,对复杂电力网络的动力学特性进行了初步探索。

Abstract：

The model of electric power network is a major part of critical infrastructure model. The topology of network structure and the principle of net function were analyzed, according to the theory of complex network, applying the agent-based modeling to design and realize the model of electric power network. Through the simulation experiment, it was demonstrated to be better ways to simulate the dynamic progress of electric transmission and the cascading failure phenomenon caused by incidental trouble. The simulation model offers an effective approach to analysis electric power network’s role in critical infrastructure, and makes a step to exploration for complex electric power network’s dynamic character.     

Robust adaptive control of hypersonic vehicle considering inlet unstart

In this paper,a model reference adaptive control(MRAC)augmentation method of a linear controller is proposed for air-breathing hypersonic vehicle(AHV)during inlet unstart.With the development of hypersonic flight technology,hypersonic vehicles have been gradually moving to the stage of weaponization.During the maneuvers,changes of attitude,Mach number and the back pressure can cause the inlet unstart phenomenon of scramjet.Inlet unstart causes significant changes in the aerodynamics of AHV,which may lead to deterioration of the tracking performance or instability of the control system.Therefore,we firstly establish the model of hypersonic vehicle considering inlet unstart,in which the changes of aerodynamics caused by inlet unstart is described as nonlinear uncertainty.Then,an MRAC augmentation method of a linear controller is proposed and the radial basis function(RBF)neural network is used to schedule the adaptive parameters of MRAC.Furthermore,the Lyapunov function is constructed to prove the stability of the proposed method.Finally,numerical simulations show that compared with the linear control method,the proposed method can stabilize the attitude of the hypersonic vehicle more quickly after the inlet unstart,which provides favorable conditions for inlet restart,thus verifying the effectiveness of the augmentation method proposed in the paper.     

Trajectory linearization control of an aerospace vehicle based on RBF neural network

An enhanced trajectory linearization control （TLC） structure based on radial basis function neural network （RBFNN） and its application on an aerospace vehicle （ASV） flight control system are presensted. The influence of unknown disturbances and uncertainties is reduced by RBFNN thanks to its approaching ability, and a robustifying itera is used to overcome the approximate error of RBFNN. The parameters adaptive adjusting laws are designed on the Lyapunov theory. The uniform ultimate boundedness of all signals of the composite closed-loop system is proved based on Lyapunov theory. Finally, the flight control system of an ASV is designed based on the proposed method. Simulation results demonstrate the effectiveness and robustness of the designed approach.     

Personal Credit Risk .Scoring Model Based on Rough Set and Neural Network

In order to improve the precision of personal credit risk assessment, applying rough set and neural network to the credit risk scoring prediction problem in an attempt to suggest a new model with better classification accuracy. To evaluate the prediction accuracy of the model, we compare its performance with those of SVM, linear discriminate analysis, logistic regression analysis, K-nearest neighbors, classification and regression tree, neural network and PCA-NN. The experimental results show the model have a very good prediction accuracy     

Reliability analysis of torpedo loading based on fractional-order optimization model

A fractional-order cumulative optimization GM(1,2)model based on grey theory is proposed to study the relationship between torpedo loading and working reliabilities. In this model, the average relative error function related to order and background value is established. Taking the average relative error function as the objective function, the optimal value of the two parameters is obtained through the optimization method,and the minimum value of the average relative error is determined. The calc...     

NETWORK ANALYSIS OF TERRORIST ACTIVITIES

This paper uses an extensive network approach to "East Turkistan" activities by building both the one-mode and the bipartite networks for these activities.In the one-mode network,centrality analysis and spectrum analysis are used to describe the importance of each vertex.On this basis,two types of core vertices——The center of communities and the intermediary vertices among communities— are distinguished.The weighted extreme optimization(WEO) algorithm is also applied to detect communities in the one-mode network.In the "terrorist-terrorist organization" bipartite network,the authors adopt centrality analysis as well as clustering analysis based on the original bipartite network in order to calculate the importance of each vertex,and apply the edge clustering coefficient algorithm to detect the communities.The comparative and empirical analysis indicates that this research has been proved to be an effective way to identify the core members,key organizations,and communities in the network of "East Turkistan" terrorist activity.The results can provide a scientific basis for the analysis of "East Turkistan" terrorist activity,and thus provide decision support for the real work of "anti-terrorism".     

一种MAV航姿估计算法及其半实物仿真

配置MIMU(Micro Inertial Measurement Unit,微惯性测量单元)中的加速度计工作在倾角仪状态,利用当地的重力加速度计算MAV(Micro Air Vehicles,微小型飞行器)的姿态角。同时利用MIMU中的陀螺仪,计算载体的姿态角。提出了一种构造加权系数的方法,可以根据MIMU的特性,构造不同性能的加权系数。通过对姿态角进行加权平均,实现惯性数据的融合,对MAV的姿态进行估计。该方法既保证了飞行器稳定飞行时姿态估计的精度,避免了姿态误差随时间的积累;又保证了姿态估计系统的动态性能,减小了系统的动态误差。基于该方法搭建的微小型AHRS(Attitude and Heading Reference System,姿态航向参考系统)体积小、重量轻、精度高,特别适用于载荷与体积都有限的载体使用。     

基于CPSO-LSSVM的单轴旋转惯导系统轴向陀螺漂移辨识

在单轴旋转惯导系统中,轴向陀螺漂移是影响系统导航精度的重要因素。为了提高惯导系统的导航精度,采用混沌粒子群算法(chaos particle swarm optimization, CPSO）优化的最小二乘支持向量机（least squares support vector machine, LSSVM）〖JP+1〗对轴向激光陀螺漂移进行辨识。利用初始对准12 h内系统纬度误差和温度变化量作为LSSVM模型的训练数据,利用CPSO对LSSVM进行参数优化,利用优化后的LSSVM模型对轴向陀螺漂移进行辨识,轴向陀螺漂移辨识精度优于0.000 2 （°）/h, 系统定位误差优于1 nm/72 h。试验结果表明,CPSO是选取LSSVM参数的有效方法,该方法能够有效地辨识轴向陀螺漂移,具有很高的辨识精度,具有很高的实际应用价值。     

数字罗盘辅助实现MIMU静基座初始对准

针对低成本MIMU中MEMS陀螺精度低,不能独立完成初始方位对准的问题,提出了基于数字罗盘辅助实现MIMU粗对准,再利用UKF方法实现精对准的低成本MIMU初始对准方法。介绍了数字罗盘的基本原理、分析了航向误差的产生原因,给出了基于最小二乘法的误差补偿方案;设计了基于数字罗盘辅助MIMU的粗对准和基于UKF的非线性精对准方案,分析了粗对准精度,建立了包含航向角误差的非线性对准模型;MIMU初始对准仿真结果表明该方法具有满意的对准性能。     

MEMS陀螺误差建模与滤波方法

从实际工程应用角度出发,探讨了微机电系统(micro-electro-mechanical systems,MEMS)陀螺误差的有效滤波降噪方法.基于随机序列时序分析法的基本原理,采用实时平均算法对陀螺原始量测数据进行常值补偿预处理,得到随机漂移信号.对去除渐进项后的差分漂移信号进行AR模型建模,并依据该模型进行改进卡尔曼滤波,在输出差分信号滤渡值的同时解算当前陀螺输出滤波值.通过对某MEMS陀螺实测数据的误差补偿结果表明,提出的滤波方法能够有效地抑制其漂移误差,提高实际应用中的测量精度.     

一种新型的灰色RBF神经网络建模方法及其应用

针对神经网络建模预测时,其建模精度往往受到数据随机性的影响,以及灰色累加生成操作(AGO)具有减小数据随机性,使数据变得有规则的特点,提出了一种新型的建模预测模型———灰色径向基(RBF)神经网络模型。此模型能够减小数据中的随机性,加快网络的建模收敛速度,使神经网络的建模精度得以提高。将此灰色RBF神经网络应用到动调陀螺仪漂移数据建模中,并将其建模验证结果和单纯使用RBF网络的建模结果进行比较,结果证明此方法是可行而有效的。     

陀螺仪漂移测试与补偿系统的研制

从硬件、软件两方面简要介绍了所研制的“陀螺仪漂移测试与补偿系统”的组成结构、控制功能及工作过程。应用多变量过程控制系统解耦理论，采用对角矩阵法选择了解耦网络，设计了一套陀螺漂移力矩反馈法测试与数据自动采集系统；提出了一种对其随机漂移进行自适应控制补偿的方法，设计了具体的控制回路对其随机漂移进行实时补偿。     

基于Elman网络结构的惯导平台漂移模型辨识方法

针对惯性导航平台漂移误差高阶非线性动态系统的特点,利用神经网络的任意逼近能力和自适应抽取系统动态信息的能力,提出基于Elman网络结构的惯性导航平台漂移模型辨识方案。首先建立惯性导航平台漂移误差模型,并选择了用于网络辩识的输入、输出量。采用动量及可变学习速率算法加速网络的收敛;在该算法的基础上,针对网络隐层,提出的扩展非线性节点函数能更好地改善网络学习效率,满足系统辨识实时性和精确性的需要。通过测得的惯性导航平台漂移误差数据对网络进行训练,获得了较为满意的辨识结果。     

基于RBF-HMM模型的时间序列在线预测

针对非线性非高斯时间序列,提出观测噪声服从隐马尔可夫模型(HMM)的径向基函数(RBF)神经网络预测模型—RBF-HMM模型,该模型具有如下两个特点:(1)用隐节点数可变的RBF神经网络对时间序列进行非线性建模;(2)用HMM对非高斯噪声进行建模.并采用序列蒙特卡罗(SMC)方法实现RBF-HMM模型参数的动态调整和时间序列的在线预测.最后采用南京禄口国际机场日旅客吞吐量数据进行实证研究,结果表明该模型的有效性.     

基于RBF神经网络与Markov组合的飞行风险预测研究

为了更加准确地对飞机飞行安全性做出评估,在径向基(radial basis function,RBF)神经网络的基础上,通过引入马尔可夫链的方式进行误差修正,建立了一种拟合程度高且无需反复调整权值的新型评估模型.并以国内某航空公司A320机型近20年来发生的飞行机械故障为基础数据对模型进行训练、拟合、修正.将修正结果与单一RBF神经网络评估方法相比较,分析二者差异后得出误差降低的结论.为管控飞机飞行风险提供了方法拓展.     

一种基于离散粒子群的自适应径向基网络模型

提出了一种基于离散粒子群的自适应RBF网络模型。即融合两个二进制编码粒子群,通过对最近邻聚类中心选择法的改进,RBF网络模型能自适应地优化隐节点数、中心向量与宽度,且在保证网络性能的前提下,使网络的结构相对简单(较少的隐层节点数)。同时为进一步提高网络性能,采用梯度下降法与递推最小二乘法混合学习策略,分别对基函数参数(中心与宽度)和输出层线性权值进行学习。仿真实验证明了该方法模型的有效性。