����Bayesian����BP�������GPS�߳�ת��

?????BP?????????????????????????????????????Bayesian?????????BP??????????????????????????GPS??????????????????L-M????????????????????????Bayesian??????BP????????????????????????????????????     

BP�����������Ǽ�������Զ��ָ����ģ�е�Ӧ��

????????????????????????????????????BP???????????????????人???????????????????????????????????BP??????????????????г????????????????????????о??????????BP????????????????????????????????????????????????     

С���������ճ�Ԥ���㷨�о�

???BP??????????????С???????????BP?????缤????????BP????????????????????????????????С???????磬??Ч?????????????????????????????????????????????????????????Ч????     

�������ɷַ�����BP���������α�Ԥ���е�Ӧ��

?????α????????????????????????????BP???????????????α??????????????????????????????????????????????????????????????????????????????????BP??????????????????Ч????     

���ڱ�׼������BP�������GPS�߳�ת��

???????BP???????????????????????????????????????????????????????????????????б?????????????????????Ч???????÷???????????????????????????????????????????緽????????????ж??ó????????????BP???????????????????????????     

����Kalman�˲���BP�����緽���ڴ�ӱ���Ԥ���е�Ӧ��

??????????????BP?????緽??????????ι???????????????????????????????BP????????,???????ж?????????С????????????????С????????,???????????????????????????,?÷??????к??????????????????????     

基于BP神经网络的企业效绩综合评价方法

为了建立企业效绩综合评价模型,在由基本的财务指标构成的财务分析体系基础上,将BP神经网络应用于效绩综合评价中.以企业历史数据作为BP神经网络的训练样本,用训练好的BP网络评价企业当期各方面的效绩.试验结果表明将BP网络应用于企业评价中,能达到令人满意的评价效果.     

����ƽ����Ϻ� BP��������Ϸ���GPS�߳�ת��

???GPS???????????????????????????y?????????BP?????編??????????????????????6????????????????????????????????????6?????????????????????????????BP?????編???????????3????????????????????     

BP�������ά����ģ���ڱ���ʱ��Ԥ���е�Ӧ��

???BP???????????????????????????????????????????????????????G??P??????????????????????????????????????D??h?????????????????m????????????????????????????????????????????????????BP????????????????????????????????????????????е??????????????????????????????????????????????????????????????????????t?????????????     

Late Quaternary sedimentary environmental evolution offshore of the Hangzhou Bay,East China——implications for sea level change and formation of Changjiang alongshore current

This study focuses on sedimentary environmental changes offshore of Hangzhou Bay,East China,since the Late Quaternary. AMS 1 4 C ages from core CJK10,lithologies,distribution of foraminifera,heavy minerals,and S and Cl elements show a f luvial terrace environment during ~23.2–11.0 cal ka BP; a littoral to tidal-f lat environment during 11.0–10.2 cal ka BP; and a shallow marine environment with a relatively low sedimentation rate(0.1–0.22 cm/a) since 4.3 cal ka BP. High depositional rates(~1.6 cm/a) from 10.9 to 10.2 cal ka BP resulted from suff icient accommodation space created by rapid sea level rise from-44 m to-33 m,from high sediment delivery by local rivers,and effective trapping of sediments by tidal-f lat vegetation. The rate of sea level rise was variable; relatively high from 10.9 to 10.6 cal ka BP(2.1 cm/a),and lower since 10.6 cal ka BP(1.2 cm/a). The Changjiang alongshore current crossed the Hangzhou Bay to form the mud wedge on the inner shelf of the East China Sea later than 9.4 cal ka BP. The CJK10 site was a tide-dominated shelf environment and experienced erosion from approximately 9.4–9.2 cal ka BP to 4.3 cal ka BP. The depositional hiatus was caused by the Changjiang alongshore current,which was relatively weak during 9.4–7.5 cal ka BP and increased in strength during ~7.5–4 cal ka BP. From ~4.3 cal ka BP,a large amount of sediment from the Changjiang River was partly deposited on the continental shelf of Hangzhou Bay with some transported southward. Therefore,this study clarif ies the history of Changjiang-derived sediment dispersal and deposition,although a detailed record of the changes in the Changjiang alongshore current since 4.3 cal ka BP is diff icult to obtain because of the scarcity of evidence.     

��������ӦUnscented Kalman�������˲�

???????????????????????????????UKF?????????????÷??????????????????Unscented Kalman????????????????????????????????????Э????????????????????Kalman?????????????Kalman?????Unscented Kalman??????????????????????UKF?????????????????????????????     

一种适用于PPP动力学模型异常的自适应Kalman滤波

为了削弱PPP参数估计中动力学模型异常对Kalman滤波解的影响,针对PPP状态向量中各类参数不符值对动力学模型异常描述特性的不同,以位置相关为条件对参数进行分类,构建分类因子自适应Kalman滤波用于PPP参数估计。选取6个IGS站点3 d的数据,使用标准Kalman滤波与构建的自适应Kalman滤波进行PPP解算分析。结果表明,相较于标准Kalman滤波,自适应Kalman滤波能通过自适应因子调节状态预测协方差,加速PPP收敛。静态模式下,平均收敛时间从28.2 min缩减到19.4 min,N、E、U方向的平均精度为1.50 cm、3.34 cm、5.55 cm,分别提高7%、14%、19%;动态模式下,构建的自适应Kalman滤波解N、E、U方向偏差的RMS值为2.7 cm、3.6 cm、6.3 cm,较标准Kalman滤波分别提高13%、28%、43%。     

����Ӧ������չKalman�����˲���������ϵ����е�Ӧ��

???????????????????????????о????????????????????????????????????????Kalman??????????????÷??????????????????Kalman????????????????????????????????????????и????????????3???????????????顢???Kalman??????????????????????????????????????????Kalman??????????????????????????????     

基于在线回声状态网络的变形数据预测分析

结合Kalman滤波与回声状态网络,将在线回声状态网络算法应用于变形数据预测。回声状态网络的输出权值通过Kalman滤波训练,直接对网络的输出权值进行在线更新,克服了传统递归网络需要收集大量样本后才能进行拟合预测的缺陷,同时也保证了预测精度。实例计算验证了该方法的有效性。     

基于卡方检验的抗差自适应Kalman滤波在变形监测中的应用

在抗差Kalman滤波的基础上引入双自适应因子,分别对动态模型不准确和观测模型存在粗差进行调节,构建双自适应因子滤波模型。针对抗差自适应Kalman滤波效率较低的缺点,通过构建基于卡方检验的抗差自适应Kalman滤波,先用卡方检验对粗差进行检验,再调用抗差自适应Kalman滤波进行处理。工程实例表明,双自适应因子滤波模型可以很好地抵御粗差,并减弱模型不精确的影响。基于卡方检验的抗差自适应Kalman滤波不仅可以削弱粗差对滤波估值的影响,而且可以提高数据处理的效率。     

��������ӦKalman�˲��㷨��GPS/INS��ϵ����е�Ӧ��

???????????????GPS/INS??????????????????仯??????????в????????????????????????????????????????????????????е??????????????????????????????????????????????????硣?????????з???GPS/INS?????????????????о???????????????????????????Ч???     

抗差自适应卡尔曼滤波模型及其在塌陷区监测中的应用

针对塌陷区等地表快速沉降区域的动力学特点及观测向量中存在的粗差对卡尔曼滤波结果的影响，设计一种抗差自适应卡尔曼滤波模型。该模型能识别稳定沉降与快速沉降2种状态，通过抗差估计减小观测向量中粗差的影响，并采用自适应因子调整动力学模型，减少状态模型的误差，提高滤波结果的精度。将该模型应用于某矿区沉降监测数据的处理，结果表明，其效果优于抗差卡尔曼滤波。     

桥梁变形监测数据小波去噪与Kalman滤波研究

针对跨海大桥三维变形监测问题,实施了测量机器人与三维激光扫描仪测量方案,二者三维监测数据最大较差不超过0.5 mm,而且测量机器人监测数据精度高,说明三维激光扫描技术应用于桥梁变形监测是可行的。根据小波分析与Kalman滤波理论,建立小波去噪后的桥梁变形监测Kalman滤波模型。实验表明,经过小波去噪的监测数据再进行Kalman滤波处理,提高了桥梁变形预测与变形分析的可靠性。通过标准Kalman滤波、自适应Kalman滤波桥梁变形观测数据处理,及平方和误差、均方误差、平均相对误差精度分析知,自适应Kalman滤波要优于标准Kalman滤波方法。     

城市复杂环境下GNSS/INS组合导航算法研究

针对城市环境下GNSS车辆导航存在卫星信号易受影响的问题,利用GNSS/INS组合算法提高复杂环境下城市车辆定位性能。基于城市环境下实测GNSS数据评估分析定位结果,使用GNSS/INS组合的常规卡尔曼滤波算法实现卫星失锁区域导航。同时,提出一种基于新息的自适应卡尔曼滤波算法,可有效增强卫星数较少及信号干扰严重区域的车辆导航定位能力。该方法利用量测与预测的关系构造自适应因子,改善定位精度。结果表明,常规卡尔曼滤波可在20 s卫星信号失锁情况下保证亚m级导航精度,自适应卡尔曼滤波算法在卫星信号受到严重干扰时,其定位精度相比于常规卡尔曼滤波算法提高30%,可满足在城市复杂环境下的高精度、高可靠性车辆导航定位服务需求。