Improving Measurement Reliability with Biased Estimation for Multi-sensor Data Fusion
-
摘要: 为了提高测量数据可靠性,多传感器数据融合在过程控制领域得到了广泛应用. 本文基于有偏估计能够减小最小二乘无偏估计方差的思想,提出采用多传感器有偏估计数据融合改善测量数据可靠性的方法. 首先,基于岭估计提出了有偏测量过程,并给出了测量数据可靠性定量表示方法,同时证明了有偏测量可靠度优于无偏测量可靠度. 其次,提出了多传感器有偏估计数据融合方法,证明了现有集中式与分布式无偏估计数据融合之间的等价性. 最后,证明了多传感器有偏估计数据融合收敛于无偏估计数据融合. 实例应用验证了方法的有效性.Abstract: The multi-sensor data fusion method is widely adopted in the process control community to improve the reliability of measured data. Based on the idea that biased estimation can reduce the variance of least square estimation, a multi-sensor biased estimation data fusion method is proposed to improve the measurement reliability. Firstly, biased measurement is presented based on ridge estimation, and a quantitative description method evaluating the reliability of the measurement data is proposed. The biased measurement reliability superior to the unbiased one is also verified. Secondly, a multi-sensor biased estimation data fusion scheme, which synthesizes the merits of the biased measurement and multi-sensor data fusion, is proposed, and the conclusion that is proven to decentralized unbiased estimation data fusion is equal to the centralized one is proved. Thirdly, the proposed fusion method is proven to converge to the unbiased estimation data fusion. Finally, a laboratory scale experiment illustrates that the presented method is effective.
-
Key words:
- Measurement /
- reliability /
- data fusion /
- biased estimation /
- ridge estimation
-
[1] Zhang Zheng-Jiang, Zeng Guo-Qiang, Shao Zhi-Jiang, Wang Ke-Xin, Chen Xi. Methodology of data reconciliation and parameter estimation for variable load in process system. CIESC Journal, 2012, 63(6): 1780-1789 (张正江, 曾国强, 邵之江, 王可心, 陈曦. 过程系统变负荷下的数据校正与参数估计方法. 化工学报, 2012, 63(6): 1780-1789) [2] Song Hai-Hua, Wang Xiu-Li, Li Hong-Hai. Measurement and prediction of interfacial area on distillation tray. CIESC Journal, 2003, 54(8): 1112-1117 (宋海华, 王秀丽, 李红海. 精馏塔板上气液相界面积的测量与预测. 化工学报, 2003, 54(8): 1112-1117) [3] E Jia-Qiang, Wang Yao-Nan, Mei Chi. Soft-sensing model of copper liquid temperature in copper refining process and its application. CIESC Journal, 2006, 57(1): 203-209 (鄂加强, 王耀南, 梅炽. 铜精炼过程铜液温度软测量模型及应用. 化工学报, 2006, 57(1): 203-209) [4] Viera A J, Garrett J M. Understanding inter-observer agreement: the kappa statistic. Family Medicine, 2005, 37(5): 360-363 [5] Ge Quan-Bo, Li Wen-Bin, Sun Ruo-Yu, Xu Zi. Centralized fusion algorithms based on EKF for multisensor. Acta Automatica Sinca, 2013, 39(6): 816-825 (葛泉波, 李文斌, 孙若愚, 徐姿. 基于EKF的集中式融合估计研究. 自动化学报, 2013, 39(6): 816-825) [6] Deng Z L, Zhang P, Qi W J, Liu J F, Gao Y. Sequential covariance intersection fusion Kalman filter. Information Sciences, 2012, 189: 293-309 [7] Carlson N A. Federated square root filter for decentralized parallel processes. IEEE Transactions on Aerospace and Electronic Systems, 1990, 26(3): 517-525 [8] Wen C B, Cai Y Z, Wen C L, Xu X M. Optimal sequential Kalman filtering with cross-correlated measurement noises. Aerospace Science and Technology, 2013, 26(1): 153-159 [9] Feng J X, Wang Z D, Zeng M. Distributed weighted robust Kalman filter fusion for uncertain systems with autocorrelated and cross-correlated noises. Information Fusion, 2013, 14(1): 78-86 [10] Sun S L, Deng Z L. Multi-sensor optimal information fusion Kalman filter. Automatica, 2004, 40(6): 1017-1023 [11] Ran C J, Deng Z L. Self-tuning weighted measurement fusion Kalman filtering algorithm. Computational Statistics & Data Analysis, 2012, 56(6): 2112-2128 [12] Mănsson K. On ridge estimators for the negative binomial regression model. Economic Modelling, 2012, 29(2): 178-184 [13] Gan Q, Harris C J. Comparison of two measurement fusion methods for Kalman-filter-based multisensor data fusion. IEEE Transactions on Aerospace and Electronic Systems, 2001, 37(1): 273-279 [14] Stein C. Inadmissibility of the usual estimator for the mean of a multivariate normal distribution. In: Proceedings of the 3rd Berkeley Symposium on Mathematical Statistics and Probability. 1956, 1: 197-206 [15] James W, Stein C. Estimation with quadratic loss. In: Proceedings of the 4th Berkeley Symposium on Mathematical Statistics and Probability. 1961, 1: 311-319 [16] Sclove S L. Improved Estimators for Coefficients in Linear Regression. Technical Report, Department of Statistics Stanford University, 1967, No.128 [17] Liu K. Using Liu-type estimator to combat collinearity. Communications in Statistics Theory and Methods, 2003, 32(5): 1009-1020 [18] Duran E A, Akdeniz F. Efficiency of the modified jackknifed Liu-type estimator. Statistical Papers, 2012, 53(2): 265-280 [19] Hoerl A E, Kennard R W. Ridge regression: biased estimation for non-orthogonal problems. Technometrics, 1970, 12(1): 55-67 [20] Zhen Zi-Yang, Wang Zhi-Sheng, Wang Dao-Bo. Predictive control based on information fusion optimal estimation for nonlinear discrete system. Acta Automatica Sinca, 2008, 34(3): 331-336 (甄子洋, 王志胜, 王道波. 基于信息融合最优估计的非线性离散系统预测控制. 自动化学报, 2008, 34(3): 331-336) [21] Du P J, Liu S C, Xia J S, Zhao Y D. Information fusion techniques for change detection from multi-temporal remote sensing images. Information Fusion, 2013, 14(1): 19-27 [22] Feng Xiao-Liang, Wen Cheng-Lin, Liu Wei-Feng, Li Xiao-Fang, Xu Li-Zhong. Sequential fusion finite horizon H∞ filtering for multisenor system. Acta Automatica Sinca, 2013, 39(9): 1523-1532 (冯肖亮, 文成林, 刘伟峰, 李晓芳, 徐立中. 基于多传感器的序贯式融合有限域H∞ 滤波方法. 自动化学报, 2013, 39(9): 1523-1532) [23] Luo Ben-Cheng, Yuan Kui, Chen Jin-Long, Zhu Hai-Bin. Uncertainty analysis based dynamic multi-sensor data fusion. Acta Automatica Sinica, 2004, 30(3): 407-415 (罗本成, 原魁, 陈晋龙, 朱海滨. 一种基于不确定分析的多传感器信息动态融合方法. 自动化学报, 2004, 30(3): 407-415)
点击查看大图
计量
- 文章访问数: 2123
- HTML全文浏览量: 143
- PDF下载量: 1659
- 被引次数: 0