传感器融合
协方差交集
融合
稳健性(进化)
卡尔曼滤波器
计算机科学
无线传感器网络
扩展卡尔曼滤波器
实时计算
数据挖掘
算法
工程类
人工智能
计算机网络
语言学
哲学
生物化学
化学
基因
作者
Hailong Li,Xinyuan Nan,Xin Cai,Sibo Xia,Haohui Chen
出处
期刊:Measurement
[Elsevier]
日期:2024-05-01
卷期号:230: 114478-114478
标识
DOI:10.1016/j.measurement.2024.114478
摘要
In the process of bio-oxidation gold extraction located in the cold and high-altitude areas, temperature monitoring in the oxidation tank is noisy and accompanied by measurement losses. A new fusion strategy is proposed for improving the temperature monitoring performance of the bio-oxidation tank. In the specific, the measurement data from the underlying sensors is preprocessed by an improved unscented Kalman filter to reduce the impact of missing data and noises on the fusion method. At the local fusion center, the data is fused by a new covariance intersection algorithm to ensure the consistency and high accuracy of the fusion results. In the global fusion center, all of the cluster head fusion results are fused by a hybrid neural network to improve the accuracy and robustness of the designed fusion algorithm. The simulation shows that the proposed fusion strategy effectively improves the temperature monitoring performance of the oxidation tank.
科研通智能强力驱动
Strongly Powered by AbleSci AI