基于多传感器数据融合的海洋牧场实时监测系统  

Real-time monitoring system of marine ranch based on multi-sensor data fusion

作　　者：刘小飞[1] 李美满[1] 詹志炬 LIU Xiaofei;LI Meiman;ZHAN Zhiju(School of artificial Intelligence,Guangdong Polytechnic Institute,Guangzhou 510091,China)

机构地区：[1]广东理工职业学院人工智能学院,广州510091

出　　处：《智能计算机与应用》2023年第6期196-202,共7页Intelligent Computer and Applications

基　　金：广东省科技创新战略专项资金项目(pdjh2020b0962);2020年广东省教育厅重点平台及科研项目(2020KTSCX401)。

摘　　要：为了提高海洋牧场环境监测数据的可靠性和环境预警的准确性,设计并实现了基于多传感数据融合的海洋牧场实时监测系统。首先设计了基于Modbus协议的数据采集控制器;其次,对同类传感器的数据利用信任度模型进行的融合,对异类传感器数据利用BP神经网络进行融合。结果表明,所设计的海洋牧场实时监测系统能提高监测数据的可靠性和预警的准确性,监测系统运行稳定。In order to improve the reliability of marine ranch environmental monitoring data and the accuracy of environmental early warning,a real-time monitoring system based on multi-sensor data fusion is designed and implemented.Firstly,a data acquisition controller based on Modbus protocol is designed.Secondly,the data of similar sensors is fused with the trust model,and the data of different sensors is fused with BP neural network.The results show that the designed marine ranch real-time monitoring system can improve the reliability of monitoring,the accuracy of early warning,and the stability of monitoring system.

关 键 词：数据融合 多传感器 海洋牧场 监测系统 神经网络 

分 类 号：TP315[自动化与计算机技术—计算机软件与理论]