基于多传感器信息融合的果蔬仓库监测算法融合

Fusion of fruit and vegetable warehouse monitoring algorithms based on fusion of multi-sensor information

适宜的仓库环境对于提高存储周期、减少腐坏率,起到至关重要的作用。将分组自适应加权融合算法与改进的BP神经网络算法相结合,应用到仓库环境监控系统中。通过使用分组自适应加权融合算法计算同类环境因素的估计值,再与异类传感器融合算法相结合,融合仓库中各种环境因子相互作用的影响,以提高输出结果的准确度。在仓库中收集温度,湿度,气味和二氧化碳浓度四个主要环境因素,并给出四类传感器数据融合处理的实时智能决策结果,从而实现对仓库环境的精确控制。结果表明,算法融合可以有效地提高数据处理精确度,节省各节点能耗,提高监测水平。

A suitable warehouse environment plays a vital role in improving the storage cycle and reducing the decay rate.This paper combines the grouped adaptive weighted fusion algorithm with the improved BP neural network algorithm and applies it to the warehouse environment monitoring system.By using the grouped adaptive weighted fusion algorithm to calculate the estimated value of environmental factors of same kind,and then combined with the heterogeneous sensor fusion algorithm,the effect of the interaction of various environmental factors in the warehouse is fused to improve the accuracy of the output result.Collect the four main environmental factors of temperature,humidity,odor and carbon dioxide concentration in the warehouse,and give real-time intelligent decision results of the fusion processing of four types of sensor data,so as to achieve accurate control of the warehouse environment.The results show that the fusion of algorithms can effectively improve the accuracy of data processing,save the energy consumption of each node,and improve the monitoring level.