论文部分内容阅读
对农作物生长环境的温度、湿度和光照等参数的实时监测和远程控制是农业生产现代化的重要手段,为此设计了一种以ARM嵌入式微处理器2410为硬件平台,结合Linux操作系统和ADS1.2集成开发环境,同时采用D-S证据理论和BP神经网络相结合算法的多传感器信息融合测控系统。BP神经网络提供一定数量的证据,D-S证据理论降低证据的不确定性,将这种基于二者的多传感器信息融合算法应用于农作物生长环境监控中,最终得到了在S1(T>30°)温度区间内93.47%的可信度,可见设计的基于多传感器信息融合的测控体系具有良好的应用前景。
Real-time monitoring and remote control of the parameters such as temperature, humidity and illumination of the crop growing environment are important means of agricultural modernization. To this end, an ARM embedded microprocessor 2410 is designed as a hardware platform, combined with Linux operating system and ADS1. 2 integrated development environment, at the same time using DS evidence theory and BP neural network combined algorithm of multi-sensor information fusion measurement and control system. BP neural network provides a certain amount of evidence, and DS evidence theory reduces the uncertainty of evidence. This two-sensor-based multi-sensor information fusion algorithm is applied to crop growth environment monitoring. Finally, The reliability of 93.47% within the temperature range shows that the designed measurement and control system based on multi-sensor information fusion has a good application prospect.