摘要
无线传感节点的能量供给问题是影响物联网长时间准确测量的重要因素之一。将太阳能转化为电能,供无线传感节点使用,是解决无线传感节点长时间工作的途径。针对太阳能受环境因素影响较大的特点,提出一种基于长短时记忆递归神经网络(LSTM-RNN)的太阳能无线传感节点能量预测方法,通过预测的能量采集信息合理规划无线传感节点的能量使用,保障无线传感节点能量供给稳定性和测量信息的准确与可靠。实验结果表明,所提出的长短时记忆递归神经网络太阳能无线传感节点能量预测方法,能够利用长时间跨度的太阳能采集历史数据,提供准确的无线传感节点能量预测结果,保障无线传感节点能量供给稳定性和测量信息的准确与可靠。
The energy supply problem of wireless sensor node is one of the important factors affecting the long-term and accurate measurement of the Internet of Things.Converting solar energy into electrical energy for use by the wireless sensor node is a way to solve the long-term operation of the wireless sensor node.Aiming at the characteristics that solar energy is greatly affected by environmental factor,this paper proposes a solar wireless sensor node energy prediction method based on long-short term memory recurrent neural network.The energy usage of the wireless sensor node is reasonably planned through the predicted energy harvesting information to ensure the energy supply stability of the wireless sensor node and the accuracy and reliability of measurement information.The experimental results show that the energy prediction method of the solar wireless sensor node based on LSTM-RNN can use the long-span solar energy to collect historical data,provide accurate wireless sensor node energy prediction result,and ensure the energy supply stability of the wireless sensor node,and the accuracy and reliability of the measurement information.
作者
崔粟晋
王雪
Cui Sujin;Wang Xue(Department of Precision Instrument ,State Key Laboratory of Precision Measurement Technology and Instruments,Tsinghua University,Beijing 100084,China)
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2018年第11期147-154,共8页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(61472216)项目资助
关键词
无线传感节点
太阳能采集预测
长短时记忆
wireless sensor node
solar energy harvesting prediction
long-short term memory (LSTM)
