摘要
为保证发电厂中发电机运行状态在线监测的实时性和准确性,针对在线监测中存在的时间同步问题,提出一种适用于电气设备在线监控、基于多参考节点的无线传感器网络(Wireless Sensor Network,WSN)时间同步算法。结合梯度同步和单向同步,同时对逻辑时间率和本地时间进行修正。参考节点间采用梯度时间同步,根据时间运行快慢设置权值,通过多个邻节点逻辑时间率的加权平均值来修正本地逻辑时间率;参考节点与普通节点间采用单向同步机制,采用线性回归法计算并修正逻辑时间率和本地时间。仿真结果表明:基于多参考节点的无线传感器网络时间同步算法具有同步误差小、通信开销低的特点,且参考节点停止同步时仍能保持网络正常同步,提高了同步稳定性。
To ensure the real-time performance and accuracy of online monitoring of generator's statues, a time synchronization algorithm with multiple reference nodes is proposed to solve the time synchronization problem existing in electrical equipment online monitoring with wireless sensor network (WSN). This algorithm combines the gradient clock synchronization and one-way synchronization, which corrects both the logical time rate and the local time. The gradient clock synchronization is adopted among reference nodes, in which the local logical time rate is corrected by calculating the weighted average value of multiple adjacent nodes' logic time rate with weight set according to the speed of time running. The one-way synchronization is adopted between reference nodes and ordinary nodes, in which linear regression is adopted to calculate and correct logical time rate and local time. Simulation results show that the proposed algorithm has the characteristics of small synchronization error and low communication overhead, and the synchronization of whole network can still be executed as normal when some of reference nodes stop synchronization, achieving a good stability.
出处
《控制工程》
CSCD
北大核心
2016年第1期12-17,共6页
Control Engineering of China
基金
国家自然科学基金面上项目(61373126)
江苏省自然科学基金(BK20131107)
关键词
无线传感器网络
时间同步
梯度时间同步
通信开销
Wireless sensor network
time synchronization
gradient clock synchronization
communication overhead