首页|基于RS-PSO-KELM算法的无线传感器网络节点故障检验技术

基于RS-PSO-KELM算法的无线传感器网络节点故障检验技术

扫码查看
当前无线传感器网络节点故障检验技术存在难以应对愈发复杂化的无线传感器网络结构的问题,因此,研究将粗糙集、粒子群算法以及核极限学习机进行结合构建组合算法,并对其进行了验证.实验结果表明,迭代次数为5时研究算法的均方根误差值为0.010,迭代次数为10为0.008,至迭代次数为50时为0.006,均低于对比算法.研究算法在4个数据集上的准确率基本高于85%,最高达到98.51%.将其应用在实际的无线传感器网络节点故障中时,故障诊断准确率维持在98%~99%,且仅存在1个诊断错误点.综合来看,研究提出的算法在无线传感器网络节点故障诊断检验具备较高性能,可以有效应用在实际的无线传感器节点故障诊断中.
Node Fault Detection Technology for Wireless Sensor Networks Based on RS-PSO-KELM Algorithm
The current node fault detection technology in wireless sensor networks is difficult to cope with the increasingly com-plex structure of wireless sensor networks.Therefore,a combination algorithm combining rough set,particle swarm optimization algo-rithm,and kernel extreme learning machine was studied and verified.The experimental results show that the root mean square error value of the research algorithm is 0.010 when the number of iterations is 5,0.008 when the number of iterations is 10,and 0.006 when the number of iterations is 50,all of which are lower than the comparison algorithm.The accuracy of the research algorithm on four datasets is basically higher than 85%,with a maximum of 98.51%.When applied to practical wireless sensor network node faults,the fault diagnosis accuracy is maintained at 98%~99%,and there is only one diagnostic error point.Overall,the algorithm proposed in the study has high performance in node fault diagnosis and verification of wireless sensor networks,and can be effectively applied in practical wireless sensor node fault diagnosis.

RS-PSO-KELMwireless sensorkernel extreme learning machinenetwork node failureaccuracy

时佳

展开 >

陕西邮电职业技术学院,陕西咸阳 712000

RS-PSO-KELM 无线传感器 核极限学习机 网络节点故障 准确率

陕西省职教学会陕西邮电职业技术学院

2023SZX432SPTC202204

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(8)