基于TensorFlow的LSTM神经网络智能电气火灾预测研究
|
摘要
随着智慧城市建设的步伐不断加快以及人工智能技术的高速发展,如何利用大数据技术对电气火灾事件实现精准预测已经成为当前火灾预防研究者特别关注的问题.本文设计并实现了基于TensorFlow智能学习系统的LSTM神经网络电气火灾预测算法,对某地区输电线路上的数据进行了分析,通过多次数据迭代和参数调优对模型进行了训练,实验结果表明,基于TensorFlow的LSTM神经网络算法对于电气火灾的预测不仅具有较高的准确度,而且泛化能力强,预测效果明显好于其他学习算法.
With the accelerating pace of smart city construction and the rapid development of artificial intelligence technology,how to use big data technology to achieve accurate prediction of electrical fire events has become a special concern of fire prevention researchers.This paper designed and implemented a LSTM neural network electrical fire prediction algorithm based on TensorFlow intelligent learning system,analyzed the data on the transmission line in a certain area and trained the model through several data iterations and parameter tuning,the experimental results show that the LSTM neural network algorithm based on TensorFlow for electrical fire prediction not only has high accuracy,and strong generalization ability,prediction effect is much better than other learning algorithms.
With the accelerating pace of smart city construction and the rapid development of artificial intelligence technology,how to use big data technology to achieve accurate prediction of electrical fire events has become a special concern of fire prevention researchers.This paper designed and implemented a LSTM neural network electrical fire prediction algorithm based on TensorFlow intelligent learning system,analyzed the data on the transmission line in a certain area and trained the model through several data iterations and parameter tuning,the experimental results show that the LSTM neural network algorithm based on TensorFlow for electrical fire prediction not only has high accuracy,and strong generalization ability,prediction effect is much better than other learning algorithms.
引文
[1]SCHMIDHUBER J.Deep learning in neural networks:an overview[J].Neural Networks,2015,61:85-117
[2]成登旺.基于GIS的太原市消防信息系统的设计与实现[D].长春:吉林大学,2016
[3]崔亮,张继权,包玉龙,等.呼伦贝尔草原火灾风险预警研究[J].草业学报,2012,21(04):282-292
[4]李杰.北京市火灾事故特征的统计分析[J].中国安全生产科学技术,2013,9(5):126-130
[5]田树仁.基于神经网络的智能电气火灾预警系统设计[J].消防科学与技术,2015,34(9):1201-1204
[6]刘海生,张鑫磊,宋丽霞.基于统计数据的全国火灾形势评价与预测[J].中国安全科学学报,2011,21(6):54-59
[7]汤昊,金静,刘义祥,等.基于二次指数平滑法的典型城市电气火灾预测[J].武警学院学报,2017,33(08):63-66
[8]李伟山,王琳,卫晨.LSTM在煤矿瓦斯预测预警系统中的应用与设计[J].西安科技大学学报,2018,38(6):1027-1035
[9]李智敏,刘一鹏,郑海峰,等.LSTM递归神经网络人体活动行为识别算法研究[J].电气技术,2018,19(11):26-30
[10]陈韫.基于LSTM深度网络的城市道路短时交通状态预测模型研究[D].福州:福建工程学院,2018
[11]松岭.基于TensorFlow的LSTM循环神经网络短期电力负荷预测[J].上海节能,2018(12):974-977
[12]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780
[2]成登旺.基于GIS的太原市消防信息系统的设计与实现[D].长春:吉林大学,2016
[3]崔亮,张继权,包玉龙,等.呼伦贝尔草原火灾风险预警研究[J].草业学报,2012,21(04):282-292
[4]李杰.北京市火灾事故特征的统计分析[J].中国安全生产科学技术,2013,9(5):126-130
[5]田树仁.基于神经网络的智能电气火灾预警系统设计[J].消防科学与技术,2015,34(9):1201-1204
[6]刘海生,张鑫磊,宋丽霞.基于统计数据的全国火灾形势评价与预测[J].中国安全科学学报,2011,21(6):54-59
[7]汤昊,金静,刘义祥,等.基于二次指数平滑法的典型城市电气火灾预测[J].武警学院学报,2017,33(08):63-66
[8]李伟山,王琳,卫晨.LSTM在煤矿瓦斯预测预警系统中的应用与设计[J].西安科技大学学报,2018,38(6):1027-1035
[9]李智敏,刘一鹏,郑海峰,等.LSTM递归神经网络人体活动行为识别算法研究[J].电气技术,2018,19(11):26-30
[10]陈韫.基于LSTM深度网络的城市道路短时交通状态预测模型研究[D].福州:福建工程学院,2018
[11]松岭.基于TensorFlow的LSTM循环神经网络短期电力负荷预测[J].上海节能,2018(12):974-977
[12]HOCHREITER S,SCHMIDHUBER J.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780