摘要
针对电缆火灾致灾定性、定量因素并存,且各因素存在时序特征与非线性关联的问题,提出了基于层次分析法与神经网络的电缆火灾致灾因素分析模型,并提高了因素分析的准确性和鲁棒性。该模型采用层次分析法对致灾因素进行逐级分析,建立了神经网络模型,对各定性、定量因素以及其特征值进行转换,并将其作为神经网络的输入进行模型训练,从而建立致灾因素与火灾之间的非线性时序模型。为了验证模型的有效性,基于真实数据对模型进行仿真实验。结果表明,采用神经网络对各因素的时序性和非线性完成建模后,可有效提高各因素与火灾预警的关联准确性。
In view of the coexistence of qualitative and quantitative factors of cable fire disaster,and the time-series characteristics and nonlinear correlation of each factor,proposes a cable fire factor analysis model based on AHP and neural network,which improves the accuracy and robustness of factor analysis.Based on the analytic hierarchy process,the model analyzes the disaster causing factors step by step;the neural network model is established to transform the qualitative and quantitative factors and their order eigenvalues,and the model is trained as the input of the neural network,so as to establish the nonlinear time series model between the disaster causing factors and the fire. In order to verify the effectiveness of the model,the simulation experiment is carried out based on real data. The results show that the neural network can effectively improve the accuracy of the correlation between various factors and fire warning after modeling the time sequence and nonlinearity of various factors.
作者
杜斌祥
赵金勇
赵子源
韩丙光
DU Binxiang;ZHAO Jinyong;ZHAO Ziyuan;HAN Bingguang(Dezhou Power Supply Company,State Grid Shandong Electric Power Company,Dezhou 253000,China)
出处
《电子设计工程》
2022年第9期107-111,共5页
Electronic Design Engineering
基金
国家电网公司2020年科技项目(2020A-040)。
关键词
电缆
火灾
层次分析法
神经网络
人工智能
cable
fire
analytic hierarchy process
neural network
artificial intelligence
