基于机器学习的电梯安全事故致因分析

Cause Analysis of Elevator Safety Accidents Based on Machine Learning Method

电梯因其方便快捷而被广泛应用于各高层建筑当中,随之而来的是电梯事故的频繁发生。本文通过Python语言搜集并处理近年来我国电梯安全事故案例及相关法律法规,围绕人为、管理、设备和环境等4个方面因素进行特征提取,总结归纳电梯事故的各个影响因素。同时用网格搜索模型、随机森林算法等,构建一种基于随机森林算法的电梯安全事故致因预测模型,并对所获预测模型的准确性进行检验,为事故后的及时归因提供依据。

Elevators are widely used in high-rise buildings because of their convenience,which is followed by the frequent occurrence of elevator accidents.This paper uses Python language to collect and deal with elevator safety accidents of our country in recent years and related laws and regulations,and extracts the characteristics around 4 factors such as people,management,equipment and environment,and summarizes the influential factors of elevator accidents.At the same time,grid search model and random forest algorithm are used to construct a prediction model of elevator safety accident cause based on random forest algorithm,and the accuracy of the prediction model is tested to provide a basis for timely attribution after accidents.