基于HHT-LSTM的冬奥会临时设施运行趋势预测方法研究  

Research on the HHT-LSTM-based operation trend prediction method of temporary facilities for the Winter Olympic Games

作　　者：常明煜 田乐 郭茂祖[1,2] CHANG Mingyu;TIAN Le;GUO Maozu(School of Electrical and Information Engineering,Beijing University of Civil Engineering and Architecture,Beijing 100044,China;Research on Intelligent Processing Method of Building Big Data Beijing Key Laboratory,Beijing University of Civil Engineering and Architecture,Beijing 100044,China)

机构地区：[1]北京建筑大学电气与信息工程学院,北京100044 [2]北京建筑大学建筑大数据智能处理方法研究北京市重点实验室,北京100044

出　　处：《智能系统学报》2024年第1期228-237,共10页CAAI Transactions on Intelligent Systems

基　　金：科技部科技冬奥重点专项(2021YFF0306303);国家自然科学基金项目(62271036)。

摘　　要：针对冬奥会延庆赛区临时设施的安全性和可使用性,本文充分结合信号处理算法与深度神经网络,提出了一种由希尔伯特黄变换(Hilbert-Huang transform,HHT)对时序数据进行信号分解和信号特征提取,长短期记忆网络(long short-term memory,LSTM)进行临时设施运行趋势预测2部分构成模型。该模型基于受严寒天气和大客流诱发的看台振动等一系列外因影响所测得的真实振动和倾角数据,实现对设施进行有效的预测,以避免发生安全问题,解决了由于受数据中一些无关特征因素的干扰导致预测准确度低的问题。论文提出的方法与循环神经网络(recurrent neural network,RNN)、门控循环网络(gated recurrent neural network,GRU)、双向RNN和双向GRU等运行趋势预测方法进行比较,验证了本文方法的可行性和有效性,实验结果也说明所提出的模型在此类任务中表现非常出色。For the safety and availability of temporary facilities in the Yanqing area of the Winter Olympic Games,by sufficiently combining signal processing algorithm and deep neural network,this paper proposes a brand-new model that consists of two parts:Hilbert-Huang transform(HHT)used for signal decomposition and extraction of signal feature for time-series data,and long short-term memory(LSTM)for prediction of the operation trend of temporary facility.Based on the real vibration and tilt angle data measured while it is affected by a series of exogenous factors such as grandstand vibration induced by severe cold weather and heavy passenger flow,the model realizes effective prediction for facilities,so as to avoid safety problems and solve the problem of low prediction accuracy due to the interference of some irrelevant feature factors in the data.By comparing with such operational trend prediction methods as recurrent neural network(RNN),gated recurrent neural network(GRU),bi-directional RNN and bi-directional GRU,the feasibility and effectiveness of the method was demonstrated.The experimental results also show that the proposed model performs very well in such tasks.

关 键 词：时间序列预测 希尔伯特黄变换 长短期记忆网络 信号处理 趋势预测 临时设施 预测方法 数据分析 自然语言处理 

分 类 号：TP911.7[自动化与计算机技术]