《机电工程》杂志社科技期刊核心期刊论文投稿机电一体化机械、仪表电工自动化、计算机浙江大学主办

Introduction

International Standard Serial Number:

ISSN 1001-4551

Sponsor:

Zhejiang University;

Zhejiang Machinery and Electrical Group

Edited by:

Editorial of Journal of Mechanical & Electrical Engineering

Chief Editor:

ZHAO Qun

Vice Chief Editor:

TANG ren-zhong,

LUO Xiang-yang

Tel:

86-571-87041360,87239525

Fax:

86-571-87239571

Add:

No.9 Gaoguannong,Daxue Road,Hangzhou,China

P.C:

310009

E-mail:

meem_contribute@163.com

Forecasting method of high-speed elevator maintenance cycle based on ForGAN

Published:2023-06-25 author:SU Wan-bin, CHEN Wei-gang, YI Can-can, et al. Browse: 1305 Check PDF documents

Forecasting method of highspeed elevator maintenance cycle

based on ForGAN

SU Wan-bin1, CHEN Wei-gang1, YI Can-can2, CHEN Qi-rui2

(1.Jiaxing Special Equipment Inspection and Testing Institute, Jiaxing 314050, China;

2.School of Mechanical Automation, Wuhan University of Science and Technology, Wuhan 430081, China)

Abstract: At present, there are still obvious deficiencies in the research on brake failure rate and maintenance decision of high-speed elevator.In order to solve the problem of insufficient accuracy and reliability of the current brake failure rate prediction of high-speed elevators, the main reasons and related parameters affecting brake failure were determined by analyzing the failure mode and mechanism of high-speed elevator brakes, and a forecasting generative adversarial networks (ForGAN)model optimized by Bayesian hyperparameters was proposed. Firstly, the performance data of the brakes were collected and normalized. Then, the theoretical failure rate was calculated using the principal component analysis method, and the Bayesian optimization-forecasting generative adversarial networks(BO+ForGAN) model was used to predict the failure rate of the brakes. Finally, the results were compared with traditional prediction models such as support vector machines (SVM) and bi-directional long short-term memory (BiLSTM).Absolute error, root mean square error and determination coefficient (R2) was selected to evaluate the accuracy of prediction results. The research results show that the BO+ForGAN-based model has the best prediction effect, the highest generalization ability, and can adapt to different experimental conditions, and the Bayesian hyperparameter optimization algorithm can find a set of optimal hyperparameters. The evaluation results show that the accuracy of the predicted value of the brake failure rate reaches 98.1%, which verifies the effectiveness of the BO+ForGAN-based model(method).

Key words: forecasting generative adversarial networks(ForGAN); Bayesian hyperparameter optimization(BO); traditional prediction model; root mean square error; generalization ability; failure rate; maintenance decision

Advanced Search

Online Office System

Chinese Core Periodicals
Chinese Sci-tech Core Periodicals
SA, INSPEC Indexed
CSA: T Indexed
UPD:Indexed