Font Size: a A A

Elevator Running Quality Analysis System Design Of Based On AR Model Support Vector Machine(SVM)

Posted on:2016-05-22Degree:MasterType:Thesis
Country:ChinaCandidate:B N ShiFull Text:PDF
GTID:2322330509959460Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
The elevator safety problem has become the hot topic in current society. Elevator running in the process of all kinds fault and tiring, and even casualties accident frequently occurred. It has brought great threat to people's life and property security, caused the bad social influence. Therefore, how to real-time monitor the elevator by quality and safety has important engineering value and social significance.In this paper, the research content mainly includes the following aspects:(1) The elevator existing by quality and safety evaluation method research. By analyzing the elevator and its reasons, will the elevator running environment of several important parameters, and the vibration acceleration, braking acceleration, average acceleration, speed as the machine running performance parameters.(2) The scene detection technology research. The above parameters of research for method and test scheme of the machine running on the performance evaluation index, and plan to collect data samples.(3) According to the principle of fault recognition of support vector machine, the intelligent analysis algorithm is designed according to the characteristics of the vibration signal of the elevator.(4) Research on line real-time 2000 time running intelligent analysis system, the system can monitor many elevators at the same time, and can save the elevator fault data, and provide data support for maintenance and accident identification.This paper will research the system used in the supervision and inspection of elevator installation, the examiner practice that this system has the advantages of simple operation, strong practicability, and is worthy of promotion.The main advantages of the system include the following aspects:(1) AR(auto regressive) model is used to train the 3D acceleration signal of the elevator. The feature kernel function is used as the input of the system, and the new data is analyzed. And it may be possible to estimate and diagnose the faults of the elevator. At the same time, it can evaluate the comprehensive quality of the elevator operation(comfort).(2) It supports 2000 runs, and the design and implementation of the automatic selection of the elevator. In the process of experiment, there is no manual operation, so as to set up a good program, the system will run according to the scheduled program, and the program will be intelligent control layer;(3) Intelligent error correcting ability. On the basis of the error to meet the national standard, and the relative error of the measurement is not more than 2%, and the elevator in the emergency situation of power down the location of the error rate is not more than 1%;(4) Simple structure, lightweight, easy to install: sensor integrated in the microcontroller.
Keywords/Search Tags:Elevator, Ride quality, Support vector machine, 2000 runs
PDF Full Text Request
Related items