Wear Prediction and Visual Decision Support of Drop Forged Rivetless Chain for Conveyors

Abstract

In order to accurately detect, predict and maintain the wear condition of drop forged rivetless chain for conveyors, the methods of wear prediction and active maintenance are developed based on gray model and visual decision support. The wear data are acquired by wear detection device based on machine vision. The obtained data are cleaned by defining and analyzing key parameters of wear condition. A wear prediction model is built based on the gray model. A decision-making support model is constructed that can obtain the optimal plan of maintenance by comprehensively evaluating the wear condition with the forecast data, historical data, and operating condition data. The wear condition can be presented in a three-dimensional and dynamic manner to quickly and accurately locate the wear chain by a visual simulation method. The experiment results show that the wear life prediction model has high fitting accuracy, low false alarm rates, and low missed detection rates of wear early-warning, which can satisfy the requirements of wear life prediction. Through the proposed wear prediction model and visual decision support method, the maintenance efficiency is significantly improved, and the risk of forced shutdown of the production line is reduced.

Keywords

• Drop forged rivetless chain;Visual decision support;Wear prediction;Gray model;Machine vision