Title

Artificial Intelligence Models for Predicting the Performance of Hydro-Pneumatic Suspension Struts in Large Capacity Dump Trucks

Abstract

Large dump trucks are being matched with large shovels to achieve bulk economic production in surface mining operations. This process results in high impact shovel loading operations (HISLO) and exposes operators to severe levels of whole-body vibrations (WBV). The performance of the hydro-pneumatic suspension struts, responsible for vibration attenuation in large dump trucks, decreases as a truck age. There is a need for a system for monitoring and predicting the performance of the suspension struts in real time. Artificial intelligence (AI) has been applied for modeling and predicting the suspension system performance for light/smaller vehicles. However, no work has been done to implement AI for modeling and predicting the performance of hydro-pneumatic struts in large dump trucks. This paper is a pioneering effort towards developing AI models for solving this problem. These AI models would incorporate the Artificial Neural Networks (ANN), Mamdani Fuzzy Logic (MFL) and a hybrid system, the Hybrid Neural Fuzzy Interference System (HyFIS), for achieving this goal. Experiments were conducted using a 3D virtual simulator for the CAT 793D in MSC.ADMAS. RMS accelerations in the vertical and horizontal directions at the operator seat were recorded as the two main outputs for the suspension system performance. Eighty percent (80%) of the total experimental data was used in training and developing the models and the remaining 20% for testing and validating the developed models. With an R2 and RMSE of 0.98168505 and 0.00852251 for the training phase, respectively, and 0.9660429 and 0.0195620 for the testing phase, HyFIS model showed the best accuracy for predicting the hydro-pneumatic suspension struts performance for dump trucks. This is the first time that AI models have been developed for dump truck suspension system performance prediction. With the implementation of these models in the dump truck, maintenance personnel can monitor the performance of the suspension system in real-time and schedule proper maintenance and/or replacement. Implementation of such a system will improve the workplace safety, operator's health and the overall system efficiency.

Department(s)

Mining and Nuclear Engineering

Keywords and Phrases

Artificial intelligence; Automobile bodies; Computer circuits; Forecasting; Fuzzy inference; Fuzzy logic; Fuzzy neural networks; Hybrid systems; Mine trucks; Mining; Neural networks; Pneumatic control; Shovels; Struts; Vibrations (mechanical); Dump trucks; HISLO; HyFIS; Mamdani fuzzy; Suspension system; Whole body vibrations; Suspensions (components); Artificial neural network; Mamdani fuzzy logic

International Standard Serial Number (ISSN)

0169-8141

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2018 Elsevier, All rights reserved.

Share

 
COinS