Artificial neural network to predict the health risk caused by whole body vibration of mining trucks

• Main Authors: Mohammad Javad Rahimdel, Mehdi Mirzaei, Javad Sattarvand, Behzad Ghodrati, Hosein Mirzaei Nasirabad
• Format: Article
• Language: English
• Published: Iranian Society of Vibration and Acoustics 2017-01-01
• Series: Journal of Theoretical and Applied Vibration and Acoustics
• Subjects: Mining trucks; Health risk; Whole body vibration; Artificial neural network
• Online Access: http://tava.isav.ir/article_24749_a55d0ff934a6430cd1fced29e552a448.pdf

<span style="font-family: 'Times New Roman','serif'; font-size: 10pt; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;">Drivers of mining trucks are exposed to whole-body vibrations (WBV) and shocks during the various working cycles. These exposures have an adversely influence on the health, comfort and also working efficiency of drivers. Determination and prediction of the vibrational health risk of the mining haul trucks at thevarious operational conditions is the main goal of this study. </span><span style="font-family: 'Times New Roman','serif'; font-size: 10pt; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: Calibri; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;">To this aim, three haul roads with low, medium and poor qualities are considered based on the ISO 8608 standard. Accordingly, the vibration of a mining truck in different speeds, weights and distribution qualities of the materials in the dump body are evaluated for each haul road quality using the Trucksim software. An artificial neural network (ANN) is used to predict the vibrational health risk. The obtained results indicate that the haul road qualities, the truck speeds and the accumulation sides of material in the truck dump body have significant effects on the root mean square (RMS) of vertical vibrations. However, there is no significant relation between the material’s weight and the RMS values. Also, the application of ANN revealed that there is a good correlation between the predicted and simulated RMS values. The performance of </span><span style="font-family: 'Times New Roman','serif'; font-size: 10pt; mso-bidi-font-size: 9.0pt; mso-fareast-font-family: 'Times New Roman'; mso-ansi-language: EN-US; mso-fareast-language: EN-US; mso-bidi-language: AR-SA;">the proposed neural network to predict the moderate and high health risk are 88.11% and 93.93% respectively</span>