An Artificial Neural Network Model for Prediction of Truck Rear Suspension Pressures in Hauling Loaded Conditions

J. Szymanski, W. Pedrycz, S. Frimpong, L. Sudak, and R. Szymanski (Canada)

Keywords

Neural networks, dynamic loading, truck haulage

Abstract

Proper maintenance and operation of equipment is essential to any effective mining operation. Operations at one of the largest oil sand producer rely heavily on truck and shovel methods. The latest generation of haul trucks is outfitted with sophisticated monitoring systems. CATERPILLAR trucks use VIMS, or Vital Information Management System. VIMS includes on-board truck measuring equipment and off-board VIMS software, which enables data download to a lap-top. A record of real time parameters gives a detailed view of what happens to a truck as it performs various functions. The input parameters of great importance are payload, ground speed, rack and pitch values off the machine. By knowing how these values influence vehicle’s rear struts pressures it is possible to assess the remaining fatigue life of the truck’s frame. In recent years neural networks have been widely applied in industry. The most commonly used model is the back propagation multilayer neural network. This technique was applied for prediction of rear cylinders pressure obtained from field testing of Cat 797 (Caterpillar) truck traveling loaded with overburden from electric shovel pit to the dump.

Important Links:



Go Back