Predictive Modeling of Hydraulic Hoses for Underground Mining

This project is about modeling flexible hydraulic hoses that are attached to mining machines and has the goal of predicting hose states under different machine configurations. For this, we will implement different modeling techniques from computational science and computer graphics to evaluate their suitability for the use-case of heavy-duty mining machines under realistic mock-up conditions. The majority of mining vehicles (e.g. drilling rigs, scoop trams, loaders) have hydraulic-driven articulated manipulators. The hydraulic fluid is transported over reinforced rubber hoses, which are often loosely attached to the outside of the machine. This puts the hoses at risk of damage and often results in hoses being torn, either by interaction with the en-vironment or the machine itself.

In this pre-study we aim at deploying different kinematic and dynamic models of a hydraulic hose: that is, a computational method of predicting the positions, velocities, accelerations, and forces acting on every component of a hose. Such models can potentially be used to predicting if a certain control to the machine would result in undesirably high strain on a hydraulic hose, and consequently should be avoided. However, if we are to use such a model to improve machine operations, we need to first evaluate how accurately such models can represent the real physical object. Thus, the primary goal and purpose of this pre-study is to perform a domain-specific evaluation of kinematic and dynamic models for modeling hydraulic hoses.