Industry 4.0 Roll-Out Strategy for Dynamic Mine Heat Load Management

Abstract

Deep-level mining is under severe financial pressure from several unique challenges. One of these is maintaining acceptable underground temperatures for humans to work in while achieving demanding production targets. As mines regularly reach new depths, additional heat is added to the system, contributing to this problem. Accurate mine heat load studies are therefore required to ensure that heat sources are actively evaluated, managed, and mitigated through adequate cooling practices. However, present heat load models are based on design parameters that cater for worst-case scenarios. Most of these models are also based on outdated empirical data taken at a time when mining differed from the present. Industry 4.0 technologies provide potential optimisation benefits when integrated with new heat load models to ensure effective monitoring, and consequently dynamic management, of heat sources. The roll-out strategy presented in this article will serve as a real alternative to earlier and outdated heat load prediction models.

Simulating the Sensitivity of Underground Ventilation Networks to Fluctuating Ambient Conditions

“…A wide range of technology is available that allows for the continuous monitoring of an underground ventilation network. Nell et al [6] described the different levels of detail to which a mine ventilation system can be instrumented. Increased levels of detail would result in the increased accuracy of the simulation.…”

“…The high cost implications of ventilating mining networks add to the financial strains and uncertainties of the future of mining operations [6]. Industry 4.0 is becoming a crucial paradigm to ensure that the mining sector remains profitable [7].…”

“…Industry 4.0 is becoming a crucial paradigm to ensure that the mining sector remains profitable [7]. Tools such as real-time climatic monitoring systems and simulations are used for ventilation system management [6], [8], [9], [10].…”

Integration of machine learning with complex industrial mining systems for reduced energy consumption

“…The outlet water temperature set-point is typically a static value based on “worst-case” design conditions and does not account for real-time ambient conditions (Maré, 2017 ; Peach et al, 2018 ). The result is an oversupply of cooling to the air going underground, resulting in energy wastage (Nell et al, 2019 ). A varying set-point will enable the system to reduce the cooling required and consequently reduce the energy consumption.…”

“…The lack of sensor infrastructure in mining prevents the implementation of semi-empirical and mathematical-based control and predictive philosophies (Nell et al, 2019 ; Prinsloo et al, 2019 ). The lack of sensors result in data points that lack features (a measureable piece of data i.e., a column in a dataset).…”