Hi Doz, good potential there I think. Most of the research I have seen is on haul trucks but it does cross into Mine-Railway... and lots of focus on operation and maintanance costs.. I think you might be onto something.
https://www.soofastaei.net/Ali Soofastaei"Soofastaei completed his Ph.D. at The University of Queensland (UQ) in the field of AI applications in mining engineering where he led a revolution in the use of deep learning (DL) and AI methods to increase energy efficiency,
reduce operation and maintenance costs, and reduce greenhouse gas emissions in surface mines."
davidpublisher.com/Public/uploads/Contribute/5e05610c65696.pdf
Energy-Efficiency Improvement in Mine-Railway Operation Using AIAbstractThe mining industry consumes an enormous amount of energy globally, the main part of which is conservable. Diesel is a key source of energy in mining operations, and mine locomotives have significant diesel consumption. Train speed has been recognizedas the primary parameter affecting locomotive fuel consumption. In this study, an artificial intelligence (AI) look-forward control is developed as an online method for energy-efficiency improvement in mine-railway operation. An AI controller will modify the desiredtrain-speed profile by accounting for the grade resistance and speed limits of the route ahead. Travel-time increment is applied as an improvement constraint. Recent models for mine-train-movement simulation have estimated locomotive fuel burn using an indirectindex. An AI-developed algorithm for mine-train-movement simulation can correctly predict locomotive diesel consumption based on the considered values of the transfer parameters in this paper. This algorithm finds the mine-locomotive subsystems, and satisfies thepractical diesel-consumption data specified in the locomotive’s manufacturer catalog. The model developed in this study has two mainsections designed to estimate locomotive fuel consumption in different situations by using an artificial neural network (ANN), and an optimization section that applies a genetic algorithm (GA) to optimize train speed for the purpose of minimizing locomotive dieselconsumption. The AI model proposed in this paper is learned and validated using real datasets collected from a mine-railway route in Western Australia. The simulation of a mine train with a commonly used locomotive in Australia General Motors SD40-2 (GM SD40-2)on a local railway track illustrates a significant reduction in diesel consumption along with a satisfactory travel-time increment. Thesimulation results also demonstrate that the AI look-forward controller has faster calculations than control systems based that usedynamic programming.
(20min delay)