William John Shipman

Mintek · Measurement and Control

Control systems require maintenance in the form of tuning their parameters in order to maximize their performance in the face of process changes in minerals processing circuits. This work focuses on using deep reinforcement learning to train an agent to perform this maintenance continuously. A generic simulation of a first-order process with a time...

Reinforcement Learning, using deep neural networks, has recently gained prominence owing to its ability to train autonomous agents that have defeated human players in various complex games. Here, Reinforcement Learning is applied to the challenge of automatically tuning a proportional-integral controller, given only the process variable, set-point,...

Process Mineralogy is the application of mineralogical techniques to the exploration of ore deposits and the design and optimisation of mineral processing flowsheets. Samples can be drill cores, rocks and milled particles, to give a few examples. X-ray microtomography has emerged as a complementary technique to the existing two-dimensional imaging...

Diffusion operators have been used extensively in engineering disciplines as well as in image processing. The numerical solution of the diffusion partial differential equation is an iterative process that can require from a few hundred iterations to several thousand iterations depending on the specific application. Furthermore, large volumetric dat...

Processing X-ray microtomography scans of ore samples to extract quantitative mineralogical information regarding composition, porosity and particle size is complicated by the presence of noise in the tomograms and artefacts resulting from non-ideal scanning conditions. In order to obtain quantitative information, one must first classify voxels int...

Linear control system theory is well developed and has lead to a number of control system types with well-defined design methods that can be applied to any linear system. Unfortunately, no system in nature is truly linear. As a result, such non-linear systems must be represented by a linear model that is accurate over some region of the operating s...

Image processing is applied to the task of characterizing the response of a miniature helicopter's main rotor to collective control inputs under static conditions. The objective is to measure the pitch of the main rotor blade in relation to collective control inputs and deduce a transfer function model from the data. The algorithm developed here ma...