MSc Robotics Final Project by Johanna Amunyela

  Mining Rate Parameters   Controlled parameter  Mining air pressure Rotation of tool   Weight on bit  Positioning  Uncontrolled Parameters  Jet water pressure  Mining Rate influencing parameters (indexes) Geology  Weather 

Autoregression  (AR) Using Past Values to predict future values  Reference: Moving Average Model      ACF vs PACF ARMA Model Coding  Conclusion  I will be looking at the SARIMA Model to understand how to correctly include seasonal effects in the model

 Abstract  "This paper explores the use of AutoRegressive Integrated Moving Average (ARIMA) forecasting on the time series data collected from various sensors from a Slitting Machine, to predict the possible failures and quality defects, thus improving the overall manufacturing process. The use of Machine Learning thus proves a vital component in IIoT having use cases in quality management and quality control, lowering the cost of maintenance, and improving the overall manufacturing process." Having spoken to a mining technical expert, I discovered that the drill bit sits on a PLC system. This means the model I plan to build will have to focus on performing inference and providing the current PLC system with the appropriate signals and or proposing a new to send the drill compensation values. the paper focuses on preventative and proactive maintenance using ARIMA. ARIMA was widely discussed and I will be using it as a start point of the Machine Learning approach once I have r...

 From geological point of view, mining rates will be most affected by the following   Gravel volumes : the thicker the gravel volumes, the longer it takes for the holes to deplete. Thicker volumes also cause surging which prolongs presentation time. Gravel texture: the coarser the material, the lower the RPMs, then the longer it will take for the hole to deplete. Homogeneous geology improves mining rates Slab frequency:  the higher the frequency, the lower the RPMs, then the longer it will take for the hole to deplete. Lithology type: courser lithologies like large cobbles slows mining rate, while softer lithologies Hard or rough terrains such as sandstone, conglomerate s outcrops or cappings make it difficult for the tool to penetrate and thus slow the mining rate. It also affects slewing speed of the boom in terms of crawler mining systems. Steep slopes: require slow rotations of the tools and thus slow m...

  I did research into the possible simulation of the drill bit. This is to help ultimately in illustrating the ML model's inference on the tool. This is the email response that I got  In the interim, I will have to see how best to generate the data to replay in the Drill Visualisation System. Below is a snap shot of how the Drill visualisation system looks and feels  DVS Drill Visualisation System

These parameters will be used to train and test the neural network. They are a few significant areas that I identify and I will be looking into a lot more. Tool - Drill Bit  Mined out Hole  Weather API Mining Process settings  Mining System    Master PC (Navigation)  Slave PC (Dilling)    Tool Drill Bit The drillers use a coak pit system to perform the drill compensation while making sure the mining system is recording all the data correctly.

 Project Initial Phase Gant Chart  Link to Gant Chart   Project Planning Notes  Goals and Objectives  devise task - what needs to be done to archive the objectives  Make a plan and update it as one works Start  Who are the people doing similar work?  What resource does one need? What additional learning material are needed?  Hits   Time based representation of project plan Split the objectives into mile stones and tasks keep task and milestones tangible and measurable  test the task outcome and report on it