Niranjan Adhikari holds a double Masters in both Statistics and Physics.

Niranjan is a PhD student at the JKMRC working on a project of the Advanced Process Prediction and Control group under Dr Gordon Forbes, Associate Professor Mohsen Yahyaei and Dr Marko Hilden.

Niranjan completed his undergraduate and Masters' degrees in theoretical physics.  He then worked for four years as a physics lecturer in a university in Nepal. Niranjan then went on to complete another Masters degree in Statistics and Operations Research in 2017 where he specialised in Machine Learning. In 2018, he worked with Advantage Data Consulting Company as a Data Scientist, before commencing his PhD in early 2020, with the working title:

Machine Learning Algorithm for Calibration of Comminution Models.

Empirical process models are frequently used in the mineral processing industry for designing a circuit, selecting equipment, testing circuit configurations, and for the process control systems. These models have been continuously used in a flowsheet simulation package such as JKSimMet. There are several parameters corresponding to these models. However, these models require calibration (updating the model parameters) in a timely manner when the physical condition of the circuit as well ore properties have changed.

In order to calibrate these models, one typically does a survey to collect relevant data then fits the model using the simulation package JKSimMet, where the entire calibration process takes at least six months. The existing calibration procedure is costly and time-consuming as it has to go through various time-consuming and costly processes such as plant survey, sample analysis, and model fitting.

Instead, the project is utilising plant operating data (ie. PI data) to extract essential information to calibrate the process models through a machine learning technique as guided by mineral process expert knowledge. There is still a requirement to do a base case survey to be able to develop an initial model, but after that, the model will update based on the current operating regime identified using PI data.

The proposed approach can improve the accuracy of the prediction of process models by correctly identifying model parameters and reduce the burden of frequent surveys. Such capability will put integrated simulation in a unique space as it will be the first simulation of this kind that provides real-time prediction capability of the process models.