An online expert for industrial minerals processes

Expert Online, as a software system, is aimed at providing plant operators in the minerals industry with effective training and on-the-job assistance. It incorporates a number of features to achieve this goal, of which the most important is its ability to explain the dynamics of the plant and control system in simple language, mimicking an informed plant expert who has online access to the inner workings of the plant and control system. This is made possible by a novel form of Artificial Intelligence, called the Interpreting Expert System. Previous implementations of the IES were only capable of explaining controller actions. Expert Online will extend this capability to also perform the more difficult task of explaining process behaviour. Expert Online will implement three main processes, namely the accumulation of data from the controllers and plant models, the interpretation, processing and filtering of that data and lastly communicating the relevant information to the user, both as written messages and in other audio-visual forms. Although Expert Online is aimed at the operator, other plant personnel, like supervisors and plant managers, will also benefit from it. This paper focuses on the structural and functional design of Expert Online for implementation on the latest version of Mintek’s open control platform, PlantStar, highlighting advances and additions since the previous implementations of the Interpreting Expert System and concluding with an example illustrating some of the principles.