Modelling WWTP Dynamics by means of Transition Networks

This paper discusses a progress report on integrating predictive skills into an integrated AI system for wastewater treatment plants (WWTP) supervision and control. Although the embedded approaches within the previously developed architecture, called DAI-DEPUR, such as numerical control knowledge, rule-based reasoning, case-based reasoning are able to cope with the overall supervision task of a plant, one feature is missed: the predictive knowledge. A methodology to induce automaton representation of change over time of a process is presented. This automaton represents the states of the process, input and output variables and actions performed to change the state. The final objective of this work is to integrate this representation as a complement of the knowledge in the system DAI-DEPUR in order to predict change over time. Keywords: Knowledge Acquisition, Machine Learning, Knowledge Discovery, Transition Networks, Industrial Application. 1 Introduction The automatic control of indus..