Integrating causal and coarse grain temporal reasoning in a model based control system

Abstract

Monitoring and controlling the processes of complex and geographically distributed systems requires a robust modeling of the behaviour of the systems in terms of causal relationships among its state variables, and the handling of temporal delays that may span an event and its causal influences throughout the system. In this paper, we explain how we have integrated the functionalities of a constraint management system and a temporal database system, to enable a model-based control of systems that exhibit large delays between the events characterizing their behaviour. Our approach has been applied to build a knowledge-based system for assisting central heating operators to optimize the efficiency and profitability of the heating process.<>