A COMBINED ROUGH-STOCHASTIC APPROACH TO PROCESS MONITORING BASED ON QUALITATIVE MODELS

摘要

A precise process monitoring is fundamental to guarantee the correct operation of a technical process. Monitoring applications are frequently based on models of the system. However, the development of such models is very time-consuming and expensive, if at all possible, due to the complexity of real process plants. This imposes the strong simplification of the models, which in turn diminishes appreciably the precision of the monitoring applications and thus their effectiveness. In this work a technique is introduced, which takes advantage of the abilities of qualitative, rough and stochastic modeling methods to enable a closer monitoring of complex real systems.