Stepwise identification of automated discrete manufacturing systems

Abstract

Abstract

Identification of industrial automation systems: Building compact and expressive Petri Net models from observable behavior

“…State evolutions non-exhibited by output evolutions will be represented by transitions between non-observable places. This has been done previously in literature [4], [7] [10] [13]. However, existing approaches show only direct event-output evolutions (Type 1).…”

“…In a previous work [10] an I/O sequence is considered to compute an IPN including cyclic behavior. Although the proposed methodology is scalable due to the algorithms efficiency, the obtained models are close to finite automata and can be huge, due to the explicit representation of observed input changes that could not be relevant to define the output evolution.…”

“…4) is to sort parcels according to their size. It has 9 signal sensors from the system: a 0 , a 1 , a 2 , b 0 , b 1 , c 0 , c 1 , k 1 , k 2 , and 4 signals to the actuators: A+, A-, B, C. This example has been used in other publications [10], [13] and we take it up again to confront previous results to the new ones.…”

Automated modelling of reactive discrete event systems from external behavioural data

“…This method has been extended and adapted for identifying actual industrial PLC-based controlled discrete manufacturing systems, which operate during a long time period performing repetitive tasks [14]; the proposed technique operates stepwise allowing updating the model when new input/output vectors are added to the sequence. In the present paper we are focussing on the implementation and application of a software tool that automates the identification method.…”