Applying REMES behavioral modeling to PLC systems

Abstract

Abstract-Programmable logic controllers (PLCs), as a specialized type of embedded systems, have been introduced to increase flexibility and reliability of the system, but at the same time to give faster response time and lower cost of implementation. In the beginning, their use brought a revolutionary change, but with the constant growth of the system complexity, it became harder to guarantee both functional and extra functional properties, as early as possible in the development process. In this paper, we show how formal methods can be applied to describe PLC-based systems and illustrate it on an example of a car wash system. First, we show how the existing behavioral modeling language REMES can be extended to model the behavior of such systems. Second, we show how REMES can be translated to networks of timed automata and priced timed automata in order to support safety and resource-wise reasoning about PLC systems. The formal verification of PLC systems is carried out in UPPAAL and UPPAAL CORA tools.

An overview of model checking practices on verification of PLC software

“…All of the studies that use FBDs are performed after 2007, and a great portion of these studies are performed after 2010. On the other hand, sequential function charts were mostly used between 2000 and 2005, they seem to be not preferred for more recent model check- CFCs [110] Simulink and data/state flowcharts [60,78] Other [107] (REMES) [45] (LUSTRE), [105] (tMPSG) [111] S (VKRC) ing studies. Another interesting point is the lack of Timed Automata studies using Petri nets.…”

“…Thapa et al [105] used timed version of their Message-Based Part Graph (MPSG) modelling language and transformed PLC models to SMV models. Vulgarakis and Causevic [107] use REMES, a modelling language for embedded systems and extend the language for interrupt support to be used in PLC model checking with UPPAAL.…”

Monte Carlo Tree Search for Priced Timed Automata

“…Applications of PTA and UPPAAL CORA are several and from a variety of areas [14], e.g. power optimization of dataflow applications [2], battery scheduling [27], planning of nano-satelites [23,30], grape harvest logistic [33], programmable logic controllers [35], smart grids [21], service oriented systems [17], and optimal multicore mapping of spreadsheets [11] to mention a few.…”