Publicação

Exploring high-level petri nets for model-driven development of digital controllers

Ver documento

Detalhes bibliográficos
Resumo:While low-level Petri nets are common for discrete-event controllers, their capacity is limited. High-level nets of-fer greater potential within Model-Driven Engineering (MDE) for ensuring correctness throughout modeling, simulation, and implementation, but this is not fully realized. This work-in-progress paper presents an approach based on a particular class of high-level nets with a threefold objective: (1) allow the design of complex controllers with data processing; (2) ease the use of interconnected Petri nets modules to build larger models; (3) test the controller with automatically generated input values. Supported by a flexible, open-source tool allowing external interaction (e.g., with code for data generating or processing), we show the methodology using a sluice gate example for offline modeling and simulation, featuring a pattern for environment-controller communication. This approach makes possible more robust model-based real-time controllers, laying the groundwork for upcoming physical device interactions.
Autores principais:Barros, João
Outros Autores:Gomes, Luís
Assunto:Petri nets Controller Reference nets Renew Simulation High-level nets Non-autonomous nets
Ano:2025
País:Portugal
Tipo de documento:comunicação em conferência
Tipo de acesso:acesso restrito
Instituição associada:Instituto Politécnico de Beja
Idioma:inglês
Origem:Repositório Institucional do IPBeja
Descrição
Resumo:While low-level Petri nets are common for discrete-event controllers, their capacity is limited. High-level nets of-fer greater potential within Model-Driven Engineering (MDE) for ensuring correctness throughout modeling, simulation, and implementation, but this is not fully realized. This work-in-progress paper presents an approach based on a particular class of high-level nets with a threefold objective: (1) allow the design of complex controllers with data processing; (2) ease the use of interconnected Petri nets modules to build larger models; (3) test the controller with automatically generated input values. Supported by a flexible, open-source tool allowing external interaction (e.g., with code for data generating or processing), we show the methodology using a sluice gate example for offline modeling and simulation, featuring a pattern for environment-controller communication. This approach makes possible more robust model-based real-time controllers, laying the groundwork for upcoming physical device interactions.