MTPPy: Open-Source AI-friendly Modular Automation

Research output: Contribution to book/conference proceedings/anthology/reportConference contributionContributedpeer-review

Contributors

  • Valentin Khaydarov - , Process Systems Engineering Group, Dresden University of Technology (Author)
  • Laura Neuendorf - , Dortmund University of Technology (Author)
  • Tobias Kock - , Equipment Design, Dortmund University of Technology (Author)
  • Norbert Kockmann - , Dortmund University of Technology (Author)
  • Leon Urbas - , Process Control Systems/Process System Engineering (Author)

Abstract

Modular Automation offers promising technologies for the process and chemical industry. It meets the requirements for greater flexibility and a shorter development cycle for production facilities by dividing the process into smaller standardized units. Key element of Modular Automation is the Module Type Package (MTP). It represents a standardized and manufacturer-independent description of the automation interface for self-contained production units, or Process Equipment Assemblies (PEA), and endows the plug-and-produce capability of PEAs. Software solutions to program MTP compatible PEAs are mostly provided by the Programmable Logic Controller (PLC) manufacturers. They are proprietary and bound to certain hardware and even programming languages. Therefore, their suitability for implementation of soft sensors based on data-driven advanced analytics is strongly limited. In this article we present an opensource Python package, MTPPy, designed for rapid prototyping of MTP-capable soft sensors with a focus on AI-based research applications. MTPPy is aimed at the accelerated deployment of soft sensors in the modular plants; thus, closing the gap between the data-driven model development and integration into the production.

Details

Original languageEnglish
Title of host publication2022 IEEE 27th International Conference on Emerging Technologies and Factory Automation (ETFA)
PublisherIEEE Computational Intelligence Society (CIS)
Pages1-7
Number of pages7
ISBN (electronic)9781665499965
ISBN (print)978-1-6654-9997-2
Publication statusPublished - 9 Sept 2022
Peer-reviewedYes

Conference

Title2022 27th IEEE International Conference on Emerging Technologies and Factory Automation
Abbreviated titleETFA 2022
Conference number27
Duration6 - 9 September 2022
Website
Degree of recognitionInternational event
LocationUniversität Stuttgart
CityStuttgart
CountryGermany

External IDs

Scopus 85141375788
Mendeley c6a5d7b6-3e0f-34ad-8119-75caaa942c79
ORCID /0000-0001-5165-4459/work/142248261

Keywords

Keywords

  • Chemical industry, Soft sensors, Raman scattering, Programmable logic devices, Rapid prototyping, Production facilities, Hardware, Advanced Analytics, Artificial Intelligence, Chemical Industry, Modular Plant, Module type package, PEA Engineering, PEA-as-Code, Process Industry, Soft sensor, VDI/VDE/NAMUR 2658