Browsing by Author "Kajola, Paavo"

Filter results by typing the first few letters
Now showing 1 - 3 of 3
  • Dynamic Adapter Connections for IEC 61499
    (2021-06-18) Kajola, Paavo; Blech, Jan Olaf; Dwi Atmojo, Udayanto; Vyatkin, Valeriy
     A4 Artikkeli konferenssijulkaisussa
    We present work on dynamic adapter connections for IEC 61499. Adapters in IEC 61499 simplify programs by encapsulating different event and data connections. Traditionally the plug and sockets are determined at design time of a system. In contrast, dynamic adapter connections do not statically link subsystems at design time, but allow to exchange interacting subsystems at runtime by retargeting plugs and sockets. This paper presents the dynamic adapter connection concept. In addition, we present an implementation concept based on OPC UA and an example.
  • IEC 61499 based distributed data collection framework for multivariate time series data
    (2024-01-22) Kajola, Paavo
     Sähkötekniikan korkeakoulu | Master's thesis
    Modern industrial systems create a vast amount of data that must be collected for analysis. This data is not necessarily easily accessible at once. Components of the system may expose their data in the network through communication protocols such as OPC UA. If the equipment hosting the OPC UA server lacks tools to actively push data to some permanent storage, such as a time series database, a data collector must be implemented to fetch the data from the endpoint and write it to the TSDB. IEC 61499 can be leveraged in creating the collector. It allows an application developer to quickly assemble and configure the collector out of modular parts, contained in Function Blocks. Service Interface function blocks can also be used to implement an interface for the analysis software, allowing the same developer to distribute analysis tasks to devices in the network using IEC 61499's device model. This is especially useful since simple data ingestion from source to TSDB may impose a relatively light computation load, but analysis will often require a powerful PC with a GPU. This work presents a data collection and analysis FB library for the 4DIAC Forte runtime environment. The FB library contains tools to route data from OPC UA servers to a TSDB, filter out unwanted data, and instantiate and configure Python based analysis algorithms and interface them via SIFBs.
  • Learning image recognition systems in factory automation
    (2017-09-26) Kajola, Paavo
     Sähkötekniikan korkeakoulu | Bachelor's thesis