Operational information system in a power plant

We present an open supervision system capable of monitoring two generation groups in a power plant (about 1200 analogic variables per group) with a sampling period of one minute. The collected data is stored in a relational database such as Oracle with HHCODE technology. As far as this subsystem is a key for process monitoring, the database operates in a secure mode and a client/server tool has been designed under tcl/tk+ oratcl to allow any kind of queries asking for any relations at any time in a graphical way. Also a visualization schema subsystem connected to the database allows to "see" the plant diagrams with real values or with function transformation values. The main advantages of this system over the classical SCADA systems are the independence of design for its storage block, improved with a RDBMS system, and the graphical query subsystem. With this kind of design it is possible to apply data mining technology in order to look for process-related dependencies and several other techniques. A powerful backup technology has been provided too, in order to have an easy path to retrieve old stored data (older than six months, which is the persistence of data on the database). Data is recovered automatically from an optical jukebox and is deleted automatically one week later.