Siemens Aktiengesellschaft

Jean Pascal John

Thomas Bierweiler

Daniel Labisch

Bernd-Markus Pfeiffer

Method and system for diagnosing messages generated during operation of a technical plant, wherein data sets and a predetermined self-organizing map are used to determine diagnostic statements on the operation of the plant, n-tuples corresponding to the data sets and containing predetermined values for process variables for error-free operation of the plant are stored at nodes of the self-organizing map, after a message has occurred minimum distances between data sets of the current time interval and data sets of the nodes of the self-organizing map are determined, and the corresponding node(s) is/are selected, symptoms are then determined by calculating differences between the data sets of the time intervals and the previously determined node(s), only differences of process variables exceeding a predefined threshold value being taken into account, and where process variables involved in previously determined symptoms are output such that a plant operator is supported in diagnosing a process message.

Method and system for diagnosing messages

A method for automatically determining a parameterization of an evaluation method via a computer includes generating a comparison data set and storing reference data in the generated comparison data set, stipulating a parameterization with which identification of particular partial data items of reference data items stored in the comparison data set is influenceable via the evaluation method, such that a relative number of partial data items identified during the evaluation in relation to all the reference data items stored in the generated comparison data set is quantitatively smaller than a predetermined threshold value, preferably smaller than 10%, preferably smaller than 5%, particularly preferably smaller than 1%, and providing the determined parameterization for performing the evaluation method to identify particular partial data items of the data set of the industrial plant.

Growing community of inventors

Munich, Germany

Jean Pascal John