European Commission logo
English English
CORDIS - EU research results
CORDIS
Content archived on 2024-05-07

Model Based Diagnosis of Rotor Systems in Power Plants

Exploitable results

All large rotating machines installed in power plant are normally equipped with vibration sensors and their signals are continuously monitored. The information in this data, however, is not fully utilized to diagnose the nature, location and severity of machine faults. A model based diagnostic system for rotating machines offers the advantage that it can make fuller use of the information in the vibration data being collected. It has the potential to give machine or plant operators more quantitative advice, allow more efficient predictive maintenance, decrease the percentage of out-of-service time of the power plants and help the operator avoid catastrophic failures, some of which have occurred both in Europe and elsewhere. The MODIAROT consortium has developed the necessary components for a model-based diagnostic system and has demonstrated its feasibility. The diagnostic system used models of the plant rotor-bearing-foundation system and models of possible faults or malfunctions. Faults are detected and quantified by comparing on-line data from the plant and simulated output from the model, using advanced computational methods. For this to be successful, it has been necessary to create models for complete rotor-bearing-foundation systems that are sufficiently accurate for the purpose. It has also been necessary to develop and verify fault models. The diagnostics is a two-stage process. First a rule-based expert system is used to rank the faults and the second stage diagnostics then determines the size and location of the most likely faults. The second stage diagnostics compares the model-with-fault data with the plant data and adjusts the fault parameters in order to minimize the difference between the model and the plant data. Three algorithms can be used for this purpose, two are based on the least-squares method and one uses fuzzy clustering. At the conclusion of the project, we can report that most of these requirements for a model based diagnostic system are in-place and a 'software diagnostics demonstrator' has been written.

Searching for OpenAIRE data...

There was an error trying to search data from OpenAIRE

No results available