Robust diagnosis with high protection to gas turbine failures identification based on a fuzzy neuro inference monitoring approach

Modern industry requires the development of new monitoring and diagnostic procedures, which enable the detection, localization, and isolation of faults. For sustainable solutions in terms of operational safety and availability, while bringing out zero accidents, zero downtime, and zero faults, for a trend acting on environmental issues. Towards this development, this work proposes solutions for the monitoring of gas turbines and their real-time implementation, in order to approximate and predict the degradation of the components of this system, by an approach of faults detection and isolation, based on an adaptive neural-fuzzy inference system. This will develop a reliable approach to maintain and monitor gas turbines, in case of failure or accident to prevent in real-time and makes it possible to achieve high power with efficiency and small footprint with High performance by operating this rotating machine. However, the application of the Adaptive Neuro-Fuzzy Inference System Observer-Based Approach, makes it possible to increase the life of the examined turbine and keep better reliability for their monitoring system and satisfy the techno-economic and environmental performance impacts. For the purpose of controlling failures and the occurrence of turbine system malfunctions, and avoiding their consequences on the safety and productivity of the installation.

Automated summary

The development of new monitoring and diagnostic procedures is essential for modern industry to enhance operational safety and availability. The proposed solution for gas turbine monitoring incorporates an adaptive neural-fuzzy inference system to predict component degradation and achieve high efficiency and reliability.