• Sumit Kumar Sar
• Dr. Ramesh Kumar

Breakdown maintenance is no longer affordable for industries as we now understand that the losses incurred on procrastinating CBM and real time preventive maintenance are too painful to be continued with. Novel techniques of the age old vibration signature analysis are being developed, such as HHT which has several advantages over the traditional FFT as it does not require the machine to be constant in its movement. Neural networks can now be Deep - trained to identify faults precisely based on the above analysis. Incorporation and simultaneous analysis of real time data enables immediate decision making using Random Forest decision trees. An assembly of the above technologies makes a significant difference in the exactness of fault diagnosis of machines.

Non-Stationary Machines, Fast Fourier transform, Hilbert Haung transform,

"SIGNAL PROCESSING FOR FAULT DETECTION IN NON-STATIONERY AND NON- LINEAR MACHINES: A COMPARISON", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 8, page no.890-894, August-2018, Available :http://www.jetir.org/papers/JETIR1808281.pdf

"SIGNAL PROCESSING FOR FAULT DETECTION IN NON-STATIONERY AND NON- LINEAR MACHINES: A COMPARISON", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 8, page no. pp890-894, August-2018, Available at : http://www.jetir.org/papers/JETIR1808281.pdf