Onset of Nucleate Boiling Detection in a Boiler Tube by Wavelet Transformation of Vibration Signals

Abstract

Condition monitoring and fault diagnosis techniques are increasingly used for assurance of machine condition and safe operation of the systems and structures. For this purpose, preliminary data about the machine performance are obtained from the system and signal processing techniques are used to extract features and final information regarding the machine performance. One type of the primary data that can be obtained from a system is the vibration responses of the system. Vibration responses of dynamical systems are usually nonstationary signals. In boilers and heat exchangers, generation and collapse of the vapor bubbles due to overheating can generate vibration in the tubes. These vibrations could be detected and identified in terms of location and time of occurrence for future maintenance and modification of the system and process. In this paper, a nonstationary vibration analysis method based on Morlet wavelet transform is presented for detection of onset of nucleate boiling (ONB) in a boiler tube. Comparison of the results by wavelet transform and visual observation of the ONB shows the capability and accuracy (less than 3% difference) of the proposed technique for detection of ONB in a boiler tube.