Impact of Fouling on Flow-Induced Vibration Characteristics in Fluid-Conveying Pipelines

Abstract

This paper has addressed the common monitoring problems in petrochemical companies, which are caused by fouling and clogging in the circulating water heat exchangers, and has introduced techniques to monitor the heat exchanger's wall vibrations for early failure detection. Due to the difficulties encountered in simulation caused by the large number of tubes inside the heat exchanger, such monitoring methods are discussed by studying the fouling of a fluid-conveying pipeline. ANSYS was used to establish the normal and fouling models of a fluid-conveying pipeline so as to analyze the changing rules of various parameters that are influenced by different inlet velocities. These parameters include flow velocity, direction, pipeline wall load, wall displacement, and the acceleration of fluid domain in a fluid-conveying pipeline. It is shown that, as the inlet velocity and fouling severity continuously increase, the wall load and the vibration acceleration increase as well, leading variations in wall vibration signals. This paper conducts extensive experiments by using straight pipes to compare the results from simulation with those from the normal fluid-conveying pipelines under the same working conditions. By such comparison, the efficacy of the simulation model has been demonstrated.INDEX TERMS Fluid-conveying pipelines, fouling impact, vibration characteristics, vibration signals.

Failure Detection Methods for Pipeline Networks: From Acoustic Sensing to Cyber-Physical Systems

“…Variations in vibration signals caused by fouling and clogging pipelines were also studied in [ 11 ], using finite element models of fluid-conveying pipelines, developed using ANSYS. In [ 71 ], a multi-feature fusion technique based on features of wavelet energy entropy, approximate entropy and fractal box dimension, extracted from acoustic signals collected from the pipelines, was proposed.…”

“…The classification of the feature sets was achieved using a SVM classifier that was optimised using the particle swarm optimisation (PSO) algorithm. Unlike the method in [ 38 ], the approach in [ 11 ] detects the presence of blockages in pipelines by distinguishing abnormal vibration signals from the normal ones, using machine learning.…”

“…The method in [ 11 ], despite being more complex computationally due to the need to collect a large data sample for training, is more cost-effective for large-scale deployment since it eliminates the need for using a large number of sensors. Both of the methods proposed in [ 11 , 38 ], respectively, have different emphases.…”

“…The method in [ 11 ], despite being more complex computationally due to the need to collect a large data sample for training, is more cost-effective for large-scale deployment since it eliminates the need for using a large number of sensors. Both of the methods proposed in [ 11 , 38 ], respectively, have different emphases. The study of [ 38 ] focused on the accurate localisation and estimation of the sizes of blockages while that of [ 11 ] gave weight to fast qualitative identification of blockages in a pipeline network by eliminating real-time signal processing.…”

An Experimental Access of Spiral Plate Heat Exchange Technique in Immiscible Liquid Mixtures

“…SHEs can attain higher convective HT rates owing to the spiral patterns which retain the turbulent flow [2]. The circulating water-HE is extensively used for exchanging heat in contemporary petro-chemical enterprises as well as accounts for forty percent of the total investment in facilities [3].…”