Search

Fluid conveying pipe used in different fields such as petrochemical, oil, gas, hydraulic structures in transmission lines.The damage caused by vibrations damaging pipes or other components of a power plant industrial plants are related to each other, one of the main reasons for the loss of production time is considered damage and ultimately fires in industrial plants. So make a way as to identify damage in the fluid conveying pipe is of particular importance.The purpose of this study, the effect of the fluid inside the pipe vibration characteristic is a theoretical approach.Then for forced vibration analysis of fluid pipe considered passes through the external moving oscillator. And then a method to detect damage in pipes recommended based on fuzzy-genetic algorithm. First, fluid conveying pipe dynamic equations, modeled interaction between external moving, pipe and fluid then solving vibration system, acceleration time history extracted from midpoint of pipe. Subsequently, a new method of damage detection in simply supported pipe is introduced based on fuzzy-genetic algorithm. The new method is capable of identifying the location and severity of the damage. This algorithm is developed to detect the damage location along the beam, which can detect the damage location based on the pattern of beam frequency variations between undamaged and damaged states. The severity of damage is assessed based on quantitative variations in frequencies.

Nowadays, increasing use of electronic instruments and nonlinear loads in Power systems, make the power quality problem as one of the most important issues. In this article, the produced data from mathematical equations and PSCAD software simultaneously have been used to simulate power quality disturbances. Because of super performance of neural networks in pattern recognition and classification, the MLP neural network for classification of power quality disturbances is used in this paper. The neural networks have been developed by simulation of nonlinear terms, and they indicated their priority for pattern recognition and classification. STFT and DWT transform to extract signal's features have been used. After classification of disturbances using MLP, the neural network robustness has been examined in different levels in presence of the noise. With presence of noise, neural network classifies all the events with 98.22 percent of accuracy. Finally, results of this article are compared with other researcher's works.