عنوان مقاله [English]
Flanged joints’ looseness is among the common causes for the failure of industrial structures with flanged joints such as wind turbines or transmission line pipes the timely detection of which can prevent the imposition of heavy, financial losses and in some cases property damage. Conventional methods for detecting this fault, such as torque control methods, have high error of measurement, or impedance-based measurement methods, have high expenses, or vibration or ultrasonic methods which lack accuracy due to the use of linear phenomena in fault detection. Vibro-acoustic modulation method is one of the nonlinear fault detection methods that detect and evaluate looseness of flanged connection with high precision through the measurement of the intensity of the vibrational and ultrasonic signals modulation applied to the structure. Although the published papers in recent years have often identified the cracking, delamination, or corrosion and decay of parts using this method, in this paper the efficiency of the vibro-acoustic modulation method in the detection and evaluation of flanged joints of simulated wind turbine towers has been numerically investigated by defining an index for modulation intensity and bond relaxation. Then, the effect of parameters such as ultrasonic and vibrational frequency, amplitude of ultrasonic and vibrational stimulation, sensors and actuators position, as well as preload force, on the method performance has been studied. Finally, in order to reduce the simulation time in ABAQUS software, the modeling of the neural network was performed using MATLAB software and the obtained results were compared with numerical results.