434 / 2022-08-12 10:00:33

Extracting Viscous Drag from Hydrodynamic Forces Based on Exponential Function

Database, Viscous drag, Exponential function, Surrogate model, Virtual simulation

Control and optimization of dynamic systems

According to Morison formula, the hydrodynamic forces of one submerged body are divided into the viscous drag and additional mass effects. The viscous drag and the additional mass force are related to the relative velocity and acceleration of the submerged body, respectively. To obtain the surrogate model of the towed vehicle for the deep-towed multichannel seismic exploration system, the database reflecting the relationship among the viscous drag, towed velocity and pitch angle of the towed vehicle should be established using an efficient and economical method. Building database is one critical process in machine learning, and the sampling data is obtained from virtual simulations here. The towed vehicle is accelerated first to derive the expected velocities, but it takes a long time for the inertia force generated by the acceleration to be faded down. The analysis results show that the hydrodynamic force fade exponentially, therefore, the exponential function is proposed to extract the viscous drag efficiently from the hydrodynamic forces herein. Two virtual scenarios, 1000 mm/s towing speed without pitch angle and 600 mm/s with 8-degree pitch angle are designed to verify the proposed method. The results show that the steady viscous drag can be extracted from merely 40% of the load history and the relative error between the extracted value and the value obtained after a long-time calculation is less than 1%. Thus, 60% of the calculation time can be saved by the exponential function method in building database. Based on the proposed exponential function method, the database is established efficiently, and the surrogate model that reflects the relation between the steady towing velocities and viscous drag of the towed vehicle is derived efficiently and economically.