PhD student
Sloshing control in a moving container by video analysis and genetic algorithm
Thesis subject : Sloshing control in a moving container by video analysis and genetic algorithm
Summery
Liquid sloshing is a phenomenon that may affect moving containers filled with liquid and cause the liquid leakage under high amplitudes. This work aims at proposing a sloshing control of a liquid-filled rigid container moving in a walking beam transfer system, applicable to any containers’ shape under condition of transparency of its walls, and without the use of a physical model. Notably, the control law of the walking beam transfer system is optimized by the Genetic Algorithm and the movement of the liquid is measured by video analysis.At first, two optimization algorithms, Gradient Descent and Genetic Algorithms, are assessed on numerical simulations obtained for a rectangular container with the aid of a simple fluid dynamics model. The objective function is defined to minimize the maximum elevation observed during the movement of the fluid. The algorithms are compared in terms of the maximum elevation of the liquid and the number of iterations required to obtain the convergence, Genetic Algorithm giving quicker convergence. Then, experimental tests are performed on a production line in the cosmetic industry. The oscillations of the liquid surface are captured by a camera and detected in real-time. The maximum wave amplitude is evaluated at each iteration and corresponds to the objective function to be minimized by the Genetic Algorithm. The results show that the optimum control law is specific for each container and reduces the liquid sloshing with respect to the common control laws used in the industry. The proposed method for sloshing control is versatile, cheap in computational time, and free from complex physical models.