| Popis: |
A device generating high-speed microjets from a liquid-filled microtube with laser-induced bubbles gets attention due to possibility of various engineering applications. In this study, we optimized the placement and bubble size of the laser-induced bubbles to maximize the efficiency of microjet generation. In the optimization, we use the Covariance Matrix Adaptation Evolution Strategy, which is one of the evolutionary algorithms. And a compressible multiphase flow solver based on the 5-equation model is used to evaluate the solution. Furthermore, data mining techniques using correlation analysis and dimensionality reduction algorithms are applied for extracting features related to the efficiency of microjet generation. Our result shows that it is effective to generate a laser-induced bubble near the tube wall. In addition, it is found that smaller volume and lower energy bubbles can provide efficient microjet generation. These results from the optimization and data mining are important for producing high-speed jets at low energy. |
