Study On Turbulence Development Of Abrasive Flow In Large Curvature Structural Surface

Large curvature structural surface with small size is very typical in mould structural surfaces. They are hard to be processed by tools because of being limited to their structural characteristics, therefore, the soft abrasive flow machining (SAFM) method for mouldstructural surface was proposed. In SAFM, the soft abrasive flow (SAF) will present different flow patterns as machining control conditions are different, thus the flow field characteristics of SAF and effective machining features are different. The thesis studied with flow characteristics and variation showed for turbulence development of SAF in U constraint flow passage under different flow patterns. Meanwhile, flow patterns formed in the process of turbulence development were identified, so as to realize the control of flow patterns and lay a foundation for making and implementing process route of SAF. According to the above research purpose, the study method for the turbulence development process of SAF based on Renormalization Group (RNG) k-εmodel and the identification method for flow patterns based on artificial intelligence were put forward. The mainly research contents are as follows:1. Based on theory of solid-liquid two-phase flow and particle kinetic, abrasive flow separation, friction loss and particle dynamic characteristics etc in large curvature constraint flow passage were studied.2. Aiming at different flow behavior in U curvature flow passage, the controllable physical model which can realize flow pattern transformation was built, and soft abrasive flow was simulated. By using Eulerian and RNG k-εmodel, the soft abrasive flow characteristics being related with processing effect were analyzed.3. In order to realize the effective control for soft abrasive flow in constraint flow passage, the flow patterns should be identified. Therefore, a method of flow pattern identification based on artificially intelligence was put forward. By means of getting pressure signal reflecting the change of flow pattern, wavelet packet analysis and probabilistic neural network, the flow pattern identification was realized.4. By putting up the experimental platform for soft abrasive flow machining, machining experiments of U large curvature structural surface were made. The soft abrasive flow in constraint flow passage was controlled in different stable flow state. Actual finishing effect under different flow patterns was obtained. Experimental results proved the validity and reliability of theory research, meanwhile, verified the feasibility of the precision machining without tools method based on soft abrasive flow.