Optimization of 5-axis freeform surfaces machining : vector field clustering approach | |
| Author | Chu Anh My |
| Call Number | AIT Diss. no.ISE-05-01 |
| Subject(s) | Vector fields Mathematical optimization Machining Cutting |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. ISE-05-01 |
| Abstract | Five-axis CNC machines have become capable and more and more popular in manufacturing freeform surfaces which can be found in wide spectra of industrial products today. To improve the production rate and the economy, the machining time - the key factor in 5-axis machining - needs to be minimized or at least decreased. Optimizing the machining time could be performed in a number of ways; in particular, it can be performed in a way that maximizes the machining strip width. Therefore, this thesis introduces a new method for tool path optimization based on vector field clustering. The vector field is composed from optimal cutting directions which maximize the machining strip. The optimal tool path follows the streamlines of the field. Unfortunately, a complicated surface produces a complicated, non-uniform vector field. In this case it is nearly impossible to construct a reasonable tool path following the prescribed vector field without many disengagement of the tool. So partitioning the vector field into clusters, on which either the zigzag pattern or the spiral pattern can be considered for generating the tool path, is needed. For identifying the vector field, a local optimization algorithm is proposed. At first, the algorithm generates a grid on the surface. At every grid point, the algorithm determines the cutting profile in order to calculate the machining strip width; then the optimal cutting direction corresponding to the maximum machining strip is recognized. This process loops for all the grid points; in the end, the vector field of optimal cutting directions is constructed. The clustering technique proposed in this thesis is based on the spectral normalized cut method combined with the analysis and recognition of the spiral clusters. The technique considers the field as a weighted graph, where the nodes of the graph correspond to the vector positions, and every weighted edge connecting a pair of nodes of the graph reflects the similarity of the vectors pair. In this manner, the vector field clustering is transformed into the graph partitioning, and eventually, it is transformed into a standard eigen problem. Thus, the clusters are yielded. The Jacobian matrix, which characterizes the deformation rate of vector field, is analyzed for identifying spiral centers and spiral clusters of the field. The spiral pattern is considered to generate the tool path on the spiral clusters, and the zigzag pattern is used to generate the tool path on the remaining clusters. The iso-scallop method is employed to calculate the final tool path. Additionally, in the case of the spiral pattern the current iso-scallop scheme requires a special modification presented in this thesis also. Finally, we consider the proposed clustering technique with the reference to traditional iso-parametric schemes. The numerical experiments complemented by the real machining show that the proposed procedure is more efficient. |
| Year | 2005 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. ISE-05-01 |
| Type | Dissertation |
| School | School of Advanced Technologies (SAT) |
| Department | Department of Industrial Systems Engineering (DISE) |
| Academic Program/FoS | Industrial Systems Engineering (ISE) |
| Chairperson(s) | Bohez, Erik L. J.; |
| Examination Committee(s) | Tabucanon, Mario T.;Huynh Ngoc Phien;Makhanov, Stanlislav S.;Lauwers, Bert; |
| Scholarship Donor(s) | Ministry of Education and Training, Vietnam; |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2005 |