Surface design using functional blending
Surface design using functional blending S Bedi
Many engineering objects have surfaces that are critical to their performance. Analysis codes are used to design these surfaces for efficient performance. The iterative procedure of design followed by analysis requires the ability to generate and modify such surfaces rapidly. In the work described in the paper, surfaces which blend together bifurcated inlets are generated. A functional blendino method based on the Bernstein polynomial is developed in this work to generate a smooth surface between algebraically defined inlets and outlets. The method also has the ability to modify the generated surface both locally or globally, thereby assisting in surface modification. The method and its abilities are illustrated with examples. Finally, an application of the method to the design of an aeroplane turboprop housing is also discussed. functional blendin#,surface design, CaD~CAM In many engineering objects, the bounding surfaces are critical to the performance of the object 1. Such surfaces can be found in ships, cars, aeroplanes etc. Advanced surface-fitting techniques such as B~zier, B-spline and rational surfaces simplify the design of critical or aesthetic surfaces 2. However, there remain a variety of surface blends which pose difficulty in design as they consume large amounts of design time, and local or global modifications required on the surface are difficult to accommodate 3. Another method of surface fitting is the generation of surfaces using functional blending. This type of surface fitting is possible as a result of advances in computers and the development of symbolic-algebra systems. The functional-Mending technique stems from the method of generating a family of conic sections from three lines dl, #2 and d3. The conic sections are contained in the triangle formed by the three lines, and they are tangent to #1 and Department of Mechanical Engineering, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 Paper received:29 August 1991. Revised:20 March 1992
volume 24 number 9 september 1 9 9 2
#2 at their points of intersection with #34'5 . The conic sections are given by #1#22 + d ( 1 - 2) = 0
0 -~
Many engineering objects have surfaces that are critical to their performance. Analysis codes are used to design these surfaces for efficient performance. The iterative procedure of design followed by analysis requires the ability to generate and modify such surfaces rapidly. In the work described in the paper, surfaces which blend together bifurcated inlets are generated. A functional blendino method based on the Bernstein polynomial is developed in this work to generate a smooth surface between algebraically defined inlets and outlets. The method also has the ability to modify the generated surface both locally or globally, thereby assisting in surface modification. The method and its abilities are illustrated with examples. Finally, an application of the method to the design of an aeroplane turboprop housing is also discussed. functional blendin#,surface design, CaD~CAM In many engineering objects, the bounding surfaces are critical to the performance of the object 1. Such surfaces can be found in ships, cars, aeroplanes etc. Advanced surface-fitting techniques such as B~zier, B-spline and rational surfaces simplify the design of critical or aesthetic surfaces 2. However, there remain a variety of surface blends which pose difficulty in design as they consume large amounts of design time, and local or global modifications required on the surface are difficult to accommodate 3. Another method of surface fitting is the generation of surfaces using functional blending. This type of surface fitting is possible as a result of advances in computers and the development of symbolic-algebra systems. The functional-Mending technique stems from the method of generating a family of conic sections from three lines dl, #2 and d3. The conic sections are contained in the triangle formed by the three lines, and they are tangent to #1 and Department of Mechanical Engineering, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1 Paper received:29 August 1991. Revised:20 March 1992
volume 24 number 9 september 1 9 9 2
#2 at their points of intersection with #34'5 . The conic sections are given by #1#22 + d ( 1 - 2) = 0
0 -~
