A neural network based tool for semi-automatic code transformation

A neural network based tool has been developed to assist in the process of code transformation. The tool offers advice on appropriate transformations within a knowledge-driven, semi-automatic parallelisation environment. We have identified the essential characteristics of codes relevant to loop transformations. A Kohonen network is used to discover structure in the characterised codes thus revealing new knowledge that may be brought to bear on the mapping between codes and transformations or transformation sequences. A transform selector based on this process has been developed and successfully applied to the parallelisation of sequential codes.

The Role of Tool/Sheet Contact in Plug-Assisted Thermoforming.

Plug-assisted thermoforming produces a wide range of polymer products through a combination of deformation by air pressure and contact with tool surfaces. In this paper the role of tool/sheet contact in determining the process output is investigated. A combination of thermoforming, friction and heat transfer tests were carried out on common tool and sheet materials. The results show that the typical friction coefficients for the material combinations are within the range 0.1 to 0.3, but the values rise sharply on approaching thermoforming temperatures. Thermal imaging tests demonstrate that all of the plug materials significantly cool the heated sheet on contact, even over very short periods of time. The temperature of the plug is very important. At low plug temperatures heat transfer effects predominate, whereas at high plug temperatures friction effects predominate. A plug temperature of approximately 100oC balances these effects and creates the most effective material distribution.

Modelling Vibrated Stone Colums in Soft Clay

The vibrated stone column technique is an economical and environmentally friendly process that treats weak ground to enable it to withstand low to moderate loading conditions. The performance of the treated ground depends on various parameters such as the strengths of the in-situ and backfill materials, and the spacing, length and diameter of the columns. In practice, vibrated stone columns are frequently used for settlement control. Studies have shown that columns can fail by bulging, bending, punching or shearing. These failure mechanisms are examined in this paper. The study involved a series of laboratory model tests on a consolidated clay bed. The tests were carried out using two different materials: (a) transparent material with ‘clay like’ properties, and (b) speswhite kaolin. The tests on the transparent material have, probably for the first time, permitted visual examination of deforming granular columns during loading. They have shown that bulging was significant in long columns, whereas punching was prominent in shorter columns. The presence of the columns also greatly improved the load-carrying capacity of the soft clay bed. However, columns longer than about six times their diameter did not lead to further increases in the load-carrying capacity. This suggests that there is an optimum column length for a given arrangement of stone columns beneath a rigid footing.