Fuzzy sets based knowledge systems and knowledge elicitation

Fuzzy sets are adequate forms of knowledge representation when the information is uncertain due to vagueness and imprecision. Knowledge structures using fuzzy sets are similar to those implemented in non-fuzzy systems. Classical knowledge elicitation methods can be used in combination with techniques to develop membership functions. The fuzzy set representation has several advantages, including flexibility in expressing uncertain knowledge during elicitation, representation of the knowledge and its uncertainty as a unique entity, easy interfacing with classical systems, and a more robust system in ill-defined domains. These advantages result in increased system reliability.     

High modulus of isotactic polypropylene attained by coextrusion with atactic polystyrene

Orientation-induced crystallization of crystallizable polymer melts can occur, under certain conditions, during flow through converging channels. Attempts have been made to achieve this phenomenon in a two-phase system, i.e., during simultaneous extrusion of a continuous concentric core of polypropylene within a polystyrene matrix through a conical duct. On one occasion, using Carlona P SY6100 (MFI = 11.0) with Hostyren N2000-V-01 (MFI = 25.0), a highly oriented polypropylene thread with a modulus of 14.6 GPa and a melting point of 178°C was extruded at a die temperature of approximately 170°C and a pressure lower than 40 MPa. It is, in principle, possible to form highly oriented, fiber-like structures as reinforcing elements in a polymer matrix.     

Quantitative determination of non-defect unstable structures in PVC through selective nucleophilic substitution with C6H5SNa

Summary The nucleophilic substitution with C₆H₅SNa, at -30°C, is compared for five PVC samples with various contents of isotactic triads. The conversion curves consist of a very fast stage followed by a steady one. All the samples are found to behave in the same way except for the content of the structures involved in the fast period. This content is estimated by extrapolating the straight lines of the steady stages to zero time. The obtained values are found to be a linear function of the content of isotactic triads. The results, together with some earlier ones, allow for the content of the labile non-defect structures in PVC to be determined.     

Heyman's assumption for reinforced concrete arches re-examined

In a previous paper [1] the well-known Heyman assumption (H.A.) for the limit design of a masonry arch was shown to be valid also for the limit design of a box-section reinforced-concrete arch. In [2] the more general case of failure under variable loads (incremental collapse or alternating plasticity) was discussed. Use of the H.A. avoids the construction of the elastic domain, that is the set, for each section, of all possible states MN in the elastic range. This is one of the great difficulties of classical analysis, if moving live loads must be considered, for example for bridge structures. The H.A. allows the stresses to be calculated at the outer fibres in the generic section, referring to only the intradosal and extradosal moments, so that it is possible to obtain the maxima of such stresses by using influence lines.Temperature changes have also been considered, as have wind and earthquake loading.Paper [2] deals with these cases, and the results agree well with the exact results. The present note is a complement to the previous work. In it, a plastic hinge is allowed to occur taking into account cracking of the concrete. The H.A. is shown to be also valid if the live load line of thrust is close to the centre-line of the arch, or the plastic safety factor of all the loads is required, dead load included.     

Energy function analysis of dynamic programming neural networks

All analytical examination of the energy function associated with a dynamic programming neural network is presented. The analysis is carried out in two steps. First, the locations and numbers of the minimum states for different components of the energy function are investigated in the extreme cases. A clearer insight into the energy function can be gained through the minimum states of different components. Secondly, the locations of the minimum states of the energy function using different parameter values are derived. It is shown that the minimum states can reside in regions which are regarded as valid solutions with certain conditions. Examples and simulation results are given to justify the validity of the theories developed.