Sunday, April 29, 2012

Fatih GÜNDÜZ 030060144 10th week Answers (Part 2)

High Energy Rate Forming : (New) (Forming)
   The susceptibility of a lattice to deformation is characterized by its tensile parameters. In general, energy densities of 50 to 1,000 J/cm3 (600 -12,000 ft • lb/in^3) suffice to alter the shape of metallic boundaries. At still higher energy densities, in the range between 10 and 100 kJ/cm^3, corresponding to pressures of 10^5 to 10^6 atm, the cohesive energy of the metallic bond is completely counteracted and the behavior of any lattice approaches that of a fluid.
   These two extreme situations define a region in which high-energy-rate forming is practically feasible. The conventionally used acronym HERF (high-energy-rate forming) implies two distinct requirements. First, we need high energy. This condition is readily visualized by the fact that expansion of a metallic boundary into a volume of 10^3 cm^3 entails typical energy levels between 50 kJ and 1 MJ. This is the typical energetic reservoir of mechanical presses operating in the 10- to 100-ton range.
   The second implication is that of high energy rate. This feature is not necessarily desirable from the viewpoint of metal forming. To the contrary, high energy rates entail increased strain rates, which in turn affect the dynamic tensile strengths of materials in a similar fashion, thus rendering the forming operation more demanding in terms of energy. However, high energy rates are expedient in two respects. First, we can relax the requirement for expensive megawatt power supplies, otherwise necessary for continuous operation. Second, the majority of methods available in high-energy forming are transient processes, as opposed to continuous processes. Hence, these methods constitute a suitable match to forming operations that are generally discontinuous and sequential (one object at a time).
   A combination of high energy and high power is therefore basic to HERF technology. We shall now analyze these energetic requirements in more detail, and we shall define the characteristic parameters of each method, together with its intrinsic limitations.
(New Trends in Materials Processin ,Taylor & Francis, p.1)


THERE İS NO OLDER DEFİNATİON !!!


Semantic ( Associative ) Networks : (New) (Knowledge Representation)
   Semantic network is one of the typical knowledge representation schemes for declarative knowledge. Special purpose computers which utilize the parallelism in semantic network have been proposed [1-3] , because computers with the Von Neumann architecture are unlikely to be candidates for processing large semantic network at a high speed. Those are focused on the declarative knowledge processing and are not intended to process procedural knowledge which is important to describe practical applications flexibly with semantic network.  
   Semantic network is a well known knowledge representation scheme. Knowledge is represented with nodes and links in a form of network. A node represents a unique concept and a link represents a relationship between two concepts. An important characteristic of semantic network is an inheritance of concept. Properties of super-class are inherited by sub-class along links (e.g. is_a links). Figure 1 shows an example of a semantic network representation. The address of computer_ division is inherited by every section along the is_a links.
   There are two basic phases in the knowledge processing system based on semantic network : (1) construction of semantic network, (2) information retrieval from semantic network. IXL is a semantic network language which supports those two phases efficiently. IXL provides two command groups. The first group is used for construction of semantic network. The second group is for queries. Semantic network construction commands have three arguments. The first argument is a link name which connects two nodes. The second and third arguments are node names which are connected each other by the first argument. For example, IXL command link(is_a, animal, dog) constructs a relation is_a between animal and dog.
(Database Machines and Knowledge Base Machines, Masaru Kitsuregawa, p.546)

THERE İS NO OLDER DEFİNATİON !!!

No comments:

Post a Comment