Search

For the development of performance based design on a proper scientific basis the use of the concept of risk is inevitable. However, the application of this concept to actual structural design is not simple because of the large ranges of probability and consequences of events which exist. This is compounded by a plethora of different actions that can be taken to reduce the probabilities of the events and also the magnitude of the consequences. It is the reduction in the magnitude of these consequences which is essentially the goal of design. This work aims to address the challenges posed by the application of the concepts of performance based design for structures in fire. Simple methodologies have been developed for the assessment of the consequences of an extreme event. These methodologies are based upon fundamental behaviour of structures in fire. A methodology has been developed which can be used to assess the capacity/deflection behaviour through the complete thermal deflection of floor slabs. This takes into account positive effects on the capacity of floor slabs of the membrane stress at the slabs boundaries at low deflections as well as the final capacity provided by the tensile membrane action of the reinforcement mesh at high deflections. For vertical stability of structures in fire, analytical equations to describe the behaviour of floor systems at the perimeter of a building are developed. From these equations, the resulting pull-in forces on external columns can be calculated as well as the resulting horizontal load applied to the column. From this, a simple stability assessment is proposed which can be used to assess the consequences of multiple floor fires on tall buildings. These analytical methodologies are brought together in a risk based frame- work for structural design which can be used to identify areas in a building or structural components which pose a high residual risk. These elements can be qualitatively ’ranked’ according to their relative risk and appropriate measures taken to reduce the risk to an acceptable level. The framework is illustrated via 2 case studies. The first is of a typical small office building, and the second is of a prestige office development.

Described is the design specification for an Artificial Neural Network (ANN) communication controller. The communication controller will provide efficient data exchange between individual ANN units in a large parallel system. Interconnections are dynamic in nature, and can be configured to implement any arbitrary feedback path. A nexus topology is used to interconnect the communication controllers in a system. Because the recursive nature of the communication protocol, the communication controllers function identically regardless of position in the hierarchy, providing scalability and simplifying VLSI design. The protocol and the communication controller's modes of operation are presented in detail.