Distributed memory systems for simulating artificial neural networks

In executing tasks involving intelligent information processing, the human brain performs better than the digital computer. The human brain derives its power from a large number [O(10¹¹)] of neurons which are interconnected by a dense interconnection network [O(10⁵) connections per neuron]. Artificial neural network (ANN) paradigms adopt the structure of the brain to try to emulate the intelligent information processing methods of the brain. ANN techniques are being employed to solve problems in areas such as pattern recognition, and robotic processing. Simulation of ANNs involves implementation of large number of neurons and a massive interconnection network. In this paper, we discuss various simulation models of ANNs and their implementation on distributed memory systems. Our investigations reveal that communication-efficient networks of distributed memory systems perform better than other topologies in implementing ANNs.     

Analysis of conduction heat transfer problems in steady, 2D regions with circular holes using a special boundary integral method

A special boundary integral method developed for two-dimensional regions containing circular holes is used to calculate temperature and heat transfer on the boundaries of several selected regions. The geometrical configuration of the region is arbitrary and convective boundary conditions are assumed. An important feature of the method is analytic representation of temperature and its normal derivative on the interior circular holes in the form of a harmonic series. This makes the application of the boundary integral method convenient and free from conditioning problems associated with small interior boundaries. Heat transfer from circular isothermal interior holes are calculated for several illustrative examples using three terms of the harmonic series representation for heat transfer at each of the circular boundaries. The results are presented and discussed.     

An integrated modeling approach to predict flooding on urban basin.

Correct prediction of flood extents in urban catchments has become a challenging issue. The traditional urban drainage models that consider only the sewerage-network are able to simulate the drainage system correctly until there is no overflow from the network inlet or manhole. When such overflows exist due to insufficient drainage capacity of downstream pipes or channels, it becomes difficult to reproduce the actual flood extents using these traditional one-phase simulation techniques. On the other hand, the traditional 2D models that simulate the surface flooding resulting from rainfall and/or levee break do not consider the sewerage network. As a result, the correct flooding situation is rarely addressed from those available traditional 1D and 2D models. This paper presents an integrated model that simultaneously simulates the sewerage network, river network and 2D mesh network to get correct flood extents. The model has been successfully applied into the Tenpaku basin (Nagoya, Japan), which experienced severe flooding with a maximum flood depth more than 1.5 m on September 11, 2000 when heavy rainfall, 580 mm in 28 hrs (return period > 100 yr), occurred over the catchments. Close agreements between the simulated flood depths and observed data ensure that the present integrated modeling approach is able to reproduce the urban flooding situation accurately, which rarely can be obtained through the traditional 1D and 2D modeling approaches.     

Optimizing the design of railway tank cars to minimize accident-caused releases

The design of vehicles transporting hazardous materials has important public safety and economic implications. Conventional wisdom among industry and government has held that a thicker tank on railroad tank cars and trucks reduces risk. However, a thicker tank increases vehicle weight and thus leads to an increase in the number of shipments required to transport the same amount of product and consequently greater exposure to accidents. In this research we develop a model that analyzes the tradeoff between increased damage resistance and greater exposure to accidents in which the objective function is minimization of the probability of release. The model accounts for the reduction in tank car release probability as a function of tank thickness, and the increased exposure to accidents that occurs due to the increased number of shipments needed for the heavier car. Three variables affecting this optimal thickness are considered in this paper: the volumetric capacity of the tank, the probability of release from other, non-tank sources, and the weight capacity of the car. Sensitivity analyses using the model indicate that for any particular configuration of tank car there is an optimal thickness. This optimal thickness is affected by several factors and there is no single optimum for all tank cars.