This paper presents a way to optimize design of large built-in self-test (BIST) programmable logic arrays (PLAs). These PLAs can be tested at clock speed with function independent test set. Hardware overhead of the design is low compared to other techniques. In the design, test pattern generators are simple shift registers connected in ring counter form. Response evaluator circuit is a signature analyzer. A two bit binary counter and two D flip-flops automate the design process and reduce the number of test control pins. The PLA can detect all stuck-at, crosspoint, bridging as well as stuck-open faults.

Experiments were done to obtain a correlation between pressure drop across a chromatographic bed packed with a compressible packing, Sephadex G100, and other measurable parameters such as flowrate, column length and diameter, particle size, and permeability. The pressure drop across the bed is obtained by subtracting the extracolumn pressure drop from the operating pressure drop. It was found that the pressure drop increased nonlinearly as the flowrate increased. There is a critical flowrate after which the pressure drop increases infinitely. The pressure drop was also found to be dependent on the column diameter within the range tested. The data were fitted well with an equation similar to the Blake-Kozeny equation but with an exponential term and a critical factor included.

Faraid is a division of inheritance according to Islamic laws. This paper describes a design and implementation of an initial logical circui for faraid calculator. The faraid calculator which has been implemented can be used to find the qualification of receiving inheritance for each 25 heirs and also the respective proportion of inheritance if qualified. The faraid solution is based on Mazhab Shafte.

The main objective of this paper is to serve as an introduction to the process of designing and training of a neural network. A simple implementation of a nonlinear separable function (e.g. exclusive OR function) using the backpropagation training algorithm and feedforward multilayer perceptron is used as an example. Finally, the paper summarizes the advantages and disadvantages of the Neural Networks. In addition, it also discusses the implementation of the neural network appraoch to voice/unvoice problems.

Hydrodynamically lubricated isotropic roughness bearing with the misalignment of the journal is analysed. A modified Reynolds’ equation which includes surface roughness and misalignment effects is solved numerically using finite difference method with the aid of successive-over-relaxation method. The average flow model is employed in evaluating hydrodynamic pressure due to isotropic roughness and misalignment effects. The bearings’ characteristics’ values is solved numerically using Simpson’s rule. The results of the numerical analysis explore the effects of both surface roughness and misalignment on the design parameter of journal bearing. The degree of misalignment and roughness structure greatly effects the performance of the hydrodynamic bearings. Controlling the roughness and degree of misalignment is an important design factor particularly for bearings operating at minimum film thickness and high eccentricity ratio.

Two novel hierarchical structures are presented which extend the applicability of previous model based double iterative loop techniques to non-convex problems. The methods incorporate integrated system optimisation and parameter estimation which utilizes process measurements to achieve real process optimality inspite of model reality differences. The double iterative loop structures of the proposed algorithms use the real process measurement within the outer loops while the inner loops involve model based computation only. This means that the algorithms use available information from the real process efficiently and a significant reduction in set-point alterations to real subprocesses is achieved. In order to cater for process complexity the inner loops are organised in the form of two level hierarchical structures. The paper presents the convergence conditions of the augmented techniques and simulation examples are provided to illustrate and compare the methods.