International Journal of Engineering Research and General Science

6-11

Authors: GIRISH M. DHOTE1, ANIL V. KALE2

In this paper a neural network-based approximation method is used to predict optimized punch entry during piercing operation. Five materials are used with varying thicknesses and punch diameters for experimentation against optimal punch entry as a measure during piercing. Rigorous experimentation was carried out and the tabulated data is used for the numerical simulation using Artificial Neural Network. Figurative mathematical model is initiated to reveal the nature of Neural Network. Scaled conjugate gradient algorithm is considered as optimization algorithms for ANN, which performs well with diversified topologies. Coefficient of correlation (R = 0.997) with the sum of square error in the range of (0.001-0.01) reflecting a better approximation of dependent variable. Key words: Piercing, Punch entry, Neural network, Diverse topologies, Coefficient of correlation

17-31

Authors: Dr. Chandrasekhar Putcha

In their seminal work, Burns and Mitchell’s (1946) definition of business cycles has two key features. The first is the co-movement or concurrence among the individual economic indicators, and the other feature is that business cycle is governed by a switching process between different phases. In general, extracting the co-movement among the economic indicators leads to the creation of the composite Economic indicators (CEI), which enables policy-makers to define the current state of the aggregate economy. That is, the creation of the CEI would inform us whether the economy is currently experiencing a slowdown, a boom, or whatever. This paper uses three important economic variables in this study - Average Income per year, (AYPY), Unemployment rate per year (URPY), and Housing Prices per year (HPPY). A new method is developed for calculation of CEI in this paper. The study is done in two phases. In phase I, these are used to calculate the Composite Economic Index (CEI) for 4 key states - Texas, New York, Florida and California and in the second phase CEI is calculated for the whole of USA.

49-55

Authors: Viraj Bhusari

Test-per-clock BIST scheme is used to determine for small test vector and short test time. This paper presents a new test-per-clock BIST method that can perform both pattern generation and response compression concurrently in the same LFSR-based design so as to reduce the area overhead. Furthermore, some internal nets are employed in two ways during test application to help reduce test time and test vector: 1) as the observation points to enhance fault detectability and 2) as the test data provider for reseeding the LFSR. These two ways lead to the benefits that all required patterns can be generated on chip and at-speed testing can be carried out without using any external or internal storage device. Experimental results show that the presented method can achieve 100% fault coverage in very short time for large ISCAS (IWLS) benchmark circuits. When compared with a conventional scan-based design, the area overhead is small considering the features of test-per-clock and no requirement of data storage element.

66-71

Authors: M. Rahman, A. Elahi, and A. Rashid

Many fire-based accidents are occurring in Industries now. Consequently, massive losses of lives, as well as money, are suffered by the owners. The conventional fire extinguishing systems are found to be not enough to reduce losses, including eliminating them. Most are conversant in the fact that the large fires originate from very small sparks or pretty little fires which are very easy to manage. So, in this research, the main objective was to easily control these small fires at the initial phase. By using the traditional fire extinguisher, it's not always possible to extinguish a small fire. One of the main reasons for this inability is that the manual handling nature. So, the authors worked with the automated extinguishing equipment and tried to enhance them to extinguish the small fire at a very initial stage and prevent huge fire which will cause massive losses. In this research, a sustainable fireball was designed and manufactured using locally available raw materials. In the ball, dry powder is used as an extinguisher referred to as ABC dry chemical powder. A stands for ash, B stands for barrel while C stands for current. The powder is typically a mixture of mono ammonium phosphate and ammonium sulfate, the previous being the active one. The combination between the two agents is typically 40–60%, 60-40%, or 90-10% counting on local standards worldwide. An innovative fireball extinguisher system has been developed using this powder to extinguish the fire quickly and conveniently. In this paper, the design and fabrication procedures of an innovative fireball extinguisher system are described along with the test result

89-96

Authors: Ms.Neetu Singh Diwakar1, Dr. J.N.Vyas2

Structural defect prediction in RC buildings, especially multi-storied buildings has long been an active area of research among structural engineers. While several standards pertaining to different geological conditions are available and are adhered to while structural design, yet it has been seen that due to external factors and non-compliance with exact standards, there exists a chance of building failure. Conventional techniques rely on pre-defined standards, but due to minute details and random nature of values, it is difficult to find regular relations or patterns among them. In this paper, standardized RC building data has been used to train an artificial neural network and subsequently is has been used to test the neural network model for predicting building failure. A total number of fifty RC buildings have been utilized for data mining. As a standard convention, 80% of the data has been used for training and remaining 20% data has been used for testing and validation. It has been found that the designed ANN structure predicts with approximately 98.51% accuracy which can be attributed to the efficacy of the Bayesian Regularization algorithm used for employing back propagation in the designed ANN structure. It may possible that the present research or proposed methodology using ANN technique may provide a guideline to detect defect or failure prediction in structural RC buildings with high accuracy.