Browsing by Author "Asafa, T. B."

Now showing 1 - 6 of 6

Aircraft Disasters- roles of materials
(2009) Asafa, T. B.; Durowoju, M. O.; Ismail, O. S.
Aircraft disaster has been in existence since air was conquered by man as a means of transportation. 487.5 million and 874.4 millions of cumulative departures and flight hours respectively have been estimated since 1959. Analysis of aircraft failure based on 5,149 on-board fatalities recorded shows that 13% of total aircraft accident was caused by mechanical failure while loss of control was responsible for over 31% of onboard fatalities. Aircraft accident is known to be most fatal during take-off and landing phase contributing about 49% while onboard fatality during cruise is about 19%. In this work, reviews of aircraft disasters were made via Fractographic examination, SEM and finite element modeling. It must be stated that few of aircraft failures which are not material related are not considered in this review. The review focused on material related failure which have been analyzed, accepted and published in reputable journals.
Artificial neural network predictive modeling of uncoated carbide tool wear when turning NST 37.2 steel
(Asian Research Publishing Network, 2012-04) Asafa, T. B.; Fadare, D. A.
We report the development of a predictive model based on Artificial Neural Network (ANN) for the estimation of flank and nose wear of uncoated carbide inserts during orthogonal turning of NST (Nigerian steel) 37.2. Turning experiments were conducted at different cutting conditions on a M300 Harrison lathe using Sandvic Coromant uncoated carbide inserts with ISO designations SNMA 120406 using full factorial design. Cutting speed (v), feed rate (f), depth of cut (d), spindle power (W), and length of cut (l) were the input parameters to both the machining experiments as well as the ANN prediction model while the flank wear (VB) and nose wear (NC) were the output variables. Nine different structures of multi-layer perceptron neural networks with feed-forward and back-propagation learning algorithms were designed using the MATLAB Neural Network Toolbox. An optimal ANN architecture of 5-12-4-2 with the Levenberg-Marquardt training algorithm and a learning rate of 0.1 was obtained using Taguchi method of experimental design. The results of ANN prediction show that the model generalized well with root mean square errors (RMSE) of 3.6% and 4.7% for flank and nose wear, respectively. With the optimized ANN architecture, parametric study was conducted to relate the effect of each turning parameters on the tool wear. The ANN predictive model captures the dynamic behaviour of the tool wear and can be deployed effectively for online monitoring process.
Detection of the point of crack initiation using multi-stage random sampling (MRS) and spatial point pattern (SPP)
(Blackwell Educational Books, 2010) Durowoju, M. O.; Asafa, T. B.; Ismail, O. S
Porosity is a major defect in cast aluminum alloys affecting in particular, the fatigue strength. The pores serve as points of stress concentration and points of crack initiation for eventual failure. In this work, fractal analysis was used to numerically characterize the pores in uni-directionally solidified Al 4.5 wt % Cu alloy micrographs, transverse section at a distance of 14mm from the metal/chill mold interface. The Spatial Point Pattern (SPP) and the Multi-stage random sampling (MRS) methods were used to determine the distribution of the pores and the point of crack initiation leading to eventual failure. The MRS method reveals that all the pores considered are of irregular shapes, i.e shrinkage pores, with sphericity β < 0.3. The "worst" of the shapes is the pore in the upper left region with β= 5.3078e-010 and D = 1.8949. The SPP method confirms the result of the MRS method because crack initiation will commence in a region with clustered pores.
Optimization of turning NST 37.2 steel with uncoated carbide cutting tools
(Nigerian Institution of Mechanical Engineers, 2010) Fadare, D. A.; Asafa, T. B.
Selection of optimimum machining parameters is an essential factor in process planning for efficient metal cutting operations. In this study, an artificial neural network based tool wear predictive model and a genetic algorithm-based optimization model were developed to determine the optimum cutting parameters for turning NST 37.2 steel with uncoated carbide cutting inserts. Multi-layer, feed-forare, back -propagation network was used in predictive model, while maximum metal removal rate (MRR) was used as the objective function and tool wear as samples NST 37.2 steel bars with 25mm diameter and 400mm length s workspiece and Sandvice Coromant® uncoated carbide inserts with International Standard Organization (ISO) designation SNMA 12406. Dry machining at different cutting conditions with cutting speed (v), feed rate (f) and depth of cut (d) ranging from 20.42-42.42 mm/min, 1.0-2.2 mm/rev and 0.2-0.8mm, respectively were carried out. Eight passes of 50mm length of cut were machined at each conediiton, the spindle power and tool wear (flank and nose) were measured during each cutting operation. Results have shown that the predictive model had acceptable accurancy and optimum cutting parameters obtained were: v=42.32mm/min, f= 2.19 mm/rev and d = 0.8mm.
Performance evaluation of uncoated carbide cutting tools in turning NST 37-2 steel.
(2009) Fadare, D. A.; Asafa, T. B.
In metal cutting operations, performance of the cutting tool is the major determinant of the productivity, functionality and production cost of the machined component. In this work, the performance of uncoated cemented carbide tools with International Standard Organisation (ISO) designation SNMA 120406 was evaluated for turning of NST 37.2 steel. Turning operations were conducted on M300 Harrison type lathe driven by 3.0 hp Kapak induction motor. The cutting conditions used were: cutting speed, 20.42, 29.06 and 42.42m/min; feed rate, 1.0, 1.8 and 2.2 mm/rev; and depth of cut, 0.2, 0.4 and 0.6 mm under dry machining. Results showed that both flank and nose wear increased with increase in number of pass, cutting speed, depth of cut and feed rate. The optical surface roughness of the machined workpiece varied from 0.658 - 0.924 and case hardening of the machined surface was observed. Segmented chips with smaller coil radii, which were less voluminous and more manageable were produced at all cutting conditions investigated. Chip breakability tends to increase with increase in cutting speed. The use of uncoated carbide tools has proved to enhance the productivity and surface quality in turning of NST 37-2 steel.
Performance ranking of artificial neural network learning algorithms in solar radiation forecast
(2010) Fadare, D. A.; Asafa, T. B.
Artificial Neural Networks (ANNs) area prormsmg alternative to conventional tools in modeling and prediction of complex and non-linear parameters. However, the selection of appropriate network parameters for optimum performance pose application challenges. In this study, the modeling and predictive performances of six backpropagation learning algorithms: Levenberg-Marquardt (LM), BFGS Quasi-Newton (BFG), Resilient Backpropagation (RP), Fletcher-Powell Conjugate Gradient (CGF), Variable Learning Rate Backpropagation (GDX) and Bayesian Reglarization (BR) in solar radiation forecast were investigated. Multilayer perceptron (MPL) neural network with five, ten and one neuron(s) in the input, hidden and output layers, respectively was designed with MATLAB® neural network toolkit and trained with the six learning algorithms using the daily global solar radiation data of Ibadan (Lat. 7.4° N; Long. 3.90 E; Alt. 227.2m), Nigeria. The network performance was ranked based on the number of iterations required for convergence, and coefficient of correlation (r-value), mean square error (MSE) and mean absolute percentage error (MAPE) between the actual and predicted values of the training and testing datasets. Results showed that the LM and BR learning algorithms are the two best algorithms to be considered for use in modeling and forecasting of solar radiation data.