Comparison of Some Neural Network and Multivariate Regression for Predicting Mechanical Properties of Investment Casting

Investment casting enables producing complex shapes with good accuracy and surface finish. A key goal for investment castings used in automobile, aerospace, chemical, biomedical and other critical applications is to be free of internal defects and to possess mechanical properties within the desired range. At present, casting quality is ascertained by destructive testing at the end of production cycle, leading to the possibility of scrapping the entire batch. In this work, the mechanical properties of investment castings have been predicted based on process parameters and chemical composition, by employing artificial neural network (ANN) and multivariate regression (MVR). The data of related process parameters (wax making, shell making, dewaxing, melting etc.), chemical composition of the alloy, and the resulting mechanical properties (ultimate tensile strength, yield strength, and percentage elongation) for 800 heats were collected in an industrial investment casting foundry. Three different ANN models: back propagation, momentum and adaptive, and Levenberg-Marquardt, with varying number of neurons in the hidden layer (from 20 to 45 in steps of 5) were trained using a portion of the data and tested with remaining data. A prediction penalty index (PPI) was developed to compare the relative predictive capability of various neural network and MVR models. It is observed that both ANN and MVR could predict the mechanical properties well, though MVR gave slightly better results. For the ANN model, better results were produced when the number of neurons in the hidden layer was equal or slightly higher than the number of input parameters.     

Effect of Laser Beam Welding on Microstructure and Mechanical Properties of Commercially Pure Titanium

Laser beam welding of commercially pure titanium sheets were carried out at different operating conditions. Laser powers of 2.0 and 2.5 kW, beam diameters of 0.18 and 0.36 mm and welding speeds of 4–8 m/min were used. The microstructure and mechanical properties of the welded samples were investigated in the present study. It was observed that the grain size of the welded samples increased with increasing laser power while it decreased with increasing welding speed and beam diameter. The sample welded at 2.5 kW laser power with 4 m/min welding speed and 0.36 mm beam diameter had comparable tensile properties with the base metal.     

Optimization of operating conditions in a single‐zone drier for two‐layer polymer coatings

In industrial dryers, hot air is blown over wet coatings either from top or from both top and bottom sides to remove the solvent. Drying of multilayer coatings is a complicated process that involves simultaneous heat and mass transfer, transport, and thermodynamic behavior of polymer solutions. Airflow and its temperature are important operating conditions in a dryer. We report optimization of the operating conditions to minimize residual solvent without inducing blisters for two‐layer coatings using mathematical modeling. Our results reveal that optimal air flow on the bottom side of the coating is always greater than or equal to that on its top side—an agreement with the common notion of “backside drying.” Furthermore, the effect of coating thickness, residence time, individual layer thickness, initial solvent concentration, and reversal of the two layers on the optimal conditions is discussed. © 2008 Wiley Periodicals, Inc. J Appl Polym Sci, 2009     

Global exponential stability of antiperiodic solutions for impulsive discrete‐time Markovian jumping stochastic BAM neural networks with additive time‐varying delays and leakage delay

This paper presents a global exponential stability of antiperiodic solution for a class of impulsive discrete‐time Markovian jumping stochastic bidirectional associative memory neural networks with additive time‐varying delays and leakage delay. By utilizing the Lyapunov‐Krasovskii functional and contraction mapping principle, several sufficient conditions and linear matrix inequalities are derived for verifying globally exponentially stable in the mean square. There is a new delay‐dependent criterion for checking the existence, uniqueness, and global stability for antiperiodic solution. Meantime, by using the numerically efficient MATLAB Toolbox, simulation examples are offered to show the effectiveness and usefulness of the obtained result.     

Impact of Martensite Spatial Distribution on Quasi-Static and Dynamic Deformation Behavior of Dual-Phase Steel

The effects of microstructure parameters of dual-phase steels on tensile high strain dynamic deformation characteristic were examined in this study. Cold-rolled steel sheets were annealed using three different annealing process parameters to obtain three different dual-phase microstructures of varied ferrite and martensite phase fraction. The volume fraction of martensite obtained in two of the steels was near identical (~ 19 pct) with a subtle difference in its spatial distribution. In the first microstructure variant, martensite was mostly found to be situated at ferrite grain boundaries and in the second variant, in addition to at grain boundaries, in-grain martensite was also observed. The third microstructure was very different from the above two with respect to martensite volume fraction (~ 67 pct) and its morphology. In this case, martensite packets were surrounded by a three-dimensional ferrite network giving an appearance of core and shell type microstructure. All the three steels were tensile deformed at strain rates ranging from 2.7 × 10^?4 (quasi-static) to 650 s^?1 (dynamic range). Field-emission scanning electron microscope was used to characterize the starting as well as post-tensile deformed microstructures. Dual-phase steel consisting of small martensite volume fraction (~ 19 pct), irrespective of its spatial distribution, demonstrated high strain rate sensitivity and on the other hand, steel with large martensite volume fraction (~ 67 pct) displayed a very little strain rate sensitivity. Interestingly, total elongation was found to increase with increasing strain rate in the dynamic regime for steel with core–shell type of microstructure containing large martensite volume fraction. The observed enhancement in plasticity in dynamic regime was attributed to adiabatic heating of specimen. To understand the evolving damage mechanism, the fracture surface and the vicinity of fracture ends were studied in all the three dual-phase steels.     

Mathematical Model Validation of Search Protocols in MP2P Networks

Broadcasting is a basic technique in Mobile ad-hoc network (MANET), and it refers to sending a packet from one node to every other node within the transmission range. Flooding is a type of broadcast where the received packet is retransmitted once by every node. The naive flooding technique, floods the network with query messages, while the random walk technique operates by contacting the subsets of every node’s neighbors at each step, thereby restricting the search space. One of the key challenges in an ad-hoc network is the resource or content discovery problem which is about locating the queried resource. Many earlier works have mainly focused on the simulation-based analysis of flooding, and its variants under a wired network. Although, there have been some empirical studies in peer-to-peer (P2P) networks, the analytical results are still lacking, especially in the context of P2P systems running over MANET. In this paper, we describe how P2P resource discovery protocols perform badly over MANETs. To address the limitations, we propose a new protocol named ABRW (Address Broadcast Random Walk), which is a lightweight search approach, designed considering the underlay topology aimed to better suit the unstructured architecture. We provide the mathematical model, measuring the performance of our proposed search scheme with different widely popular benchmarked search techniques. Further, we also derive three relevant search performance metrics, i.e., mean no. of steps needed to find a resource, the probability of finding a resource, and the mean no. of message overhead. We validated the analytical expressions through simulations. The simulation results closely matched with our analytical model, justifying our findings. Our proposed search algorithm under such highly dynamic self-evolving networks performed better, as it reduced the search latency, decreased the overall message overhead, and still equally had a good success rate.     

Evolution of Structural and Mechanical Properties of TiN Films on SS 304LN

We investigate the effect of orientation and residual stress on mechanical properties of reactive magnetron‐sputtered TiN thin films on SS 304 LN with a function of substrate temperature. All these films are polycrystalline with a preferred orientation (200). Residual stress of these films were calculated by sin²Ψ technique and found to be in the range of ?2.6 to ?4.5 GPa. The hardness and modulus of these films ranged between 24–29 GPa and 326–388 GPa, respectively. Temperature‐dependent orientation change is clearly observed and this in turn influenced the residual stress. Hardness and modulus of these films exhibited dependence on the orientation and residual stress.