Optimisation of turning parameters by integrating genetic algorithm with support vector regression and artificial neural networks

This paper focuses on optimisation of process parameters of the turning operation, using artificial intelligence techniques such as support vector regression (SVR) and artificial neural networks (ANN) integrated with genetic algorithm (GA). The model is trained using the turning parameters as the input and corresponding surface roughness, tool wear and power required as the output. Data, obtained from conducting experiments is analysed using support vector machine (SVM) and artificial neural network. SVM, a nonlinear model, is learned by linear learning machine by mapping into high-dimensional kernel-induced feature space. The genetic algorithm is integrated with these to find the optimum from the response surface generated. The results are compared with those obtained by integrating GA with traditional models like response surface methodology (RSM) and regression analysis (RA). This paper illustrates the impact that techniques based on artificial intelligence have on optimising processes.     

Cloning, Soluble Expression and Purification of High Yield Recombinant hGMCSF in Escherichia coli

Expression of human granulocyte macrophage colony stimulating factor (hGMCSF), a cytokine of therapeutic importance, as a thioredoxin (TRX) fusion has been investigated in Escherichia coli BL21 (DE3) codon plus cells. The expression of this protein was low when cloned under the T7 promoter without any fusion tags. High yield of GMCSF was achieved (～88 mg/L of fermentation broth) in the shake flask when the gene was fused to the E. coli TRX gene. The protein was purified using a single step Ni(2+)-NTA affinity chromatography and the column bound fusion tag was removed by on-column cleavage with enterokinase. The recombinant hGMCSF was expressed as a soluble and biologically active protein in E. coli, and upon purification, the final yield was ～44 mg/L in shake flask with a specific activity of 2.3 × 10(8) U/mg. The results of Western blot and RP-HPLC analyses, along with biological activity using the TF-1 cell line, established the identity of the purified hGMCSF. In this paper, we report the highest yield of hGMCSF expressed in E. coli. The bioreactor study shows that the yield of hGMCSF could be easily scalable with a yield of ～400 mg/L, opening up new opportunities for large scale production hGMCSF in E. coli.     

Validation of arrayed imaging reflectometry biosensor response for protein-antibody interactions: cross-correlation of theory, experiment, and complementary techniques

One of the critical steps in the development of an analytical technique is to confirm that its experimental response correlates with predictions derived from the theoretical framework on which it is based. This validates the technique quantitatively and, in the case of a biosensor, facilitates a correlation of the sensor's output signal to the concentration of the analyte being tested. Herein we report studies demonstrating that the quantitative response of arrayed imaging reflectometry (AIR), a highly sensitive label-free biosensing method, is a predictable function of the probe and analyte properties. We first incorporated a standard one-site Langmuir binding model describing probe-analyte interactions at the surface into the theoretical model for thickness-dependent reflectance in AIR. This established a hypothetical correlation between the analyte concentration and the AIR response. Spectroscopic ellipsometry, surface plasmon resonance, and AIR were then used to validate this model for two biomedically important proteins, fibroblast growth factor-2 and vascular endothelial growth factor. While our studies demonstrated that the 1:1 one-site Langmuir model accurately described the observed response of macrospot AIR arrays, either a two-site Langmuir model or a Sips isotherm better described the behavior of AIR microarrays. These studies confirmed the quantitative performance of AIR across a range of probe-analyte affinities. Furthermore, the methodology developed here can be extended to other label-free biosensing platforms, thus facilitating a more accurate and quantitative interpretation of the sensor response.     

Structural and Chemical Evolution of the Near-Apex Region of an Atomic Force Microscope Tip Subject to Sliding

Atomic force microscopy and molecular dynamics simulation are used to study the nanoscale wear of a silicon dioxide tip sliding on a copper substrate. Wear is characterized in terms of structural and chemical evolution of the system where the latter is possible experimentally using atom probe tomography of the slid tips. Comparison of the experimentally observed and simulation-predicted wear reveals that adhesive wear is dominant in the short sliding distances of the simulation at any applied load, while the sliding distances in the experiments are long enough to observe load-induced transitions between adhesive-dominated and abrasive-dominated wear.     

Physico-chemical studies of polyetherurethaneurea in phosphate buffer solution

Polyetherurethaneureas are increasingly used as implants in the human body, where they come in contact with body fluid. Most of the reported properties of polyetherurethaneureas have been determined in the dry state. The present work deals with the properties of these materials in aqueous systems. A number of linear polyetherurethaneureas based on polyethoxyglycol and polypropoxyglycol of different molecular weights, 4,4′-diphenylmethane diisocynate, 1,3-propanediamine and 1,6-hexanediamine have been prepared. Their swelling characteristics and phase composition in phosphate buffer solution are evaluated. The role of water in changing various properties of polyetherurethaneureas is discussed.     

Preferred link based delay-constrained least-cost routing in wide area networks

Multimedia applications involving digital audio and/or digital video transmissions require strict QoS constraints (end-to-end delay bound, bandwidth availability, packet loss rate, etc.) to be met by the network. To guarantee the real-time delivery of packets satisfying these constraints, a real-time channel (D. Ferrari and D.C. Verma, A scheme for real-time channel establishment in wide-area networks. IEEE JSAC, 8(3), 368–379, 1990) needs to be established before the transmission of packets of a connection can begin. The establishment of such channels requires the development of efficient route selection algorithms that are designed to take into account the QoS constraints.

The general problem of determining a least-cost delay-constrained route in a given communication network has been proved to be NP-hard (M.R. Garey and D.S. Johnson, Computers and Intractability: a guide to the theory of NP-completeness, W.H. Freeman, 1979). In this paper, we describe a preferred link approach to distributed delay-constrained least-cost routing in order to establish real-time channels. The approach attempts to combine the benefits of probing and backtracking based algorithms (better adaptiveness and wider search) with the advantages of distance-vector type algorithms (lower setup time). The scheme is flexible in that a variety of heuristics can be employed to order the neighbouring links of any given node. Three heuristics are proposed and their performance is studied through simulation experiments. The simulation results indicate that the proposed heuristics provide better performance than other preferred neighbour methods, in terms of increased call acceptance rate and lower average route cost. The heuristics are also shown to adapt much better to dynamic variations in network and link characteristics.