Improved Kalman filter initialisation using neurofuzzy estimation

It is traditional to initialise Kalman filters and extended Kalman filters with estimates of the states calculated directly from the observed (raw) noisy inputs, but unfortunately their performance is extremely sensitive to state initialisation accuracy: good initial state estimates ensure fast convergence whereas poor estimates may give rise to slow convergence or even filter divergence. Divergence is generally due to excessive observation noise and leads to error magnitudes that quickly become unbounded (R.J. Fitzgerald, 1971). When a filter diverges, it must be re initialised but because the observations are extremely poor, re initialised states will have poor estimates. The paper proposes that if neurofuzzy estimators produce more accurate state estimates than those calculated from the observed noisy inputs (using the known state model), then neurofuzzy estimates can be used to initialise the states of Kalman and extended Kalman filters. Filters whose states have been initialised with neurofuzzy estimates should give improved performance by way of faster convergence when the filter is initialised, and when a filter is re started after divergence

Experimental and statistical investigation of Australian native plants for second-generation biodiesel production

This work explores the potential of Australian native plants as a source of second-generation biodiesel for internal combustion engines application. Biodiesels were evaluated from a number of non-edible oil seeds which are grow naturally in Queensland, Australia. The quality of the produced biodiesels has been investigated by several experimental and numerical methods. The research methodology and numerical model developed in this study can be used for a broad range of biodiesel feedstocks and for the future development of renewable native biodiesel in Australia.

Integration over song classification replicates: song variant analysis in the hihi

Human expert analyses are commonly used in bioacoustic studies and can potentially limit the reproducibility of these results. In this paper, a machine learning method is presented to statistically classify avian vocalizations. Automated approaches were applied to isolate bird songs from long field recordings, assess song similarities, and classify songs into distinct variants. Because no positive controls were available to assess the true classification of variants, multiple replicates of automatic classification of song variants were analyzed to investigate clustering uncertainty. The automatic classifications were more similar to the expert classifications than expected by chance. Application of these methods demonstrated the presence of discrete song variants in an island population of the New Zealand hihi (Notiomystis cincta). The geographic patterns of song variation were then revealed by integrating over classification replicates. Because this automated approach considers variation in song variant classification, it reduces potential human bias and facilitates the reproducibility of the results.

Simultaneous voltammetric determination of four triazine herbicides in water samples with the aid of chemometrics

A novel differential pulse voltammetry (DPV) method was developed for the simultaneous analysis of herbicides in water. A mixture of four herbicides, atrazine, simazine, propazine and terbuthylazine was analyzed simultaneously and the complex, overlapping DPV voltammograms were resolved by several chemometrics methods such as partial least squares (PLS), principal component regression (PCR) and principal component–artificial networks (PC–ANN). The complex profiles of the voltammograms collected from a synthetic set of samples were best resolved with the use of the PC–ANN method, and the best predictions of the concentrations of the analytes were obtained with the PC-ANN model (%RPET = 6.1 and average %Recovery = 99.0). The new method was also used for analysis of real samples, and the obtained results were compared well with those from the GC-MS technique. Such conclusions suggest that the novel method is a viable alternative to the other commonly used methods such as GC, HPLC and GC-MS.

Frequency response function based damage identification using principal component analysis and pattern recognition technique

Pattern recognition is a promising approach for the identification of structural damage using measured dynamic data. Much of the research on pattern recognition has employed artificial neural networks (ANNs) and genetic algorithms as systematic ways of matching pattern features. The selection of a damage-sensitive and noise-insensitive pattern feature is important for all structural damage identification methods. Accordingly, a neural networks-based damage detection method using frequency response function (FRF) data is presented in this paper. This method can effectively consider uncertainties of measured data from which training patterns are generated. The proposed method reduces the dimension of the initial FRF data and transforms it into new damage indices and employs an ANN method for the actual damage localization and quantification using recognized damage patterns from the algorithm. In civil engineering applications, the measurement of dynamic response under field conditions always contains noise components from environmental factors. In order to evaluate the performance of the proposed strategy with noise polluted data, noise contaminated measurements are also introduced to the proposed algorithm. ANNs with optimal architecture give minimum training and testing errors and provide precise damage detection results. In order to maximize damage detection results, the optimal architecture of ANN is identified by defining the number of hidden layers and the number of neurons per hidden layer by a trial and error method. In real testing, the number of measurement points and the measurement locations to obtain the structure response are critical for damage detection. Therefore, optimal sensor placement to improve damage identification is also investigated herein. A finite element model of a two storey framed structure is used to train the neural network. It shows accurate performance and gives low error with simulated and noise-contaminated data for single and multiple damage cases. As a result, the proposed method can be used for structural health monitoring and damage detection, particularly for cases where the measurement data is very large. Furthermore, it is suggested that an optimal ANN architecture can detect damage occurrence with good accuracy and can provide damage quantification with reasonable accuracy under varying levels of damage.

Quasi-static and dynamic strain sensing using carbon nanotube/epoxy nanocomposite thin films

Thin films are developed by dispersing carbon black nanoparticles and carbon nanotubes (CNTs) in an epoxy polymer. The films show a large variation in electrical resistance when subjected to quasi-static and dynamic mechanical loading. This phenomenon is attributed to the change in the band-gap of the CNTs due to the applied strain, and also to the change in the volume fraction of the constituent phases in the percolation network. Under quasi-static loading, the films show a nonlinear response. This nonlinearity in the response of the films is primarily attributed to the pre-yield softening of the epoxy polymer. The electrical resistance of the films is found to be strongly dependent on the magnitude and frequency of the applied dynamic strain, induced by a piezoelectric substrate. Interestingly, the resistance variation is found to be a linear function of frequency and dynamic strain. Samples with a small concentration of just 0.57% of CNT show a sensitivity as high as 2.5% MPa-1 for static mechanical loading. A mathematical model based on Bruggeman's effective medium theory is developed to better understand the experimental results. Dynamic mechanical loading experiments reveal a sensitivity as high as 0.007% Hz(-1) at a constant small-amplitude vibration and up to 0.13%/mu-strain at 0-500 Hz vibration. Potential applications of such thin films include highly sensitive strain sensors, accelerometers, artificial neural networks, artificial skin and polymer electronics.

Characterization of Failure Modes in CFRP Composites - An ANN Approach

Carbon fiber reinforced polymer (CFRP) composite specimens with different thickness, geometry, and stacking sequences were subjected to fatigue spectrum loading in stages. Another set of specimens was subjected to static compression load. On-line acoustic Emission (AE) monitoring was carried out during these tests. Two artificial neural networks, Kohonen-self organizing feature map (KSOM), and multi-layer perceptron (MLP) have been developed for AE signal analysis. AE signals from specimens were clustered using the unsupervised learning KSOM. These clusters were correlated to the failure modes using available a priori information such as AE signal amplitude distributions, time of occurrence of signals, ultrasonic imaging, design of the laminates (stacking sequences, orientation of fibers), and AE parametric plots. Thereafter, AE signals generated from the rest of the specimens were classified by supervised learning MLP. The network developed is made suitable for on-line monitoring of AE signals in the presence of noise, which can be used for detection and identification of failure modes and their growth. The results indicate that the characteristics of AE signals from different failure modes in CFRP remain largely unaffected by the type of load, fiber orientation, and stacking sequences, they being representatives of the type of failure phenomena. The type of loading can have effect only on the extent of damage allowed before the specimens fail and hence on the number of AE signals during the test. The artificial neural networks (ANN) developed and the methods and procedures adopted show significant success in AE signal characterization under noisy environment (detection and identification of failure modes and their growth).

Cow's milk allergy and the development of tolerance

Aims: To gain insight on the immunological processes behind cow’s milk allergy (CMA) and the development of oral tolerance. To furthermore investigate the associations of HLA II and filaggrin genotypes with humoral responses to early oral antigens. Methods: The study population was from a cohort of 6209 healthy, full-term infants who in a double-blind randomized trial received supplementary feeding at maternity hospitals (mean duration 4 days): cow’s milk (CM) formula, extensively hydrolyzed whey formula or donor breast milk. Infants who developed CM associated symptoms that subsided during elimination diet (n=223) underwent an open oral CM challenge (at mean age 7 months). The challenge was negative in 112, and in 111 it confirmed CMA, which was IgE-mediated in 83. Patients with CMA were followed until recovery, and 94 of them participated in a follow-up study at age 8-9 years. We investigated serum samples at diagnosis (mean age 7 months, n=111), one year later (19 months, n=101) and at follow-up (8.6 years, n=85). At follow-up, also 76 children randomly selected from the original cohort and without CM associated symptoms were included. We measured CM specific IgE levels with UniCAP (Phadia, Uppsala, Sweden), and β-lactoglobulin, α-casein and ovalbumin specific IgA, IgG1, IgG4 and IgG levels with enzyme-linked immunosorbent assay in sera. We applied a microarray based immunoassay to measure the binding of IgE, IgG4 and IgA serum antibodies to sequential epitopes derived from five major CM proteins at the three time points in 11 patients with active IgE-mediated CMA at age 8-9 years and in 12 patients who had recovered from IgE-mediated CMA by age 3 years. We used bioinformatic methods to analyze the microarray data. We studied T cell expression profile in peripheral blood mononuclear cell (PBMC) samples from 57 children aged 5-12 years (median 8.3): 16 with active CMA, 20 who had recovered from CMA by age 3 years, 21 non-atopic control subjects. Following in vitro β-lactoglobulin stimulation, we measured the mRNA expression in PBMCs of 12 T-cell markers (T-bet, GATA-3, IFN-γ, CTLA4, IL-10, IL-16, TGF-β, FOXP3, Nfat-C2, TIM3, TIM4, STIM-1) with quantitative real time polymerase chain reaction, and the protein expression of CD4, CD25, CD127, FoxP3 with flow cytometry. To optimally distinguish the three study groups, we performed artificial neural networks with exhaustive search for all marker combinations. For genetic associations with specific humoral responses, we analyzed 14 HLA class II haplotypes, the PTPN22 1858 SNP (R620W allele) and 5 known filaggrin null mutations from blood samples of 87 patients with CMA and 76 control subjects (age 8.0-9.3 years). Results: High IgG and IgG4 levels to β-lactoglobulin and α-casein were associated with the HLA (DR15)-DQB1*0602 haplotype in patients with CMA, but not in control subjects. Conversely, (DR1/10)-DQB1*0501 was associated with lower IgG and IgG4 levels to these CM antigens, and to ovalbumin, most significantly among control subjects. Infants with IgE-mediated CMA had lower β -lactoglobulin and α-casein specific IgG1, IgG4 and IgG levels (p<0.05) at diagnosis than infants with non-IgE-mediated CMA or control subjects. When CMA persisted beyond age 8 years, CM specific IgE levels were higher at all three time points investigated and IgE epitope binding pattern remained stable (p<0.001) compared with recovery from CMA by age 3 years. Patients with persisting CMA at 8-9 years had lower serum IgA levels to β-lactoglobulin at diagnosis (p=0.01), and lower IgG4 levels to β-lactoglobulin (p=0.04) and α-casein (p=0.05) at follow-up compared with patients who recovered by age 3 years. In early recovery, signal of IgG4 epitope binding increased while that of IgE decreased over time, and binding patterns of IgE and IgG4 overlapped. In T cell expression profile in response to β –lactoglobulin, the combination of markers FoxP3, Nfat-C2, IL-16, GATA-3 distinguished patients with persisting CMA most accurately from patients who had become tolerant and from non-atopic subjects. FoxP3 expression at both RNA and protein level was higher in children with CMA compared with non-atopic children. Conclusions: Genetic factors (the HLA II genotype) are associated with humoral responses to early food allergens. High CM specific IgE levels predict persistence of CMA. Development of tolerance is associated with higher specific IgA and IgG4 levels and lower specific IgE levels, with decreased CM epitope binding by IgE and concurrent increase in corresponding epitope binding by IgG4. Both Th2 and Treg pathways are activated upon CM antigen stimulation in patients with CMA. In the clinical management of CMA, HLA II or filaggrin genotyping are not applicable, whereas the measurement of CM specific antibodies may assist in estimating the prognosis.

Geant4 Hadronic Cascade Models and CMS Data Analysis : Computational Challenges in the LHC era

This work belongs to the field of computational high-energy physics (HEP). The key methods used in this thesis work to meet the challenges raised by the Large Hadron Collider (LHC) era experiments are object-orientation with software engineering, Monte Carlo simulation, the computer technology of clusters, and artificial neural networks. The first aspect discussed is the development of hadronic cascade models, used for the accurate simulation of medium-energy hadron-nucleus reactions, up to 10 GeV. These models are typically needed in hadronic calorimeter studies and in the estimation of radiation backgrounds. Various applications outside HEP include the medical field (such as hadron treatment simulations), space science (satellite shielding), and nuclear physics (spallation studies). Validation results are presented for several significant improvements released in Geant4 simulation tool, and the significance of the new models for computing in the Large Hadron Collider era is estimated. In particular, we estimate the ability of the Bertini cascade to simulate Compact Muon Solenoid (CMS) hadron calorimeter HCAL. LHC test beam activity has a tightly coupled cycle of simulation-to-data analysis. Typically, a Geant4 computer experiment is used to understand test beam measurements. Thus an another aspect of this thesis is a description of studies related to developing new CMS H2 test beam data analysis tools and performing data analysis on the basis of CMS Monte Carlo events. These events have been simulated in detail using Geant4 physics models, full CMS detector description, and event reconstruction. Using the ROOT data analysis framework we have developed an offline ANN-based approach to tag b-jets associated with heavy neutral Higgs particles, and we show that this kind of NN methodology can be successfully used to separate the Higgs signal from the background in the CMS experiment.