Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications.

Filter design using radial basis function neural network and genetic algorithm for improved operational health monitoring

The problem of denoising damage indicator signals for improved operational health monitoring of systems is addressed by applying soft computing methods to design filters. Since measured data in operational settings is contaminated with noise and outliers, pattern recognition algorithms for fault detection and isolation can give false alarms. A direct approach to improving the fault detection and isolation is to remove noise and outliers from time series of measured data or damage indicators before performing fault detection and isolation. Many popular signal-processing approaches do not work well with damage indicator signals, which can contain sudden changes due to abrupt faults and non-Gaussian outliers. Signal-processing algorithms based on radial basis function (RBF) neural network and weighted recursive median (WRM) filters are explored for denoising simulated time series. The RBF neural network filter is developed using a K-means clustering algorithm and is much less computationally expensive to develop than feedforward neural networks trained using backpropagation. The nonlinear multimodal integer-programming problem of selecting optimal integer weights of the WRM filter is solved using genetic algorithm. Numerical results are obtained for helicopter rotor structural damage indicators based on simulated frequencies. Test signals consider low order polynomial growth of damage indicators with time to simulate gradual or incipient faults and step changes in the signal to simulate abrupt faults. Noise and outliers are added to the test signals. The WRM and RBF filters result in a noise reduction of 54 - 71 and 59 - 73% for the test signals considered in this study, respectively. Their performance is much better than the moving average FIR filter, which causes significant feature distortion and has poor outlier removal capabilities and shows the potential of soft computing methods for specific signal-processing applications. (C) 2005 Elsevier B. V. All rights reserved.

Mathematical model and rule extraction for tool wear monitoring problem using nature inspired techniques

In this paper, pattern classification problem in tool wear monitoring is solved using nature inspired techniques such as Genetic Programming(GP) and Ant-Miner (AM). The main advantage of GP and AM is their ability to learn the underlying data relationships and express them in the form of mathematical equation or simple rules. The extraction of knowledge from the training data set using GP and AM are in the form of Genetic Programming Classifier Expression (GPCE) and rules respectively. The GPCE and AM extracted rules are then applied to set of data in the testing/validation set to obtain the classification accuracy. A major attraction in GP evolved GPCE and AM based classification is the possibility of obtaining an expert system like rules that can be directly applied subsequently by the user in his/her application. The performance of the data classification using GP and AM is as good as the classification accuracy obtained in the earlier study.

Fuzzy-logic-based health monitoring and residual-life prediction for composite helicopter rotor

A health-monitoring and life-estimation strategy for composite rotor blades is developed in this work. The cross-sectional stiffness reduction obtained by physics-based models is expressed as a function of the life of the structure using a recent phenomenological damage model. This stiffness reduction is further used to study the behavior of measurable system parameters such as blade deflections, loads, and strains of a composite rotor blade in static analysis and forward flight. The simulated measurements are obtained using an aeroelastic analysis of the composite rotor blade based on the finite element in space and time with physics-based damage modes that are then linked to the life consumption of the blade. The model-based measurements are contaminated with noise to simulate real data. Genetic fuzzy systems are developed for global online prediction of physical damage and life consumption using displacement- and force-based measurement deviations between damaged and undamaged conditions. Furthermore, local online prediction of physical damage and life consumption is done using strains measured along the blade length. It is observed that the life consumption in the matrix-cracking zone is about 12-15% and life consumption in debonding/delamination zone is about 45-55% of the total life of the blade. It is also observed that the success rate of the genetic fuzzy systems depends upon the number of measurements, type of measurements and training, and the testing noise level. The genetic fuzzy systems work quite well with noisy data and are recommended for online structural health monitoring of composite helicopter rotor blades.

Tool wear monitoring using ant behaviour

In this paper we show the applicability of Ant Colony Optimisation (ACO) techniques for pattern classification problem that arises in tool wear monitoring. In an earlier study, artificial neural networks and genetic programming have been successfully applied to tool wear monitoring problem. ACO is a recent addition to evolutionary computation technique that has gained attention for its ability to extract the underlying data relationships and express them in form of simple rules. Rules are extracted for data classification using training set of data points. These rules are then applied to set of data in the testing/validation set to obtain the classification accuracy. A major attraction in ACO based classification is the possibility of obtaining an expert system like rules that can be directly applied subsequently by the user in his/her application. The classification accuracy obtained in ACO based approach is as good as obtained in other biologically inspired techniques.

Identification of delamination in composite beams using spectral estimation and a genetic algorithm

An efficient strategy for identification of delamination in composite beams and connected structures is presented. A spectral finite-element model consisting of a damaged spectral element is used for model-based prediction of the damaged structural response in the frequency domain. A genetic algorithm (GA) specially tailored for damage identification is derived and is integrated with finite-element code for automation. For best application of the GA, sensitivities of various objective functions with respect to delamination parameters are studied and important conclusions are presented. Model-based simulations of increasing complexity illustrate some of the attractive features of the strategy in terms of accuracy as well as computational cost. This shows the possibility of using such strategies for the development of smart structural health monitoring softwares and systems.

A new intelligent algorithm for online voltage stability assessment and monitoring

This paper presents a new approach for assessing power system voltage stability based on artificial feed forward neural network (FFNN). The approach uses real and reactive power, as well as voltage vectors for generators and load buses to train the neural net (NN). The input properties of the NN are generated from offline training data with various simulated loading conditions using a conventional voltage stability algorithm based on the L-index. The performance of the trained NN is investigated on two systems under various voltage stability assessment conditions. Main advantage is that the proposed approach is fast, robust, accurate and can be used online for predicting the L-indices of all the power system buses simultaneously. The method can also be effectively used to determining local and global stability margin for further improvement measures.

A linear programming approach for estimating the structure of a sparse linear genetic network from transcript profiling data

Background: A genetic network can be represented as a directed graph in which a node corresponds to a gene and a directed edge specifies the direction of influence of one gene on another. The reconstruction of such networks from transcript profiling data remains an important yet challenging endeavor. A transcript profile specifies the abundances of many genes in a biological sample of interest. Prevailing strategies for learning the structure of a genetic network from high-dimensional transcript profiling data assume sparsity and linearity. Many methods consider relatively small directed graphs, inferring graphs with up to a few hundred nodes. This work examines large undirected graphs representations of genetic networks, graphs with many thousands of nodes where an undirected edge between two nodes does not indicate the direction of influence, and the problem of estimating the structure of such a sparse linear genetic network (SLGN) from transcript profiling data. Results: The structure learning task is cast as a sparse linear regression problem which is then posed as a LASSO (l1-constrained fitting) problem and solved finally by formulating a Linear Program (LP). A bound on the Generalization Error of this approach is given in terms of the Leave-One-Out Error. The accuracy and utility of LP-SLGNs is assessed quantitatively and qualitatively using simulated and real data. The Dialogue for Reverse Engineering Assessments and Methods (DREAM) initiative provides gold standard data sets and evaluation metrics that enable and facilitate the comparison of algorithms for deducing the structure of networks. The structures of LP-SLGNs estimated from the INSILICO1, INSILICO2 and INSILICO3 simulated DREAM2 data sets are comparable to those proposed by the first and/or second ranked teams in the DREAM2 competition. The structures of LP-SLGNs estimated from two published Saccharomyces cerevisae cell cycle transcript profiling data sets capture known regulatory associations. In each S. cerevisiae LP-SLGN, the number of nodes with a particular degree follows an approximate power law suggesting that its degree distributions is similar to that observed in real-world networks. Inspection of these LP-SLGNs suggests biological hypotheses amenable to experimental verification. Conclusion: A statistically robust and computationally efficient LP-based method for estimating the topology of a large sparse undirected graph from high-dimensional data yields representations of genetic networks that are biologically plausible and useful abstractions of the structures of real genetic networks. Analysis of the statistical and topological properties of learned LP-SLGNs may have practical value; for example, genes with high random walk betweenness, a measure of the centrality of a node in a graph, are good candidates for intervention studies and hence integrated computational – experimental investigations designed to infer more realistic and sophisticated probabilistic directed graphical model representations of genetic networks. The LP-based solutions of the sparse linear regression problem described here may provide a method for learning the structure of transcription factor networks from transcript profiling and transcription factor binding motif data.

Timing the Emergence of Resistance to Anti-HIV Drugs with Large Genetic Barriers

New antiretroviral drugs that offer large genetic barriers to resistance, such as the recently approved inhibitors of HIV-1 protease, tipranavir and darunavir, present promising weapons to avert the failure of current therapies for HIV infection. Optimal treatment strategies with the new drugs, however, are yet to be established. A key limitation is the poor understanding of the process by which HIV surmounts large genetic barriers to resistance. Extant models of HIV dynamics are predicated on the predominance of deterministic forces underlying the emergence of resistant genomes. In contrast, stochastic forces may dominate, especially when the genetic barrier is large, and delay the emergence of resistant genomes. We develop a mathematical model of HIV dynamics under the influence of an antiretroviral drug to predict the waiting time for the emergence of genomes that carry the requisite mutations to overcome the genetic barrier of the drug. We apply our model to describe the development of resistance to tipranavir in in vitro serial passage experiments. Model predictions of the times of emergence of different mutant genomes with increasing resistance to tipranavir are in quantitative agreement with experiments, indicating that our model captures the dynamics of the development of resistance to antiretroviral drugs accurately. Further, model predictions provide insights into the influence of underlying evolutionary processes such as recombination on the development of resistance, and suggest guidelines for drug design: drugs that offer large genetic barriers to resistance with resistance sites tightly localized on the viral genome and exhibiting positive epistatic interactions maximally inhibit the emergence of resistant genomes.

Structural Basis of Drug Resistance by Genetic Variants of HIV Type 1 Clade C Protease from India

Using computer modeling of three-dimensional structures and structural information available on the crystal structures of HIV-1 protease, we investigated the structural effects of mutations, in treatment-naive and treatment-exposed individuals from India and postulated mechanisms of resistance in clade C variants. A large number of models (14) have been generated by computational mutation of the available crystal structures of drug bound proteases. Localized energy minimization was carried out in and around the sites of mutation in order to optimize the geometry of interactions present. Most of the mutations result in structural differences at the flap that favors the semiopen state of the enzyme. Some of the mutations were also found to confer resistance by affecting the geometry of the active site. The E35D mutation affects the flap structure in clade B strains and E35N and E35K mutation, seen in our modeled strains, have a more profound effect. Common polymorphisms at positions 36 and 63 in clade C also affected flap structure. Apart from a few other residues Gln-58, Asn-83, Asn-88, and Gln-92 and their interactions are important for the transition from the closed to the open state. Development of protease inhibitors by structure-based design requires investigation of mechanisms operative for clade C to improve the efficacy of therapy.

Genetic analysis of foot-and-mouth disease virus serotype A of Indian origin and detection of positive selection and recombination in leader protease- and capsid-coding regions

The leader protease (L-pro) and capsid-coding sequences (P1) constitute approximately 3 kb of the foot-and-mouth disease virus (FMDV). We studied the phylogenetic relationship of 46 FMDV serotype A isolates of Indian origin collected during the period 1968-2005 and also eight vaccine strains using the neighbour-joining tree and Bayesian tree methods. The viruses were categorized under three major groups - Asian, Euro-South American and European. The Indian isolates formed a distinct genetic group among the Asian isolates. The Indian isolates were further classified into different genetic subgroups (<5% divergence). Post-1995 isolates were divided into two subgroups while a few isolates which originated in the year 2005 from Andhra Pradesh formed a separate group. These isolates were closely related to the isolates of the 1970s. The FMDV isolates seem to undergo reverse mutation or onvergent evolution wherein sequences identical to the ancestors are present in the isolates in circulation. The eight vaccine strains included in the study were not related to each other and belonged to different genetic groups. Recombination was detected in the L-pro region in one isolate (A IND 20/82) and in the VP1 coding 1D region in another isolate (A RAJ 21/96). Positive selection was identified at aa positions 23 in the L-pro (P<0.05; 0.046*) and at aa 171 in the capsid protein VP1 (P<0.01; 0.003**).

Inbreeding and the incidence of childhood genetic disorders in Karnataka, South India

Consanguineous marriages are strongly favoured among the populations of South India. In a study conducted on 407 infants and children, a total of 35 genetic diseases was diagnosed in 63 persons: 44 with single gene defects, 12 with polygenic disorders, and seven with Down's syndrome. The coefficient of inbreeding of the total study group, F = 0.0414, was significantly higher than that previously calculated for the general population, F = 0.0271, and autosomal recessive disorders formed the largest single disease category diagnosed. The results suggest that long term inbreeding may not have resulted in appreciable elimination of recessive lethals and sub-lethals from the gene pool.

trans gene regulation in adaptive evolution: A genetic algorithm model

This is a continuation of earlier studies on the evolution of infinite populations of haploid genotypes within a genetic algorithm framework. We had previously explored the evolutionary consequences of the existence of indeterminate—“plastic”—loci, where a plastic locus had a finite probability in each generation of functioning (being switched “on”) or not functioning (being switched “off”). The relative probabilities of the two outcomes were assigned on a stochastic basis. The present paper examines what happens when the transition probabilities are biased by the presence of regulatory genes. We find that under certain conditions regulatory genes can improve the adaptation of the population and speed up the rate of evolution (on occasion at the cost of lowering the degree of adaptation). Also, the existence of regulatory loci potentiates selection in favour of plasticity. There is a synergistic effect of regulatory genes on plastic alleles: the frequency of such alleles increases when regulatory loci are present. Thus, phenotypic selection alone can be a potentiating factor in a favour of better adaptation.

Genetic studies of the PRP17 gene of Saccharomyces cerevisiae: a domain essential for function maps to a nonconserved region of the protein

The PRP17 gene product is required for the second step of pre-mRNA splicing reactions. The C-terminal half of this protein bears four repeat units with homology to the beta transducin repeat. Missense mutations in three temperature-sensitive prp17 mutants map to a region in the N-terminal half of the protein. We have generated, in vitro, 11 missense alleles at the beta transducin repeat units and find that only one affects function in vivo. A phenotypically silent missense allele at the fourth repeat unit enhances the slow-growing phenotype conferred by an allele at the third repeat, suggesting an interaction between these domains. Although many missense mutations in highly conserved amino acids lack phenotypic effects, deletion analysis suggests an essential role for these units. Only mutations in the N-terminal nonconserved domain of PRP17 are synthetically lethal in combination with mutations in PRP16 and PRP18, two other gene products required for the second splicing reaction. A mutually allele-specific interaction between Prp17 and snr7, with mutations in U5 snRNA, was observed. We therefore suggest that the functional region of Prp17p that interacts with Prp18p, Prp16p, and U5 snRNA is the N terminal region of the protein.

Serum proteomics of glioma: methods and applications

The prognosis of patients with glioblastoma, the most malignant adult glial brain tumor, remains poor in spite of advances in treatment procedures, including surgical resection, irradiation and chemotherapy.Genetic heterogeneity of glioblastoma warrants extensive studies in order to gain a thorough understanding of the biology of this tumor. While there have been several studies of global transcript profiling of glioma with the identification of gene signatures for diagnosis and disease management, translation into clinics is yet to happen. Serum biomarkers have the potential to revolutionize the process of cancer diagnosis, grading, prognostication and treatment response monitoring. Besides having the advantage that serum can be obtained through a less invasive procedure, it contains molecules at an extraordinary dynamic range of ten orders of magnitude in terms of their concentrations. While the conventional methods, such as 2DE, have been in use for many years, the ability to identify the proteins through mass spectrometry techniques such as MALDI-TOF led to an explosion of interest in proteomics. Relatively new high-throughput proteomics methods such as SELDI-TOF and protein microarrays are expected to hasten the process of serum biomarker discovery. This review will highlight the recent advances in the proteomics platform in discovering serum biomarkers and the current status of glioma serum markers. We aim to provide the principles and potential of the latest proteomic approaches and their applications in the biomarker discovery process. Besides providing a comprehensive list of available serum biomarkers of glioma, we will also propose how these markers will revolutionize the clinical management of glioma patients.