A neural network enhanced generalized minimum variance self-tuning proportional, integral and derivative control algorithm for complex dynamic systems

A neural network enhanced proportional, integral and derivative (PID) controller is presented that combines the attributes of neural network learning with a generalized minimum-variance self-tuning control (STC) strategy. The neuro PID controller is structured with plant model identification and PID parameter tuning. The plants to be controlled are approximated by an equivalent model composed of a simple linear submodel to approximate plant dynamics around operating points, plus an error agent to accommodate the errors induced by linear submodel inaccuracy due to non-linearities and other complexities. A generalized recursive least-squares algorithm is used to identify the linear submodel, and a layered neural network is used to detect the error agent in which the weights are updated on the basis of the error between the plant output and the output from the linear submodel. The procedure for controller design is based on the equivalent model, and therefore the error agent is naturally functioned within the control law. In this way the controller can deal not only with a wide range of linear dynamic plants but also with those complex plants characterized by severe non-linearity, uncertainties and non-minimum phase behaviours. Two simulation studies are provided to demonstrate the effectiveness of the controller design procedure.

System identification using partitioned least squares

A novel partitioned least squares (PLS) algorithm is presented, in which estimates from several simple system models are combined by means of a Bayesian methodology of pooling partial knowledge. The method has the added advantage that, when the simple models are of a similar structure, it lends itself directly to parallel processing procedures, thereby speeding up the entire parameter estimation process by several factors.

Neurofuzzy state identification using prefiltering

A new state estimator algorithm is based on a neurofuzzy network and the Kalman filter algorithm. The major contribution of the paper is recognition of a bias problem in the parameter estimation of the state-space model and the introduction of a simple, effective prefiltering method to achieve unbiased parameter estimates in the state-space model, which will then be applied for state estimation using the Kalman filtering algorithm. Fundamental to this method is a simple prefiltering procedure using a nonlinear principal component analysis method based on the neurofuzzy basis set. This prefiltering can be performed without prior system structure knowledge. Numerical examples demonstrate the effectiveness of the new approach.

Modelling niches of arbitrary shape in genetic algorithms using Niche Linkage in the Dynamic Niche Clustering framework

We present some additions to a fuzzy variable radius niche technique called Dynamic Niche Clustering (DNC) (Gan and Warwick, 1999; 2000; 2001) that enable the identification and creation of niches of arbitrary shape through a mechanism called Niche Linkage. We show that by using this mechanism it is possible to attain better feature extraction from the underlying population.

A variable radius niching technique for speciation in genetic algorithms

In this paper, a continuation of a variable radius niche technique called Dynamic Niche Clustering developed by (Gan & Warwick, 1999) is presented. The technique employs a separate dynamic population of overlapping niches that coexists alongside the normal population. An empirical analysis of the updated methodology on a large group of standard optimisation test-bed functions is also given. The technique is shown to perform almost as well as standard fitness sharing with regards to stability and the accuracy of peak identification, but it outperforms standard fitness sharing with regards to time complexity. It is also shown that the technique is capable of forming niches of varying size depending on the characteristics of the underlying peak that the niche is populating.

Wiener System identification using B-spline functions with De Boor recursion

A simple and effective algorithm is introduced for the system identification of Wiener system based on the observational input/output data. The B-spline neural network is used to approximate the nonlinear static function in the Wiener system. We incorporate the Gauss-Newton algorithm with De Boor algorithm (both curve and the first order derivatives) for the parameter estimation of the Wiener model, together with the use of a parameter initialization scheme. The efficacy of the proposed approach is demonstrated using an illustrative example.

Identification and characterisation of five novel miniature inverted-repeat transposable elements (MITEs) in amphioxus (Branchiostoma floridae)

As the sister group to vertebrates, amphioxus is consistently used as a model of genome evolution for understanding the invertebrate/vertebrate transition. The amphioxus genome has not undergone massive duplications like those in the vertebrates or disruptive rearrangements like in the genome of Ciona, a urochordate, making it an ideal evolutionary model. Transposable elements have been linked to many genomic evolutionary changes including increased genome size, modified gene expression, massive gene rearrangements, and possibly intron evolution. Despite their importance in genome evolution, few previous examples of transposable elements have been identified in amphioxus. We report five novel Miniature Inverted-repeat Transposable Elements (MITEs) identified by an analysis of amphioxus DNA sequence, which we have named LanceleTn-1, LanceleTn-2, LanceleTn-3a, LanceleTn-3b and LanceleTn-4. Several of the LanceleTn elements were identified in the amphioxus ParaHox cluster, and we suggest these have had important implications for the evolution of this highly conserved gene cluster. The estimated high copy numbers of these elements implies that MITEs are probably the most abundant type of mobile element in amphioxus, and are thus likely to have been of fundamental importance in shaping the evolution of the amphioxus genome.

System identification of Wiener systems with B-spline functions using De Boor recursion

In this article a simple and effective algorithm is introduced for the system identification of the Wiener system using observational input/output data. The nonlinear static function in the Wiener system is modelled using a B-spline neural network. The Gauss–Newton algorithm is combined with De Boor algorithm (both curve and the first order derivatives) for the parameter estimation of the Wiener model, together with the use of a parameter initialisation scheme. Numerical examples are utilised to demonstrate the efficacy of the proposed approach.

An elastic net orthogonal forward regression algorithm

In this paper we propose an efficient two-level model identification method for a large class of linear-in-the-parameters models from the observational data. A new elastic net orthogonal forward regression (ENOFR) algorithm is employed at the lower level to carry out simultaneous model selection and elastic net parameter estimation. The two regularization parameters in the elastic net are optimized using a particle swarm optimization (PSO) algorithm at the upper level by minimizing the leave one out (LOO) mean square error (LOOMSE). Illustrative examples are included to demonstrate the effectiveness of the new approaches.

Identification of genetic and phenotypic differences associated with prevalent and non-prevalent Salmonella Enteritidis phage types: analysis of variation in amino acid transport

In this study, differences at the genetic level of 37 Salmonella Enteritidis strains from five phage types (PTs) were compared using comparative genomic hybridization (CGH) to assess differences between PTs. There were approximately 400 genes that differentiated prevalent (4, 6, 8 and 13a) and sporadic (11) PTs, of which 35 were unique to prevalent PTs, including six plasmid-borne genes, pefA, B, C, D, srgC and rck, and four chromosomal genes encoding putative amino acid transporters. Phenotype array studies also demonstrated that strains from prevalent PTs were less susceptible to urea stress and utilized L-histidine, L-glutamine, L-proline, L-aspartic acid, gly-asn and gly-gln more efficiently than PT11 strains. Complementation of a PT11 strain with the transporter genes from PT4 resulted in a significant increase in utilization of the amino acids and reduced susceptibility to urea stress. In epithelial cell association assays, PT11 strains were less invasive than other prevalent PTs. Most strains from prevalent PTs were better biofilm formers at 37 degrees C than at 28 degrees C, whilst the converse was true for PT11 strains. Collectively, the results indicate that genetic and corresponding phenotypic differences exist between strains of the prevalent PTs 4, 6, 8 and 13a and non-prevalent PT11 strains that are likely to provide a selective advantage for strains from the former PTs and could help them to enter the food chain and cause salmonellosis.

Genetic diversity among Escherichia coli O157 : H7 isolates and identification of genes linked to human infections

An Escherichia coli oligonucleotide microarray based on three sequenced genomes was validated for comparative genomic microarray hybridization and used to study the diversity of E. coli O157 isolates from human infections and food and animal sources. Among 26 test strains, 24 (including both Shiga toxin [Stx]-positive and -negative strains) were found to be related to the two sequenced E. coli O157:117 strains, EDL933 and Sakai. However, these strains showed much greater genetic diversity than those reported previously, and most of them could not be categorized as either lineage I or H. Some genes were found more often in isolates from human than from nonhuman sources; e.g., ECs1202 and ECs2976, associated with stx2AB and stx1AB, were in all isolates from human sources but in only 40% of those from nonhuman sources. Some (but not all) lineage I-specific or -dominant genes were also more frequently associated with isolates from human. The results suggested that it might be more effective to concentrate our efforts on finding markers that are directly related to infection rather than those specific to certain lineages. In addition, two Stx-negative O157 cattle isolates (one confirmed to be 117) were significantly different from other Stx-positive and -negative E. coli O157:117 strains and were more similar to MG1655 in their gene content. This work demonstrates that not all E. coli O157:117 strains belong to the same clonal group, and those that were similar to E. coli K-12 might be less virulent.

Are greenhouse gas signals of Northern Hemisphere winter extra-tropical cyclone activity dependent on the identification and tracking algorithm?

For Northern Hemisphere extra-tropical cyclone activity, the dependency of a potential anthropogenic climate change signal on the identification method applied is analysed. This study investigates the impact of the used algorithm on the changing signal, not the robustness of the climate change signal itself. Using one single transient AOGCM simulation as standard input for eleven state-of-the-art identification methods, the patterns of model simulated present day climatologies are found to be close to those computed from re-analysis, independent of the method applied. Although differences in the total number of cyclones identified exist, the climate change signals (IPCC SRES A1B) in the model run considered are largely similar between methods for all cyclones. Taking into account all tracks, decreasing numbers are found in the Mediterranean, the Arctic in the Barents and Greenland Seas, the mid-latitude Pacific and North America. Changing patterns are even more similar, if only the most severe systems are considered: the methods reveal a coherent statistically significant increase in frequency over the eastern North Atlantic and North Pacific. We found that the differences between the methods considered are largely due to the different role of weaker systems in the specific methods.

A mathematical model of the sterol regulatory element binding protein 2 cholesterol biosynthesis pathway

Cholesterol is one of the key constituents for maintaining the cellular membrane and thus the integrity of the cell itself. In contrast high levels of cholesterol in the blood are known to be a major risk factor in the development of cardiovascular disease. We formulate a deterministic nonlinear ordinary differential equation model of the sterol regulatory element binding protein 2 (SREBP-2) cholesterol genetic regulatory pathway in an hepatocyte. The mathematical model includes a description of genetic transcription by SREBP-2 which is subsequently translated to mRNA leading to the formation of 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMGCR), a main precursor of cholesterol synthesis. Cholesterol synthesis subsequently leads to the regulation of SREBP-2 via a negative feedback formulation. Parameterised with data from the literature, the model is used to understand how SREBP-2 transcription and regulation affects cellular cholesterol concentration. Model stability analysis shows that the only positive steady-state of the system exhibits purely oscillatory, damped oscillatory or monotic behaviour under certain parameter conditions. In light of our findings we postulate how cholesterol homestasis is maintained within the cell and the advantages of our model formulation are discussed with respect to other models of genetic regulation within the literature.

Fast identification algorithms for Gaussian process model

A class identification algorithms is introduced for Gaussian process(GP)models.The fundamental approach is to propose a new kernel function which leads to a covariance matrix with low rank,a property that is consequently exploited for computational efficiency for both model parameter estimation and model predictions.The objective of either maximizing the marginal likelihood or the Kullback–Leibler (K–L) divergence between the estimated output probability density function(pdf)and the true pdf has been used as respective cost functions.For each cost function,an efficient coordinate descent algorithm is proposed to estimate the kernel parameters using a one dimensional derivative free search, and noise variance using a fast gradient descent algorithm. Numerical examples are included to demonstrate the effectiveness of the new identification approaches.