A systematic approach to error isolation in computerized wastewater simulation models

Activated sludge models are used extensively in the study of wastewater treatment processes. While various commercial implementations of these models are available, there are many people who need to code models themselves using the simulation packages available to them, Quality assurance of such models is difficult. While benchmarking problems have been developed and are available, the comparison of simulation data with that of commercial models leads only to the detection, not the isolation of errors. To identify the errors in the code is time-consuming. In this paper, we address the problem by developing a systematic and largely automated approach to the isolation of coding errors. There are three steps: firstly, possible errors are classified according to their place in the model structure and a feature matrix is established for each class of errors. Secondly, an observer is designed to generate residuals, such that each class of errors imposes a subspace, spanned by its feature matrix, on the residuals. Finally. localising the residuals in a subspace isolates coding errors. The algorithm proved capable of rapidly and reliably isolating a variety of single and simultaneous errors in a case study using the ASM 1 activated sludge model. In this paper a newly coded model was verified against a known implementation. The method is also applicable to simultaneous verification of any two independent implementations, hence is useful in commercial model development.

Propagation of nonstationary curved and stretched premixed flames with multistep reaction mechanisms

The propagation speed of a thin premixed flame disturbed by an unsteady fluid flow of a larger scale is considered. The flame may also have a general shape but the reaction zone is assumed to be thin compared to the flame thickness. Unlike in preceding publications, the presented asymptotic analysis is performed for a general multistep reaction mechanism and, at the same time, the flame front is curved by the fluid flow. The resulting equations define the propagation speed of disturbed flames in terms of the properties of undisturbed planar flames and the flame stretch. Special attention is paid to the near-equidiffusion limit. In this case, the flame propagation speed is shown to depend on the effective Zeldovich number Z(f) , and the flame stretch. Unlike the conventional Zeldovich number, the effective Zeldovich number is not necessarily linked directly to the activation energies of the reactions. Several examples of determining the effective Zeldovich number for reduced combustion mechanisms are given while, for realistic reactions, the effective Zeldovich number is determined from experiments. Another feature of the present approach is represented by the relatively simple asymptotic technique based on the adaptive generalized curvilinear system of coordinates attached to the flame (i.e., intrinsic disturbed flame equations [IDFE]).

Empirical evidence for ultrametric structure in multi-layer perceptron error surfaces

Combinatorial optimization problems share an interesting property with spin glass systems in that their state spaces can exhibit ultrametric structure. We use sampling methods to analyse the error surfaces of feedforward multi-layer perceptron neural networks learning encoder problems. The third order statistics of these points of attraction are examined and found to be arranged in a highly ultrametric way. This is a unique result for a finite, continuous parameter space. The implications of this result are discussed.

Chebyshev real wave packet propagation: H+O-2 (J=0) state-to-state reactive scattering calculations

In this paper we explore the relative performance of two recently developed wave packet methodologies for reactive scattering, namely the real wave packet Chebyshev domain propagation of Gray and Balint-Kurti [J. Chem. Phys. 108, 950 (1998)] and the Lanczos subspace wave packet approach of Smith [J. Chem. Phys. 116, 2354 (2002); Chem. Phys. Lett. 336, 149 (2001)]. In the former method, a modified Schrodinger equation is employed to propagate the real part of the wave packet via the well-known Chebyshev iteration. While the time-dependent wave packet from the modified Schrodinger equation is different from that obtained using the standard Schrodinger equation, time-to-energy Fourier transformation yields wave functions which differ only trivially by normalization. In the Lanczos subspace approach the linear system of equations defining the action of the Green operator may be solved via either time-dependent or time-independent methods, both of which are extremely efficient due to the simple tridiagonal structure of the Hamiltonian in the Lanczos representation. The two different wave packet methods are applied to three dimensional reactive scattering of H+O-2 (total J=0). State-to-state reaction probabilities, product state distributions, as well as initial-state-resolved cumulative reaction probabilities are examined. (C) 2002 American Institute of Physics.

The impact of genotyping error on haplotype reconstruction and frequency estimation

The choice of genotyping families vs unrelated individuals is a critical factor in any large-scale linkage disequilibrium (LD) study. The use of unrelated individuals for such studies is promising, but in contrast to family designs, unrelated samples do not facilitate detection of genotyping errors, which have been shown to be of great importance for LD and linkage studies and may be even more important in genotyping collaborations across laboratories. Here we employ some of the most commonly-used analysis methods to examine the relative accuracy of haplotype estimation using families vs unrelateds in the presence of genotyping error. The results suggest that even slight amounts of genotyping error can significantly decrease haplotype frequency and reconstruction accuracy, that the ability to detect such errors in large families is essential when the number/complexity of haplotypes is high (low LD/common alleles). In contrast, in situations of low haplotype complexity (high LD and/or many rare alleles) unrelated individuals offer such a high degree of accuracy that there is little reason for less efficient family designs. Moreover, parent-child trios, which comprise the most popular family design and the most efficient in terms of the number of founder chromosomes per genotype but which contain little information for error detection, offer little or no gain over unrelated samples in nearly all cases, and thus do not seem a useful sampling compromise between unrelated individuals and large families. The implications of these results are discussed in the context of large-scale LD mapping projects such as the proposed genome-wide haplotype map.

Towards the clonal propagation of coconut

Past studies from our laboratory have shown that whole immature, or mature sliced, zygotic embryos are a very good starting explant for coconut somatic embryogenesis. The highest rate of somatic embryogenesis was obtained when certain polyamines were added into the culture medium as well as activated charcoal (AC) to absorb unwanted phenolics. These past studies also showed that the development and maturation of the somatic embryos produced could be improved by the addition of abscisic acid (ABA), alone or with one of several osmotically active agents, into the culture medium. In the present study this well characterised somatic embryogenic system for zygotic tissues is being modified and applied to somatic tissues. This recent approach should be a better method for the rapid production of clonal, true-to-type coconut palms. The present research approach is focused on young leaf section explants which have been found to be very responsive to callus production. Young leaf sections produced optimum callus when cultured on media containing 2,4-D (150 μM) and the amount produced could be increased by soaking the sections in sterile water (15 to 60 minutes) or ascorbic acid (15 to 30 minutes) prior to culturing. Further improvement in callus production, as well as a reduction in the time taken for callogenesis was obtained when casein hydrolysate and/or certain polyamines were added to the callus induction medium. The development of the somatic embryos was improved by using ABA and polyethylene glycol (PEG) in the maturation medium. Despite these initial successes in improving coconut somatic embryogenesis, further studies are now being considered to shorten the time to achieve somatic embryogenesis, to better germinate somatic embryos and to improve the rate of somatic seedling conversion into plantlets.

Predicting cave initiation and propagation

The most widely used method for predicting the onset of continuous caving is Laubscher's caving chart. A detailed examination of this method was undertaken which concluded that it had limitations which may impact on results, particularly when dealing with stronger rock masses that are outside current experience. These limitations relate to inadequate guidelines for adjustment factors to rock mass rating (RMR), concerns about the position on the chart of critical case history data, undocumented changes to the method and an inadequate number of data points to be confident of stability boundaries. A review was undertaken on the application and reliability of a numerical method of assessing cavability. The review highlighted a number of issues, which at this stage, make numerical continuum methods problematic for predicting cavability. This is in particular reference to sensitivity to input parameters that are difficult to determine accurately and mesh dependency. An extended version of the Mathews method for open stope design was developed as an alternative method of predicting the onset of continuous caving. A number of caving case histories were collected and analyzed and a caving boundary delineated statistically on the Mathews stability graph. The definition of the caving boundary was aided by the existence of a large and wide-ranging stability database from non-caving mines. A caving rate model was extrapolated from the extended Mathews stability graph but could only be partially validated due to a lack of reliable data.

Fast, scalable master equation solution algorithms. III. Direct time propagation accelerated by a diffusion approximation preconditioned iterative solver

In this paper we propose a novel fast and linearly scalable method for solving master equations arising in the context of gas-phase reactive systems, based on an existent stiff ordinary differential equation integrator. The required solution of a linear system involving the Jacobian matrix is achieved using the GMRES iteration preconditioned using the diffusion approximation to the master equation. In this way we avoid the cubic scaling of traditional master equation solution methods and maintain the low temperature robustness of numerical integration. The method is tested using a master equation modelling the formation of propargyl from the reaction of singlet methylene with acetylene, proceeding through long lived isomerizing intermediates. (C) 2003 American Institute of Physics.

Simultaneous FT-NIR and ESR analyses to yield propagation rate coefficients for polymerization of methyl methacrylate based monomers

A hyphenated instrumental approach has been used to obtain reliable values for the propagation rate coefficients as a function of conversion for polymerizations of methyl methacrylate (MMA) and a mixture of MMA and ethyleneglycol dimethacrylate (EGDMA) with a 1:1 concentration of double bonds, from near the onset of the Trommsdorf region into the glass region. ESR spectroscopy was used to measure the radical concentration while FT-NIR fibre-optic spectroscopy was employed to measure instantaneously the double-bond concentration within the temperature-controlled cavity of the ESR instrument during polymerization. The advantage of this approach to the measurement of the rate coefficient is that it is equally applicable to branching and linear polymerizations. For the polymerization of methyl methacrylate, the values of the rate coefficient at the lowest conversions at which reliable values could be obtained were in agreement with recently reported values obtained by the PLP-SEC method. For the lowest conversions, the values obtained were 403 1 mol(-1) s(-1) at 306 K for MMA and 5201 mol(-1) s(-1) at 310 K for a 1:1 mixture of MMA and EGDMA. (C) 2003 Society of Chemical Industry.

Influence of leaf retention on cutting propagation of Lavandula dentata L.

Vegetative propagation of lavender offers several advantages over sexual propagation, among them crop homogeneity and yield of higher quality essential oil. However, Lavandula species have been propagated mostly by seeds and are said to be recalcitrant to rooting when propagated by cuttings. During cutting propagation, one of the important variables that influence the rooting capacity of cuttings is the leaf retention. The objective of this work was to evaluate the influence of leaf retention on rooting of L. dentata cuttings. Apical cuttings of L. dentata of 10 cm in length, keeping approximately 1/3, 1/2 or 2/3 of their leaves were planted in commercial substrate Plantmax HT® under intermittent mist. After two months, averages of root number, length of the longest root, root fresh and dry weights, and the survival percentage were evaluated. Root length and fresh weight were statistically greater with 2/3 of leaf retention and when fewer leaves were kept on the cuttings, lower means of root dry weight was observed. Under the conditions applied in this study, greater leaf retention was better for rooting of L. dentata cuttings.

Error bounds for low-rank approximations of the first exponential integral kernel

A hierarchical matrix is an efficient data-sparse representation of a matrix, especially useful for large dimensional problems. It consists of low-rank subblocks leading to low memory requirements as well as inexpensive computational costs. In this work, we discuss the use of the hierarchical matrix technique in the numerical solution of a large scale eigenvalue problem arising from a finite rank discretization of an integral operator. The operator is of convolution type, it is defined through the first exponential-integral function and, hence, it is weakly singular. We develop analytical expressions for the approximate degenerate kernels and deduce error upper bounds for these approximations. Some computational results illustrating the efficiency and robustness of the approach are presented.