A Simplified Evaluation Procedure of the Scale of Fluctuation Combined Recurrence Averaging and Weighted Linear Regression

The divergence of properties from one location to another within a soil mass is termed spatial variability, which traditionally includes three parameters the mean, the standard deviation, and the scale of fluctuation, in order to stochastically describe a soil property. Among them, determining the scale of fluctuation in the evaluation of spatial variability of soil profiles is not easy due to soil condition complexity. A simplified procedure is presented in the paper to determine the scale of fluctuation combined recurrence averaging and weighted linear regression. The alternative approach utilizes widely usable spreadsheet to solve the problem more directly and efficiently.

The mean value concept in mono-linear regression of multi-variables and its application to trace studies in geosciences

IEECAS SKLLQG

Stocking models of Chinese mitten crab (Eriocheir japonica sinensis) in Yangtze lakes

The fanning of Chinese mitten crab, a quality aquatic product in China and neighbouring Asian countries, has been developing rapidly in China since last decade. It reached a total yield of 3.4 X 10(5) tonnes in 2002. Due to the successive over-stocking year after year, many lakes in the mid-lower Yangtze Basin, the main farming area, are under deterioration, leading to a reduction of crab yield and quality, and, subsequently, a loss of fanning profits. Aiming at a normal development of crab culture and the sustainable use of lakes, an annual investigation dealing with lake environmental factors in relation to stocked crab populations was carried out at 20 farms in 4 lakes. The results show that the submersed macrophyte biomass (B-Mac) is the key factor affecting annual crab yield (CY). Using the ratio of Secchi depth to mean depth (Z(SD)/Z(M)), an easily measured parameter closely correlated to BMac, as driving variable, 10 regression models of maximal crab yields were generated (r(2) ranging 0.49-0.81). Based on the theory of MSY (Maximum Sustainable Yield), in combination with body-weight (BW) and recapture rate (RR) of adult crabs, a general optimal stocking model was eventually formulated. All models are simple and easy to operate. Comments on their applications and prospects are given in brief. (c) 2006 Elsevier B.V. All rights reserved.

Correcting the systematic error of the density functional theory calculation: the alternate combination approach of genetic algorithm and neural network

The alternate combinational approach of genetic algorithm and neural network (AGANN) has been presented to correct the systematic error of the density functional theory (DFT) calculation. It treats the DFT as a black box and models the error through external statistical information. As a demonstration, the AGANN method has been applied in the correction of the lattice energies from the DFT calculation for 72 metal halides and hydrides. Through the AGANN correction, the mean absolute value of the relative errors of the calculated lattice energies to the experimental values decreases from 4.93% to 1.20% in the testing set. For comparison, the neural network approach reduces the mean value to 2.56%. And for the common combinational approach of genetic algorithm and neural network, the value drops to 2.15%. The multiple linear regression method almost has no correction effect here.

Pressure drops of single and two-phase flows through T-type microchannel mixers

In this paper, preliminary experimental results are presented on pressure drop characteristics of single and two-phase flows through two T-type rectangular microchannel mixers with hydraulic diameters of 528 and 333 mum, respectively. It is shown that both N-2 and water single-phase laminar flows in microchannels, with consideration of experimental uncertainties, are consistent with classic theory, if additional effects, such as entrance effects that will interfere with the interpretation of experimental results, are eliminated by carefully designing the experiments. The obtained pressure drop data of N-2-water two-phase flow in micromixers are analyzed and compared with existing flow pattern-independent models. It is found that the Lockhart-Martinelli method generally underpredicts the frictional pressure drop. Thereafter, a modified correlation of C value in the Chisholm's equation based on linear regression of experimental data is proposed to provide a better prediction of the two-phase frictional pressure drop. Also among the homogeneous flow models investigated, the viscosity correlation of McAdams indicates the best performance in correlating the frictional pressure drop data (mean deviations within +/-20% for two micromixers both). Finally it is suggested that systematic studies are still required to accurately predict two-phase frictional performance in microchannels. (C) 2004 Elsevier B.V. All rights reserved.

Using HPLC retention parameters to estimate fish bioconcentration factors of organic compounds

Reversed-phase high performance liquid chromatography (RP-HPLC) was employed to develop predictive models for fish bioconcentration factors (BCF) of organic compounds. Estimation of BCF from RP-HPLC retention parameters on octadecyl-bonded silica gel (ODS), cyanopropyl-bonded silica gel (CN), and phenyl-bonded silica gel (Ph) columns were investigated. The results show that, for a set of compounds belonging to different chemical classes, the CN stationary phase is the best one among the three columns and better than n-octanol/water model for BCF estimation. A multi-column RP-HPLC model, using the retention parameters on the CN and Ph columns as the variables of multiple linear regression equations, was further evaluated to estimate BCF of organic compounds belonging to different chemical classes, and the results show that the multi-column RP-HPLC model is better than that of any single RP-HPLC column for BCF estimation.

Study on QSPR of alcohols with a novel edge connectivity index F-m

A novel edge degree f(i) for heteroatom and multiple bonds in molecular graph is derived on the basis of the edge degree delta(e(r)). A novel edge connectivity index F-m is introduced. The multiple linear regression by using the edge connectivity index F-m and alcohol-type parameter delta, alcohol-distance parameter L can provide high-quality QSPR models for the normal boiling points (BPs), molar volumes (MVs), molar refraction (MRs), water solubility(log(1/S)) and octanol/water partition (logP) of alcohols with up to 17 non-hydrogen atoms. The results imply that these physical properties may be expressed as a liner combination of the edge connectivity index and alcohol-type parameter, 6, alcohol-distance parameter, L. For the models of the five properties, the correlation coefficient r and the standard errors are 0.9969,3.022; 0.9993, 1.504; 0.9992, 0.446; 0.9924,0.129 and 0.9973,0.123 for BPs, MVs, MRs, log(1/S) and logP, respectively. The cross-validation by using the leave-one-out method demonstrates the models to be highly reliable from the point of view of statistics.

Studies on the method of leaps and bounds regression

In this paper, we introduce the method of leaps and bounds regression which can be used to select variables quickly and obtain the best regression models. These models contain one variable, two variables, three variables and so on. The results obtained by using leaps and bounds regression were compared with those achieved by using stepwise regression to lead to the conclusion that leaps and bounds regression is an effective method.