The Region-Dividing Coupling finite element method and weighted residual method in the solution of field proplems is advanced in this paper──the different regions of the field are handled alone with finite element method or with weighted residual method, then synthesizes them with the concerted conditions on the boundary lines between the regoins.

For each sample, weight residuals (WR) and absolute WR were calculated. Results Median weighted residual (MDWR) of Schuttler model was less significantly than that of all other models in ≥65y group. Predicted- measured concentration with Schuttler model had best goodness of fit in all patients during infusion (P<0.01).

Weight residuals (WR) and absolute WR, median weight residu-als(MDWR)and median absolute weight residuals (MDAWR), objection functions were calculated.

Methods Sixteen patients, ASA graded Ⅰ-Ⅱ, received propofol 60 ml/h (≥65y) or 75 ml/h (<65y) respectively. Arterial blood samples were collected during and after infusion for measurement of plasma concentration. Using four documented pharmacokinetic models to predict the plasma concentrations of propofol.

In Chapter 5, a newly developed method, an alternating normalization-weighted error (ANWE), was utilized to three-way excitation-emission matrix fluorescence data analysis for second-order calibration.

A combined detection logic, including weigted residual calculation and predicted residual, and a ill data identifying algorithm based on predicted residual are used for detection of ill value, and the evaluating precision is assured.

We substitute weighted sum of squares for sum of squares what filter eigenvalue's method , which is obviously amended the systemic error,and greatly raised the precision of model in principal regression;

New models of film flows down inclined planes have been derived by combining a gradient expansion at first or second order to weighted residual techniques with polynomials as test functions.

The Principle of Minimum Potential Energy for the problem is obtained by means of weighted residual method, and its Generalized Variational Principle by means of identified Lagrange multiplier method.

A weighted residual method for elastic-plastic analysis near a crack tip and the calculation of the plastic stress intensity fac

In this paper, a weighted residual method for the elastic-plastic analysis near a crack tip is systematically given by taking the model of power-law hardening under plane strain condition as a sample.

On such a basis, a perturbed-weighted residual finite element model for small Deborah number situations is formulated.

Based on the common form of the two-dimensional Saint-Venant equations, some improvements in the Galerkin finite element method, one of the weighted residual methods, have been made. Besides the triangular elementswith three or six nodes, two new types——one with four nodes and theother with one curve side are proposed in this paper. These four types of elements can be combined in use to describe the surface of water body, using a quadratic spline to fit its curvilinear contour more satisfactorily. La order to...

Based on the common form of the two-dimensional Saint-Venant equations, some improvements in the Galerkin finite element method, one of the weighted residual methods, have been made. Besides the triangular elementswith three or six nodes, two new types——one with four nodes and theother with one curve side are proposed in this paper. These four types of elements can be combined in use to describe the surface of water body, using a quadratic spline to fit its curvilinear contour more satisfactorily. La order to calculate rapidly the integrals in the Galerkin equations, five influence coefficients have been introduced for each type of elements. Tables of the coefficients for straight-sided elements are given through analytical integration to avoid numerical calculation. Various types of boundary conditions and their treatment for either the fixed or the movable boundaries interfaced both with land and with neighboring water bodies, on the latter of which exchanges of mass and momentum take place lumpily or distributively, are proposed so as to represent practical problems as well as possible. Then the wellposedness and accuracy of problems are analyzed. In addition, skills aimed at increasing the speed of solution and decreasing the requirement of computer memory are listed, and a general-purpose program package has been worked out on PDP-11/23 computer to facilitate the utilization. Finally, some primary experience for improving numerical stability is presented through an example.

Two dynamic models, one of which was used to simulate the transient behavior in binary distillation columns and another for rectifying and stripping section separately, are applied to a ternary mixture, benzene-toluene-ethylben-zene, in a large column with 48 trays. Changing the relux ratio by ±5% and ±10%,reboiler duty by ±2.5% and ±5%, the feed flowrate by ±5% and the concentration of benzene, light key component, in the feed stream from 0.5 to 0.45, while the concentration of toluene, heavy key component,...

Two dynamic models, one of which was used to simulate the transient behavior in binary distillation columns and another for rectifying and stripping section separately, are applied to a ternary mixture, benzene-toluene-ethylben-zene, in a large column with 48 trays. Changing the relux ratio by ±5% and ±10%,reboiler duty by ±2.5% and ±5%, the feed flowrate by ±5% and the concentration of benzene, light key component, in the feed stream from 0.5 to 0.45, while the concentration of toluene, heavy key component, changes from 0.4 to 0.45, and using the method of weighted residuals to treat the set of differential equations, the compositions and other variables in the column are solved when the column reaches steady-state operation. The computational results are satisfactory since the RMS errors of liquii compositions in the column are less than 4×10-3 for most conditions and the computer time is reduced by 60%.

Dynamic models for distillation columns generally involve a large number of dependent varibles, which severely limits thier usefulness. The author,in cooperation with Prof. E.H. Wissler in University of Texas at Austin,employed natural spline functions of degree three as basic functions to simulate the composition profiles in the distillation column fully for whole one and separately for the rectifying and stripping section. In order to reduce the number of variables the method of weighted residuals was used...

Dynamic models for distillation columns generally involve a large number of dependent varibles, which severely limits thier usefulness. The author,in cooperation with Prof. E.H. Wissler in University of Texas at Austin,employed natural spline functions of degree three as basic functions to simulate the composition profiles in the distillation column fully for whole one and separately for the rectifying and stripping section. In order to reduce the number of variables the method of weighted residuals was used to treat obtained set of non-linear differential equations. The results were obtained by means of numerical method. These two models were applied to the binary and ternary distillation columns with 48 and 39 trays, respectively. By changing reflux ratio by ±5%,±10%, he duty of reboiler by ±2.5%,±5%, the flowrate of feed by±5% and the composition of the key components in feed by 10%, the compositions and other variables at each tray were obtained when the column reached steady-state. It was shown from computational results that the computer time was saved by more than 60% with satisfied accuracy.