Linear programming computer codes

In real-life problems ia the process iadustry, aeady always there is a nonlinear objective fuactioa. The gradieats deteroiiaed at any particular poiat ia the space of the variables to be optimized can be used to approximate the objective function at that poiat as a linear fuactioa similar techniques can be used to represent nonlinear constraints as linear approximations. The linear programming code can then be used to find an optimum for the linearized problem. At this optimum poiat, the objective can be reevaluated, the gradients can be recomputed, and a new linearized problem can be generated. The new problem can be solved and the optimum found. If the new optimum is the same as the previous one then the computations are terminated. 

In the pocket in the back of this book are four simple Fortran generic codes that can be used to solve sets of linear and nonlinear equations on microcomputers or mainframes. Because these programs are simple, and thus may on occasion fell to solve your problem, you may want to use more polished computer codes that are available in your computer center software library. Such codes are more robust, but of course it takes more of your time to understand how to use them proficiently. 

One-electron cluster calculations are carried out based on the local density functional approach using the DV-Xtt computation code developed by Adachi et al. (5). The exchange-correlation energy is taken into account using the Slater s XCl potential (6). In the present work, a is set at 0.7, which is found to be the most appropriate value in many cases (7). The MOs are obtained as linear combinations of atomic orbitals (LCAO). The most remarkable feature of our program is that the atomic orbitals (AOs) are numerically calculated in each iteration and optimized for the chemical environment. The details of this program have been discribed in ref. (5)... 

Following the computation of the factors [L] and [U] it is a simple matter to obtain the column matrix (x) as follows. The first step (forward substitution) is to compute a column matrix (z) = [t/](x) by solving the linear system [LKz) = (b The second step (backward substitution) involves the computation of (%) from [U](x) = (z). We shall not go into the details of the decomposition process here. The interested reader can find them in almost any book on numerical methods. The entire method can be programmed in only a few lines of computer code (see, e.g.. Press et al., 1992). 

For many computer tasks and for the transfer of structiural information from one computer program to another, a linear representation of the chemical structure may be more suitable. " A popular linear representation is the SMILES notation. Part of its appeal is that for acyclic structures the SMILES is similar to the traditional linear diagram. For example, ethane is denoted by CC and ethylene C=C. Examples of additional SMILES are given in Figure 4. SMILES is the basis of a chemical information system, and this notation provides a convenient framework for more sophisticated computer coding of chemistry described below. For some internal computer functions, structures encoded in a linear notation may be converted to connection tables. 

At present there are numerous computer programs available for analyzing bonded joints. However, most of these computer codes incorporate linearly elastic material behavior, and some allow for nonlinearly elastic and plastic behavior. Computer programs which incorporate viscoelastic material behavior are quite often limited to the simple spring-dashpot type of model for linear materials. Such inaccurate modeling of the constitutive behavior of the structure can seriously compromise the accuracy of the analytical predictions. 

Thus the problem has been converted to one of linear programming, in solving which a simplex code is used with advantage. To economize the operative store of the computer and speed up the calculation, the following strategy is recommended In the first approximation, calculate values of the function Ui = In for all values of i and three values of 0.0 0.5 and 1.0. In the following approximations, narrow down the interval to one half each time, which represents the calculation of another value of Uj. Halving of the interval with each new approximation continues, until the required accuracy has been achieved. 

Chapters 3 and 4 on Nonlinear Equations and Solution of Sets of Equations provide an introductory approach to the methods used throughout the book for approaching nonlinear engineering models. This is referred to as the linearize and iterate approach and forms the basis for all the nonlinear examples in die book. These chapters include useful computer code that can be used to solve systems of nonlinear equations and these programs provide a basis for much of the remainder of the book. 

Nonlinear circuit problems such as those discussed in these two examples are usually solved in Electrical Engineering by use of the SPICE circuit analysis program. For complicated circuits, this should certainly be the means of solving such problems. This program has built-in models for all standard electronic devices and is very advanced in approaches to achieve convergence. However, at the heart of the SPICE program is an approach very similar to that of the much simpler nsolv() program used here. SPICE will automatically set up the equation set to be solved, but uses first-order linearization and iteration to solve the nonlinear equations just as employed in nsolvQ. While SPICE is the preferred tool for its domain of application, a tool such an nsolv() can be readily embedded into other computer code for specialized solutions to problems not appropriate for an electronic simulation. 

Hence, through the LCAO expansion we have translated the non-linear optimization problem, which required a set of difficult to tackle coupled integro-differential equations, into a linear one, which can be expressed in the language of standard linear algebra and can easily be coded into efficient computer programs. 

---