Programmed optimization procedures

One of the most reliable direct search methods is the LJ optimization procedure (Luus and Jaakola, 1973). This procedure uses random search points and systematic contraction of the search region. The method is easy to program and handles the problem of multiple optima with high reliability (Wang and Luus, 1977, 1978). A important advantage of the method is its ability to handle multiple nonlinear constraints. 

Another method for solving nonlinear programming problems is based on quadratic programming (QP)1. Quadratic programming is an optimization procedure that minimizes a quadratic objective function subject to linear inequality or equality (or both types) of constraints. For example, a quadratic function of two variables x and X2 would be of the general form ... 

The count median diameter, d, and the geometric standard deviation, Og, were calculated by using the experimental diffusion battery penetration values, gj, as input for a computer program based on the optimization procedure described earlier (Busigin et al., 1980). 

Problem 4.1 is nonlinear if one or more of the functions/, gv...,gm are nonlinear. It is unconstrained if there are no constraint functions g, and no bounds on the jc,., and it is bound-constrained if only the xt are bounded. In linearly constrained problems all constraint functions g, are linear, and the objective/is nonlinear. There are special NLP algorithms and software for unconstrained and bound-constrained problems, and we describe these in Chapters 6 and 8. Methods and software for solving constrained NLPs use many ideas from the unconstrained case. Most modem software can handle nonlinear constraints, and is especially efficient on linearly constrained problems. A linearly constrained problem with a quadratic objective is called a quadratic program (QP). Special methods exist for solving QPs, and these iare often faster than general purpose optimization procedures. 

The MCSCF optimization process is only the last step in the computational procedure that leads to the MCSCF wave function. Normally the calculation starts with the selection of an atomic orbital (AO) basis set, in which the molecular orbitals are expanded. The first computational step is then to calculate and save the one- and two-electron integrals. These integrals are commonly processed in different ways. Most MCSCF programs use a supermatrix (as defined in the closed shell HF operator) in order to simplify the evaluation of the energy and different matrix elements. The second step is then the construction of this super-matrix from the list of two-electron integrals. The MCSCF optimization procedure includes a step, where these AO integrals are transformed to MO basis. This transformation is most effectively performed with a symmetry blocked and ordered list of AO integrals. Step... 

The octasaceharide which carries an additional A -acetylglucosamine at the reducing end in comparison to the heptasaccharide was prepared by an independent synthesis. This compound shows a different conformational behavior than the heptasaccharide [122]. The minimum energy conformation, as calculated by the GESA program, shows the trisaccharide at the 6-position of the P-mannose bent back towards the reducing end. This particular effect is in qualitative agreement with the observation of the conformational difference between the tri- and the pentasaccharide described above. The NMR analysis of this octasaceharide confirms the calculated structure. One important fact for the correct prediction of the conformational data was the simultaneous treatment of all independent parameters in the optimization procedure (Table 8). 

By varying the geometries in the program, an optimum design can be achieved. The choice of column diameter was a critical decision in the optimization procedure. In order to evaluate the effect of changing the column diameter, various column diameters were used and the number of trays required was determined. The results of these test runs are shown in Table G.l... 

Issaq and McNitt [585] published a computer program for peak recognition on the basis of peak areas. They investigated the reproducibility of the area of some well-separated peaks for three solutes (anthraquinone, methyl anthraquinone and ethyl anthraquinone) in the 10 solvents used for their optimization procedure. The solvents included binary, ternary and quaternary mixtures of water with methanol, acetonitrile and THF. The areas were found to be reproducible within about 2 percent. The wavelength used for the UV detector in this study was not reported. 

Stan and Steinbach [613] have described a sequential optimization procedure for programmed temperature GC that searches for an optimum multisegment program in a systematic way. This procedure can be divided into three different stages ... 

In this optimization procedure a segment usually refers to a part of the temperature program at which heating takes place. Such segments may be separated by isothermal periods, during which the temperature is kept constant. In the present discussion we will refer to the two kinds of segments as non-isothermal and isothermal, respectively. 

The response surface for the optimization of the primary (program) parameters in programmed temperature GC is less convoluted than a typical response surface obtained in selectivity optimization procedures (see section 5.1). This will increase the possibility of a Simplex procedure locating the global optimum. 

A systematic sequential optimization procedure may be used to establish an optimum multisegment temperature program. 

As with programmed temperature GC, the application of the Simplex optimization procedure to programmed solvent LC is relatively straightforward. The same procedure can be used both for isocratic and for gradient optimization, as long as an appropriate criterion is selected for each case. ... 

An indication of this latter effect can be found in figure 6.11, which shows the result of the Simplex optimization procedure applied to the programmed solvent LC separation of three antioxidants [621]. 

Figures 6.16e and 6.16f show two chromatograms using gradients which were predicted to yield insufficient separation. Using the optimization procedure of Jandera et al., a number of gradient programs can be tested by calculating the resulting chromatograms, so that the number of experiments required can be greatly reduced.

In isocratic analysis, the general motivation is that the larger the supply of a particular kind of sample, the more optimization effort is warranted. In programmed analysis this is not true. In that case, the larger the supply of samples, the larger the urge to look for alternative methods. Therefore, gradient optimization procedures are only relevant if they represent a limited effort. It yet remains to be established just how far the word limited will reach. 

The inclusion of programming options (temperature programming in GC, solvent programming in LC) in the instrument may also be helpful, not only if a programmed analysis may be the result of the optimization procedure (chapter 6), but also to provide a scanning (or scouting ) facility for unknown samples (section 5.4). 

---