Simple Unknown Problems

The oxypurines can be determined by their ultraviolet absorption, changes in absorption following reaction with enzymes, colorimetric reactions, or combinations of these methods. The assay of large quantities of reasonably pure oxypurines is quite simple practical problems, however, arise in the determination of small quantities in the presence of unknown impurities. No completely satisfactory methods for the determination of these compounds have been developed. One of the difiBculties in interpreting some of the results in the literature is to decide whether the cerebrations of the investigators exceed the limitations of their methods. 

This is illustrated by a simple synthetic problem in which bromohydrin 2 is prepared from cyclohexanone (1), which requires only functional group transformations. The synthetic relationship between 1 and 2 is represented by the diagram 2=> 1. This representation is meant to show that the bromohydrin can be formed from the ketone by an as yet unknown number of synthetic steps. Corey and co-workersi 2 called this representation a transform and used it to define retrosynthetic relationships (sec. 1.2). If we analyze the 2 => 1... 

This is a type 2 problem. The equation contains two unknowns, V and / therefore, to solve it, we need an additional equation or relationship among the variables listed. The second relationship is provided by Fig. 6.10, which relates / and V. If jwe could represent Fig. 6.10 by some simple equation, we could use it to eliminate f ox V from Eq. 6.20 and thus proceed to an analytic solution. However,I the data on Fig. 6.10 have not been successfully represented by any such simple equation, so it is most convenient to proceed by trial and error. Later we will see that this trial and error and the next one are fairly simple computer problems. 

This chapter contains relatively simple NMR unknown problems. The molecular weight or empirical formula of the molecule is sometimes given, and it is up to the reader to elucidate the structure of the unknown. If you are an instructor and wish to confirm that you have the correct structure, you may contact the author via e-mail at j simpson mit. edu. 

For implicit solutions and steady state problems the initial guess of the unknown fields are given physical values, as close to the expected solution as possible. For very simple flow problems a uniform flow field may suffice in which some of the velocity components are set to zero. 

Combining tliis witli the Omstein-Zemike equation, we have two equations and tluee unknowns h(r),c(r) and B(r) for a given pair potential u r). The problem then is to calculate or approximate the bridge fiinctions for which there is no simple general relation, although some progress for particular classes of systems has been made recently. 

An illustrative example generates a 2 x 2 calibration matrix from which we can determine the concentrations xi and X2 of dichromate and permanganate ions simultaneously by making spectrophotometric measurements yi and j2 at different wavelengths on an aqueous mixture of the unknowns. The advantage of this simple two-component analytical problem in 3-space is that one can envision the plane representing absorbance A as a linear function of two concentration variables A =f xuX2). 

Minerals generally present difficult problems in chemical analysis, and these problems grow more serious when the elements being determined are as difficult to separate as are those named above. The time and effort that x-ray emission spectrography can save are therefore great, but there are obstacles to be surmounted. Among these are (1) Absorption and enhancement effects are often serious. (2) The element of interest may be present at low concentration in a matrix that is unknown and variable. (3) Satisfactory standards are not always easy to obtain. (4) Simple equipment sometimes does not resolve important analytical lines- completely. (5) Sample preparation and particle size often influence the intensities of analytical lines Class II deviations (7.8) can be particularly serious with minerals. 

Simple material-balance problems involving only a few streams and with a few unknowns can usually be solved by simple direct methods. The relationship between the unknown quantities and the information given can usually be clearly seen. For more complex problems, and for problems with several processing steps, a more formal algebraic approach can be used. The procedure is involved, and often tedious if the calculations have to be done manually, but should result in a solution to even the most intractable problems, providing sufficient information is known. 

In Section II,C we have deliberately chosen a simple set of problem specifications for our steady-state pipeline network formulation. The specification of the pressure at one vertex and a consistent set of inputs and outputs (satisfying the overall material balance) to the network seems intuitively reasonable. However, such a choice may not correspond to the engineering requirements in many applications. For instance, in analyzing an existing network we may wish to determine certain input and output flow rates from a knowledge of pressure distribution in the network, or to compute the parameters in the network element models on the basis of flow and pressure measurements. Clearly, the specified and the unknown variables will be different in these cases. For any pipeline network how many variables must be specified And what constitutes an admissible set of specifications in... 

Indeed, if the problem is simple enough that the connection weights can be found by a few moments work with pencil and paper, there are other computational tools that would be more appropriate than neural networks. It is in more complex problems, in which the relationships that exist between data points are unknown so that it is not possible to determine the connection weights by hand, that an ANN comes into its own. The ANN must then discover the connection weights for itself through a process of supervised learning. 

Programming a CAM for fluorometry is far more complex than for spectrophotometry. Spectrophotometry is simple because it is based on the ratio of light in to light out. But fluorometry creates many of the problems associated with true radiometry—measuring the emission spectrum of an unknown source. The logic may become circular. Radiometry to determine an emission spectrum requires the relative spectral sensitivity of the photometer to be known, but how can this be determined without a source with a known emission spectrum Fortunately, physicists in our national standardization organizations provide us with calibrated sources and photometers. 

Three algorithms have been implemented in both single and multiperspective environments. In this way any bias introduced by a single algorithm should be removed. The first is the statistical Naive Bayesian Classifier, ft reduces the decision-making problem to simple calculations of feature probabilities, ft is based on Bayes theorem and calculates the posterior probability of classes conditioned on the given unknown feature... 

---