Extrapolation, Interpolation, and Smoothing of Stationary Time Series
Extrapolation, Interpolation, and Smoothing of Stationary Time Series
Hi-index | 0.00 |
In the design and development of controlled processes there are two areas of intense activity---the theoretical simulation of the total plant and empirical simulations involving the collection of reliable experimental data to assist in the construction of a special purpose model. The simulation of a newly proposed process, even from the best available theoretical basis, is usually only approximate. It requires use of large and expensive computer installations, either analog and/or digital, so that by direct programming of the design criteria, the optimum plant operating conditions may be determined. However, even if such a model were extremely reliable (pilot plant), the scaleup problems are quite complex. In some cases, it might be well-nigh impossible to maintain the exact relationships between certain intrinsic parameters, surface tension, heat transfer, etc.---and, of course, few theoretical models can anticipate all the significant process characteristics. As a result, many full scale plants are in need of partial redesign. The authors feel that techniques relying on plant data are indispensable for accomplishing this end.