Information structures and order concepts for chemical process computations

A study was made to determine how to carry out chemical process computations on a digital computer in a more effective manner. The concepts and algorithms developed during this study were then used in the development of a more powerful and more efficient chemical process simulation system. More effective use of computer memory was desired, along with a minimal sacrifice in computational efficiency. This was achieved by storing process data on secondary storage devices and retrieving this data only as needed. The retrieval and storage function was done simultaneously while computations were proceeding. This feature minimizes the computational delays normally associated with the retrieval of data from the comparatively slow secondary storage devices. An algorithm was developed for ordering process computations in such a way to minimize the number of recycle parameters. There is strong evidence to indicate that computational efficiency can be significantly improved by ordering process computations in this manner. A different approach was used in devising this algorithm than that used by previous investigators. This approach is intuitively similar to the signal flow diagram concept used in control system theory. The algorithm which resulted from this study is effective and conceptually much simpler than other existing algorithms. Lastly, a method was proposed to permit a user of a process simulator to vary design parameters during the actual execution of a simulation. This will permit a good design engineer limited interaction with the simulator and will be of considerable value.Chemical and Biomolecular Engineering, Department o