Dynamic programming: for chemical engineering applications

Dynamic programming's (DP) versatility and generality, especially its ability to handle nonlinear and nonanalytic functions, make it useful in applications where other techniques will not work. The use of DP in a process controller application is illustrated. The first step in a DP optimization...

Full description

Saved in:
Bibliographic Details
Published in:Chemical engineering (New York) 1987-12, Vol.94 (18), p.113-119
Main Authors: RHINEHART, R. R, BEASLEY, J. D
Format: Magazinearticle
Language:English
Subjects:
Advantages
Applications
Applications of mathematics to chemical engineering. Modeling. Simulation. Optimization
Applied sciences
Chemical engineering
Controllers
Dynamic programming
Exact sciences and technology
Mathematical analysis
Optimization
Process controls
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Dynamic programming's (DP) versatility and generality, especially its ability to handle nonlinear and nonanalytic functions, make it useful in applications where other techniques will not work. The use of DP in a process controller application is illustrated. The first step in a DP optimization is to quantify the situation and the relevant interrelationships. A visual illustration of the problem will be useful in this regard. The 2nd step produces the state transformation function. DP optimization includes a recursive fixing procedure, which analyzes the last stage first. Determining the optimal policy to minimize the total cumulative cost requires working backward from the initial stage. Four calculational tips are: 1. Calculate a single, stationary stage contribution applied to the cumulative objective function value matrix. 2. Search only over rational states. 3. Specify the decision. 4. Reduce the size of the grid.
ISSN:0009-2460
1945-368X